BOTANY OF THE FÆROES BASED UPON DANISH INVESTIGATIONS PART III ILLUSTRATED WITH 12 PLATES, AND 51 FIGUHES IN THE TEXT (PUBLISHED BY THE AID OF THE CARLSBERG FUND) COPENHAGEN AND CHRISTIANIA GYLDENDALSKE BOGHANDEL • NORDISK FORLAG LONDON JOHN WHELDON & CO. 1908 PKINTED BY H.H.THIELE CONTENTS Page F. Borgesen : The Algæ-Vegetation of the Færoese Coasts wlth remarks on the Phyto-Geography (583 C. H. Ostenkeld: Additions and Corrections of the List of the Phanerogamæ and Pteridophyta of the Færoes 835 Gazet Patursson : List of Popular Plant Names from the Færoes 864 C. H. Ostenkeld : The Land-Vegetation of the Færoes 867 F. Borgesen : Gardening and Tree-planting 1027 P. Feilberg: Some Notes on the Agriculture of the Færoes 1044 EuG. Warming: Field-Notes on the Biologj- of the Flowers of the Færoes.... 1055 1. C. Nielsen : The Insect-Fauna of the Færoes 1066 APPENDIX. F. Borgesen and Helgi Jonsson: The Distribution of the Marine Algæ of the Arctic Sea and of the Northernmost Part of the Atlantic p. I— XXVIII CORRIGENDA. Page 711, line 12 from bottom. for two feet read 100 feet 711, - 6 - — , for eight feet read eighty feet BOTANY OF THE FÆROES BASED UPON DANISH INVESTIGATIONS ILLUSTRATED WITH 24 PLATES, AND 202 FIGURES IN THE TEXT (PUBLISHED BY THE AID OF THE CARLSBERG FUND) COPENHAGEN AND CHRISTIANIA GYLDENDALSKE BOGHANDEL • NORDISK FORLAG LONDON JOHN WHELDON & CO. 1901-1908 PRINTED BY H. H. THIELE PREFACE. WITH the third volume, the present work on the Botany of the Færoes is brought to an end. I had originally hoped thai the work wouid have been completed three years ago; but more time than was expected had to be expended upon the last volume. There is, of course, in the vegetation of the Færoes, much which still requires investigation and deeper consideration; and this is more particularly the case in connection with the lower piants and the plant-biology. But these matters must be left for future research, because other more pressing tasks, notably the botanical investigation of Iceland, are now claiming the attention of Danish botanists. It is well known that Danish investigations have been cairied on in Greenland for a number of consecutive years, and they will be continued; reference can be made to the standard work, the series Meddelelser om Grønland, published by the Commission appointed for the Geographical and Geological Investigation of Greenland, in which will be found all the Danish investigations made since 1876. The Danish West Indies, also, are being botanically investigated with great energy (see especially Botanisk Tidsskrift, published by the Danish Botanical Society). But among the dependencies and colonies of Denmark, Iceland is the one with regard to which botanical investigation is least ad- vanced; and it is the one which, owing also to its situation be- tween Greenland and the Færoes, will now present the most interesting field for research. Il is my hope Ihal a more Ihorough investigation will be commenced there as early as the summer of 1909. My heartiest thanks are due to the many contributors to this work, for the warm interest they have shown, and for the devoted labour and unwearying trouble they have given to the subject. EUG. WARMING. Botanical Garden, Copenhagen, November 3rd, 1908. GONTENTS. Vol. I (1901). ^ ' Page EUG. WARMING: Historical Notes on the Botanical Investigation of the Færoes 1 C. H. OSTENFELD : Geography and Topography 6 — Industrial Conditions 20 — Geology 24 — Climate 32 — Phanerogamæ and Pteridophyta 41 — Phyto-geographical Stndies based upon observations of »Phanerogamæ and Pteridophyta« 100 G. JENSEN : Bryophyta 120 — Phj'to-geographical Studies based upon the Bryoplij'ta 185 F. BORGESEN : Freshwater Algæ ... - 198 ERNST OSTRUP : Freshwater Diatonis 260 — Phyto-geographical Studies based upon the Freshwater Diatonis 291 E. ROSTRUP: Fungi 304 J. S. DEICHMANN BRANTH : Lichenes 317 Vol. II (1903). F. BORGESEN : Marine Algæ 339 E. OSTRUP: Diatoms from the Marine Algæ of the Færoes 533 C. H. OSTENFELD: Phytoplankton from the Sea around the Færoes 558 F. BORGESEN and C. H. OSTENFELD: Phytoplankton of the Lakes of the Færoes 613 H. DAHLSTEDT : The Hieracia from the Færoes 625 EUG. WARMING: The History of the Flora of the Færoes 660 Vol. III (1905—1908). F. BORGESEN : The Algæ- Vegetation of the Færoese Coasts with remarks on the Phyto-Geography 683 C. H. OSTENFELD: Additions and Corrections of the List of the Phanero- gamæ and Pteridophyta of the Færoes 835 GAZET PATURSSON: List of Popular Plant Names from the Færoes 864 ^' ^^> c. H. 0STP:NFELD: The Land-Vegetation of the Færoes 867 F. BOHGESEN ; Gardening and Trec-planting 1027 P. FEILHERG; Some Notes on the Agricultuie of the Færoes 1044 EUG. WARMING: Field-Notes on tiie Biology of the Flowers of the Færoes 1055 I. C. NIELSEN : The Insect-Fauna of the Færoes 1066 APPENDIX. F. BORGESEN and HELGI JONSSON: The Distrihution of the Marine Algæ of the Arctic Sea and of the Northenimost Part of the Atlantic p. I— XXVIII Separate copies were published of pp. 1 — 197 Februarj' 15, 1901; — of pp. 198-259 April 3; — of pp. 260— 303 April 20, — of pp. 304-338 April 30, 1901; — of pp. 339-532 Nov. 15, 1902; — of pp. 533—557 March, 1903; — of pp. 558— 612 Fehr. 25, 1903;— of pp. 613—624 August 15, 1903; — of pp. 625— 659 Aug. 31, 1903; — of pp. 659— 681 September 12, 1903; — of pp. 683-834 November 1, 1905; — of pp. 835—864 November 13. 1907; — of pp. 867-1026 Febr. 3, 1908; — of pp. 1027—1043 Febr. 3, 1908; — of pp. 1043-1070 December, 1908; — and of the Appendi.x Aug. 28, 1905. CORRIGENDA. Page 711, line 12 from bottom, for two feet read 100 feet 711, - 6 - - , for eight feet read eighty feet THE ALGÆ -VEGETATION OF THE FÆROESE COASTS WITH REMARKS ON THE PHYTO-GEOGRAPHY BY F. BØRGESEN. PREFACE. This paper contains the final report on my six journeys to the Færoes in the period 1895 — 1902. During my first three visits (1895, 1896 and 1898) I travelled about, partlyby boat, partly by the local steamer »Smiril«, but the last three times I visited the Islands, (1899, 1900 and 1902), I stayed on board the vessels stationed at the Færoes for the protection of the fisheries and for surveying pur- poses, the first two years on the gunboat »Guldborgsund«, in 1902 on the torpedoboat »Beskytteren«. In this way I have been able to obtain a far more thorough knowledge of the algæ-vegetation than would otherwise have been possible. I have for instance visited all the Islands, except Fuglø, and have been on a great many islets and cliffs, and on many remote coasts, where access might otherwise have been difficult. Further, the vessels of the Navy have not only helped me to make wide excursions, but their assistance has also been of advantage to my work as I then had the opportunity of working with larger appa- ratus than I could have used from a small boat. Thus in 1899 I had the use of a small trawl which helped me to obtain a rich material, and in 1902 a very large iron scraper was used, also with good results. INTRODUCTION. As is w^ell known, J. G. Agardh was the first to try to give a description of the different regions of the algal vegetation on the Scandinavian coasts in his w-ell- known work, »Novitiæ Floræ Sveciæ« (2)^ In this work he distinguishes between three regions : 1. Regniim Algarum Zoospermariim. 2. Regniim Alganim OUvacearum and 3. Regniim Algarum Floridearum, which are further divided into subregions. ^ This and the followlng figures refer to the Index to literature. Botany of the Færoes. 44 684 At the same period (1836) Lyngbye was dealing with the algæ- vegetalion in a paper entitled »Rariora Codana«, which was only piiblished much later (1879) by Professor Warming. Lyngbye makes a division into three Zones : firstly that of the Ulvaceæ, reaching from the surface of the sea to a depth of 30 feet, secondly that of the Florideæ, as far down as from 30 to 60 feet, and finally that of the Laminar iaceæ, from 50 to 90 feet. For each zone he mentions a number of specifically characteristic species, partly from Denmark, partly from the Færoes and from Norway. Lyngbye's division does not however seem quite satisfactory to me. It would have been most natural to determine the zone of the Florideæ as the lowest one, as in J.Agardh's division. Besides, several species (I refer only to the Færoese specimens mentioned by Lyngbye) do not occur in the depths mentioned. Thus Fiiciis loreiis, Scytosiplwn fdum lomentarius and Callithamnion arbiiscula are mentioned as belonging to the zone of the Florideæ; they are, however, all littoral. Lami- naria agarum is given as in the zone of the Laminariaceæ , but it has never been fonnd in the Færoes; to this zone he also refers Callithamnion arbuscula. It ~ seems as if his memory had been failing him, and most likely his strength had already grown less as it was but a short time before his death (compare Warming' s introduction [60, page 3]). It must be remembered also, that he wrote his paper many years after his visit to the Færoes. In 1844 Ørsted published his well-known work, »De Re- gion ibu s Marinis«, in wdiich, relying on the above-mentioned work of J. Agardh, he gives a detailed description of the algal vegetation of the Sound. He subdivides this into firstly, »Regio Algarum viridium s. Chlorospermearum« , reaching from the surface of the sea to a depth of 2 to 5 fathoms and embracing: »Snbregio Oscillatorinearum« above, »Snbregio Uluacearum« below; secondly, »/?e- gio Algarum olivacearum s. Melanospermear nm c<,conshi'mg of »Snbregio Fncoidearnm et Zosteræ marinæ« and »Snbregio Laminariearnm« below it, and whose upper limit Hes in depths of 3 to 5 fathoms and lower in 7 to 8 fathoms; fmally, Regio Algarnm pnrpnrearnm s. Rhodospermearnm, in depths of 8 to 20 fathoms. Ørsted seeks the chief explanation for this distribution of the Algæ in the varying degrees of penetration of the hght through the water, and, in connec- tion with this, in the different colour of the light at different depths. In addition to these authors, Areschoug, Kleen, Ekman and others have also contributed to our knowledge of the algæ of 685 the Scandinavian coasts, but it is only in different works by Kjell- man, especially in his fundamental work: »UeberAlgen region en und Algenformat ionen im ostlichen Skager Rack« (45) that a more precise limitation and terminology of the particular algal regions and communities are first introduced. Several papers on the algæ-vegetation have been published later; amongst the more impor- tant of these, which are at the same time interesting as a means of comparison with the Færoese algæ-vegetation, I need only men- tion those dealing with the Norwegian algæ by Boye (6), Gran (36, 37) and Hansteen (38), and the detailed description of the Greenland algal vegetation by Rosen vinge (71). With regard especially to the Færoes, Landt (57) is, so far as I know, the first to give some particulars of the occurrence and growth of some few algæ, in his list of Færoese algæ. Some short notes are also given by P. A. Holm (40, p. 203, 204). Rostrup, who States (72, p. 16 — 17) that many algæ are so abundant as to form regular forests of seaweeds, gives some information as to the habits of the algæ; thus he states, that Porphyra laciniata occurs abundantly on rocks washed at high tide. The paper by Sim- mons: Zur Kenntniss der Meeresalgen-Flora der Fær- oer«, published in 1897, is the first important contribution to our knowledge of the Færoese algæ-vegetation. In accordance with Kjell man, Simmons makes a division be- tween the littoral, the sublittoral and the elittoral regions, and gives a description of different formations belonging to these regions. Moreover he tries to compare the algal vegetation of the Færoes with that of the coasts of the adjacent countries, and his main re- sults are briefly as follows: the Færoese algal vegetation is cha- racterized by its strong concentration in the littoral and sublittoral regions and by its disappearance at no very great depth; secondly, an algal vegetation closely connected with that of the Færoese seems to occur in Nordland on the western coast of Norway, which lies far more towards the north than the Færoes. As to the first statement, however, I do not consider it quite correct; as a vigorous growth of Laminaria hyperborea is found even at a depth of about 15 fathoms (as to this I shall give further information later), and on the whole the algal vegetation is found down to a depth of 25 fathoms, wherever circumstances are favourable to its occurrence. This agrees with the observations made in adjacent countries. Further investigations of the algal vegetation of Nordland 44* 686 will probably pro ve, that Ihere is no little resemblance be- tween Ibe algal vegetation of Nordland and that of the Færoes. Kl een has expressed the same view (51, p. 6— 7, Note); but on the other hånd, Kleen's descriptions are too brief and too few to allow any final conclusion lo be drawn from them. Moreover his statement (page 9) that »on steep, overhanging cliffs, in immediate vicinity of the sea, no sorj; of luxuriant vegetation is found above the lowest water mark, where mostly nothing but small tufts of Ceramiiim acanthonotiim and Callithanmion arhnscnla occur« ^ is quite incompatible with the luxuriant littoral vegetation of the Færoes. My investigations on the algal vegetation of the Færoes were published in Danish in the spring of 1904, and the present issue in English is based essentially on the Danish edition. My work has been sharply attacked by Porsild and Simmons, to whose criticisms I have already replied in the »Botaniska Notiser«. Referring to these notices for particulars, I may briefly mention here, that the main criticisms of these authors are directed against my view, that it is quite possible, that the Færoese algal flora may be transported over the sea to these islands. Thus, Porsild maintains that it is a »physical impossibility« for algæ, for example, to be carried from Ireland or the west coast of Scotland across the Gulf Stream to the Færoes. In his opinion algæ from the west coast of Norway, from the Shetlands, Orkneys and Scotland, must be carried far into the Arctic Ocean before they could reach the Færoes; a journey taking at least 3 years, probably more! Porsild further maintains, that but few of the Færoese algæ can float; also, that 26,7 ^/o of all the P^æroese species of algæ is what he calls »strictly sublittoral«, meaning thereby those algæ which are never found, in his opinion, so high up as low water mark, and which therefore can never have the chance to fix themselves on floating timber or the hke and thus be carried about in the sea. Finally, 12,4% according to Porsild are calcareous algæ, either living in calcareous shells, or incrusting stones, etc. »These algæ can certainly not float, and by far the majority of them are sublittoral forms, which can also not fix themselves to stones or shells borne by algæ which can float«. As to the biological conditions for the algæ d uring their drift to the Færoes, Porsild maintains in opposition to my 1 »På branta, ytterst mot hafvet belågna klippor saknas all rikarc vegetation ofvenfor lågsta vattenmårket; och man tråffar derstådes vanllgen blott små tufvor af Ceraminm acanthonotum och Callithamnion arbusciila:. 687 view: 1) that the biological conditions in the currents, which lead to the Færoes, are very variable and different from those at the Færoese coasts, 2) that marine algæ are very sensitive to changes of temperature and salinity, and also, 3) that the white light of the surface may have a destructive influence especially on the sublittoral algæ, 4) that the algæ could neither fix themselves again after the drift across the sea, nor develop new reproductive organs, these being lost presumably on the way. Consequently, even though cases are known of marine algæ having drifted a long way, yet the pro- bability of any effective distribution occurring in this way is ex- tremely small, and so far as known it has never been observed. Further Porsild opposes the possibility that algoid spores may be distributed widely by the oceanic currents, and considers navigation of no importance as a means of distribution. In faet, according to Porsild, the algal flora of the Færoes can only have arisenthrough some postglacial communication with land. Simmons entirely agrees wåth Porsild 's view-. He also endea- vours moreover, to criticise my description of the algal commu- nities and my comparison of the Færoese flora with that of the neighbouring lands, and, for this purpose gives a list of the algal forms of the North Atlantic as well as of the Arctic Ocean. I shall nol enter upon further details here, however, but merely mention that his list is not drawn up with sufficient accuracy to enable it o form the basis of a thorough comparison. I. ON THE EXTERNAL CONDITIONS AFFECTING THE ALGAL VEGETATION ON THE FÆROESE COASTS. 1. Climatic and hydrographic conditions. a. Temperature and Salinity of the Sea. The Færoes, lie in the Atlantic Ocean in 61 <^ 23'— 62'' 24' N. Lat. and 6° 14' — 7*^41' W. Long. The chmate is markedly insular, the temperature very uniform, in summer low, but in winter rela- tively high. Rain and fogs are frequent, and the climate on the whole stormy and rough. Consequently the sea is most often in motion, and on exposed coasts there is much surf. This very rough climate is mostly due to the faet that the boundary line between the warm Gulf Stream and the cold East Icelandic Polar Current is found at, or at anv rate near bv, the 688 Færoes. The boundary line between these Iwo currents is in- fluenced by the predominant direction of the wind, and thus goes somelimes nort band east, sometimes west and south, of the Færoes (see Martin Knudsen 52). In a morere cent paper (»Havets Natur- lære« 53, p. 29) Knudsen savs: »It seems as if the Polar Current hardly ever reaches so far down as to surround the Færoes, but it may happen. On the other band the Polar Current rarely re- cedes far northward from the Færoes«. Whilst the Gulf Stream, as mentioned by Knudsen (52, p. 158), bas a rather bigh temperature and a salinity above 35,25 %o, the Arctic current is of a much lower temperature, which may even reach the freezing point of the sea in winter and spring, and its salinity is below 35,25^/oo. In summer and autumn a thin layer on the surface of the Polar Current will, however, attain a temperature almost equal to that of the adjacent water of the Atlantic Ocean. The Ingolf expedition passed the dividing line between the two currents several times, and it was then observed: »that in passing from the Atlantic Ocean to the East Icelandic Polar Current the salinity of the surface was reduced from more than 35,25 Voo, to less than this. A slight, but still perceptible, fall of temperature was likewise observed«. In spite of the proximity of the colder Polar Current, the in- fluence of the Gulf Stream predominates, causing the temperature of the sea to be very uniform all the year round, and in winter especially very high. When Simmons writes (78, p. 263), that it is rather low, his statement must onh' be applied to the temperature of summer, which is indeed much below that of the west coast of Norway in the same latitude, where an almost corresponding temperature is only met with much farther north, in Nordland. If we look at a map of the Northern Ocean, showing the tempera- ture of the surface in summer (see for instance Hjort, Nordgaard and Gran: Report on Norwegian Marine Investigations 1895 to 1897, Bergen 1899), we observe that the isotherm of 12*^ C. is drawn midway across the Færoes. The isotherm first bends a little south east towards the Shetland Isles, and then north west to the centre of Nordland. The isotherm of 11*^ C, which passes close by the west and south coasts of Iceland, exlends northward round the Færoes at no great distance from the Nordreoerne, and ends on the west coast of Norway as far up as the centre of Lofoten. The isotherm of 13" C. passes at a somewhat great 689 distance southward round the Færoes, goes between the Shel- land Isles and Ihe Orkneys, and ends on the west coast of Nor- way, a httle north of Trondhjem Fjord. In winter and spring on the other hånd the water is warmer round the Færoes than on the west coast of Norway. According to Mohn: »Den Norske Nordhavs Expedition«, 2 yoI. Christiania 1883, the isotherm of 6*^0. runs in March midway across the Færoes, then it bends north- ward, then southward, and stops almost in the centre of the east coast of Scotland. According to observations made near Thorshavn (see W i 1 1 a u m e- Jantzen 84, p. 29), the annual mean temperature of the surface of the sea is 7,8° C; from January to March its temperature is 5,5° C, and from July to September 10 — 10,5° C, in exact accordance with the average observations of 20 years. The lowest mean tempera- ture of the surface of the sea was in one monlh (March) 4,25° C, and the highest in an equally long period (August) 11,75° C. These figures clearly show the very small difference of temperature in the sea round the Færoes, where the greatest difference observed between the mean temperature of the warmest and the coldest months thus only amounts to 7,50° C. Ice is also unknown, and it is only in the inmost part of the larger fjords such as Skaale- fjord, litlle affected by the tides, and where much fresh water streams into the sea, that the water may sometimes be covered by a thin crust of ice. The above-mentioned figures only refer to the surface of the sea, but as the marine algæ are found down to a depth of about 25 fathoms, (for the Færoes I believe that this may be stated as the lowest limit), it would of course be of interest to our subject to know the temperature as far down as this depth. Such observa- tions, however, do not exist, so far as I know, but on my application the officers of the »Guldborgsund« in 1899—1900 were kind enough to obtain some hydrographical data by aid of a reversing thermometer lent me by the Metereological Institute. The accompanying table shows the results of these observations and indicates that the differ- ence of temperature between the surface of the sea and the lower layers of water is exceedingly small. During the warmer period of the year the temperature of the surface rises slightly, yet seldom more than V2°, whereas in winter the water is somewhat warmer deep down than on the surface. In the fjords only, there may be a greater difference (see e. g. the observation "/2 Klaksvig) : it is due 690 Deep-water temperatures in the sea around the Færoes 1S99-1900. (The temperature is given in degrees Celsius and is everywhere -|-)- Locality Sur- 5 10 15 20 25 lime j.^^^ fathoms fathoms fatlioms fatlioms fathoms Fuglefjord 1900 p. m. '«/i 1 iVa 6.5 6.80 6.80 Bottom 7.00 Klaksvig^ 1900 10/2 noon 12 1.80 Bottom 5.50 Sjov (Stromo) , . . 1900 ^V2 p. m. 6 4.0° 4.30 4.50 4.90 Bottom 5.10 Sorvaag 1900 28/2 a. m. 10 4.8" 4.30 5.30 Bottom 5.40 Kvannesund .... 1900 "/2 p. m. 3 5.50 5.50 Bottom 5.50 Lopra 1899 "/e 9.00 8.80 Bottom 8.70 Viderejde 1899 "/7 9.20 8.80 8.70 8.8 8.90 Bottom 8.80 Thorshavn 1899 9.40 9.80 Bottom 9.2 Trangisvaag 1899 9.40 9.10 9.00 Bottom 9.00 Tværaa 1900 Vs p. m. 4V2 10.3 10.0 10.0 Bottom 9.8 Porkere 1900 p. m. 3 10.3 10.0 Bottom 9.9 Tværaa 1900 2«/8 p. m. 10.8 10.2 10.0 Bottom 9.9 Tværaa 1900 p. m. 7 10.4 10.0 10.0 Bottom 10.0 Thorshavn 1900 V9 a. m. 7V2 10.2 10.0 Bottom 9.9 Vestmanhavn . . . 1899 p. m. 2V2 10.10 Bottom 10.00 Thorshavn 1899 ^79 a. m. 9.90 9.90 Bottom 9.8 Kongshavn 1899 p. m. 3 9.80 9.80 9.30 9.80 Bottom 9.80 Trangisvaag 1899 a. m. 7 9.00 8.50 9.10 Bottom 9.10 1 i Fundingsbotten. . 1899 «/io a. m. IV2 9.50 9.30 9.2 N. f. Ennebjærg (Viderejde)- . . . 1899 Vu a. m. IOV2 ' 9.6« 9.70 ] 9.8 ^ Tlie surface of the sea covered with snow. Calm. * Change of current; depth, 45 fathoms, the thermometer was broken at 15 fathoms. 691 liere to the more stagnant water, the surface of which may be some- what cooled at a low temperature ol" the air, whereas the water of the open sea and of the sounds is constantly renewed by the strong current and mixed with the lower water-layers, so Ihat it does not have time to be cooled to any degree worth mentioning. The algæ growing in deep water hve consequently at more uniform tempera- tures than tliose growing in shallow water. Near the shore among the seaweeds and in tide pools, the sun of course may warm the water to a considerable extent, so as to make it almost lukewarm; thus in a somewhat large water-basin at Skuo, 25° C. was observed on June Uth, 1900, and in smaller pools the temperature may certainly rise still higher. As to the salinity, it varies, as already remarked, according to whether the salter water of the Atlantic Ocean (above 35,25 Voo) or the less salt water of the Polar Current reaches the coasts; the difference however is never very great. Fresh water streams into the sea from almost all parts of the Færoese coasts, but on account of the strong oceanic currents it has hardly any effect in comparison with the large volume of cir- culating seawater, and as a rule it is only of local importance to the growth of algæ at the outlets of brooks and waterfalls. But at the bottom of larger fjords, where the water is more seldom re- newed, the fresh water from the land is perceptible to a greater extent, and here the water often becomes more or less brackish. This has no small influence on the algal vegetation, which in these piaces is extremely poor in species. b. Tides. Oceanic Currents. Tides occur almost everywhere on the coasts of the Færoes. The difference between high and low tide is not great, as the tide does not rise to anything like the heights reached on the coasts of Norway and Scotland. On the western side of the istands, where the tidal wave is strongest, the difference between ebb and flood wili at most amount to 8 — 10 feet during the spring-tides, on the eastern side to hardly more than 4 — 5 feet. The accompanying observations on the tides, made in the sum- mer of 1900 and kindly placed at my disposal by the officers of the »Guldborgsund« further illustrate this. The observations extend 692 over quite a monlh, during which period Ihe weather was calm, so that storms had no disturbing effects. Vaagfjord (open to tlie east). Almost at tlie end of tlie fjord, tlie difference between liigh and low water is a little more tlian 3 feet wlien the current is strongest (spring-tide); with »fair« eurrent (neap-tide) only ^/4 feet. Higli and low water occur very regularly in tliis fjord. Trangi.svaagfjord (open to the east). High and low water do not occur so regularly in this fjord as in Vaagfjord. When the current is strongest the difference is generally about 3 feet, but one day during the period of observation it reached 6 feet; with »fair« current the dif- ference is only 1 foot. Sorvaagfjord (open to the westi. Regular tides occur. At spring-tide there is a difference of ca. 6 feet, at neap-tide of ca. 3 feet. Midvaagfjord. The fjord is open to the south-east, but on account of the western situation of the island the difference is considerable. Re- gular tides occur. At spring-tide there is a difference of ca. 5V2 feet, at neap-tide of 2V3— 3 feet. Vestmanhavn. The fjord is open to the west, and at Vestmanhavn- sund to the north-west and south-west. Regular tides occur. At spring-tide there is a difference of ca. 6 feet, at neap-tide of ca. 2 feet Sandsbugt (open to the south-west). Regular tides occur. At spring- tide there is a difference of about 5V2, but it may reach to about 7 feet; when the current is »fair« the difference is hardly 2 feet. As appears from these observations, the greatest difference noted with certainly is about 7 feet; but there is no reason to doubt that there are piaces where a somewhat greater difference may occur. This is probably the case on the western side of Sydero and on the north-western side of Stromo. At F'amien on Sydero, the in- habitants have told me that the difference may amount to from 8 — 10 feet. These figures are however small in themselves and would only mean a littoral algal vegetation of relatively limited extent compared with that of the western coasts of Norway and of the British Isles, were it not that other circumstances aid tlie algæ of exposed coasts to grow even far above the highest water mark. The tidal wave, or, as they say in the Færoes, the »West- fall« and the »Eastfall«, produces a very rapid current, changing regularly every 6th hour^ in open fjords and especially in the sounds. At spring-tide, wdien the current is most rapid, it may in cer- tain piaces, such as Vestmanhavnsund, flow as fast as 8—10 miles an hour, and in such piaces the larger algæ are of course exposed to a very great strain. Whilst on all exposed coasts and in most fjords and sounds 1 Nolsofjord fiowever excepted; on account of peculiar circumstances the cur- rent tliere runs 8 hours one way and 4 liours tlie other. 693 there are tides and consequently cunenls, there is a large tract of water without either tides or currents, or with at any rate almost imperceptible ones. This is found between Stronio and Ostero, bounded on the south by a line drawn from Hojvig to Næs on Ostero, reaching on the north as far as Kvalvig in Sundelaget, thus including the three large fjords Skaalefjord in Ostero, Kaibak- and Kollefjord in Stromo, besides »the Sounds« between both Islands. This almost stagnant area of water naturally shelters an algal vegetation of a peculiar kind. It is e. g. the true habitat of Laminaria færoensis, and this characteristic alga reaches here its highest development. c. Action of the Waves. Exposed Coasts; sheltered Coasts. As already mentioned, the surf breaks more or less vigorously on all the coasts of the Færoes during the greater part of the year. The heavy, unbroken waves of the Atlantic roll in on the mostly steep and rocky coasts, and during the storms of winter the waves may be strong enough to move rocks, as was the case at Bosdalafos on the western coast of Vaago, about 80 feet above the level of the sea. It may even happen, that the sea rushes into Sorvaags Lake itself, to which Bosdalafos is the outlet. It is not unusual for the surf to reach 100 feet up the cliffs, and it has even been said to reach several hundred feet up. Even in summer, when the weather is fme, there is almost always some surf produced by the swell, when it meets the land. Still periods may occur, espe- cially in summer, when the weather is so fine and the sea so calm that the surf almost disappears. When the algal vegetation of a countrj'^ is investigated, it will soon be observed, that the more or less exposed condition of the localities is of great importance. The open coasts, constantly ex- posed to the surf, are covered by an algal vegetation differing greatly from that on sheltered coasts. The difference is so great, that with some few exceptions the species growhig on exposed coasts are of quite another kind than those growing on sheltered coasts. This is quite easy to explain, as the mechanical influence of the waves is not the only influence affecting the algæ of open coasts in contrast to those on sheltered coasts; it is also probable, that the water is more rich in oxygen, more nutritive and often salter than in sheltered piaces, and further the water coming in con- tact with the algæ is more frequently renewed by the dashing of the 694 waves on exposed coasts than is Ihe case on sheltered coasts, where the change of substance is likely to occur more slowly. From the composition of the algæ-vegelation one may tell at once whether the coast on which they grow is exposed or sheltered, and if we come from the open sea towards a sheltered place we may no- tice how the algal vegetation changes in character. The change often comes on gradually, but sometimes a sudden bend in the coast may as suddenly change the character of the vegetation. This faet may often be observed, and by way of illustration I may mention an investigation I made from the mouth to the head of Kalbakfjord. I landed at short distances along the southern side of this fjord , starting from Hvidenæs and noting the most impor- tant species of algæ. Tides and current are liere scarcely perceptible. At Hvidenæs there is a perfect »open sea« algal vegetation. Some small creeks with calcareous gravel excepted, the coast is here mostlj' steep, almost vertical, and as is usual in such piaces is covered from low water mark downwards by a dense Alaria- vegetation, and above it by the Balamis-he\\., which supports several of the smaller red and brown algæ, e. g. Ceramium acanthonotum, Polysiphonia urceolata, Callithamnion arbiisciila, Ectocarpiis litoralis, etc. The littoral vegetation is however rather scanty on these ver- tical cliffs, which is probably due to the faet, that the sea is often relatively calm at this place, especially in summer, Nolso as well as Ostero yielding it some shelter. A little west of Hvidenæs (the place is marked 1 on the accompanjdng sketch, Fig. 151) the following species were the most prominent, passing upwards from the sea level: Alaria esciilenta, Himanthalia lorea, Gigartina mamillosa, Cera- mium acanthonotum, Callithamnion arbuscula, Scijtosiphon lomentarius in small numbers in shallow pools, Fucus spiralis f. nana and Por- phijra umbilicalis; as may be seen, exactly the vegetation of exposed coasts. A little farther on, at 2, where the shore is likewise steep, but with scattered rocks fallen from the overhanging clifTs, almost the same algæ were found: Alaria esculenta, Himanthalia lorea, Gi- gartina mamillosa, Fucus spiralis f. nana, Porphijra, Enteromorpha intestinalis var. genuina and var. micrococca. We thus have still the algal vegetation of exposed coasts, which was also found at 8, al- though the somewhat larger forms of Fucus indicate a somewhat more sheltered spot, viz. : Alaria esculenta and Laminaria digitata, Himanthalia, Polysiphonia urceoluta, Gigartina mamillosa, Ceramium acanthonotum and C. rubrum, Callithamnion arbuscula, rather large 695 specimens of Faciis inflatiis, Fiicns spiralis f. nana, Porphyra iimbili- calis and Enteromorpha. At 4, the algal vegetation has also the character of algæ be- longing to somewhat exposed coasts. The shore was here rather steep with almost flat rocks in front close to the water's edge; here were found Laminaria digitata, Alaria esciilenta with a sub- growth of Corallina, Gigartina mamillosa, Chordaria flagelliformis, Scytosiphon lomentariiis, Ceramium nibnim, large specimens of Fiicus inftatiis with Elachista fiicicola, Porphyra iimbilicalis, Enteromorpha intestmaUs var. micrococca and Ectocarpus litoralis. At 5, the flora has still almost the same character; here were found Laminaria Fig. 151. Sketch of Kalbakfjord after the niap of the generalstafl'. digitata, L. saccharina, Alaria esculenta, Himanthalia lorea, Rhody- menia palmata, Gigartina mamillosa, Ceramium rubrnm, Dumontia fdiformis, Acrosiphonia spec. (albescens?), Fncns inftatus (large) with Elachista fiicicola , Fuciis spiralis (large) and Porphyra. Close by the village of Sand there is a small and low mass of rocks (6). On its eastern side, towards the mouth of the fjord, the following species were found: Laminaria digitata, Alaria esculenta, Himanthalia, Rho- dymenia palmata, Gigartina mamillosa and Ceramium acanthonotiim, Acrosiphonia, Fiiciis inftatus (large), Porphyra and Enteromorpha in- estinalis, the principal vegetation thus resembling that of exposed coasts; on the other hånd, the western side of the rock, facing landwards, was covered by the typical flora of sheltered coasts, with Fiiciis uesiciilosus at the top and next to that large luxuriant masses of Ascophyllum nodosum covering the almost horizontal sur- several pools of different sizes filled with water, the bottoms covered by Corallina officinalis, and Leathesia difformis was growing epiphy- tically on the latter in great profusion. Monostroma fusciim, M. Gre- 696 villei, Chordaria flag elli for mis , Ceramium rnhriim and Acrosiphonia spec. were also tbund. From Sund a little farther up, the fjord inclines slightly towards the north, with Ihe result that the algal vegetation remains si- milar to that of somewhat exposed coasts; at 7, where the coast is fringed hy steep rocks, were found : Alaria esciilenta, Laminaria digitata, Rhodymenia palmata, Fiicus inflatiis (large), Ceramium rii- brum, Porphyra together with Bangia and Urospora pemcilliformis, and still farther, at 8, Laminaria digitata, Alaria esculenta, Coral- lina officinalis, Gigartina, Cladophora riipestris, Fucus inflatiis (large), Fiiciis spiralis (large), Porphyra iimhilicalis, Enteromorpha, and so on, that is, a vegetation typical on the whole of somewhat exposed coasts, but large and vigorous specimens of Fiiciis spiralis and Fucus inflatiis, up to 2 feet long, now occur. That Ihe waves may wash heavily at times even so far up the fjord as this, is clear from the faet that there was some surf even on the day of my visit, in spite of the great calmness of the sea. The fjord now bends again more towards the west, and the vegetation is henceforward typically that of sheltered coasts. At 9, the beach slopes gently and is covered with stones of different sizes on which are found sublittorally Laminaria saccharina and Chorda filiim, [Alaria is now no longer found), Himanthalia lorea and Fucus inflatiis (on open shores, the two last are always littoral), the lastmentioned still continuing far enough upwards to become littoral, and to form large clumps on the stones; then come Fucus spiralis and some few Porphyra iimbilicalis. A little farther, at 10, in water one foot deep, were found: Chorda filum, Laminaria sac- charina, Ceramium riibrum, Monostroma fuscum, Enteromorpha Linza, Ectocarpus litoralis (in large, thick masses), Dictyosiphon foeniculaceus, Stictyosiphon tortilis, Scytosiphon lomentarius; in shallower water: large clumps o^ Fucus inflatiis; and at the surface: Fucus vesiculosus and Fucus spiralis, the latter partly dried. Still farther, at 11, were noted: Laminaria saccharina, Ceramium rubrum, Acrosiphonia, En- teromorpha Linza, Monostroma fuscum, Dumontia filiformis, Fucus inflatiis, all sublittoral ; and at the w^ater's edge and a little above : Ascophyllum nodosum, Fucus vesiculosus and Fucus spiralis. Al the head of the fjord, Enteromorpha clathrata, E. intestinalis, Ectocarpus litoralis, Ceramium rubrum and Fucus inflatiis were found at low water; farther also, Fucus vesiculosus and Fucus spiralis lit- torally, and highest of all Pelvetia canaliciilata. Much fresh water 697 flows into the sea liere, which is therefore often very brackish, and at certain piaces almost fresh. Even at the mouths of the large effluents only Enteromorpha intestinalis and a few Ectocarpiis litoralis are found, but at a short distance from the exit specimens of Fiiciis, Enteromorpha ckiihrata, Ceramium rubrum and others. Thus it is evident, and examples can be met with everywhere in the Færoes, that in passing from exposed to sheltered coasts the flora of these two localitics will be found to vary greatly. Further illustration of this will be given in the section concerning communities of algæ. Flowing or stagnant water are likewise of great importance to the algal vegetation and give it a widely varying character. The current constantly carries fresh sea-water, and thus new nutritive substances, to the algæ, whereas the vital functions must naturally proceed more slowly in stagnant water, where the supply of fresh sea-water is much smaller. If we examine the same forms of algal vegetation, e. g. the Laminaria hyperborea-associaiion, we likewise find it much more vigorously developed than in piaces where the Avater is stagnant. Most likely this is also due to the mechanical influence of the current; at any rate we find a very well developed algal vegetation in piaces where there are rapid currents. Just as a current causes fresh-water piants to become elongated, many algæ growing in piaces much exposed to currents have long and rather narrow fronds. I have seen very fine examples of this at Gotebue near Mjovenæs on Ostero, on dredging in ca. 10 fathoms of water. Here a Laminaria hyperborea -asaociaiion was found, in which the main plant was strong and robust, but the lamina had narrow laps; and growing epiphytically on this, were some Deles- seria alata a foot in length, and some still longer Delesseria sinuosa and Odonthalia dentata, all with very narrow fronds. On the other band, the thallus of other species e. g. Euthora cristata and Rhodo- phyllis dichotoma was broad and robust in this place as elsewhere in rapid currents and on exposed coasts. The same species of algæ is often very different in its appear- ance according to its habitat. Thus it seems, that many algæ in sheltered piaces are inclined to form proliferations, that is, long, thin appendages from the otherwise broad thallus; fine examples of this are shown by Delesseria sinuosa, which in sheltered piaces has a peculiar form, f. lingulata, remarkable for its numerous narrow proliferations, and Enteromorpha intestinalis, a variety of 698 which, var. prolifera, has by maiiy authors been designated a separate species, but which is most likely nothing but strongly proHferating individuals, ofteii loose and lloating, growing in sheltered piaces. In other species, such as Rhodop hyllis dichotoma, the thailus becomes almost filiform wheri it grows at the bottom of fjords, but may become a centimeter broad in the open sea, as mentioned above. Kjellman has given a description and sketch of this peculiar narrow form (48, table XII, fig. 3). Exactly the same may be said with reference to Euthora cristata, which, as mentioned above, is robust with rather broad fronds in the open sea, whilst these are almost filiform in fjords. According to Rosen vinge (71, p. 227), similar conditions have been observed in Greenland. Other species are also more or less transformed in stagnant water, e. g. Himanthalia lorea, the receptacles of which may be- come irregularly swollen; Laminaria digitata, where the lamina, as is well known, is split into a great many lobes, and which on ex- posed coasts with rapid currents has a form known as f. stenophylla, considered a separate species by many algologists, is often found in piaces of stagnant water with its lamina undivided and more or less sinuous and often urceolate in shape (f ciiciillata). The fronds of Alaria escnlenta and Laminaria saccharina likewise change ac- cording to the habitat. A special form of the latter, f linearis, is found on exposed coasts and is remarkable for a narrow, but thick and robust thailus (see fig. 85 of my flora), whereas its broad forms, f. biillata and f. grandis, are found in sheltered piaces or in deep water: Alaria too becomes broad in sheltered piaces without currents (the narrow form from exposed coasts is drawn in my flora p. 449, fig. 84). It is most peculiar, that similiar conditions may aet quite differently on different species of algæ, as may be seen from the preceding. I am unable to explain this satisfactorily. d. Temperature and Humidity of the Air. So far as the littoral algæ are concerned, and especially in the Færoes where the littoral algal vegetation often reaches far above the highest water mark, the warmth and humidity of the air are naturally of great importance and may therefore be briefly men- tioned liere. With regard firstly to the amount of heat, the average tempe- rature is 6,5 '^ C. according to Willaume- Jantzen. January, the coldest month, has a temperature of 3,2'^', July, the hottest, 10,8° C, 699 and according to observations diiring 25 years the lowest tempe- rature is — 11,6*', the highest 21,2°, but these extremes are verj^ rare and on the whole of very short duration when they occur. Thus it is evident, that the temperature of the Færoes is rela- tively high in winter, low in summer, and very uniform all the year round. It is of course the direct influence of the sun that is of special importance to the algæ, and on hot, sunshiny days the amount of heat and consequently the evaporation and drying in- crease considerably. But days of really hot sunshine are as a rule few, (the Færoes have no so-called summer days, according to Willaume-Jantzen); moreover, the sun's warmth acts directly, as a rule, for but a short period and is soon replaced by rain, fogs or clouds, so that the danger of the algæ on the Færoese coasts being exposed to high temperature and consequently to evaporation is greatly diminished. The amount of humidity in the air is rather high in the Færoes; for the whole year it is 82, in summer a little more, about 8o, in spring a little less, about 80. Deviations naturally may occur, but are generally of short duration. The very uniform and low mean temperature of the air to- gether with the high degree of humidity, both varying but a little from the normal, are certainly of great importance to the littoral algal vegetation, the very luxuriant growth of which is certainly due for the most part to these conditions. This view has partly also been put forward by Si m mons (75). On page 263, he writes: »Das Auftreten vieler Arten in weit hoherem Niveau als sonst, kann deshalb, was die fåroische (und wohl auch Nordlands) Algenvege- tation betrifft, nur oder wenigstens am besten durch die Temperatur- verhåltnisse erklårt werden.« In addition to the temperature, the little direct sunshine, the weaker light on the whole, the humid air with much rain and fog, and especially the rough sea are naturally likewise imporlant factors. e. Light. The climate of the Færoes is, as already mentioned, rich in fogs, rain, and cloudy weather, and on the other hånd rather wanting in sunshine. According to Willaume-Jantzen, the amount of clouds at Thorshavn is 7,4 (O = clear sky; 10 = over- cast); and the Færoes have in all 178 »cloudy days«, that is, Botany of the Færoes. 45 700 days when the amount of clouds stands at 8. We must add that the amount of clouds is the greatest and fogs the most frequent at the brightest time of the year. The direct influence of the sun on the algæ- vegetation is thereby still more diminished just at the time of the year, when it niight be most efTective; at the darker time of the year the sun is so low in the sky, that ils influence is but very slight. Thus the faet, that the algæ -vegetation on the coasts of the Færoes does not grow at any greater depth, may most likely be ac- counted for by the small amount of sun and light. By dredging in a depth of 25 fathoms, I have sometimes found some tole- rably well developed specimens of some Florideæ, but no vegeta- tion whatever will by any means be found belov^^ 25 — 30 fathoms. This is in agreement with Ro sen vinge (71, p. 233), when he says: »As to the Arctic and the northern part of the Atlantic Ocean, all investigators seem to agree in this, that below 20 fathoms nothing but a scanty algæ- vegetation is in any case to be found, whereas it is very usual for the vegetation to reach as far down as this.« In more southern countries the algæ -vegetation will, however, be found at much greater depths. At Capri in the Bay of Naples, in clear water, Berthold (5, p. 414) found a luxuriant algæ-vegetation of deep-water forms at a depth of 120—130 metres, that is, almost three times as far down as at the Færoes ^ It is true that algæ have been said to be found in very deep water, even in arctic districts, but these statements are surely in so far incorrect, as the algæ fished up in deep water were certainly floating and not fixed to any substratum, which has been pointed out by Kjellman and Rosenvinge. On the whole, the faet, that the intensity of light diminishes and its colour changes as depth increases, is one of the most impor- tant factors as to the distribution of the algæ. As to the Bay of Naples, Berthold points out (5, p. 415), that among algæ growing in greater depths the Florideæ are certainly predominant, whereas only a small number of Florideæ and Chlorophyceæ but especially the bulk of brown algæ seek direct sunshine in shallow water. This is, however, almost in agreement with what has been observed at the Færoes. I have found that the Laminariæ and other larger, 1 According to Sauvageau (77, p. 234—5, note), Rodriguez observed that the extreme limit to which the algæ -flora reached was 160 metres in the middle of the Mediterranean Sea, near Minorca. 701 brown algæ gencially disappear at a depth of a])oul 15 — 20 fa- thoms, leaving an almost pure vegetation of Florideæ, with the exception of some green, and bluish-green algæ, living in testa, and, however, often turning reddish liere. On the beach, especially in its lower part, as well as in the upper part of the sublittoral region, where light is still intense, a vigorous vegetation of lirown algæ is found. To tliese must be added several algæ, especially some green and bluish-green at the upper part of the beach, and at the lower part several red algæ. That this distribution is principally, perhaps solely due to light, its quantity as well as its quality, may easily be proved by an investigation of the algæ -vegetation of one of the numerous caves on the coasts of the Færoes. When rowing into such a cave, it will be observed, that the species of algæ, common at the entrance, gradually disappear and are replaced by a great many difTerent sublittoral Florideæ, directh^ under the surface of the w^ater^. This will be more thoroughly explained later on, when the cave vegetation is dealt with. This faet has also been men- tioned by several investigators, e. g. Berthold and Falkenberg (18, p. 220). But whilst Berthold only seeks the cause of the distribution of algæ in the intensity of light, Gaidukow^ main- tains that it is entirely due to its quality. Founding his opinion on Engel mann's well known investigations as w^ell as on his own, Gaidukow has recently in an interesting and instructive paper (35) clearly shown, that the distribution of the algæ is not so much due to the intensity as to the colour of the light. The faet is, that the green and bluish-green algæ grow uppermost, as the red ' In this connection it should however not be forgotten, that it seems as if some sublittoral species on the shores of the Færoes are only to be found at a certain depth. never directly under the surface of the sea, not even in dark piaces, e. g. several species of Lithothamnion, CallophylUs laciniata, RhodophijUis dichotoma, Antithamnion Pliimnla, Desmarestia aculeata and D. viridis, etc. As to Greenland, some species are likewise nientioned by Rosen vinge (71, p. 228) as never occurring near the surface of the sea. He thinks that this is due to the faet, that temperature and salinity both become less variable as the depth increases. Several of these algæ may however be found littorally in other districts. Kjellman for instance states, that Desmarestia aculeata and D. viridis can occur in the littoral region, on the west coast of Norwa}-, and Antithamnion boreale is sometimes met with littorally in the Norwegian Polar sea. Le Jolis (58) writes that CallophylUs lacinata is to be found at the lowest part of the beach. As these algæ do not appear on the shores of the Færoes at such a high level, even in the caves, it is most probabU' due to the faet that they cannot bear the strong surf here. 45* 702 ravs which they especially need, already become fainter at a small depth. The red algæ grow deepest, as they need the green rays, occurring at greater depths, the most; fmally the brown algæ are specifically fitted for growing at a depth between that of the green and the bluish-green algæ on one side, and that of the red algæ on the other side. He calls our attention to the faet, that the algæ are very well able to accommodate themselves to the colour of the light of their habitats. He refers to the interesting observation made by Nadson, that certain Cyanophyceæ and Chlorophijceæ are represented by green specimens, near the surface of the sea, and by red, in deep water. This may also happen on the coasts of the Færoes, as I have already men- tioned. Nadson thinks for instance, that Conchocelis rosea is a deep water form of Ostreobium Oiiekettii. Bert hold and 01 tm an ns determine the Florideæ as shade piants, but this is criticised by Gaidukow, who asserts that Olt- mann's experiment only confirms Engelmann's theory, that the Florideæ of deep water have the typical colour of the Florideæ, be- cause they grow in green and blue light, whereas they turn brown and yellow near the surface in the white light ^ The Engelman n- Gaidukow theory on the whole coincides with my observations ; still it seems to me that the quantitj^ of the white light must practically be of no small importance not only to the colour of the algæ, but also to their distribution. On the coasts of the Færoes, on the beach, and even above it, a great many Florideæ of a deep red colour are found, perhaps some- tiraes of a more reddish-brown, but I have hardly ever seen them turning almost light yellow as the same species do, for instance, on the west coasts of Norway or on our own coasts^. No doubt this is due to the weaker light at the Færoes, where fogs are fre- quent, and the sky much overcast; and as a vegetation of Flo- rideæ, typically red and including some of the forms usually belonging to great depths, is often found in small inlets between Thorshavn and Arge, in water not even 2 fathoms deep and overshadowed by Laminariæ, this is certainly also due to the ^ In the West Indies Florideæ growing in shallow water, exposed to direct siinlight, are often perfectly green, or bluish-green, e. g. Grateloiipa filicina. Svedelius has observed the same faet on the shore of Ceylon (compare: Bot. Notiser. 1905, p. 181). * On one occasion I found Porphijra nnibiUcalis turning 3-ellow, in tlie sounds north of Kvalvig. 703 weaker, but yet ralher white light, which reaches it. The Florideæ growing near the surface of the sea, in the often very lofty caves on the Færoes, must Hkewise be supposed to get a some perhaps ra- ther weak, but still white light, besides the still weaker, reflected, blue or green light, that comes up from the sea. As the epiphytes on stipes of Laminaria hyperborea have always been found to be growing in a precise order, from the top down- ward (more will be said of this in connection with the Laminaria hijperborea-assoc'mWon), it is certainly solely due to a regard for light. The faet is, that those at the top are algæ wanting much light, and those further down, overshadowed by the latter, are algæ generally found in greater depths. Similar observations have been made by Berthold in the Mediterranean (5, p. 421— 422), showing that the epiphytic algæ on a stem of Cystosira are likewise arranged according to their requirements of light. Light is likewise, as before mentioned, of no small importance to the colour of the algæ, especially to the Florideæ. Florideæ growing in deep water or in the shade are of pure, bright red colour; most Florideæ growing littorally are of a dark reddish-brown often of an almost blackish shade, e. g. Callithamnion arbiiscula, Ceramium acanthonotiim, Gigartina mamillosa, Polysiphonia iirceolata, Porphyra umbilicalis, Chondrus crispus and others, whereas the same species, when they find themselves in the shade, will recover their bright F/oric/eæ-colour. According to Berthold (5, p. 416—417) the same thing has been observed in the Bay of Naples. This author likewise points out, that many more Florideæ are found on exposed than on sheltered coasts. This is in perfect agreement with my observations made on the coasts of the Færoes. That the Flori- deæ are thus found growing littorally in great numbers on ex- posed coasts, must naturally in the first instance be attributed to the faet that, in conformity with the Gaidukow^-Engelmann's theory, they are of a reddish-browai colour. There may surely, however, also be some truth in the explanation given by Berthold, who supposes that, as the algæ are constantly stirred by the surge, and thus incessantly turning other sides to the light which reaches them, the influence of the light can only be of short duration on each part of their thailus; moreover the foam of the surf also lends some shade. Possibly many of these algæ are in different degrees adapted to stand the intense light; several species, e. g. Chondrus crispus, Rhodymenia palmata, Odonthalia dentata become 704 beautifully iridescent, as Ihey are able to rellect certain rays of the light that reaches them, and it has been proved by an experiment made by Berthold (1. c. p. 419), that a vigorously iridizing specimen of Chijlocladia kaliformis lost its iridizing power by being placed in the shade, and recovered it by being once more exposed to the light. A rich profusion of hairs is likewise probably of some im- portance to the algæ, as a means of protection from the intensity of the light. Several of the algæ growing in shallow water, especi- ally in sheltered piaces, e. g. most of the species belonging to the Stictijosij)hon-i\ss,oc'n\iion, are rich in hairs. It is however most probable, that the hairs are in the first instance meant to serve as organs of absorption and respiration, as pointed out by Rosen- vinge^. Whether they are likewise capable of defending the algæ against the dashing of the waves, as suggested by Henckel (Scripta botanica XX, p. 105), I cannot tell. From what precedes it will be seen that I do not fully agree wåth Simmons when he writes (p. 262): »Was den Einfluss der Lichtintensitåt betrifft, so bin ich geneigt anzunehmen, dass man besonders, vas die Florideen angeht, derselben zu grosse Bedeu- tung hat zutheilen wollen. Wie will man nåmlich erklåren, dass im nordwestlichen Norwegen , wo doch im Sommer nicht von ge- ringer Lichtintensitåt die Rede sein kann, die noch dazu wåhrend der Ebbe trocken begende Litoralregion so viele Florideen beherbergt?« There are always some Florideæ wiiich require a great deal of light, and in the Færoes, where the light is far from strong, and the direct insolation particularly slight, a great many Florideæ may therefore easily thrive on the beach, even above the highest water mark, without fading, whilst they fade in districts with more light, are less numerous, and do not grow so far above the level of the sea, except where local circumstances are especially favourable. During a journey in Norway in 1904, I stayed a few^ days at Christianssund, thus getting an op})ortunity of observing the algæ-flora there. It was interesting to observe, how much less de- veloped the littoral algæ-vegetation was here than on the Færoes, the floral composition of the two vegetations, how^ever, being much the same. The Florideæ were faded, light yellow to yellowish- ^ Rosen vi lige, L. Kolderup: Sur les organes piliformes des Rhodomelacées (Oversigt over det kgl. danslie Videnskabernes Selskabs Forhandlinger, 1903, Nr. 4, p. 447—449). 705 brown, Ihe vegetation was scattered, and with the exception of some faded fragments of the Bangia, it did not reach l)eyond high water mark. It was moreover most instructive to notice the difference between the httoral algæ-vegetation of coasts facing the North or facing the South. On exposed shores of the former piaces a rather luxuriant algæ-vegetation grew, not only littorally, but even above the highest water mark, whereas the latter piaces were almost de- stitute of algæ. I have not met with a similar difference on the coasts of the Færoes. 2. Nature of the Coast. The Færoese coast is particularly favourable to algal vegeta- tion. We meet with more or less broken clif!s everywhere on the open shore, especially on the northern and western sides of the is- lands. The rocks are of basalt and similar readily crumbling ma- terials, which the wearing of the sea and weather make very un- even, thereby offering a very good hold for the algæ. Sometimes the walls of the cliff descend almost vertically to considerable depths, sometimes they slope down gently and evenly, allowing the surf to wash up the slope constantly. Or long stretches of the coast, as at Myggenæs, on the western side of Sydero and the northern sides of Stromo and Vaago, are wild and rugged, with numerous large and small rocks shelving into the sea and with masses of fallen rocks piled on one another; in this way the most varying habitats are produced, from the much exposed to those relatively sheltered behind the sea cliffs; the beach too offers many different degrees of light to the algæ, daylight sometimes coming straight down on the algæ, at other times obscured by overhanging or fallen rocks. Finally there are many ravines and caves along the coasts, where even sublittoral algæ, occurring elsewhere only in deep water, find suitable light at the very surface of the water. At the foot of these almost vertical rocks there is often an al- most horizontal or but slightly sloping foreshore, which is some- times very broad, and more or less covered at high tide, but per- fectly dry at ebb-tide. In such piaces a very luxuriant algal vegetation is often found, and on account of the sometimes very slight slope, the different algæ can spread widely in a horizontal direction, whilst those on vertical rocks are often reduced to a minimum , as many species only occur within a very limited regi- 706 onal level out of which they never thrive or only exceptionally. We therefore find the same belt-like mode of growlh in the water- pools on the beach and practically all round the coast, one algal community gradually replacing another, exactly as an inland tlora grows in belts round lakes and pools. Lastly, numbers of large and small pools and basins fdled with water are found on the beach at every possible level above the sea, sometimes so high up, that they are only filled with salt water by the storms of winter and consequently get more or less brackish and polluted in the course of summer, sometimes so low tliat they are cut ofF from the sea only for a short time at low ebb; the great variety of the algal vegetation in these shore-pools is naturally due to these varying habitats. In fjords and sounds the bottom fre- quently consists of stones or gravel forming a developing ground for many algæ, but most often the bottom is soft, covered with sand or mud which, apart from various free-floating algæ, is per- fectty barren, as Characeæ are wanting, and Zostera, which at other piaces generally covers such regions of somewhat shallow water and often shelters a rich epiphytical algal vegetation, is but rarely found in the Færoes; it has only been met with in the innermost parts of Vaagfjord on Sydero, covering the bottom of a small lo- cality where the water was from 1 to 2 fathoms deep. On the whole, however, the rocky coasts of the Færoes must be considered especially favourable to algal vegetation; wherever one approaches the coast, it is found covered by a luxuriant vege- tation, and such barren rocks as on the coasts of Greenland, men- tioned by Rosenvinge (71, p. 152), are nowhere found on the coasts of the Færoes. As to the conditions at the bottom, these are not every where favourable to the algæ. In the fjords, the bottom in shallow water often consists of sand, and in deep water, as in the sounds, of mud. Yet, large regions are covered with stones or rocks offering a fa- vourable habitat for the growth of the algæ. On account of the great depth sometimes met with in the immediate vicinity of the land, the algal vegetation naturally disappears at such piaces close to land, the depth-limit of the sublittoral region being soon reached here. According to the most recent survey, the curve of 25 fathoms is found at a mean distance of about 6000 feet from the coast in southern islands, but only 2000 feet in the northern, especially in the Sounds. Occasionallv it lies several miles from the coast. 707 II. ALGÆ-REGIONS AND ALGÆ- FORMATIONS ON THE COASTS OF THE FÆROES. In his well-known paper on the algæ- flora of the Murmann Sea (44, p. 57) Kjell man divides the flora: »in drei Gebiete und zwar in das litorale, sublitorale und elitorale«. By littoral »Gebiet« he means the part of the bottom of the sea which is left dry by the ebb of the spring tide, that is, from the highest tide mark to the lowest ebb mark. The sublittoral »Gebiet« extends from di- rectly below the littoral »Gebiet« down to a depth of 20 fathoms, and finally the elittoral »Gebiet« consists of the bottom lying below yet overgrown by algæ. In a later paper (45) on the algæ -flora of the Skagerak, Kjell man introduces the name »region« instead of »Gebiet«. He asserts that the limit betwcen the two first named regions is distinct and natural. The vegetation of the littoral region is essentially difterent from that of the sublittoral region, evidently a natural consequence of the very difterent local circumstances pre- vailing in each of these two regions. The limit between the sub- littoral and the ehttoral regions is, however, less distinct. With more or less variation of the limits of these regions, later authors have used this division by Kjellman as a basis for their division of the algæ -vegetation of other tracts of the sea. I shall do the same, with some alterations. Smaller types of vegetation belonging to the algæ -regions are named »Algenformationen« by Kjellman (45, p. 10); his definition is as follows: »Unter einer Algenformation sollte folglich ein Ab- schitt der ganzen Algenvegetation verstanden werden , der durch ein eigenthiimliches Vegetations-Gepråge ausgezeichnet ist. Im All- gemeinen erhalten diese Abschnitte dadurch dieses Gepråge, dass eine oder einige Algenarten die Hauptmasse ihrer Bestandtheile ausmachen.« I propose the name of association for these types of vegeta- tion. These associations are often united in a natural way in larger communities, where they live together under the same or very si- milar biologicai and ecological conditions. I propose the name of formation for these more comprehensive groups. If any further subdivision is particularly wanted, we may use the word facies, 708 applied by Lorenz in his interesting and instructive paper^ Ac- cepted in the sense applied by him to Ihis word, it, however, mainly coincides \vith the term »formation«. The species which form these formations and associations, often (litTer widely from each other in outer habit, their growth- form may vary mnch; but ecologically they must of course as a rule demand the same physical conditions (Hght, the temperature of the air, the temperature and salinity of the sea, the dashing of the waves, etc). The large characteristic algæ of the formations do not grow under the same conditions as the smaller epiphytes which hnd their home on them. Epiphyles growing under the thailus of larger algæ are exposed to a w^eaker light, and on the beach larger algæ not only shade smaller epiphytes by their thallus, but also protect them from being dried or heated by the sun, from the dashing of the waves etc. Protected by the larger algæ, the smal- ler, often more delicate species thus obtain suitable surroundings; they are of course only found in localities answering to their re- quirements. I. The Littoral Region. It has been pointed out by Rosen vinge in his report on the algæ-flora of the coast of Greenland (71, p. 189), that Kjell man 's determination of the limits of the littoral region as being re- spectively the highest tide mark and the lowest ebb mark, is not quite satisfaclory, and in so far as the Færoes are concerned, Kjell man's determination is certainly not quite suitable. As to the uppermost limit, the algæ-vegetation on the exposed coasts of the Færoe Isles really extends far beyond, in more exposed piaces most likely even more than 100 feet beyond, the highest water mark. In sheltered piaces, subject to the influence of the tide, the uppermost tide- mark will be almost identical with the upper- most limit of the algæ-vegetation. According to Rosen vinge, the sea at springtide rises beyond the uppermost limit of the algæ-vege- tation in Greenland; this will hardly ever happen even in the most sheltered piaces in the Færoes. It is true, that sheltered piaces with ebb and flood are but small in numbers here, so that my ^ Lorenz, I. R., Phj'sicalische Verhaltiiisse und Vertheilung der Organismen im Quarnerischen Golfe. Wien 1863. Page 188 lie defines tlie facies as follows: »Die Facies sind also Vegetations- (oder Thier-) Formationen, oder Unterabtheilungen (Typen) derselben, betrachtet vom Standpunkte der Location und der bed i ugen- den physicalischen Agentien.i: 709 observations are relatively few; but according to my observations in Vestmanbavn, Klaksvig, and Trangisvaagfjord, Pelvetia is found a Httle below as well as a little above tbe bighest water-mark, and the £n/e/'omo/-jo/ja-association, growing above that and indeed preferring piaces where fresh water oozes from the rocks, is met witb far above the bigbest water-mark. I therefore fully agree witb Rosen vinge in tbinking that tbe upper limit of tbe littoral region should be stated at tbe level wbere tlie algal vegetation begins^; but wben Rosen vinge declares tbis bmit to be identical witb neap-tide mark at bigh water in Greenland, it is evident from what bas already been said, that tbis greatly dif- fers from the observations made on the Færoes, at any rate on tbeir exposed coasts. On account of the very luxuriant algal vegetation whicb ex- tends, as before mentioned, far beyond the bigbest water-mark, it might perhaps be justifiable to introduce a special supralittoral region, as Lorentz has done (1. c. p. 193) and Warming suggested; (compare Si m mons, 66, p. 173). I shall briefly state the reasons wby I do not side witb tbis view. I take for granted that tbe bigbest water-mark is tbe lowest limit of such a region ; but on investigating an exposed rocky coast, the algal vegetation will be found to offer not the sligbtest traces whatever of anj' limit. The specificallj' littoral region, wbich might tbus be stated to be below tbe bigbest water-mark, merges imper- ceptibly into that of tbe algal vegetation above it; a line drawn at the bigbest water-mark on the rocks would certainly cut through several formations, causing tbe same formation to belong to both regions. As to tbe littoral and the sublittoral regions circumstances are quite different, tlie limit being often for miles sbarp and distinct, tbough irregularities may indeed occur in the algal vegetation. Moreover, tbe more or less exposed situation of the coast might cause tbe same formation to belong, now to the supralittoral, now to the littoral region. It therefore seems most natural to me to say, that tbe littoral region includes also tbe algal vegetation growing beyond the bigbest water-mark, just as in what follows 1 say that the sublittoral includes tbe elittoral region, no dividing line being found between them. As to tbe determination of the lowest limit of the httoral region, ^ sat den ovre Grænse for denne Region bor sættes der, hvor Algevegetationen begynder«. 710 I likewise agree with Rosen vinge in stating, that it ouglit to be drawn somewhat above the lowest ebb-mark. On the coasts of the Færoes, the ////7ja/?//ja//a- association of tliis region distinctly marks the limit. At very low tide a rather large portion, up to one or two feet of the i4 /ar/a-vegetation growing under Himanthalia, may sometimes be dried, but this undoubtedly belongs to the sub- littoral region; Rosenvinge moreover reporls, that portions of the Lomz/?a/'za-vegetation, which naturally belong to the Laminaria-^orma- tion growing below that, are likewise uncovered at low tide in Green- land. The lower limit of the littoral region must therefore be fixed at about the ebb mark at neap-tide. On tidal coasts it is therefore not very difficult to make a natural division of the algæ- vegetation into a littoral and a sub- littoral region, but on non-tidal coasts the determination of these regions is much more difficult. Attempts have been made to define the limits in other ways; for Tonsbergtjord Gran, for the western part of the Baltic Sea Reinke, for Bohuslån ^ Kjell man have made use of certain characteristic species growing within well- defined limits in the localities in question. In his valuable work on »Ostersjon's Hafsalger«, Svedelius (79) is however certainly right in objecting that this division is anything but satisfactory, as one and the same species may grow at very difTerent depths in dilTerent parts of the sea. It is for instance a well-known faet, pointed out by Kjell- man, that many algæ which are littoral on the west coast of Nor- way are sublittoral on those of Bohuslån, where the fresher surface current does not reach them. Svedelius therefore is certainly right, when he warns us against drawing a parallel between the regions belonging to heterogeneous flora domains, as the factors which in one place justify a very distinct division into regions and make a limit for the occurrence of a species, may be totally wanting in another. As I have already explained in the introduction, there is in the Færoes a small tract of the sea between Ostero and Stromo where the tides are almost imperceptible. Here the question is where to fix the limit between the two regions. If the limit be fixed at the beginning of the Laminariæ as Kjellman has done on the coasts of Bohuslån, it would be very easy to point out a very distinct line between the two regions within the area mentioned, as the Laminaria /æyoenszs- association 711 begins at a depth of about 3 to 4 feet; the -S//c/{/os//>/jo/i-association growing above that would theii be reckoned littoral. I have not, however, done this. A comparison between this algal vegetation and that of a sheltered tidal locality e. g. Vest- manhavn, proves that under the Fncaceæ-iormat'ion just below the lowest water mark, a perfectly similar vegetation, the Stidyosiphon- association, is found, whicli is here sublittoral. In this area of the sea the littoral region is therefore but slightly developed, only consisting of some scattered Pelvetia, Porphyra um- bilicalis, Fnciis vesicnlosns and a few others, besides some crusts of bluish-green algæ, Hildenbrandia and Rhodochorlon. The algæ-vegetation found on exposed coasts differs as a rule considerably from that found on sheltered coasts; as in the former a great many species occur that are not always found in the latter piaces, and as the species common to both piaces are often re- presented by varying forms in each place I have divided the algæ- vegetation into that of exposed and that of sheltered coasts. The Norwegian investigators Hansteen and Boye have dealt in the same way with the algal vegetation of the west coast of Norway. We will first consider the vegetation of exposed coasts. a. Exposed Coasts. The Hildenbrandia-Formation, or the formation of the crustlike algæ as well as of lichens is widely spread along the exposed coasts of the Færoes. It covers the rocks with a dense mal of various colours to a considerably height, i. e. up to more than two feet above the level of the sea and down to the Coral- /z/?a-formation. The uppermost part of it mostly consists of lichens, which, according to the Rev. Deichmann Branth, belong to dif- ferent species of Verrucaria, but crusts of bluish-green algæ soon appear, especially Calothrix scopiilorum, Riviilaria atra etc. and the crimson Hildenbrandia rosea; the last mentioned algæ I have found on the west coast of Vaago, about eight feet above the level of the sea. At a shorter distance from the sea, Hildenbrandia becomes more and more predominant, still crusts of lichens and bluish-green algæ are intermingled with it, and another crust-alga, Ralfsia vernicosa, can now be found; this however prefers littoral pools where the water is constantly renewed by the surf. The walls of such basins 712 are almost ahvaj^s covered by brown, red, and black crusts, re- spectively consisting of Ralfsia, Hildenbrandia and Lichens. These algæ, with the exception o^ Ralfsia, must necessarily be able not only to live for a long time witbout sea water and consequently to be dried up, but also to be wasbed by fresh water. They are moreover excellently fitted for resisting the surf, as they are very firmly attached to the rocks, of which they may ahnost be said to form a part. Lastly, they must be well fitted to stand intense light and great changes of temperature, as the southern side of rocks is often much heated by the sun. On the west coast of Sydero at Vaag's Ejde 1 have seen extensive growths of Ralfsia, which are here found in large flat littoral pools with only a few inches of water. Whilst Ralfsia thrives hest in littoral pools with shallow water, Hildenbrandia wdll thrive excel- lently in the latter, as w^ell as on steep rocks which are dry in calm weather. Among other algæ contributing to form this formation, Pilinia maritima, Pleurococcns spec, Codiolum gregarium, etc, which I have found intermingled with crusts of bluish-green algæ, ought likewise to be mentioned. This formation is exposed to clear daylight and to the varying influence of the weather in its elevated habitats; in more low-lying regions it will often be found as a subvegetation covered by different larger algæ of the place. Almost in the middle of the beach or somewhat farther down, this formation is replaced partly by Corallina officinalis, partly by Phymatolithon polymorphiim, both spreading widely downwards into the sublittoral region. Simmons briefly describes a »Hildenbrandtia-formaiion«, but he says that it is only to be found »An ruhigeren Oertlichkeiten«, and that »die Steine und Pelsen in der Wasserflache von dieser Formation bedeckt sind.« Later on he gives (66, p. 173) a further explanation of this formation, defining it as purely littoral. The Chlorophyceæ-Formation . The characteristic algæ of this formation are: Prasiola crispa subspec. marina, different forms oi ,Enteromorpha intestinalis, Rhizo- cloninm riparium, Prasiola stipitata, Prasiola furfiiracea and a few other species. This formation stretches like a green belt along the exposed 713 coasts of the Færoes. Ceiiain bluish -green algæ and perhaps Hildenhrandia excepled, this is the algæ -vegetation which extends the farthest upwards on bare rocks, on which, with the exception of lichens, an inland vegetation mav only be found at a consider- able height, in very exposed piaces, never less than several feet above the level of the sea. On exposed coast this formation there- fore may be found many feet above the level of the sea; on the west coast of Vaago, at Bosdalafos, the outlet of Sorvaagsvandet, at the time of year when the sea is calmest, that is, in June and July, above the cataract, consequently about 80 feet above the level of the sea, in moist cracks in the rocks shaded from the sun, I have gathered forms of Enteromorpha intestinalis and Rhizo- cloniiim ripariuin, which are characteristic of the formation under discussion. As a matter of course the algæ belonging to this formation must be able to do without water, and to be washed by fresh water. Some, but not quite all these species, may also be found under certain forms growing in more or less brackish water. They generally grow so far above the level of the sea, that they are only reached by the surf and the spray, and if the sea be calm for some length of time, which may sometimes happen, especially in summer in the very changeable climate even of the Færoes, they are totally prevented from being moistened by salt water. On the contrary they are at times soaked by fogs and rain to such a degree, that they are most probably completely deprived of salt. Some few of the algæ belonging to this formation moreover grow^ in localities where fresh water oozes from the rocks. On the other band several of these algæ may in dry weather and sunshine appear to dr}' up, some of them, e. g. Prasiola crispa subsp. marina, Prasiola stipitata and Enteromorpha intestinalis even so as to be easily pul- verised. As to Rhizocloninm riparium, 1 have not found it dry; it generally occurs on wet rocks and thus always keeps rather moist. The algæ belonging to this formation must moreover be able to stand the full day and sunlight, as they grow on rocks facing south as well as north. On the other band they are less in danger of being injured by the surf, as they most often grow so far up that only the spray can reach them. When they grow farther down, it may be supposed that the low, dense, often very gregarious growth of, for instance, Prasiola stipitata and P. fiirfiiracea and Enteromorpha, serves as a means of protection from the dashing 714 of the waves. Prasiola crispa '■'marina and Rhizoclonium may like- wise be protected by the great llexibility and elasticity of their Ihallus. The upper limit of this algæ-vegetation changes as the locaHty becomes more or less exposed, the lower limit likewise gradually moving downward as the locality becomes more sheltered. Several of these algæ, for instance, Prasiola stipitata and Entero- niorpha, may be found rather far down in the littoral region, in more sheltered piaces. This green algæ- formation may be divided into several asso- ciations according to the predominant characteristic algæ. Naming them from the top downward: a Prasiola crzs/?a-association, a Rhi- zoc/o77f«/7i-association, an Enteromorpha intestinalis-associai'ion and a Prasiola stipil ata-assoc'iaiion. The Prasiola cnspa- association is found at least 40 feet above the level of the sea^ It prefers steep, precipitous rocks, on which it is found as a dense cover, turning light yellowish-green when it dries up. This light colour makes it stand out against the dark rocks to which the thallus is fixed. When moist it turns dark green, each thread becomes free and sways with the waves. As subvegetation we find blackish lichen crusts, which have been de- fined by the Rev. Deichmann Branth as Verrucaria mergacea, V. maiira, V. mucosa and V. striatula. The i?/izroc/o/7»i777- association may be found at the same level above the sea, but it prefers moist localities, for instance, such where fresh water is constantly dripping down, or pools and caves, where the air is very damp. I never found this alga perfectly dry; it is evident that it cannot stand to be dried up. It often forms large, unmixed growths and is easily discernible on account of its light- green colour and its often crisp, almost curly, appearance. In some piaces, at Glivernæs for instance, I have found Vaucheria coronata at about the same level in moist, but more sheltered little depres- sions in the rock. The £'77/e/-on7orjD/7«-association is generally found farther down, but on an exposed coast it may reach to about the same height, that is ca. 40 feet. It consists for the most part of Enteromorpha intestinalis var. micrococca , and smaller forms of var. gennina, especially f. cornucopiæ. Enteromorpha prefers piaces where fresh water oozes from the rocks, but also grows in dry piaces and ^ As before mentioned, several of these algæ, ho^\evcr, may l)c found growing scattered at least twice as high up on the rocks. 715 is sometimes found perfectly dry. This association may be found covering large stretches of rocks as a dense, moss-like mat, some- times only consisting of Enteromorphæ, sometimes of Enteromorphæ mingled with Prasiola stipitata etc. In some piaces the latter algæ, however, grow quite unmixed with any other species, and often carpet the rock wilh a short, dark Fig. 152. Prasiola s(ipi7a(a-as.sociation. From rocky coast near Hojvig. (F. B. phot.). green covering, as shown in fig. 152. Prasiola stipitata grows as a rule on horizontal or slightly sloping sides of rocks, and prefers piaces where birds live, especially the summits of rocks under fowling cliffs. This alga may also be found quite dry. Simmons mentions (78, p. 251) an Enteromorpha -iormaiion which, judging from the species mentioned by him, must at any rate partly belong to the £'n/e/'omo/-jD/ia- association. But on the other hånd he also mentions some Cladophora-species as belonging to this association; these species, according to my observations, be- long to algæ-associations which grow farlher down. Botanj' of the Færoes. 46 716 The Porphijra- Association . This association is foiind under the Chlorophycece-iormaVwn, but still on exposed coasts far above tlie level of the sea, 40—50 feet, perhaps even more. It is widely spread along the coasts of the Færoes, on which it extends as a helt often many feet broad. In much exposed piaces it extends to far above the level of the sea ; at Bosdalafos on the west coast of Vaago to at least 40 feet. On the north western side of Videro, I visited a rock (see plate XIII) about 30 feet high which was quite covered by Porphijra umbilicalis. On the islet of Myggenæs, where the southern side of the rocky coast slopes gently upwards, Porphyra certainly extends up to at least 50 feet above the level of the sea. These observations were moreover all made in summer time; in winter, when the surf reaches much farther up the coast, it is most probable, that Por- phyra also grows at a greater height. With the exception of the Hildenbrandia-formnt'ion, which here often forms a subvegetation, and of some species belonging to the C/i/orojo/iyceæ-formation, Porphyra umbilicalis is almost predominant in this association. Porphyra umbilicalis covers the rocks, sometimes densely, sometimes more sparsely. It grows in small tufts, 2 to 3 indies high, whose convolutions gives them the appearance of crumpled paper (see Wi Ile 85, p. 38). The colour of this vegetation is a dark reddish-brown ; sometimes, however, more purpie, sometimes more tawny. On account of its crumpled nature Porphyra umbilicalis can retain sea-water for some length of time among the many folds of its thailus. If the exterior of the plant looks dry, there is, however, often some moisture left between the interior parts of the thailus and those parts which turn downwards. Neverthe- less it can stand to become so dry, that it will creak when trodden upon. It is not only liable to be dried up, but also to being soaked in fresh water, i. e. by rain, for a long time. It is probably protected from the surf by the elasticity and toughness of its thailus, and by the lubricity of its surface. In more sheltered piaces Porphyra umbilicalis becomes higher, often to one foot or more, and its large, curling membranes then cover the rocks or hang down vertical rock-sides like long tufts swaying in the wind (see plate XVII). In sheltered piaces of this kind it likewise grows far down into the littoral region. As before mentioned, this association generally only consists of 717 this alga together \vith Hildenhrandia and some species of Verrii- caria as subvegetation, but different forms of Enteromorpha intesti- nalis, Edocarpiis littoralis and others are, however, often intermingled ; the Fucaceæ- formation mentioned below (page 720) likewise some- times extends into tlie Porp/jyro-association. The Porphyra -\ege\a\ion was well developed in the months, April to August, when I visited the Færoes; Mag. H. Jonsson like- wise tells me, that he has found it vigorously developed when he visited the Islands from October to December, 1897. Thus it seems to be well developed all the year round, and is not, as Simmons says (1. c. p. 250), mostly a »Winterformation« on ex- posed coasts, even if it may be supposed to be still more vigorously developed in the winter, judging by its growth in other piaces. On coasts especially exposed to the surf it may also be supposed to extend still further up above the level of the sea. In more southern districts the Porjo/jyra- association is really a characteristic winter vegetation. Thus when Kjell man (45) visited the »Skærgaard« of Bohuslån from Dec. 1874 to Jan. 1875, he found a luxuriant Porp /jy/'a-vegetation on the outer rocks facing the sea, whilst this vegetation is totally wanting in summer, which I have also observed during a visit to Lysekil in 1899. Neither was this alga found on the rocks in the Firth of Forth at North Berwick, where I had an opportunity of observing the local algal vegetation one day in July 1900. In winter it is most likely vigorously developed liere also. On the west coast of Norway, Boye points out (6, p. 20 — 21), that Sognefjorden seems to be the southern limit of its occurrence ; he founds this opinion on his own as well as on Hansteen's observations. This is, however, scarcely correct. Mr. E. Norum, teacher at Haugesund, has for a long time investigated the marine algæ growing on this part of the west coast of Norway, and kindly reports, that Porphyra umbilicalis is met with as an unmixed vegetation in much exposed piaces at Haugesund and Utsire, an island about 3 miles west of Hauge- sund. At Utsire, where Mr. Norum made his observations in July, the Por/;/7y/a-vegetation reaches to a height of 2 to 3 metres above the »Balamis-heli« on the southern and eastern sides of the island, and it may be supposed that it reaches still further on its northern side. At Christianssund, I have myself seen Porphyra at the beginning of August, growing some few feet above the highest water mark on an exposed coast ; but the association was indeed only slightly developed here, compared with that of the coasts of the Færoes. 46* 718 A summer vegetation of Porphyra as liixuriant as that of the Færoes will most likely only be fonnd in more nortliern parts of Norway. As to Nordland, Kleen for instance writes, that Por- phyra iimbilicalis is found at the highest level, often far above the highest water mark. That it may be found luxuriant even in summer, in the Færoes, is certainly due to the faet that the sea is almost always stormy. Even in calm weather there is most often a swell and therefore surf along the coast. And even when it happens that the sea is quite calm for some time, the littoral algæ-vegetation is not in great danger of being dried up, because the air is very moist, the weather often cloudy, and fogs and rain frequent. That it, however, may happen, that the littoral algæ-ve- getation, to which the Porp/jf/ra-association belongs, may be found dried on sunshiny days has already been mentioned. The lowest limit of the Po /-jo/jz/ra-association is determined by Boye to lie at the highest usual water mark, which is almost identical with the sharply-marked, white line formed by the Bala- nus, as he likewise points out. As may be seen in my illustration from Midvaag (see plate XVII), his view agrees with observations from the Færoes, although only in somewhat more sheltered piaces. In more exposed piaces, the Porjo/iyra-association is replaced here by other algæ, and is not found until far above the highest water mark. In these exposed piaces it may extend far in a ver- tical direction, sometimes even to many metres, thus Simmons' remark, that the Porp/iyra-association has »nur eine recht geringe vertikale Verbreitung«, holds good only in the case of somewhat sheltered piaces. The Rhodochorton- Association. The P/?oG?oc/70/'/o/j-association commonly forms a kind of close felt, reddish-brown to crimson, on stones and rocks, but in piaces where it is apparently more exposed to be dried up and to the surf, it may be found on the rocks in small firm lumps resembling peas (f. globosa). It often grows on the underside of rocks and stones, in clefts in the rocks, and in the caves (about which more will be said presently); but it may also be found growing in full daylight. It is found partly in the littoral region, especially in its higher part, and partly far above the highest water mark, especially in the caves. 719 The association almost exclusively consists of Rhodochorton Ro- tliii, yet Sphacelaria britannica, Ectocarpus littoralis, Callithamnion, etc. mav be found intermingled with it, and, epiphytically on Rhodochorton, Pleurocapsa amethijstea var. is oflen found in great numbers. Rhodochorton Rothii can stand to be dried up to a considerable degree, and may even be found apparently quite dry. It also very well stands fresh water, not only rain, but also that of small cataracts, growing sometimes at the very spot where the water falls on the rock. The i?/joc/oc/7or/on- association has been briefly described by Simmons (p. 250), and Hansteen (p. 348) and Boye (p. 30) have mentioned its occurrence on the west coast of Norway. The Bangia- Urospora- Association. On exposed coasls, prominent points and steep rocks are often covered by a dense and matted reddish-brown vegetation, which in dry condition shines almost like silk, unless it is curled up like wool, as is often the case. It is Bangia fusropurpurea, an alga very common on the coasts of the Færoes. It is found like a dense covering on the rocks, often at a very great height, 5 — 6 metres or more, above the highest water mark. Consequently it can stand being dried up for a long time, as well as being soaked with fresh- water. Intermingled with Bangia, a luxuriant vegetation of Urospora mirabilis is likewise often found; its growth is quite similar to that o( Bangia; it also bears to be dried up, though perhaps to a lesser extent. It forms a similar covering, on the rocks, only dark green. They are both algæ growing typically where the waves dash on the coast (cf. Gran, 37, p. 9). Of these two algæ, Bangia certainly grows highest above the sea level, often forming liere a pure Bangia- facies; in much exposed piaces it extends far into the Porphyra- association; Urospora however may also be found far above the highest water mark. On Myggenæs Holm, for instance, this alga was found at a height of at least 20 feet above the sea level, partly as a subvegetation under and among larger algæ, such as Fucus inftatus f. disticha, Rhodymenia pahnata, etc. Hansteen, who, Strange to say, has not observed the Bangia- association on the west coast of Norway, writes that it has been found by Wille at Mandal. According to Wil le, it formed an 720 almost continuous, light yellow belt of dried up threads of Bangia, on rocks and stones above the average high water mark. According to what Mr. E. Norum kindly informs me, Bangia is found in the neighbourhood of Haugesund on exposed coasts, to a height of 6 feet above the Balamis- region, often covering sloping rocks abundantly. Mr. Norum supposes, that its total absence in some years from long stretches of coast, may be the reason why Hansteen, as mentioned above, has not observed this species on the west coast of Norway. At Christianssund I saw Bangia myself; it grew on an exposed coast up to a few feet above the highest water mark, and was easily distinguishable because of its Hght yellow colour. Similar light yellow growths of Bangia may fairly often be found in Denmark and on »Kullen« at the entrance to the Sound. In the Færoes Bangia however always keeps its dark, reddish-brown colour. According to Gran, Bangia is the alga growing at the greatest height above the sea-level in Christiania Fjord; it is not so in the Færoes, where several algæ are found at a much greater height. However much the two algæ treated in this section may differ, they agree very well biologically and form a very characteristic association, whether growing in the same locality, or, what is also frequent, each in its place forming a Bangia-, and an Urospora- facies. Intermingled with Urospora a great many Ulothrix flacca and other species of Ulothrix are found. According to H. Jonsson, Urospora is in autumn and winter partly replaced by species of Ulothrix and Codiolum, which may each of them form extensive associations. On the coasts of Greenland, Rosenvinge (71, p. 201) has ob- served a very similar association, composed of Monostroma groen- lundicum, Ulothrix flacca, Urospora mirabilis and Bangia fusco- purpurea. Thus it is also here formed by thread-like, branchless algæ, as pointed out by Ro s en vinge. He calls it the Monostroma g roenlandicum- formation. The Fucaceæ-Formation on exposed coasts. This formation is found spread, but very commonly on all the exposed coasts of the Færoes, even in the most exposed piaces, as for instance Store and Lille Dimon, Mj'ggenæsholm, the west coast of Sudero, Viderejde, Sumbo Holm, etc. It prefers 721 sloping rocks and especially recesses in Ihese rocks that may be reached by the sea. In much exposed piaces this formation may extend far above Ihe highest water mark, and the algæ growing here are consequently in danger of being dried up for rather a long time; this they, however, greatly avoid by means of their low, tufthke growth, and especially by their great amount of mucus, in which Fiiciis iuflatiis f. disticha is particularly rich. Fig. 153. Fucus spiralis f. steep rocky coast near Viderejde. The formation is typically formed by two species of Fiicus, that is, Fucus spiralis f. nana and Fucns inflatus f. disticha. Fucus spiralis grows uppermost, often a couple of feet higher up than Fucus inflatus. They almost always grow together, but as Fucus spiralis seems to be able to stand being dried up better than Fucus inflatus, the former is found in somewhat more sheltered piaces, where the sea may happen to be calm for some length of time. On the other band, it is most often only Fucus inflatus f. disticha which is found in particularly exposed piaces, and which seems thus to be the best fitted for resisting the surf. I have found vigorous specimens of it, 2—4 inches long, on Myggenæs Holm and Muletangen at 722 Vaag's Ejde, whilst Fiiciis spiralis only was found less exposed in the latter place. I mvist nevertheless point out, that I have found both species growing togetlier on the west coast of Lille Dimon, which must also be considered a particularly exposed place. The F«c»s-planls hang in small, dense tufts, a few indies long, down the sloping rocks, often covering them so closely, that it is only by removing them that we discover the red subvegetation of Hildenbrandia which is common here, together with the inter- mingled lichens. Forms of Hctocarpiis littoralis, s\^ecies of Ulothrix, besides Elachi- sta fiicicola, and a few others frequcntly grow epiphytically on the F«c«s-plants. Among the latter, Porphyra umhilicalis, Enteromorpha intestinalis and other littoral algæ are often met v^ith. Whilst these small specimens, only a few inches high, are cha- racteristic of the most exposed piaces (see fig. 153), the Fncns-plants gradually bccome larger as the localily becomes more sheltered (see plate XIV), and the most regular gradations may then be found, from the typical large form, f. edentata (fig. 158), to the dwarf form, f. disticha. Sometimes they may even be found growing in the same locality, the smaller form still growing uppermost, at or above the highest water mark, whilst large, well-developed specimens be- longing to the main form grow at the lowest part of the beach. This may easily be observed along the coast, between Arge and Glivers«æs, on the east of Stromo (see plate XIV), a locality where Nolso really affords some shelter, but which must be considered rather exposed. A very easy transition may be found here from the FwcHS- vegetation of exposed coasts to that characteristic of a sheltered coast, as not only the two species of Fucus growing on exposed coasts gradually become larger, but they are joined by Pelvetia canaliciilata , Ascophijlhim nodosum and Fucus uesiculosus, I have observed that these three algæ appear in the order I have named them. Pelvetia may be found in rather exposed piaces, grow- ing abundantly for instance on rocks on the w^est coast of Sydero, at Vaag's Ejde, at a height of about 5 metres above the sea-level (see fig. 154). The place w^here it grows is, however, always some- what sheltered. It is, for inslance, never found on rocks facing the open sea, but on rocks facing the land, or at least in piaces sufficiently sheltered from the strengest surf. At Vaag's Ejde, a very exposed place, Ascophyllum nodosum is also found on the most shel- tered, inmost rocks, but I have not seen Fucus vesiculosus there. 723 Even the most sheltered piaces here are not calm enough for this alga. When the wind blows on the shore, the sea washes over the rocks, in hard weather, flowing over the »Ejde« itself. I have not met with any description of such a Fwcus-vegetation growing on an exposed coast. Si mm o ns, who found Fuciis disti- chiis in its well-known habitat, the west coast of Sj^dero, writes (1. c. p. 254): »Wo an der Westkiiste von Sudero Fiicus distichus auftritt, Fig. 154 phot.). konnte man vielleicht am richtigstcn eine besondere Formation unter- scheiden, die dann die Fiicus distichus-Porphyra-iormaXion zu nennen wåre. Fiiciis distichus bewohnt nåmhch mit Vorhebe abschiissige Pel- sen in der Nåhe der Ebbengrenze und wåchst da mit Porphyra laci- niata vermischt.« Judging by these words, Simmons seems hardly to have got any distinct idea of this characteristic formation. Nor have I met with any description of it from Norway, where it may, however, be supposed to grow. It has therefore been very inter- esting to me, to receive the following report from Mr. Norum in Haugesund. »A F«CHs-formation corresponding exactly with those 724 from the Færoes is likewise found here in the most exposed piaces. It consists of Fiiciis spiralis (most probably f. nana) and Fucus in- flatus f. disticha^. They are most often attached to the Balanus. They grow partly intermingled, partly with Fucus spiralis extend- ing a little higher up than Fucus inflatus. Fucus spiralis disap- pears where the surf is the most vigorous. Thus Fucus inflatus is also here the Fucus which is the hest fitted for resisting the dash- ing of the waves.« According to this description, this Norwegian vegetation seems to correspond perfectly with that of the Færoes. It is true that Boye (l.c. p. 25) speaks of a Fwcus-formation grow- ing on the outer side of the rocks; he says, that »even in the most ex- posed localities a very hixuriant Fucus-vegetation is usuaUy found.« As members of this vegetation, he mentions, however, Fucus serratus f. elongata, and two forms of Fucus vesiculosus, which makes it quite clear, that his FncHS-vegetation is widely different from the Færoese. Fucus serratus is nowhere found in the Færoes, and Fucus vesi- culosus grows here essentially on the inner side of rocks; it is the species of Fucus which I have mentioned before as disappearing first in passing from sheltered to exposed coasts. I am on the whole rather dubious as to the correctness of Boye's report. During a journey in Norway in 1904 I stayed for a few days at Christians- sund, in order to observe the algæ-vegetation, and I found Fucus vesiculosus only in sheltered piaces. Mr. Norum also kindly in- forms me, that Fucus vesiculosus is never found in very exposed piaces in the neighbourhood of Haugesund, whilst Fucus serratus can grow on a very exposed coast immediately below the Balanus-helt. Nor doesKleen (51) mention any such Fzzczzs- vegetation ; yet it must be supposed that a formation similar to that of the Færoes is found in Nordland. Mag. Jonsson reports, that a Fticus-formation, very much like this, is found in Iceland on exposed coasts, and Stromfelt writes (80, p. 10): »On the clifFs of Seley which are constantly washed over by the surf, a particular Fwcacé-formation is found. It consists of Fucus spiralis, Fucus evanescens *dendroides, which are through many intermediate forms connected with f. nana of *arcticus and t contracta of '-^'norvegicus that are likewise found here^«. This vege- ^ Mr. Norum has kindly presented me with specimens of both species, fully agreeing with tliose from tlie Færoes. * På de af ståndiga branningar ofverspolade strandlilipporna på Seley upp- trådde en egendomlig Fucacé-formation, bildad af Fucus spii-alis, Fucus evanescens 725 tation must certainly greatly resemble that of the Færoes, as all the forms named by S trom felt are, according to my definition of species, similar to or at least closely connected with Fucus inflatas f. disticha (compare Jonsson 41, p. 184). As to Greenland, Rosenvinge writes (71, p. 199) as follows: Where the coast is exposed to the dashing of the waves, the vege- tation bears a totally different stamp. The Fucaceæ are either entirely absent or limited to deeper clefts in the rocks or to pools (Fucus inflatus)«. It has since been said by Si m mons (66, p. 178), that this circumstance may be due to the ice. This may possibly be true in some cases, but probably not in all. In describing the algæ vegetation of the Murman Sea, Kjell- man writes (44, p. 59), after pointing out that all littoral algæ are low: »Fucus evanescens, welche am håufigsten innerhalb des lito- ralen Gebietes angetroffen wird, scheint mir mit Recht den Namen {f. 77077«) zu verdienen, den ich ihr gegeben, denn sie ist selten uber 6 ctmr. hoch und immer sehr schmal«. There must likewise be some resemblance between this and the Færoese vegetation. It may be supposed, that Fucus spiralis is found on the Ork- neys in localities resembling those in the Færoes (see Traill, 81, p. 316, who calls it Fucus platycarpus). If Fucus anceps, as I think\ ought to be regarded as a form of Fucus inflatus f. disticha, the two characteristic forms of the Færoese formation are found on the west coast of Ireland. At Biarritz Sauvageau^ has found some quite small specimens of Fucus spiralis, which certainly grew on an exposed coast there. On the northern side of Kullen, on steep rocks a little above the sea level, I have met with a dwarfish form of Fucus spiralis, that is, a very similar vegetation, only that Fucus inflatus f. disticha is also wanting here. The Callithamnion- Association. The above -mentioned types of vegetation have their habitats above the easily distinguishable white line formed by Balanus and bivalves. We shall deal next with an association growing in and about this »animal association«, which is found at about the highest \vater *dendroides genom en mångd former forbunden med de åfvenledes derstådes upp- trådande f nana af ^arcticus och f contracta af "iiorvegicus. ^ Judging by a few small specimens kindly sent me by Mr. Batters. ^ Sauvageau, C. Note préliminaire sur les algues marines du golfe de Gas- cogne, p 6—7 and p. 22 — 23. (Extrait du Journal de Botanique, XI, 1897). 726 mark in the Færoes, and, according to Boye, at the same height on the west coasl of Norway. This association is typically represented by Callithamnion ar- biisciila and Ceramiiim acanthonotum, which are almost always found together on exposed coasts and which are about equally com- mon. On more sheltered coasts, Callithamnion arbnscula is often wanting, as it clearly prefers an exposed coast; and Si m mons, who declares himself (78, p. 250 and 273), that he has not had much op- portunity of investigating exposed coasts, and who has therefore not met with Callithamnion arbnscula, consequently calls this asso- ciation the Ceram/H/n-formation. These algæ form small, dense , reddish-brown tiifts, 2 to 4 indies long, and are attached either to the Balanus or to the rock itseif, now forming a scattered growth, now a dense covering. On account of their ample ramification and their dense, compact growth, they look almost spongeous, and at high tide they really absorb water which is retained by their capillary action during low tide. After having been laid bare for several hours, they are still so full of water that it can be wrung from them as from a sponge. Kjellman has already (46, p. 479) briefly mentioned this faet. Berthold speaks in a similar way about Callithamnion granulatum, which has been found a few times on the coasts of the Færoes, together with C. arbnscula and much resembling the latter in its habits. He wTites in the following way about it (5, p. 406): »Bemerkenswerth sind in dieser Hinsicht die dichten Thallome von Callithamnion gra- nulatum mit sparrig gespreizten Åsten, welche sich wie ein Schwamm mit gr6sserenWassermengenYollsaugen,wodurcheinAustrocknenauch nur der peripherischen zarten Spitzen vollståndig verhindert wird.« In less exposed piaces, this association hardly reaches above the highest tide mark, but in much exposed piaces it may extend far above it. At Viderejde on a rock lying off the landing place, I found a well-developed Co//z7/7am/?zo/j-association reaching to several feet above the extreme tide mark (see plate XV). Here it grew on steep, sloping rocks; besides the two characteristic algæ, a few Por- phyra umbilicalis and several Himanthalia lorea grew here, the latter however being fructiferous only at the lower part of the association. There were, moreover, Acrosiphonia albescens and Polysiphonia urceo- lata, both fitted for retaining water by their dense, felt-like growth, together with Ceramium rubrum, Dumontia filiformis, Scytosiphon lomentarius, Phyllitis zosterifolia, a few small Alaria esculenta and. 727 as subvegetation, CoralUna officinalis al the lowest part of the association. All these species and a few others are rather frequently foiind intermingled in this association. A Ca/W/jom/jzon-association probably similar to this Færoese association has been found by Boye on the part of the west coast of Norway which has been especially investigated by him. Judging by his brief description it is probably quite similar to the Færoese one. According to Hansteen (38, p. 346), the Callilhamnion-SLSSocia- tion in the neighbourhood of Bergen is formed of Callithamnion arhiisciila, whilst nothing is said oi Ceramiiim acanthonotam ; there is however no reason to think that this species should be wanting there. At Christianssund, I have found a vegetation corresponding exactly with that of the Færoes, formed by the two characteristic algæ, and growing on steep rocks facing the open sea. It grew almost at the highest water mark. It has already been described by Ek- man (17, p. 4) as being found in this place. Finally it may be sup- posed that a vegetation well agreeing with that of the Færoes may be found in Nordland (see Kleen 51, p. 9). I have also met with this association on exposed coasts in Shetland, at Muckle Holm in Yell Sound. As before mentioned, Simmons (p. 250) calls this association the Ceranirum-formation, or, with regard to Nordland, the Ceramium- Callithanmion -formaiion. As that of the Færoes however corre- sponds with that of Norway, I have not considered it justifiable to change the name. The Rhodymenia- Association. A low, dense, dark, brownish-red vegetation, made up almost exclusively of Rhodymenia palmata, is found on sloping rocks on exposed coasts from a little below, or even a little above, high tide mark, down to a little above lowest tide mark. It grows very densely on the rocks, forming low tufts only a few inches high (see plate XVI and XVII). It is mostly a narrow-lobed form, related to f. sarniensis, which is found here, yet the typical form also occurs. Generally this association is not intermingled with other larger algæ, but a few Fucus inftatus f. disticha, Acrosiphonia albescens, etc. may be met with. On the other band, there are many smaller epiphytes on the thailus of Rhodymenia, among which Ectocarpus Htoralis, E.tomentosus and E.fasciculatus, Myrionemaceæ, Chantransia virgatula, C. seciindata and others should be pointed out. 728 The /?/}0(iy/rjen/a- association is widely spread on the Færoes. Il seems lo prefer piaces where fresh water oozes from the rocks, which is verj' common on the coasts. Even in piaces where small waterfalls fall from vertical or beetling rocks down on the lit- toral rocks, Rhodymenia grows abundantly. In such piaces il is of course soaked in fresh water at low tide, and this alga must be well adapled to resist great differences of salinity, for at high tide it is more or less flooded by the sea. The specimens are well de- veloped even in such piaces; they are only of a paler colour, which perhaps indicates Ihat circumstances are less favourable to them there. I imagine that Rhodymenia scarcely stands complete drying up and Rosenvinge is of the same opinion (71, p. 202). When Rhody- menia, how^ever, is found on the coasts of the Færoes, rather far up on the beach, sometimes even above the highest water mark, the reason is, that it grows gregariously, and that it is ahvays kept moist al ebb lide by the fresh water oozing from the rocks. I have not found any description of a i?/joG?ymenfa-association quite agreeing with this vegetation which is so widely spread on the coasts of the Færoes. Still I feel inclined to believe that it will be found to be rather common on the coasts of the North Atlantic. According to Kieen (1. c. p. 9 and 17), it is probably also found in Nordland. It is true that Boye (1. c. p. 28) speaks of a Rhodymenia- formation on sheltered coasts growing in the AscophyUum-Fucus- association, and on the the shellered coasts of the Færoes Rhody- menia really oflen grows abundantly among and under the Fucus bushes. Lastly Kj eliman menlions a sublittoral »Rhodymenia-region« (44, p. 67) on the coasts of Novaya Semlya and Spitzbergen. This agrees well wålh the faet that many littoral algæ else where become sublittoral in Arctic countries. According to Rosen vinge (71, p. 202), il may however be found in the lowest part of the beach in Green- land, but usually only in small numbers. Si m mon s does not men- lion this association. Resides forming this littoral association growång on rocks, Rho- dymenia makes a characleristic littoral association close to, yet above the lowest water mark. On the parts of the Laminaria hy- /)erfco/-ea-association wdiich grows in such shallow^ water that the tops of the stipes rise above the surface of the sea at low^ tide, Rhody- menia palmata is found attached to the uppermost part of the La- minaria slalk, and oflen in such numbers, that the brown leaves of the Laminariæ are almost covered by the large, dark red Rhody- 729 me/jza-thallus. In the illustration (fig. 160, p. 756), a great part of the vegetation which is seen above the surface of the sea, is Rhody- menia. So this is really a littoral, but epiphytical association. Rho- dymenia is often 2 feet long here and not short as when it grows on the rocks. Its thailus, especially the older parts of it, shelter a luxuriant vegetation of epiphytes, mostly species of Ectocarpns and Myrionema, Chantransia, etc. S i m mon s also mentions this littoral association (1. c. p. 256). In Trangisvaagfjord, he has found Panc- taria latifolia^ in numbers on the Laminæ oi Laminaria hyperborea and saccharina ; to this epiphytical association he moreover refers a great many other algæ, e. g. Dictyosiphon foeniculaceus, D. hippuriodes, species of Ectocarpns, and Scytosiphon lomentariiis. They may cer- tainly be found here exceptionally in more sheltered localities, but facing the open sea they are not common in this association, ac- cording to my observations. The littoral Corallina-Forincdion or the Lomentaria-Corallina-formaiion, will in much exposed piaces reach upwards into the Callithanmion-association, thus even passing the highest water mark. It however belongs mostly to the lower part of the littoral region, and Corallina is found together with sub- littoral species far down in this region. In more sheltered piaces, Corallina hardly passes the lowest water mark, and is here limited to the more low-lying, littoral pools which it covers with a dense, pale-pink covering. As before mentioned, Corallina grows on the most exposed coasts, e. g. Lille Dimon and Store Dimon, Muletangen at Vaag's Ejde, Viderejde, Sumbo Holm, etc. In this respect, the occurrence of Corallina (48, p. 116) does not bear out Kjellman's words, that this alga »prefers sheltered piaces« ^. In the more or less dense Cora//ma-vegetation, a great many small and scattered growing algæ are often sheltered. Larger algæ also find a favourable habitat here, occuring sometimes in such great num- bers that they form their own associations, which will be described ^ Cailed P. plantag inea bj^ Simmons (compare mj^ Algæ-Flora, 7, p. 436). ^ It has been said by Wille (85, p. 14), that the chalkj' incrustation which he is undoubtedh' right in considering as a means of protection from herbivorous animals, can hardl}^ be supposed to afford anj' protection from the dashing of the waves. He founds this opinion on the observations of Kjellman and others. I cannot fully adopt this view. It seems to me that the firm construction of the thailus, together with the low, gregarious growth of the plant, must necessarily protect it from the dashing of the waves (cf. Wille p. 37). 730 laler on (seetableXXI). Of the many algæ-species which arefound in the Corrt////?a-formation, only some of the most important will be named liere. One of the comnionest and most characteristic is Lomentaria articLilata, which grows in dense, dark reddish tufts, intermingled with Corallina; moreover, several forms of Ceramiiim nibriim; Porphyra leucosticta, Polysiphonia iirceolata, and, especially in small littoral pools, also P. Brodiæi; moreover Chondrus crispiis, Scyfosiphon lo- mentarhis, Phyllitis fascia, Diimontia filiformis, Laiirencia pinnatifida, Acrosiphonia, Hiinanthalia, Gigartina, Monostroma, and many more. DernmtoUthon macrocarpiim f. Corallinæ is fonnd epiphytically on Co- rallina; in more sheltered piaces Leathesia difformis is found; the latter is not common, but it is abundant in the few piaces where it grows, for instance at »Sundskær« in Kalbakfjorden. In this more sheltered place, Chordaria flagelliformis, Dictyosiphon foeniculaceiis and D. hippiiroides were also found. It seems as if Corallina does not stand to be dried for a long time. This is clearly shown by the faet that in more sheltered piaces it grows luxuriantlj^ in low-lying pools, but never outside of these. I have already mentioned that on a much exposed coast, e. g.. Muletangen at Viderejde, it can stretch far up inlo the Callithamnion-associaiion on sloping rocks, and even sometimes pass the highest water mark. But if the weather and the sea are calni in summer, it is also here discoloured, and it assuredly dies, if it is dried up for a long time. In such piaces Corallina is, however, protected from being dried, because it often forms a subvegetation under larger algæ. It is likewise not well fitted to stand fresh water; hence it is not found in piaces where fresh w^ater oozes from the rocks; here the /?/joc/f//J7enza-association occurs. As before mentioned, the Coro//z/?a-formation grows rather far dowai in the sublittoral re- gion ; this will be farther explained later on. Boye (6, p. 26) seems to have found a very similar Corallina- vegetation on the part of the coast of Norway investigated by him. It is to be supposed that a littoral Co/a/Zzna- formation, very similar to this, is common along large stretches of the west coast of Norway; compare Hansteen's: »Broget-pelagiske- formation« (p. 348), wdiich, as pointed out by Simmons (p. 260), includes about the same species as the formation from the Færoes. According to Kl een (51), it may also be supposed that a very similar Coral- Zzna-formation is found in Nordland, Lomentaria articulata however being rare and but badly developed Ihere. 731 As to the Færoes, this formation has already been described by Simmons (78, p. 252), who points out that it is widely distri- buted here. To the Corrt//ma- formation I refer the following four charac- teristic and universally destributed associations. The Monost roma Grevillei- Association. On horizontal rocks or in very shallow, httoral pools where only very little water is left at low tide, Corallina will frequently be fonnd at a height of a few feet above the lowest water mark, covered by a very dense, low spinach-green matting of the Monostroma Grevillei. Where Corallina sometimes is wanting, Monostroma Gre- villei covers the rock itself. Its growth is so dense, that it is only exceptionally intermingled with other species, such as Phyllitis fascia, Scytosiphon lomentarius, Dumontia filiformis, etc. On account of its dense and gregarious growth, the Monostroma is always kept moist by the great quantity of sea- water which it can retain, even when it grows on a slightly sloping substratum. As Monostroma Grevillei is a spring and summer alga, this asso- ciation disappears later in summer and the underlying Corallina- vegetation is now clearly seen or is covered by other algæ. The Monos/ro/na- association is very common along the coasts of the Færoes and is found even in very exposed piaces, in spite of its apparently slight power of resistance against the surf. The Acrosiphonia-Polysiphonia- Association. This association which is formed of species of Acrosiphonia, especially A. albescens, and of Polysiphonia urceolata reaches from al- most the lowest water mark to a few feet above it (part of this asso- ciation is seen at the bottom of plate XVII). It grows on sloping rocks, which it covers with a very dense matting, generally green at the top and reddish-brown at the bottom, these two algæ frequently making unmixed facies, an Acrosiphonia-facies uppermost and a Polysiphonia- facies below it. They may however also be found intermingled. On account of their very ample ramification ^ they are both able to retain a great amount of sea-water at low tide, which protects them from becoming dried up. ^ The littoral form of Polysiphonia urceolata is far more ramified than the piants that grow sublittorally (cf. Grau 37, p. 10). Botany of the Færoes. 47 732 Mingled with the characteristic algæ, we find e. g. Ceramiiim ru- briini, Cladophora riipestris, and very rarely Cladostephiis spongiosus. Corallina is usually absent in this association, as the dense growtli of both species leaves no room for a subvegetation. This association, \vhich is very common in the Færoes, has not been described by Si m mon s, nor by any Norwegian investi- galor. It is however undoubtedly closely connected with Han- steen's »Broget-pclagiske-formation« (38, p. 348) and partly related to Boye's »Gigartinaformation« (6, p. 22 — 23). The Gigartina- Association. Somewhat above the lowest water mark, the height varying ac- cording to the exposure of the locahty, a vegetation of Gigartina ma- niillosa (see plate XVIII and XXI), often dense and up to 6 inches high, is found in the Com//ma-formation. This association often extends far horizontally, and on almost horizontal rocks, found within the level of this association, it is often widely distributed (see plate XVIII). Vertically it usually reaches only to a height of one or two feet; the height being the greatest on specially exposed coasts. Epiphytes common on Gigartina are: Chantransia virgatnla, very often forming a dense, velvety covering on the stalky part of the thailus, and Dermatolithon macrocarpum f, færoensis which grows on it in smaller or larger pale pink incrustations. Ceramium ru- lyrnm, Chondrus crispus, Polysiphonia urceolata, Acrosiphonia alhe- scens, and frequently some younger or older piants of Himanthalia lorea, besides Corallina officinalis, are found mingled with Gigartina. But generally these algæ are only few in numbers, or they may be totally wanting, so that the dark reddish-brown colour of the Gigar- tina predominates. Gigartina mamillosa grows luxuriantly in the most exposed piaces. Its tough, cartilaginous character probably makes it well fitted for resisting the surf. It is hardly exposed to any drying process, as it generally grows at the Iower part of the littoral region. According to Hansteen (38, p. 348), and Boye (6, p. 22), a si- milar G/g'ar/ma-association is found along the west coast of Nor- way. It seems to differ from the Færoese association only in its colour. Boye writes: »It (the formation) is easily distinguishable at a long distance as a light yellowish green belt, immediately above the lowest water mark, because the algæ inviduals are faded by the 733 sun«^. But on the coasts of the Færoes, Gigartina ahvays keeps its dark reddish-brown colour. Simmons does not set apart any special G/^'ar/z/ja-association. In reference to Boye's G/g-ar/ma-formation, he writes (p. 260): »Die letztere konnte wohl auch auf den Fåroern unterschieden werden, ich bin aber geneigt, sie nur als eine lokale und zufållige Ausbil- dung der Corallinaformation zu belrachten, wo Gigartina iiberwie- gend ist.« Opinions may of course dilTer on this matter, but as Gigartina can be found as an unmixed growth and often of consider- able extent, I agree with Boye in maintaining a special Gigartina- association. I have found the G/(7ar///7a- association beautifuUy developed on the northern side of the entrance of Trangisvaagfjord, where Gigartina (see plate XVIII) forms extensive growths on horizon- tal rocks. On Sumbo Holm, the GzV/ar///?a-association was found on a very exposed coast, at a height of 2—4 feet, and on Myggenæs Holm I have seen it at a height of at least 10 feet above the lowest water mark. It is, on the whole, one of the commonest associations on the coasts of the Færoes. Directly below this association, The Himanthalia- Association is found. The peculiar and characteristic Himanthalia /orea-associa- tion (see plate XIX) is very common on the exposed coasts of the Færoes. It grows from immediately above the lowest water mark down to the sublittoral A/ar/rt-association. On flat rocks, the Himan- //?a//a-association will stretch very far horizontally, but vertically, only to one foot as a rule. On sloping rocks which are constantly washed over by the sea, it may reach somewhat higher, sometimes even to the Callithamnion-associaiion. For instance, on rocks at Viderejde I have observed well-developed specimens with receptacles (see plate XV), at a height of at least 6—7 feet above low water mark. On the other band, the characteristic buttonlike or often globular, inflated, yellowish-brown, young piants are frequently found rather far above its special level, sometimes even above high water mark. But they ^ »Den (formationen) er meget ioinefaldende paa lang afstand som et hst gulgrønt belte straks over nederste vandstandsmærke; de enkelte algeindivider er nemlig aldeles afblegede i sollyset. 47* 734 cannot produce receptacles in this nnfavourable habitat. The Hi- /rja7j//ja//a- association is so characteristic on exposed coasts, that a luxuriant growth of this alga is a sure sign of the presence of surf. It is an association particularly fitted for resisting the dashing of the waves; the surf constantly washes over the association at low tide, and the metre- long, phable, elastic receptacles of the Himanthalia follow resistlessly the movement of the sea. Himan- thalia cannot stand to be dry for any length of time; when found Fig. 155. The //iniGrotta del Tuono« in the neighbourhood of Naples, with only very shal- low water, which was constantly renewed by its connection with the sea. He says: »Trotz des niedrigen Wasserstandes in die- sem Bassin finden sich hier an den dunkelsten Stellen Algen, die sonst als charakteristische Pflanzen einer Tiefe von etwa 50 — 60 Meter im Golfe sich finden.« And after having named a certain number of the algæ found in this place, he continues: »So zeigt diese kleine Grotte mit flachstem Wasserstande in ihrer Algen- vegetation je nach dem grosseren oder geringeren Grade von Dun- kelheit einen åhnlichen Wechsel, wie er im freiem Meere bei dem Uebergang aus grosserer in geringere Tiefe stattfindet«. In the section on the influence of the different degrees of the intensity of light on the distribution of the marine algæ, Bert hold points out that the algæ without exception must, as independent assi- milating plan Is, require a certain intensity of light, yet different 742 for each species. He further writes: (5, p. 414) »In den beschat- telen Grotten, in welchen Lithophijllum Lenormandi, Callithamnion elegans, Derbesia Lamourouxii die åussersten Grenzen der Vegetation bezeichnen, verschwinden diese Formen schon voUståndig in ge- ringen Entfernungen vom Eingange. Ist es erlaubt aus den Befunden an der Oberflåche auf das Verhalten in grosseren Tiefe zu schliessen — die Berechtigung dazu diirfte kaum bestritlen werden konnen, denn die Algenformen sind an den entsprechenden Ortlichkeiten theils ganz dieselben, theils nahe verwandt — , so muss die Licht- intensitåt auch hier bei vorhandenem Pflanzenwuchs noch eine ziemlich betråchtliche sein«. If it were possible to measure in some way the amount of light that reaches the algæ growing in the inner- most parts of the caves, this might certainly, as suggested by Bert hold, serve as a measure of the intensity of the light at the depth where the algæ-vegetation disappears^ b. Sheltered Coasts. Th e Ch lorophyceæ-Form a tion . Here this formation is chiefly represented by an Enteromorpha- association, but sometimes we also find others of the different asso- ciations belonging to this formation on exposed coasts e. g. the Prasiola stipitata-associaiion. The E nteromor pha-association grows highest, often somewhat far above the highest water mark and stretches down to a little be- low it. It is often found on moist rocks and in clefts and cracks, where fresh water oozes from the rock, and consequently it does not form any continuous belt, but larger or smaller patches in these piaces. It may moreover be found at the inner end of bays and in- lets, where the bottom is covered with gravel or stones, and here it may often partly replace the Fucacece- formation. It is, for in- stance, richly developed in the innermost part of Kaibak- and Kolle- fjord, in Skaalefjord elc. Different forms of Enteromorpha intestinalis ^ In order to try if any result might possibly be obtained, I used in 1900 Wynne's actinometer, which serves for photographic purposes to indicate the exact term of exposure in a certain place (cf. Joh s. Schmidt: Uidrag til Kend- skabet om Skuddene hos den gamle Verdens Mangrovetræer, København 1903, p. 11). But in these caves, where the liglit is faint and the term of exposure consequently long, the instrument proved less useful. With an in.strument constructed by Dr. Steenstrup (cf. Meddelelser om Grønland. 25, page 1) to measure the intensity of the light, a satisfactory result might possibly be obtained. 743 are met with here: var. micrococca uppermost, next to that var. geniiina, especially the formå cornucopiæ. Together with Entero- morpba, Ectocarpus litoralis is also often found, especially the form riipincola. This association is closelj' related to the Enteromorpha-assocm- tion, which has been mentioned as characteristic on exposed coast (p. 714), and to the Enteromorpha-vegetalion which is found in the highest littoral pools. It seems lo correspond exactly with the »Ente- romorjo/ia- formation« found by Hansteen (38, p. 358) at the head of the Vindefjord. The Fucaceæ- Formation. On sheltered coasts, especially in the littoral region, a very characteristic and luxuriant vegetation is found both on rocks and stones and consists almost entirely of the Fucaceæ: Peluetia canali- ciilata uppermost, next to that Fuciis spiralis, F. vesiculosiis, Asco- phijlhim nodosiim, and lastly Fiicus inflatus, which grows a little above and sometimes a little below the lowest water mark. As these five Fucaceæ are differently adapted to resist the dashing of the waves or the supply of fresh water, they are not always found together. In the most sheltered piaces Peluetia, Fucus vesiculosus and Ascophijllum will generally predominate, whilst Fucus spiralis and F. inflatus are absent entirely or only represented by a few scattered individuals. Where the coast begins to be more exposed, the two last-mentioned algæ appear. At the same time, Fucus vesi- culosus, and later Ascophyllum totally disappear. Peluetia may still be found in rather exposed piaces, but in the most exposed it like- wise disappears, so that only Fucus spiralis and Fucus inflatus are left. These two algæ may be typically developed on a somewhat exposed coast, but the more exposed it is, the smaller the Fucus- species. Thus they pass easily into the above -mentioned Fucacé- formalion found even on the most exposed coasts. With regard to the species found on the coasts of Greenland, Fucus vesiculosus, Ascophyllum nodosum and Fucus inflatus, these observations agree well with Rosen vi nge's report (71, p. 197). I also entirely agree with Rose uvinge, when he declares, that Fucus vesiculosus grows the farthest into the fjords, where the water may be almost fresh because of the inflow of rivers. Yet Ascophyllum may sometimes be found growing at the outlet of rivulets, where the alga at low-tide is irrigaled by fresh water, but 744 this takes place generally nearer the nioiiths of the fjords, where the sea-water inundates its habitat at high tide. As mentioned already Pelvetia grows uppermost. On a sheltered coast and at or a little above the highest water mark, it constitutes an association which may spread far horizontally, but which is vertically most often only one foot broad (see plate XXII). When moist, it is of a light yellowish, olive-brown hue, but in calm weather it is often liable to be dried; Pelvetia may be found quite dry and stiff, and is then of a blackish hue. In tideless bays, it is of course very liable to be drained, and circumstances are on the whole less favourable to it here than in piaces which are subject to a regular tide. This is quite clear from its slighter development; specimens from the Kalbakfjord for instance, especially those growing highest, were often but one inch long and consisted almost entirely of receptacles, a form which has been called f. minima by Simmons. Yet this dwarlish growth here in the fjords, may perhaps be partly due to the great amount of fresh water which flows constantly from the rivers, and which makes the surface-water with which Pel- vetia comes in contact, very fresh. On coasts subject to the tide, a much more vigorous and better developed Pe/f;e/m-association is found. Other algæ are but rarely found in the 7^e/pe//«-association as Simmons (p. 251) has pointed out, probably because only a few algæ are able to grow so far above the level of the sea in sheltered placfes. Ulothrix flacca and Ectocarpiis litoralis are however some- times found epiphytically on Pelvetia, on the rock we fmd a sub- vegetation of Verrncaria and Hildenbrandia rosea, and in cracks in the rock, Rhodochorton Rothii and Ectocarpiis litoralis. Below this association, a Fiiciis vesiciilosns-Ascophylliim nodosiim- association grows on a sheltered coast. It is very common and on slightly sloping rocks it is widely distributed horizontally; a specially luxuriant growth of these algæ seems to be found on coasts with boulders and fragments of rocks of different sizes. On these stones, Fiiciis vesiciilosus generally grows uppermost, in large dense clumps and facing landward; below, Ascophylliim hangs down the sides of the stones (see plate XXIV). Yet Ascoplujllum may also be found growing above Fiiciis vesiciilosus. This I have seen in the Vestman- havnbugt, for instance. Strømfelt likewise writes (80, p. 10), that Ascophylliim grows uppermost, and Fucus vesiculosiis farther down ^, whilst Rosen vinge reports that F«r«s vesiciilosus grows uppermost ^ According to Jonsson's observations (41, p. 192 — 193), it is usually tlie reverse. 745 in Greenland. It seems to me most likely that the somewhat dif- ferent occurrence of these two species, also in the Færoes, may be explained thus, that Finns nesicnlosns, as mentioned before, re(|uires to be more sholtered than the other; therefore, this alga in a somewhat exposed place grows facing landward and upper- most, whilst it may grow nearer to tlie sea and farther down in a locaUty as shehered as Vestmanhavn. Kig. 157. FiiciiK /ji//(i/;i.s association al a small, low-lying pool on tlie beacli. Illustrations from rocky coast at Thorshavn. (F. B. phot.). Elachista fncicola is a very common epiphyte on Fncus vesicu- losus, and Polysiphonia fastigiata is hardly ever sought in vain on Ascophyllum nodosum. Besides these, other epiphytes are sometimes found e. g. Ectocarpns litoralis, Edocarpus tomentosus, Ceramium rnhrnm, Rhodymenia palmata etc. on both algæ. Below the dense tiifts of seaweeds several red algæ are often found, e. g. Gigartina mamillosa, Chondrns crispns, Hildenbrandia rosea, Rhodochorton Rothii, etc. On a somewhat more exposed coast, Fucus vesicnlosns, as be- fore mentioned, soon disappears and is replaced by Fucus spiralis which forms an association at about the same height as Fucus vesicnlosns, or a little higher. Below Ascophyllum nodosum, Fucus 746 inflaius occurs. On a more exposed coast, the latter forms an extensive association (figs. 157 and 158). Both in the Fiicus spiralis- and the Fiiciis z/?//rt///s-associations, Elachisla fiicicola is a very com- mon epiphyte. On Fiiciis inflatus, several species of Ectocarpus e. g. • Ectocarpiis tomentosiis, Ectocarpus fasciciilatus, etc. are often found. On the rocks and stones, below and among these algæ, especially in the lowest Fiicns inflalns-associalion, one may find a great many difTerent algæ, e. g. Porplujra umbilicalis, Urospora mirabilis, Clado- phora riipestris, Acrosiphonia nlbescens, Pohisiphonia iirceolata, Cera- mium riibriim, Gigarlina mamillosa, Chondrus crispiis, etc. In order to characterize the Fiicaceæ-formaiion more closely, I shall give a description of some of the localities in which I have met it at its greatesl development. In Vestmanhavn , and a little north and south of this gulf, in the Veslniannasund, between Vaago and Stromo, a specifically well-developed Fucaceæ-formalion is found, on sheltered coasts. The tide rises rather high here, about 6 — 8 feet, and at low tide the formation is laid perfectly bare. We find Pelvetia uppermost as a light yellow helt, almost a foot broad, showing up the high water mark with an almost perfectly straight line. Below this, we have some Fiicus spiralis; this alga is abundant only on somewhat more exposed coasts. The Fnciis-AscophyUiim- association, which comes next, is however very luxuriant. As shown in plate XXIII, the coast in certain piaces consists of large stones which are completely covered by these algæ. This vegetation is found as far down as the lowest water mark, where some few Fiicus inflatus may occur. On the east coast of Stromo, between Thorshavn and Hojvig, the Fnmceæ-formation is of a somewhat different appearance. The coast is here rather exposed and rugged, with several larger or smaller islands girding the coast. In one of the inlets, called Volga- ravik, which forms a little natural harbour, having but a small opening towards the sea, a well-developed Fncaceæ-vegetation was found (see plate XXIV). Near the land, the coast was composed of boulders of different size which were covered by a dense Fucus- vegetation; a little farther out, in a more exposed position, Fucus vesiculosus disappeared making way for a luxuriant Ascophyllum- vegetation, which was replaced by a well-developed Fucus inflatus- vegetation at the greatest distance from the shore and at a lower level. Here, on the stones lying farthest from the shore, the position is rather exposed, and in a strong sea making towards the land the surf 747 dashes heavily into the inlet. This Fwcacecp -vegetation was widely distributed horizontally, but not vertically, as was easily observed on the perpendicular walls of the inlet, where all three species grew together, forming a helt hardly more than a few feet broad. It should not be forgotten, however, that ebb and ilow aie scarcely perceptible on this part of the coast. 158. Fuciis in/}atus. A small part of the Fiicus in/!(i(us-association illustrated in lig. 157. (F. B. phot.). From Thorshavn, which has rather an exposed situation, south- ward to Arge, a F»rrtcea? -vegetation is moreover found, with Fucus spiralis uppermost, Fiiciis inflatiis lowest (see figs. 157 and 158). To these must be added AscophijUiim nodosiim in somewhat sheltered piaces, and Fiicns vesicnlosiis in the most sheltered. At Glivernæs, still more toward the South, and further on in the same direction, the coast becomes more exposed and only the dwarfish forms of Fiiciis spiralis and F. inflatiis are found. Examples of the above-mentioned Fncacecp- formation will be Botany of the Færoes. 48 748 foiind almost anywhere on the coasts of the Færoes. The formation constantly adapts itself to the more or less exposed situation of the locahty and by aid of the Fucaceæ- vegetation , it may as a rule be easily determined whether a certain locahty is sheUered or exposed. The Fiicaceæ- (ormaiion of the Færoes seems to agree weh with Rosenvinge's of the coasts of Greenland. According to Strom- felt's soniewhat scanty description (80, p. 10), a very similar vege- tation must be found on the coasts of Iceland. If we compare the F/zcrtcece-formation of the coast of Norway, described among others by Hansteen (38, p. 351— 52) and Boye (6, p. 27), we at once ob- serve that they are essentially different from a floristical point of view, as Fiicus serratus, which is very common on the west coast of Norway, is absent on the Færoes; æcologically there likewise seems to be a great difference between them. Boye writes (1. c, p. 25): »Even in the most exposed localities, a very luxuriant Fucus-vegetation is usually found«; he mentions moreover two forms of Fucus vesiculosiis, besides a f. elongata of Fiicus serratus as being found there. This is quite contrary to what we observe on the Færoes, where Fiiciis vesiculosiis is absent on exposed coasts. The only Fz/r«s- species that will thrive here are Fucus spiralis and F. inflatus. As before-mentioned (p. 724) I doubt, however, whether Boye 's observations are perfectly correct. In the northern part of Norway, where Fucus inflatus is com- mon, the Fwcizs-vegetation hardly differs from that of the Færoes in anything but this, that Fucus serratus is frequent. The Færoese Fwrus-vegetation also much resembles that of the Shetland Isles, ex- cept that a luxuriant growth of Fucus serratus occurs on the shores of the latter. What makes the resemblance still greater is, that Fucus inflatus has been found here (cfr. 9, p. 5). Simmons mentions (78, p. 251 and 253) a Pe/fe/m- formation and a Fucus -Ascophyllum- formation found on the Færoes. 2. The Sublittora! Region. As mentioned before, I agree with Rosen vinge in determining the upper boundary line of this region at about low water mark, at neap-tide. In the Færoes the boundary line is almost identical with the distinct upper line of the A/a/*/« -association. Kjell man determines the lower limit of this region at the curve of 20 fathoms, and he calls the area covered by algæ below this, the elittoral re- 749 gion. The reason why he chooses this curve for the lower limit is, that the algæ-vegetation at and below 20 fathoms becomes uniform and poor in species, as the greater part of the subHttoral algæ will not thrive at so great a depth. He himseif points out, that this Hmit is »weniger scharf oder wenigstens nicht so auffallend« between these regions as between the hitoral and the sublittoral. In agreement with Hose uvinge (45, p. 2.37), I therefore think, that it is unadvisable to determine the boundary line between two regions so arbitrarily, as the number of species and the luxuriancy of the vegetation are in reality reduced regularly and gradually from a certain depth downward. Moreover the sublittoral species growing at the lowest depth are likewise often found in the upper part of the sublittoral region. No other species are in- Iroduced here that might change the appearance of the vegetation. When moving downwards from the surface of the sea, the brown algæ first disappear, and at last hardly anything but Florideæ are left^ Finally, the curve of 20 fathoms may serve as a boundary line only in the northern seas, wdiere the light does not reach far down. In the Mediterranean Sea, a vigorous vegetation is still found at a depth of more than 40 fathoms, and in the Tropics we may be sure to find a rich algæ-vegetation still farther down. It does not seem to me natural therefore to maintain this boundary line, as the algæ-vegetation is luxuriant down to a certain depth, (in the Færoes to about 15 fathoms, as far as my observations go) and then leaves off slowly and gradually, as the depth increases. When Kjellman objects (45, p. 8 — 9) to Ørsted 's division of the algæ-vegetation in the Sound, that the boundary line between the brown and the red algæ is indistinct, it seems to me that the same objection might be applied to Kjellman 's division into a sublittoral and elittoral region. If the sublittoral region must be divided, it is probably Ørsted 's division that ought to be maintained, as it would seem * In connectlon with this I shall refer to Rosenvinge's summaiy of the species (71, p. 231) found at a depth of at least 10 fathoms; algæ found at a still greater depth, at 15 fathoms and more, are printed in thick type. As pointed out b}' Rosen- vinge (1. c. p. 232), it is evident from the list that the red algæ increase in number with the depth. There is hardly any doubt, that here also an almost pure Florideæ- formation will be found, shortly before the vegetation ceases. It is true that Rosen- vinge in the same connection (p. 236) speaks of a Desmarestia aciileata, found at a depth of about 30 fathoms, near Upernivik. But I doubt whether this alga was adhering to any substratum. 48* 750 most natural to determine the boundary line where the brown algæ leave off, and where only a pure or alniost pure Florideæ- vegetation is left. But while we still know so little of the algæ- vegetation at greater depths, it seems to me most natural, as men- tioned before to determine the sublittoral region as one, and, in agree- ment with Rosenvinge (1. c. p. 237), »to move the lower boundary line of the sublittoral region from the curve of 20 fathoms to the limit where the algæ-vegetation disappears. Thus the elittoral region will in- clude that part of the bottom of the sea which is destitute of algæ«^ a. Exposed Coasts. The Phymatolithon-Association. This association begins at about the lowest water mark, or a little above, and probably extends down to a depth of several fathoms. My observations are, however, not exact, as Phijmatolithon poly- morphum which forms this association is almost always covered by the Alaria, etc. and is therefore not easily observed. Still I have seen that it may be found in caves several fathoms below low water mark as well as at and even over the highest water mark. It covers the rock with a whitish or slightly pink , glabrous crust, but as mentioned before it is mostly covered by various larger algæ, especially the Alaria. This association grows on the most exposed coasts. Its stone-like consistency, and its firm hold on the rocks, serve to protect it from the surf. Simmons mentions a Lithothamnion- formation (p. 251) which probably agrees with this association. It is true that, ac- cording to his report, it is found almost only in the littoral region (»ausnahmsweise bis an die Fluthgrenze hinaufreichend und sich bis an oder zuweilen etwas unter die Ebbengrenze streckend,«) but later he has corrected this. The sublittoral Corallina-Formation. The littoral Cora//mrt-formation stretches sublittorally, as men- tioned before, down to a depth of several fathoms^. Here the Co- ^ »at flytte den lavere Grænse for den sublitorale Region fra 20-Favne Kurven til den nedre Grænse for Forekomst af Algcvegetation overhovedet, saa at den eli- torale Region kommer til at omfatte den Del af Havbunden, der er blottet for Ve- getation«. - In my flora (8, p. 402), I have stated this depth as 1—2 fathoms. According to my later observations, it may however reach to at least 4 fathoms. 751 rallina is however often replaced by the above-mentioned Phyma- tolithon polymorphiim-assoc'ialion. Like the littoral Cora //zno-formation, the subHttoral shelters a great many difTerent algæ, either epiphytically on the Corallina, or on the rock among the hitter. Some of these are e. g. Chondrus crispiis, Delesserin alata, D. siniiosa, Chætomorpha Melagoniuni, Lo- mentaria ckwellosa, Ptilota phimosa, Porphijra leucosticta, P. miniata, etc. In smaller group of rocks, where the habitat is somewhat shel- tered but the coast still exposed, e. g. on the east coast of Stromo, from the redoubt near Thorshavn northward, to Hvidenæs, or at Ejde on Naalso, large associations of the gregarious Fiircellaria fa- stigiata are often found in the Co/'a//ma-formation, besides large or small associations of Ahnfeltia plicata. A conimon epiphyte on the Corallina in these piaces is Ceramhim riibrum siibf. seciindata. Ulua Lactiica and Monostroma fiisciim are likewise found here. The Co- rrt///na-formation is moreover partly the subvegetation of the Alaria- association. The Laminar iaceæ-Formation. This, the most widely distributed formation on the coasts of the Færoes is naturally divided into several well-defined associations. On an exposed coast, the A/orfa-association is uppermost, which is sometimes replaced by a Laminaria digitata-a.ssociaiion; a Laminaria safr/zrt/zVjrt-association is more seldom found, and lowest of all a L. hyperboreaassocial'ion. On a sheltered coast we moreover fmd a Laminaria /"æroens/s-association , but of this and the Laminariaceæ- formation as a whole we will give further details later on. This La/nznanaceæ- formation corresponds with Kjell man 's Laminariaceen-formation (45, p. 34). The Laminaria digitata- Association. This association, which gradually merges into the Laminaria hyperborea-assoc'iai'ion, growing below it, is found on open coasts on slightly sloping rocks. In localities where the sea is particularly turbulent and therefore constantly washing the rocks, this association may reach to several feet above the lowest water mark. On the Southern and eastern coasts of Sydero, from Sumbo Holm to Vaag- fjord, and on Vaago and Troldhoved on Sando, I have seen this association beautifully developed (see plate XX). The Laminaria digitata is well adapted to resist the violence of the waves. Bolh leaf and stalk are of a tough, yet pliable character 752 and the slalk is, as poinled out by Wil le (85, p.9 — 10), very exlensible, in contrast lo Laminaria hijperhorea which has a very stiff stalk. The stalk being often miich compressed at the top, adds to its llexibility, which is also mentioned by Wille (p. 6). The plant is moreover firmly fastened to the rock by vigorous haptera, and if one tries to pull it up, generally only the stalk is broken, whilst the haptera are left. On coasts where the rocks project vertically into the sea, this association, however, is usually entirely absent, or a single Laininarin digitata at most may now and then be found. It is the form stenopijUa which is represented here, but where the association is well developed, all transitional forms besides the main form are found. As a subvegetation, we find Corallina, Gigartina, Polysiphonia nrceo- lata, Cladophora riipestris, PhymatoUtlion polymorphum, elc. and on the Laminariæ themselves, some smaller epiphytes which grow on the stipes or sometimes also on the margin of the lobes of the Ihallus. Among the latter Ectocarpiis fasciciilatiis, E. tomentosiis and E. Hincksiæ are the most important. Early in spring a dense covering of the parasitical Ectocarpiis tomentosoides is invariably found (com- pare 8, p. 415). The Laminaria digitata -assocmlion grows at about the same height as the Himanthalia-nssocintion which is usually absent where the former is well developed. In connection with this I may say that I have also, though but rarely, found a Laminaria saccharina- associalion growing a little above and a little below the lowest water mark. On Sumbo Holm, a little above the lowest water mark, I have found a vegetation of L. saccharina represented by a form which I think might be refer- red to f. linearis (see the illustration of this alga in my flora, 8, p. 453). This form was smaller, the thallus narrow, thick, leathery and very wrinkled, probably an adaption to its very exposed habitat. On the lamina a great many epiphytes were often found, espe- cially Pogotrichiim filiforme , wdiich made the alga resemble an ostrich feather when floating. The Ålaria- Association. This association generally grows uppermost in the sublittoral re- gion directly below the /fm7a/j//?rt//a-formation, as pointed out bj^ Han- steen. Thus it begins at about the lowest water mark at neap-lide; at spring-tide, part of it, one foot or more, is laid bare. In very exposed piaces, in clefts, or on sloping rocks that are constautly Fig. 159. Ataria-associalion on the beach at Myggenæs Holr (F. B. phot.). 754 washed by Ihe sea, Alaria esciilenta may exceptionally be found up to several feet above the lowest water mark, as at Myggenæs Holm, wheie it gro\YS up to at least 10 feet above tbe lowest water mark, tbat is to say, it passes above the highest water mark (see fig. 159). Other piaces where it grows under the same conditions, are e. g. Lille Dimon, Molen at Ejde, etc. (see also plate XXI). The A/ar/a-association prefers steep or perfectly vertical rocks in the most exposed piaces, and covers these rocks with a dense covering, often many feet thick, at varying but hardly ever very great depths. Yet the Alaria may be found at a de])th of several fathoms, as in sounds where there is a rapid current. Here the Alaria is often richly represented in the Laminaria hyperborea-associaiion. The Alaria is excellently adapted to the violence of the waves; its elastic leaves constantly follow the motion of the water, as their great flexibility makes them utterly unresisling. In spite of this, the upper- most part of its lamina clearly shows traces of the force of the sea on exposed coasts, as it is torn and rent into small lobes, and lastly nothing but the bare, thin, much-worn midrib is left (see the illustration of this alga in my flora, p. 449). Generally the association consists solely of the Alaria esciilenta; the Alaria Pylaii seems to prefer somewhat more sheltered coasts, and it may be supposed that this species is rare, as only a few specimens of it have been found. It is therefore of no importance in reference to the composition of the A/arza-association. On the other band, the Litosiphon Laminariæ, which is almost always found abundantly on the leaves of the Alaria, and the Chantransia Alariæ, which covers the lamina with a very dense, purpie, velvet mat, are very characteristic in the association. I have however only seen the latter a few times on the coasts of the Færoes. On old stems of Alaria a number of different epiphytes are found; some of these are Ectocarpiis fasciciilatiis , E. Hincksiæ and E. tomentosiis , Rhody- meiiia palmata, Porphyra iimbilicalis, Ulothrix flacca, etc. Larger algæ but rarely belong to the association; however, La- minaria saccharina f. linearis may be found more or less abundantly, as well as Laminaria digitata f. stenophyUa, which may in some piaces be found on slightly sloping rocks even far up in the litto- ral region, as will be mentioned below. In narrow inlets of shal- low water lying between rocks and small islands, »Skærgaarde«, e. g. those mentioned before on the east coast of Stromo, north of the redoubt, large numbers of Laminaria saccharina f. bullata 755 are often. found intermingled with the Alaria. Circumstances are, however, always somewhat altered where other species are inter- mingled. In piaces favourable to the A/ar/a-association, that is, on steep or vertical rocks on exposed coast, this association may be found almost unmixed for miles. The A/ano- association in the Færoes which has already been shortly described by Si m mons (78, p. 254) seems to agree well with Hansteen's (l.c.) and Boye's (6, p.4 — 5) A/aWa-formation. Yet we may suppose according to Boye that »Laminaria stenophyUa« Harv. must be somewhat more common in the association on the coast of Norway than at the Færoes. The Laminaria hyperborea- Association. This characteristic algæ-association is widely distributed on the coasts of the Færoes. It grows abundantly from a depth of about 1 fathom down to 10 — 15 fathoms, and it is most luxuriant at 5—10 fathoms, forming large submarine forests. It prefers hori- zontal or slightly sloping rocks, on stony ground. In such locali- ties, the very dense vegetation of Laminaria hyperborea , with its brown, pendulous, fan-like laminæ swaying with the waves, may be seen when the water is clear. Glose to the land, in gulfs and inlets, it often grows in such shallow water that the uppermost part of the lamina are seen above the surface of the sea at low tide (see fig. 160). It is here that the above -mentioned epiphytical Hhodymenia- vegetation has its habitat. On account of its vigorous haptera, the Laminaria hyperborea is able to cling to even the most exposed piaces. This algæ is how- ever not much exposed to the surf; it generally grows at too great a depth, and when it grows in shallow water it is most often in rather sheltered inlets and gulfs, as mentioned before. Its solid and comparatively not very flexible stalk would certainly make it unfit for resisting a strong surge. On the other band, it is much exposed to and also excellently fitted for resisting the strong pull of the rolling waves. When the sea is rough , the effect of this pull is felt far down, as is seen from the faet that a stormy sea can break at more than 10 fathoms of water. On a small rock outside of the »redoubt« near Thorshavn there is a dense vegetation of Laminaria hyperborea at a depth of 3 — 4 fathoms. During a lempest from the south and a rough sea, I have 756 seen how the lamina of the Laminnrin hijperboren was sucked into the waves and kept swaying to and fro when they rushed over the rock; it is certainly exposed to a very strong pvill here. But Laminaria hijperborea is excellently constructed for resisting this attack of the sea, its stalk being bolh thick and stiff at the bottom but becoming thinner and more elastic towards the apex, so that the flexible lamina can easily follow the motion of the water. It is firmly atlached to the bottom by aid of the strong haplera. It is ■,m^ '^åF Fig. 160. Laminaria hiipcrborca with nunicrous epiphytical lUiodijmenia palmata rising above the surface of the sea at low tide. Illustration from the neighbourhood of Thorshavn. (F. B. phot.) a well-known faet to anyone who has dredged among the Lami- naria hijperborea, that only by a very strong pull, or by the teeth ot the dredge cutting through the haptera more often through the stalk itself, can the plant be torn from the bottom. I have often Iried from a boat to pull up the planls that grew in shallow water, but I have hardly ever succeeded in loosening them without cutting through the haptera. Thus the plant may resist a very strong pull without being detached from the bottom. Its gregarious growth also helps to protect it from the attack of the sea. This association covers large areas with an almost unmixed growth of its characteristic alga. It would therefore look very uni- form but for the very luxuriant subvegetation of epiphyles, mainly the 757 Florideæ. It is the haptera and especially llie stalk which are almost densely covered by large or small algæ (see fig. 161). The lamina however is mostly destitute of epiphytes, most likely because it is changed every year. Various species, however, e. g. Rhodymenia palmata, Edocarpus tomentosoides, which would probably be very common in spring, E. tomentosiis , E. fascicnlatiis, and some others, may be found growing on the lamina, especially in shallow water. Among the many epiphytes that grow on the stalk and the roots, the following species may be pointed out: Pohjsiphonia iir- ceolata, Chantransia Daviesii, CaUophyllis lacininta, Eiithora cristata, Rhodophyllis dichotoma, Lomentaria claoellosa, L. rosea, Plocamiiim coccineiim, Delesseria alata , D. siiuwsa, D. sangiiinea, Polysiphonin elongata, Odonthalia dentala, Ptilola pliimosa, Pt. pectinata, Antitham- nion floccosum, Ceramiiim rubnim, Rhodochorton Rothii, Criwria pel- lita, Lithophylhim Crouani, Dermatolithon macrocarpiiin f. Laniinariæ, Peyssonelia Diibyi, Splmcelaria fiircigera, Desmarestin aciileatn, D. vi- ridis; that is to say, scarcely any but Florideæ. Most of these epi- phytes may be found even in rather shallow water in the semi- darkness under the Laniinariæ which also afford some protection from the violence of the waves. In describing the algæ-vegetation of the Mediterranean Sea, Ber- thold points out (5, p. 422) that the epiphytes on a Cystosira granii- lata grow in a cerlain fixed order. Those that require more light grow on the top of the Cystosira, those that want less grow under the latter. The epiphytes on the stalks of the Laminaria hyperborea likewise grow in a certain fixed order from the top downward, according to the degree of light required by each alga; those that want much light grow uppermost, the others lower down. The colour of the w^ater can hardly be of any consequence where the distance from the top to the bottom oraly means the Icnglh of the stipes. At the top of the stipes we find Ceramiiim riibrum, Pohj- siphonia iirceolata, Delesseria alata, Polysiphonia elongata, and from these algæ downward, we find Odonthalia dentata, Delesseria simiosa, Plocamiiim coccineiim, Ptilota plumosa, Lomentaria clavellosa, Splmce- laria fiircigera, Rhodochorton Rothii, Criwria pellita, Dermatolithon macrocarpiim f. Laniinariæ etc. On the lowest part of the stipes, and especially on the haptera, CaUophyllis laciniata, Rhodophyllis di- chotoma and Eiithora cristata are found. The Laminaria hyperhorea-assoc'ial'ion reaches its finest develop- ment in the course of the summer. The lamina is changed in winter, 758 and the old fronds are thrown off in spring, at which lime of the 3'ear they are washed up on the beach in great numbers by the heavy storms. Even in May and June, however, specimens may be found which still have the old lamina attached. Fig. 161. Stipes and haptera o( Laminaria liyperbnrea with a dense covering of epiphytes. On the stipes, Plo- camium coccineiim is prominent; on Ihehaplera, CallophylUs laciniata and Eutlioracristata. Uppcrniost on the stipes we find Ccramium rubrum, I'olijsiplionia urceolata, Delesseria alata, Polysiphonia elongata. (F. B. phot. The Laminaria hijperborea association grows more or less densely in all suitable localities down to a depth of 10 — 15 fathoms^ (Sim- mons gave 25 meters). Besides the above mentioned epiphytes, ^ It is hardly probable that the Laminaria hijperborea should be foiiiul forming an association below tliis depth; but a single specimen may well be dredged now and then from a still greater depth. 759 whicli are common and peculiar to the Laminaria lujperborea it- self, a number of large and some small algæ are found in the Laminaria hij per bo rea-associaiion, where the latter is less abundant. Thus a rich growth of the Alaria escnienta is often found in piaces with a rapid ciirrenl, down to a depth of about 5 — 6 fathoms. At about the same depth, I have found enormous specimens, a fa- thom long, of Chorda tomentosa in the rapid current where Haralds- sund is narrowest. On stones and rocks below and among the Laminaria hijper- borea, we often meet with different red algæ e. g. Porphyra miniata, Phijllophora Brodiæi, Odonthalia dentata, and several of the algæ that grow epiphytically on Laminaria hijperborea. Lastly Desmarestia aculeata and D. viridis may be found abundantly down to a depth of about 10 — 15 fathoms. So far as may be judged from the dred- gings, these algæ form often an almost pure Des/nares//a-association. The Laminaria hyperborea- dissociation has already been men- tioned in connection with the Færoes by Simmons, who likewise writes that the »Laminaria hy perborea-f ormaiion« is the most com- mon on the coast of these Isles. It is indeed only absent where the bottom is sandy right up to the beach. With the exceplion that the epiphytic vegetation presents slight differences, the Laminaria-yege- tation of the Færoes corresponds exactly with the Laminaria-forma- tion on the west coast of Norway described by Boye (6, p. 5) and Hans teen (38, p. 351); compare also Ekman (17,p. 4— 5 and p. 10). On the Shetland Isles I have had an opportunity of seeing a Lami- nar ia-\ege{aiion which corresponds with this association. S t rom- felt does not give any special account of the Icelandic Lami- naria hyperborea- associaiion; but most likely a similar vegetation must be found on the southern and western coasts of Iceland. The Lithoderma- Association. On the coasts of the Færoes, I have not seen this association so typically developed as has been described by Kjellman (23, p. 66) and Rosen vinge (45, p. 223). The reason is, probably, that the bottom in deeper water is ot- ten of sand or mud and therefore destitute of piants. In the piaces where I found this association, the bottom was covered by peb- bles. Lithoderma fatiscens grew here together with some other algæ, especially Florideæ , which made the vegetation look somewhat variegated. 760 Near Gliversnæs, at a depth of about 10 — ISfathoms, on a bot- tom of pebbles and shells, a vegetation was found consisting of the following algæ: AntUhamnion Plumnla, Polijides rotiindus, Porphyra miniata, Phyllophora Brodiæi, Callophyllis laciniata, Ulva Lactuca, Ectocarpus Hincksiæ, Desmarestia viridis, Laminaria saccharina, and Ibe following crust-like algæ: Lithoderma fatiscens, Peyssonnelia Dii- byi, and less abundantly, Lithothamnion læve and Phymatolithon lævi- gatiim. The calcareons shells found among the pebbles had heie, as everywhere in the subliltoral region, a green or reddish hue from the Conchocelis rosea, Gomonlia polyrhiza, Ostreobiiim Qiieketti and Hyella. At Hvidenæs, at a depth of about 10 fathoms, a similar vegetation was found; yet the following algæ: Odonthalia dentata, Delesseria simiosa, Derbesia marina, Polysiphonia Brodiæi and P. elon- gata occurred in addition to the majority of the above-mentioned algæ from Gliversnæs. This vegetation seems to correspond with Kjellman's and Rosenvinge's descriptions except for some flori- stical differences. The characteristic alga is however not so com- mon on the coasts of the Færoes as in Greenland and in the »Murman Sea«. In connection with this, we must also point out that the Færoes seem to be perfectly destitute of the Lithothamnion-formation else- where very common in the northern seas. With the exception of Phymatolithon polymorpham, which is very common on rocks from the lowest water mark downward into the sublittoral re- gion, the Lithothamnion-species are relatively few on the Færoes. The specimens found there are, according to what Mr. Foslie kindly tells me, often stunted and badly developed. Most likely the reason is, that the bottom is in deep water most often of sand or mud, and therefore not suitable to the requirements of these algæ. Si m mons, who once dredged with me at Gliversnæs, gives a description of some of the algæ from that place (p. 257) yet with- out drawing any final conclusion as to the character of this vege- tation. The snbUttoral Florideæ-For mation. The brown algæ which grow deepest e. g. Laminaria hyper- borea and Desmarestia acnleata, already become scarce at 15 fa- thoms, and most likely disappear at 20 fathoms. At this depth hardly anything but Florideæ are found. At the curve of 20 fathoms 761 the bottom on the coasts of the Færoes is most often of soft material, either sand or mud, and therefore destitute of piants. I have however happened to meet with sonie small stretches covered with plant-life. In Nolsofjord, for instance, outside Gliversnæs, at a depth of 20 — 25 fathoms, I found a somewhat luxuriant vegetation on larger or smaller stones, quantities of which canie up in the dredge together with the algæ. The following species were found here: Delesseria siniwsa, D. san- giiinea, CallophijUis laciniata, Rhodophyllis dichotoma, Euthora cristata and Ptilota phimosa. Among the algæ were several Brijozoa which had a more or less reddish hue from the RIwdochorton inembra- naceum and the Rh. penicilliforme. Small crusts of species of Litho- thamnion also grew on the stones, but only imperfectly developed. The calcareous shells scattered among the stones had likewise a reddish hue from the Conchocelis rosea and the Hyella cæspitosa var. nitida, and some had become green from the Ostreobium Qiie- ketti. As may be seen from this enumeration, this vegetation was rather rich in species. It must be pointed out further, that all the algæ, with the exception of the Ostreobium, were red, which agrees with the theory of Engelmann, (ep. Gaidukow, 35). If Nadson is right, as seems likely, in declaring that Conchocelis rosea — Ostreobium Queketti, all the algæ found here may justly be cailed red algæ, as it seems quite natural, that an alga which takes on, now a red, now a green hue, according to the surrounding conditions, may also be found in both forms at one time. Below^ 25 fathoms the individuals became very weak and were overgrown by Bryozoa, and at a depth of 30 fathoms no algæ were found. In Klaksvig, I have found Phyllophora Brodiæi on pebbles, at a depth of about 20 fathoms. Moreover it must not be forgotten that all these algæ are com- mon North Atlantic sublittoral species, growing especially in the deeper parts of this region. No species characteristic for these depths have been found either here or anywhere else. This association agrees with the sublittoral F/orWeæ-formation found by Rosen vinge in Greenland (p. 222); floristically however it differs somewhat. 762 b. Sheltered Coast. The Stictyosiphon- Association. At Ihe low water mark and below, and on non-tidal coasts di- rectly below Ihe surface of the sea to a depth of from 4 to 6 feet, in some piaces even still farther down, \ve meet with an associa- tion which consists of a great many difFerent species of algæ. It corresponds to the Co/a///na-formalion on exposed coasts; Ihe Coral- lina officinalis is moreover often found scattered in the vegetation. On account of the great difference between the algæ belonging to this association, the colour of the vegetation varies greatly according to which algæ predominate at each place. It may be sometimes the Ulvaceæ, sometimes the brown, sometimes the red algæ. Their ample ramifications and bushy form are characleristic of the algæ which grow here. Where these marks are absent, they at least grow gregariously, in dense tufts. This tuft-like growth is also due perhaps to the faet, that this association grows on a stony bottom, especially on pebbles. Most often one stone carries one alga, its neighbour another. Green algæ typical of this association are: Enteromorpha Linza, Acrosiphonia albescens, Urospora Wormskioldii, Enteromorpha intesti- nalis, Monostroma fnscum etc. ; brown algæ : Chordaria flagelliformis, Dictyosiphon foeniculaceus , Stictyosiphon tortilis, Scytosiphon lomen- tarius, Phyllitis fascia, Punctaria plantaginea, different forms of Ecto- carpus litoralis; Ectocarpus siliculosns, Castagnea virescens, Chorda fdum, Fucus inflatus, and sometimes Laminaria saccharina and Leathesia difformis epiphytically on Corallina officinalis. Amongst the red algæ are: Ceraminm rnbrnm, Cystocloniiim purpurascens, Rhodomela subfusca and R. lycopodioides, Polysiphonia Brodiæi and P. nrceolata, Dnmontia fdiformis, Chondruscrispns,RhodymeniapaImata, and, more rarely,Ha/o- saccion ramentaceum, Antithamnion floccosum and Laurencia pinnatifida. At Strender in Skaalefjord, which is not subject to high and low tides, this association is beautifully developed. There is a large flat area covered with shallow water, from 2 to 6 feet deep. In the imme- diate vicinity of the land, I found a rather broad helt of Acrosiphonia albescens, which was partly laid bare on the day when I visited the place, the water being extremely low and the weather fine and calm. Next to that, there came a variety of algæ of difFerent colours, with the Corallina officinalis as a fairly frequent subvegetation. There were large, red tufts of Ceraminm rnbrum, Rhodomela lycopodioides, Dumontia fdiformis, Cystoclonium purpurascens etc, together with some 763 brown tufts of Dictijosiphon foeniculaceiis , Chordaria flagelliformis, Stictyosiphon tortilis, Ectocarpus siliculosus and E. litoralis together with gregarious growths of Phyllitis fascia and Scijtosiphon. Among these darkish algæ the fresh green Enteromorpha Linza lightened up the vegetation, in which Urospora Wormskioldii, Monostroma fusciim and other green algæ were still intermingled. A little farther out in a few feet of water large specimens of Laminar ia saccharina were found mixed with the association. On the banks of the »Sundene« between Thorsvig and Kvalvig, a very similar vegetation is found. The bottom here is of sand with large stones and pebbles. From the lowest water mark down to a depth of from 3 to 6 feet we find an assemblage of algæ of many colours e. g. Cystocloniiim purpiirascens . Castagnea virescens, Punctaria plantaginea, Ectocarpus siliculosus and E. litoralis, Dumontia filiformis, Chorda fdum, Enteromorpha Linza and several of the above mentioned species. At a short distance from the land, where the water is from 4 to 6 feet deep, the Laminaria fau-oensis-assoc'm- tion covers the bottom. Neilher of the two above mentioned tracts are tidal, but a tidal locality with a similar vegetation from the lowest water-mark down- wards to a depth of from 4 to 5 feet is found on the coast of the Vestmanhavn. This vegetation includes: Monostroma fuscum, Ce- ramium rnbrum, Porphyra miniata, Rhodymenia palmata, large »clouds« of Ectocarpus litoralis, Enteromorpha clathrata, Dumontia fdiformis, Chorda fdum, the upper part of whose thailus floats on the surface of the water, and many more. In Vestmanhavn and likewise in Klaksvig, where the vegeta- tions much resemble each other, Halosaccion ramentaceum, otherwise very rare in the Færoes, is found. In Klaksvig it grows abund- antly on stones to a depth of a few feet at low water, and it is most often much overgrown by green algæ e. g. Enteromorpha, Acrosiphonia, etc. In the Danish edition I have called this association »The varie- gated Association«, because of its very motley appearance, due to the many different species which belong to it. But as this name has already, as pointed out by me (10, p. 70), been used by Kjell- man (45, p. 24) I have found it more convenient to change the name, since the association of the Færoes is quite different from that of Bohuslån. Botany of the Færoes. 49 764 According to m}^ definition of this association, it may perhaps partly agree with the Ulvaceenformation described by Simmons (p. 251). He at least says so himself (66, p. 174); but it seems to me, that his formation agrees mostly with the Monostroma-Entero- Tnorp/ia- association mentioned below. Simmons declares, how- ever, that his formation belongs to the Httoral region, though he remarks that it may be found as far down as about 2 meters below ebb mark. Referring here to my statement on p. 710 I need only mention, that it seems very unnatural to call a vegetation Ht- toral, when it is found below the lowest water mark in a locality subject to ebb and tide, and on the whole below the littoral algæ- vegetation, even though most of these algæ may also be found in the littoral region. According to the descriptions of the algæ-associations on the coasts of Norway, I think, that this association shows no small similarity to the Dictijosiphon-, Spermatochnus- and Cora////?a-forma- tions, described by Boye. The Monostroma-Enteromorpha-Association. This association reaches from the surface of the sea down to a depth of a few fathoms. The transition from the previously men- tioned association is often smooth, as the Chiorophyceæ under cer- tain conditions gradually become predominant. The association is especially luxuriant in the bottom of the fjords, where the outflow of the rivers turns the water more or less brackish. It mostly grows on a bottom of stones or gravel in shallow water, but it may also be found drifting on the surface of the sea. Among the species characteristic of this association Monostroma fuscum ought to be mentioned, as it is often predominant and may be found as a pure facies covering large spaces of the bottom. Usu- ally, various species of Enteromorpha , especially ^. clathrata and forms of E. intestinalis, åre, however, intermingled. The latter species may grow in the fresbest water, especially var. genuina and var. proUfera which often form large entangled masses at the mouths of rivers, for instance in Kalbakbotn. Further, varieties of Monostroma undiilatum are found in the association. In Klaksvig, in the direct vicinity of the land, great numbers of Chætomorpha tortuosa and an Acrosiphonia spec. are found, besides the above-mentioned algæ. 765 Together with Ectocarpns liloralis these algæ form tough, en- tangled, loose-lying masses in shallow water. Besides these algæ we find Chorda filiim, Scijtosiphon lomentariiis, Ceramiiim ruhriim, Halo- saccion rdinentaceiim and somelimes also other species belonging to the above mentioned Stictyosiphon-assoc'iaVion. As already mentioned this association together with the Stictij()sij)hon-association perhaps answers to Si m mons's Ulvaceenformation. Th e Ha lidrys-Associa tion . As this association is but rarely found at the Færoes, it is of minor importance there. It has only been found attached to the bottom within a very limited region at Glibre in Skaalefjord on Østero, but as this alga is ralher abundant in this locality and agrees quite well with Hansteen's (38, p. 353), Boye's (6, p. 8) and Gran's (36, p. 19) descriptions of their //a/zf/rys- formation, I think that it ought to be mentioned here. It grows here on a stony bot- tom at a depth of from 1/2 to 1 fathom, forming a narrow belt at no great distance from the coast. The specimens are enormous, almost one fathom long. It grows together with scattered specimens of Laminaria digitata, L. hyperborea and L. færoensis and Chorda filiim. When I visited the tract in May 1898, I found several small specimens of Piindaria latifolia, Chordaria flagelliformis, Ectocarpns litoralis, Stictyosiphon foeniculaceus , Dictyosiphon and a great many Diatoms growing epiphytically on the Halidrys. In August (?) Si m mons found it densely overgrown by Dic- tyosiphon hippiiroides ; he therefore calls it »the Hcdidrys-Dictyosiphon- formation«. The Laminariaceæ-Formation on sheltered coasts. The Lammanaceæ-formation varies much on sheltered coasts. Where there is no current, Laminaria færoensis grows abundantly, forming widely spread associations. But where there is a rapid cur- rent, as in most sounds between the islands, L. færoensis is wanting and is replaced chiefly by L. hyperborea. Let us first consider the Laminaria færoensis-association whicli grows on sheltered coasts. This alga necessarily requires a habitat where the sea is never disturbed. The large lamina which is often several fathoms long and easily broken would soon be torn by the surf. The slender haptera and the brittle stalk would not long be able to resist the strain caused by the current. 49* 76() I have found this association at its finest development in the Sounds between Stromo and Ostero, and especially between Thorsvig and Kvalvig. The tides are wanting here, consequently there is no current, and even during a storm the water is only slighlly troubled in this narrow sound. Huge specimens of L. færoensis grow abundantly here in shallow water near the coast forming dense »forests«. The stalk which is often nearly two fathoms long and thin below, is fastened by several haptera to stones on the bottom. Its uppermost part is thick, hollow and full of air; it often measures some inches in diameter and serves as an air-bladder. These may be seen side by side on the surface of the water, the huge lamina hanging down from them. This agrees perfectly with Rosenvinge's description (71, p. 211) of L. longicriiris on the coasts of Greenland, which species in the whole mode of its life seenis to be closely related to L. færoensis. L. færoensis fructifies in summer, and probably then reaches ils highest development. The changing of the lamina is hardly restricted to any special season. According to my observations it may be supposed, that the lamina grows more or less evenly for the greater part of the year, and gradually decays at the apex. As mentioned before, Laminaria færoensis grows down to a depth of about 10 fathoms; in deep water it is, however, most often some- what smaller. Far out in the fjords, where there is some current and more troubled water, the lamina becomes narrower, almost lanceolate and the stalk less swelled. I call this form f. snccharini- formis, as it is very much like L. saccharinn, from which it may, however, be easily distinguished by its hollow stalk. The stalks of L. færoensis are generally without epiphytes. Ac- cording to Rosen vinge (1. c. p. 212), this is also most often the case with L. longicraris in Greenland. In »Sundene«, in shallow water, I have however sometimes met with piants with the stalks den- sely overgrown with some green and brown algæ, e. g. Monostroma fuscum, Ectocarpus litoralis, Ectocarpus spec, etc. The stones on which it grows are likewise often covered by large, dark red crusts of Cruoria pellita. In deep water various red epiphytes, e. g. Polij- siphonia nrceolata, Lomentaria clavellosa, Ptilota plumosa, Delesseria sinuosa, etc. are also found on the stalk. It very commonly oc- curs also, that young specimens of Laminaria færoensis are attached to the stalks of old specimens. In deeper water the Laminaria 767 hijperhorea is often found in the association. The lamina of L. hijperborea in this sheltered place is often only slightly divided, sometimes even perfectly wliole as with f. ciicullata of L. digitata, whicli species may also be found belonging to the association especially in shallower water. The Laminaria /'ceroe/js/s- association has also been found at the bottom of Trangisvaag and Vaagfjord on Sj^dero, and in Kal- bakfjord and Kollefjord on Stromo, and in Skaalefjord on Ostero, that is, in the most sheltered localities, but nowhere so luxuriantly developed as in the »Sundene«. This vegetation is however quite wanting in piaces where there is a current or surf from the sea. Here we find a Laminaria hyperborea -associaiion which essentially resembles that of exposed coasts. The rapid current makes up for the troubled sea and con- stantly carries clean water to the algæ. In these conditions we therefore find an epiphytical vegetation on the stalk of L. hyper- borea, as luxuriant and well developed as on exposed coasts. La- minaria digitata is often found intermingled in the upper part of the association, and both here and lower down Laminaria saccharina and Alaria escnienta may be found. Si m mons mentions a Laminaria longicrnris-formaiion (78, p. 256) which he calls this association, following my former determination of Laminaria færoensis. The Desmarestia- Association. Here and there, where the bottom is stony, at a depth of about 10 fathoms, we find an association which mainly consists of Des- marestia acnleata with Desmarestia viridis intermingled. To judge from the dredgings, this association was well developed at the entrance of the gulf at Klaksvig, where it has also been observed by Si m mon s. Phyllophora Brodiæi, Polyides rotnndns, Porphyra miniata, Antithamnion Plnmnla, and some few Laminaria saccharina and L. hyperborea were also found here on stones. Si m mons men- tions Ph. rubens and Ph. membranifolia in this locality, but my obser- vations (8, p. 358) indicate that this must certainly be a mistake. In the »Sundene« I have met with a similar Desmaresfza-association ; and Porphyra coccinea was found growing on Desmarestia acnleata which may be perfectly hidden by this beautiful, little rosy alga. On the coasts of the Færoes this /)(?s/27ores//a-association is at- tached to the bottom, even in sheltered piaces. Some loose speci- 76cS mens mav of course be sometimes found, but not in great numbers, what is, however, the case on the coasts of Greenland, according to Rosenvinge (71, p. 218). The name »formation of loose-lying Algæ« may in the Færoes only be used of the above mentioned loose masses of algæ, which mainly consist of species of Enteromorpha, Chætomorpha tortiiosa, Acrosiphonia, Monostroma, etc. The Vegetation on soft bottom. The Zostera- Association. With the exception of the Characeæ, which may be found in brackish water, algæ from the northern seas will not usually thrive on a bottom of sand or mud^, which is therefore destitute of vegetation. Only here and there, where stones are found, may algæ be met with, but as a soft, loose bottom is rather common in the sounds between the islands and in the fjords, at any rate in deep water, somewhat large, naked regions occur with only a few loose drifting algæ. What makes the tracts still more naked is, that the Zostera marina, which has its habitat in shallow water (down to a depth of from 6 to 7 fathoms in our seas) in this kind of ground with loose bottom, is almost totally wan- ting at the Færoes. It has only been found at the bottom of Vaag Fjord. In a small limited territory it makes an association here at a depth of about 1 fathom, and at low tide, its leaves may be seen floating on the surface of the sea. The Zostera here as elsewhere shelters a number of different epiphytes, e. g. species of Enteromorpha, Ectocarpns litoralis, Chantransia virgatnla and C. secnn- data, Microsijphar Zosteræ, Scytosiphon lomentarins, and so forth. 3, The Iower limit of the Algæ-vegetation. The elittoral Region. As mentioned before, I agree with Ro sen vinge in determining the elittoral region as that part of the bottom of the sea where no vegetation grows on the bottom ; I shall therefore leave this re- gion out of consideration. At far greater depths than those at which the fixed algæ grow, we may of course find loose-lying algæ which have been carried by the current away from their habitats. Such portions of algæ may perhaps live for some time by con- suming their reserve stores, but when these are used up they must ^ In the tropics however we find an algæ-vegetation which is often qnite dense and fastened on a bottom of sand or mud. 769 of course perish. They are, however, of no importance for the deter- mination of the lowest deptli at whicli the algæ-vegetation mav be found; in this connection we are only interested with the fixed algæ- vegetation. It is, however, difficult to determine how deep this limit is. It varies much in the ditTerent seas and depends on how far down a sufficient amount of light may reach. The light reaches farthest down in the tropics, where the rays of the sun strike the surface of the sea almost vertically. In higher latitudes the rays strike the surface more and more slantingly, and therefore do not reach so far down. According to Berthold, a rich vegetation of deep w^ater algæ was still found at a depth of about from 120 to 130 metres in the gulf at Naples, and according to Rodriguez, the very lowest limit of the vegetation is 160 meters at Minorca (com- pare Sauvageau: Remarques sur les Sphacélariacées, p. 235). In the tropics the algæ-vegetation probably grows at a still greater depth. According to Ros en vinge, the lower limit of the algæ-vegeta- tion in Greenland is about the curve of from 20 to 30 fathoms. As I have before indicated, this almost agrees with what has been observed at the coasts of the Færoes. Naturally, local conditions often cause the vegetation to disappear above this depth ; there may for instance be plenty of Plankton, or the bottom may at a certain depth turn into mud or sand, which makes it unfit for the algæ. This is perhaps the case on the coasts of Bohuslån, where the algæ-vegetation, according to Kjell man, hardly grows any farther down than a depth of 20 fathoms. III. GENERAL CHARACTER OF THE ALGÆ-VEGETATION. Upon the whole the algal vegetation on the Færoese coasts must be considered specially vigorous and luxuriant. That stunted forms may be found in unfavourable localities, such as the upper part of the beach or the heads of inlets, where the water becomes brack- ish or almost fresh, is of no consequence in comparison with the luxuriance which otherwise characterizes the vegetation. We could not easily imagine any locality better fitted for developing a vigo- rous algæ-vegetation than the Færoes, where the nature of the coast, the situation in the midst of a large ocean, and the climate, all serve to facilitate the growth of the algæ. 770 Exposed coast Sheltered 12 13 14 15 16 17 18 19 20 21 - 22 23 24 25 26 27 28 29 frtt Explanation of the diagram: The accompanying diagram will serve to display the occurrence of the different algæ-formations and algæ-associations on the coasts of the Færoes. The numbers to the left indicate the number of feet respectively above 771 or belovv the sea level at middle height of the water (0). The beach, which is represented by the distance between the two horizontal lines, is soniewhat enlarged, in order to make it more distinct. The vertical lines indicate the level at which each algæ-forniation and association mostly occurs, the dottcd parts of the lines that they may be found beyond this level, though more scattered. The nunibers above the vertical lines indicate as foUows: 1. HildenbraiuUa-Formation with lichens. 2. Chlorophyceæ-Formation on exposed coast. Prasiola cr/spa-Association. i?/j/zoc/onåzm-Association. Enteromorpha- Association. Pixisiola stipitata-As^ociation. 3. Po/'pAi/ra-Association. 4. /?/jodoc/?or/on-Association. 5. Bangia- Urospora Association. 6. Fucac^æ-Formation on exposed coast. Fiiciis spiralis-AssocmXion. Fiiciis m/Za/us-Association. 7. Ca//;7/jam/HO/j-Association. 8. i?/jorfymen/a-Association. 9. The littoral Co/-a////ja-Formation. 10. Monos/;-oma-G/-efH7/c /-Association. 1 ^, „ ■ ^- u .. , • , • r, ; • L • i • +• These four associations be- 11. Acrosinhonia-Polijsiphonia-AssociaUon. , ^ xi i-x^ i /- ; .„ ^. ;. . ... long to the littoral Coral- 12. Giqa rtina- Asaociahon ,. °^ .o TT ,/;•*•♦• /z;jrt-Formation. 13. //zmo/7//jo//a-Association. | 1 4. Phynuitolithon- Association. 15. The sublittoral Co/'a//ma-Formation. 16. Laminaria dia itata- Associaiion ] r • ■ • ^ ,, . . ... Lamznanaceæ-t' ormation 17. A/arza-Association. 18. Laminaria hy per bo rea- Associaiion ) °" exposed coast. 19. Z)esmorcs//a-Association on exposed coast. 20. Lz7/jode/-ma-Association. 21. The sublittoral HoWrfeæ-Formation. 22. C/i/o/op/zz/ceæ-Formation on sheltered coast. Fn/eromo/yj/ia-Association. 23. Fucaceæ-Formation on sheltered coast. Pe/z'e//a-Association. Fiiciis vesiciilosus-Ascophylliim-Associalion. Fu- ciis inflatas-Associaiion. 24. S/zc/yos/p/io/j-Association. 25. Monostroma-Enteromorpha-Associaiion. 26. //a//rfrr/s-Association. 27. Lam manoceæ- P'ormation on sheltered coast. Laminaria /ceroe/isw-Association. Laminaria hyperborea- Association. 28. Desmaresfm-Association on sheltered coast. (29. Zostera-Associatioij.) Lastly, it should be pointed out, that the heights refer to the coast as a whole, not to any special locality; thus the space within which an association may be found is represented larger than would be the case if we examined any single locality. 772 From even far above the highest water mark down to a depth of about 15 fathoms, we may often find a dense, well developed algal vegetation , which is divided naturally by the tides into two regions: the littoral and the sublittoral. At few piaces is there, I think, a littoral vegetation so luxuriant as that of the Færoes. This is certainly due in the first place to the very favourable cli- mate, secondly to the situation in the open sea, where the surf constantly washes on the coast and thus enables the algæ to grow far above the highest water mark. Nevertheless the vegetation of exposed coasts hardly ever becomes as luxuriant as the Fiicaceæ- vegetation in more sheltered piaces. But even this abundant, lit- toral vegetation is far surpassed by the sublittoral vegetation of the Laminariaceæ which covers a large area of the bottom of the sea with »forests« of almost a man's height, and a vigorous undergrowth of epiphytes. As almost everywhere in the cold temperate seas, the brown algæ predominate in size as well as in number on the coasts of the Færoes, both in the sublittoral region with its Laminariaceæ and other algæ sheltered by these, and also in the littoral. As to the latter, this is only the case, however, so far as the Fucaceæ-vegeta- tion of sheltered coasts is concerned, on exposed coasts the vege- tation becomes more varied, both red, brown and green algæ crowd together, and sometimes green algæ, e. g. Enteromorpha and Acrosiphonia , sometimes red, e. g. Porphyra and Rliodymenia pre- dominate over large tracts. Even in the size of the piants, the red algæ may rival the brown algæ. With the exception of some Laminariæ and Alariæ which in favourable piaces sometimes ex- tend into the littoral region, only Himanthalia attains any consider- able height. In connection with this brief description of the general charac- ter of the algal vegetation I endeavour to give in the diagram above a general view of the different algæ-formations and algæ-associa- lions, the different levels at which they occur, and the manner in which they replace each other; thus, it will be clearly seen that the number of algæ-associations reaches its height in the littoral region and then gradually decreases on each side. 773 IV. THE PHYTO-GEOGRAPHICAL POSITION OF THE ALGÆ- FLORA OF THE FÆROES. I. The constituent parts of the Færoese Algæ-flora. Before entering on a comparison between the marine algæ-flora of the Færoes and that of neighbouring countries, we mav first of all review the species occurring at the Færoes. In my paper on the Færoese marine algæ 212 species are men- tioned, with 4 whose genus only was determined. On renewed in- vestigation, however, I have come to the conclusion that one of the latter, Chilionema spec. may be referred to Ch. reptans. Thus we have 213 + 3 species. From this number, again, Dermatoli- tohn Corallinæ (see my »Flora« p. 402) must be withdrawn as, ac- cording to Mr. Foslie's latest investigations, kindly placed at my disposal, it should be referred to Dermatolithon macrocarpum, or, as he now calls it, LitophyUiim macrocarpum. Moreover Acrosiphonia Binderi must be omitted, renewed examination of the present material having proved that some imperfectly developed curved branches occur, though but rarely, and this convinces me that it should be referred to A. alhescens. Thus the number is reduced to 211 + 3, but to these must be added Choreolax Polysiphonice, which I have discovered later^ In order to give a general view of the 212 + 3 species, and in order to discuss more closely the elements of which the Færoese algæ-flora consists, I have divided tliem, in accordance with the divi- sion made by Jonsson and myself (12, page XV — XXI), into the 4 groups: the subarctic, the boreal-arctic, the cold-boreal and the warm-boreal group (the ar et i c group is entirely wanting in the Færoes). As to the hmitation of these groups, I may refer the reader to the above-menlioned paper. It should not be forgotten, however, that any such division must inevitably suffer from a certain ' Besides these alterations, the name of the species, or of the genus, or of both have been altered in the following list of Færoese algæ in the case of the following species: Clathromorphiim circiimscriptiim which, according to Mr. Foslie, should be referred to PhymatoUthon compactiim, Acrosiphonia flaccida, which I refer to A. hijstrix (Stromf.) Borgs., as already pointed out bj' me (8, p. 512). The form which I gave provisionalh' as Spongomorpha lanosa, and which I supposed to be related to Sp. bombycina (Kjellm.), I have found on renewed examination to be Spongomorpha vernalis. 774 arbilrariness, parlly because we know so very little as yet about the distribution of many algæ, partly because several algæ are distri- buted so as to make it almost equally juslifiable to refer them to different groups. Whilst Reinke and Kuckuck leave the Chloro- pliyceæ and the Cyanophijceæ out of consideration , as our know- ledge of these piants is especially defective, I side with Rosen- vinge 's opinion, however, that it is more correct to mention them here, as the species belonging to these orders are as important from a phytogeographical point of view as those of the Rhodo- phyceæ and Phæophyceæ. The species hitherto found only on the Færoes are all referred to the cold-boreal group. SUMMARY OF THE FÆROESE MARINE ALGÆ. I. Rhodophyceæ. 83 species. A. The arctic group. O species. B. The siibarctic group. Subdivision 1. 6 species. Rhodophyllis dichotoma. Halosaccion ramentaceum. Ptilota pectinata. Rhodochorton penicilliforme. Phymatolithon compactum. Lithothamnion læve. Subdivision 2. 14 species. Porphyra miniata. Conchocelis rosea. Chantransia efflorescens. — virgatula. Harveyella mirabilis. Phyllophora Brodiæi. Actinococcus subcutaneus. Euthora cristata. Rhodymenia palmata. Delesseria sinuosa. Rhodomela lycopodioides. Odonthalia dentata. Ptilota plumosa. Lithothamnion glaciale. C. The boreal-arctic group. 10 species. Bangia fuscopurpurea. Porphyra umbilicalis. Chantransia secundata. Ahnfeltia plicata. Polysiphonia elongata. Antithamnion Plumula. Ceramium rubrum. Rhodochorton membranaceum. — Rothii. Hildenbrandia rosea. D. The cold-boreal group. 38 species. Erythrotrichia ceramicola. Porphyra coccinea. Chantransia Alariæ. — Daviesii. Choreocolax Polysiphoniæ. Gigartina mamillosa. Phyllophora membranifolia. 775 Callocolax neglectus. Sterrocolax decipiens. Cystoclonium purpurascens. Lomentaria rosea. Delesseria alata. — sangiiinea. Polj'siphonia urceolata. — violacea. — fastigiata. — atrorubescens. — nigrescens. Rhodomela subfusca. Callithamnion polj^spermum. — arbiiscula. Plumaria elegans. Antithamnion floccosum. Ceramium acanthonotum. Rhodocborton seiriolanum. Dumontia fiHformis. Furcellaria fastigiata. Polyides rotundus. Cruoriella Diibjå. Rbododermis elegans. PhymatoHthon lævigatum. — polymorphum. Lithotbamnion Lenormandi. Lithopbyllum Crouani. — incrustans. — macrocarpum. — bapaHdioides. CoralHna officinalis. E. The warm-boreal groiip. 15 species. Porpbyra leucosticta. Cbondrus crispus. Callopbyllis laciniata. Lomentaria clavellosa. — articulata. Plocamium coccineum. Nitopbyllum laceratum. Laurencia pinnatifida. Polysipbonia Brodiæi. Pterosiphonia parasitica. Griffithsia setacea. Callitbamnion scopulorum. — corymbosum. — granulatum. Cruoria pellita. II. Phæophyceæ. 72 species. A. The ar die group. O species. B. The subarctic group. Subdivision 1. 8 species. Lithoderina fatiscens. Sorapion Kjellmani. Myrionema globosum. Ectocarpus æcidioides. Spacelaria britannica. Cbætopteris plumosa. Laminaria færoensis. Alaria Pylaii. Subdivision 2. 17 Arter. Ralfsia clavata. Ectocarpus tomentosoides. Leptonema fasciculatum. Elachista fucicola. Punctaria plantaginea. Istbmoplea spbæropbora. Litosipbon filiforme. Stictyosipbon tortilis. 776 Dictyosiphon hippuroides. — Ibeniculaceus. Desmaresta viridis. — aculeata. Chordaria flagelliformis. Chorda filum. — tomentosa. Laminaria digitata. Fucus inflatus. C. The boreal-arctic group. 7 species. Ectocarpus littoralis. — confervoides. — siliculosus. Scytosiphon lomentarius. Phyllitis fascia. Fucus vesiculosus. Ascophyllum nodosum. D. The cold-boreal group. 35 species. Petroderma maculiforme. Ralfsia verrucosa. Myrionema vulgare. — Corunnæ. — foecundum. — færoense. — speciosum. Chilionema reptans. Microsyphar Polysiphoniæ. — Zosteræ. Ectocarpus Stilophoræ. Ectocarpus velutinus. — lucifugus. — tomentosus. — dasycarpus. — fasciculatus. — granulosus. — Hincksiæ. Elachista scutulata. Sphacelaria cæspitula. — furcigera. — cirrhosa. Cladostephus spongiosus. Desmotrichuni undulatum, Litosiphon Laminariæ. Phæostroma parasiticum. Phyllitis zoslerifolius. Dictyosiphon Ekmani. Castagnea virescens. Laminaria saccharina. — hyperborea. Alaria esculenta. Fucus spiralis. Pelvetia canaliculata. Himanthalia lorea. E. The warm-boreal group. 5 species. Punctaria latifolia. Asperococcus echinatus. Desmarestia ligulata. Leathesia difformis. Halidrys siliquosa. III. Chlorophyceæ. 44 species. A. The arctic group. O species. B. The subarctic group. Subdivision 1. 6 species. Chlorochytrium inclusum. Monostroma undulatum. — fuscum. Ulothrix consociata. — pseudoflacca. Acrosiphonia hystrix. 777 Subdivision 2. 18 species. Codiolum gregarium. Percursaria percursa. Monostroma Grevillei. Ulothrix flacca. Bolbocoleoii piliferum. Pilinia maritima. Ulvella fucicola. — confliiens. Pringsheimia scutata. Urospora mirabilis. — Wormskioldii. Chætomorpha Melagonium. Spongomorpha vernalis. Acrosiphonia incurva. Cladophora riipestris. — sericea. Ostrobium Queketti. Vaucheria coronata. C. The boreal-arctic groiip. 7 species. Enteromorpha inlestinalis. — clathrata. Ulva Lactuca. Chætomorpha tortuosa. Rhizoclonium riparium. Cladophora gracilis. Gomontia polyrhiza. D. The cold-boreal group. 11 species. Codiolum pusillum. Prasiola crispa •= marina. — furfuracea. — stipitata. Endoderma Wittrockii. Acrochæte repens. Acrosiphonia albescens. — flagellata. — grandis. Derbesia marina. Valonia ovalis. E. The warm-horeal group. 2 species. Enteromorpha Linza. Bryopsis plumosa. IV. Cyanophyceæ. 13 species. Å. The ar die group. O species. B. The siibarctic group. Subdivision 1. O species. Subdivision 2. 1 species. Pleurocapsa amethystea. C. The boreal-arctic group. 3 species. Spirulina subsalsa. Calothrix scopulorum. Rivularia atra. D. The cold-boreal group. 2 species. Dermocarpa Farlowii. Hyella endophytica. E. The warm-boreal group. 7 species. Chlorogloea tuberculosa. Dermocarpa violacea. Hyella cæspitosa. Phormidium autumnale. Lyngbya lutea. Microcoleus tenerrimus. Calothrix æruginea. 778 It is clear from this summary that the Rhodophyceæ are most numerous, contributing almost Vs of all the species collected. The Phæophyceæ are a little less numerous, aboul ^3 of all the marine algæ- species of the Færoes; next come the Chlorophyceæ, which only amount to a little above V5, and lastly the Cyanophyceæ, of which only a few species have been collected. Ro se uvinge (71, p. 173) has given a summary in per centages of the number of species belonging to each of the four classes of algæ, as they occur in the algæ-floræ of Greenland, the British Isles and in the Spanish-Canary region. If the numbers that represent these groups in the Færoese algæ-flora are placed in this summary, and if, at the same time, the numbers representing the Greenland algæ are altered to agree with the latest contributions concerning the latter, and the numbers from the British Isles are calculaled from Bat- ter s's latest list (4), the result will be as follows: Greenland The Færoes ^'^^ ^"^ish j The Spanish- ; Isles rCanarj' Reg. Rhodophyceæ 26.0 % 38.6 »/o 42.3 Vo \ 60.4 "/o Phæophyceæ j 40.0 "/o 34.0 % ; 27.8 % 18.7 «/o Chlorophyceæ 30.0% I 20.9 7o ! 17.9 7o , 14.8 7o Cyanophyceæ 1 6.0 7o 6.5 «/o 12.0 "/o 6.1 Vo II J \ \ Total number of species. . [i 169 215 744 492 As clearly shown by the figures, the Færoes hold an inter- mediate position between Greenland and the British Isles; Ihe Rhodo- phyceæ, which are in the majority in England, to say nolhing of the Spanish-Canary region, are still predominant on the Færoese coasts; whereas the Phæophyceæ, which are vastly in the majority in Greenland almost reach to the same number. The Chlorophyceæ likewise hold an intermediate position as regards number of species between the British Isles and Greenland, where they are relatively more numerous; the same may be said of the Cyanophyceæ, only inversely, as these are more numerous in the British Isles and less numerous in Greenland. That only a re- latively small number of the latter have been found in the Spanish- • Canary region, is certainly due to the faet that this region (as regards the Cyanophyceæ) has hitherto been but very little investigated. It is, on the whole, a well-known faet that the Phæophyceæ 779 predominate in tlie northern countries both in growth^ and in nuniber of species (see Rosenvinge's table 71, p. 174); whether the same mav be said of tiie Chlorophyceæ , as the table above seems to indicate, appears to me, bowever, rather doubtful. At any rate they are very numerous in the littoral and the upper sublit- toral regions of the West-Indies and seemingly predominate there^. After these remarks we may consider how the Færoese species are grouped in the above- mentioned five groups. We liave the following result: O arctic species or O °/o 70 subarctic species or ... . 32.55 Vo 27 boreal-arctic - - 12.55 Vo 89 cold-boreal - - .... 41.4 % 29 warm-boreal - - 13.5 »/o It is clear from this table that the cold-boreal species are in the majority in the Færoes, as not far from one half of the Færoese algæ belongs to this group. Next to that come the subarctic species, including nearly one third of all the Færoese algæ. Of species oc- curring in more southern regions, the warm-boreal species, the Færoes have a little more than 13*^/o; almost the same number of species must be referred to the boreal-arctic group, which is, bow- ever, of less interest from this point of view, as all these species are widely spread and often ubiquitous, and at any rate common in all five groups. If we leave the Chlorophyceæ and Cyanophyceæ out of consi- deration, as they possibly make the result less reliable, the result is nevertheless much the same. 156 red and brown algæ in all are found at the Færoes; that is: 45 subarctic species or 28.9 Vo 17 boreal-arctic - - 10.9 % 74 cold-boreal - - 47.4 Vo 20 warm-boreal - - 12.8% In order to show more plainly in which of the surrounding countries we meet witli an algæ-flora having a composition very closely connected to that of the Færoes, and to demonstrate the ^ Cf. Kjell man, F". R., Ur polarvaxternes Hf (Nordenskiold, Studier och Forsk- ningar, p. 544). * Cf. Kjellman, F. R., 1. c. p. 540 where he saj^s: »Ett utmårkande drag for algvegetationen i Ishafvet år fattigdomen på grona alger«. Botany of the Færoes. 50 780 phyto-geographical position of Ihis flora on tiie whole, I give a summary oF the Rhodophyceæ and Phæophyceæ of these countries in the following table; the numbers are taken from the table pre- pared by Jonsson and myself (12, p. XXII)^. Rhodophyceæ and Phæophyceæ 73 e o 1 1 II 1- S c rtT3 11 ■z H L Number of Spe- / cies \ 184+128 = 312 1014 89 = 190 49+38 = 87 83-t72 = 155 70+53 = 123 65+60 = 125 65+60 = 125 45+55 = 100 37+35 = 72 41+63 = 104 arctic ... . 3.8 0.5 3.16 « 8.9 0.8 9.7 1.6 15.2 5.5 16.6 8.9 18.8 15.3 18.0 17.3 subarctic 1 26.0 — 2 .... 9.9 14.2 17.2 20.0 16.2 24.0 23.8 29.7 50.0 27.8 boreal-arctic . . . 5.7 8.4 16.0 11.0 13.0 13.6 14.3 15.8 18.0 16.3 cold-boreal 25.8 38.0 40.0 i7.1 /t9.6 5S.4 36.5 25.7 18.0 12.5 warm-l)oreal 1 . 6.7 9.47 18.4 13.0 10.6 6.4 4.0 0.9 » » — 2 . 28A 26.3 8.0 ^ » 0.8 » » 3> » — 3 . 19.0 * " * " " * ^ " » It is evident from this table, that the Færoes come next to the Shetland Isles and Nordland. Scotland has a consider- able percentage of warm-boreal species, no arctic and only very fevsr subarctic species. West Norway has a much smaller number of warm-boreal, Va *^/o of arctic [Tnrnerella Pennyi found in Trondhjems" fjord by Foslie), and a little higher percentage of subarctic species, The Shetland Isles likewise have a smaller number of warm- boreal species, but apparently an equally large number of subarctic species as West Norway; this is, however, certainly due to the faet, that these Isles have not been sufficiently investigated, for it is clear, that if the subarctic group 1 is represented by 3.8 *^/o in Scotland, it must reach to the same amount at least in the Shetland Isles, where none of these species have been found hitherto. Of the Færoese algæ, only 13 ^/o are warm-boreal, almost one half are cold boreal and 29^0 subarctic. Nordland has not quite so many warm-boreal spe- cies, that is, only a little more than 10%; one half of the species are cold-boreal ; the subarctic group is a little smaller than at the Fær- oes, but on the other band a small arctic element is found here. In * The numbers belonging to the Færoes differ a little from those above, as Ectocarpus spec. is left out of consideration here. 781 South West Iceland, the warm-boreal group is again reduced (to a little more than 7 Vo) and the cold-boreal group is Hkewise dimi- nished; on the other hånd the subarctic group is much increased amounting to more than 39 °/o, and here we meet with ahnost 2 ^/o of arctic species. If only South Iceland is taken into con- sideration, however, the agreement with the Færoes becomes still more evident, as arctic species are totally wanting and the subarctic group is much reduced in numbers. As Jonsson, however, is preparing a paper on the Icelandic algæ-vegetation in which he in- tends to give a further account of the different smaller groups into which the algæ-flora of Iceland is naturally divided , I shall not enter on this matter here, the more so, as this examination re- quires a very Ihorough knowledge of the algal vegetation of Ice- land. In Finmark the warm-boreal group is further reduced (to 4%) and the cold-boreal is likewise a little reduced; the subarctic group is only a little richer in species than that of South West Ice- land, but here we meet with 5^/2 °/o of arctic species. In North East Iceland, we see that the warm-boreal species do not even reach to 1^/0: the cold-boreal species are considerably reduced; the sub- arctic group amounts to almost one half (48.5 Vo) of the algæ of this area, and lastly, almost 9 *^/o of arctic species are found here. On the coasts of the Murman Sea, no warm-boreal species are found, and only 18% of cold-boreal species; almost one half (48 %) are subarctic, and Hiere are more than 15 ^/o of arctic species. Lastly, as to West Greenland, warm-boreal species are likewise totally wanting; there are only 12.5 Vo of cold-boreal species, more than one half of all the species are subarctic (53.8%), and more than 17% arctic. As will be clearly seen from the table, the Færoes and Nord- land must be called the habitat of the cold-boreal algæ, as in both piaces almost one half of the algæ belong to this group; next to these countries must be named, on one side the Shetland Isles, on the other South West Iceland. 2. Comparison with the adjacent Countries. A more thorough comparison shows us a good many diffi- culties of various kinds, especially caused by the faet, that the algal vegetations of the different countries have not all been equally thoroughly investigated. It is no wonder, therefore, that the results obtained may suffer from small and large errors. 50* 782 In order to show some of Ihe most important causes of errors, I shall point out, that several of the species, which have been found on the coasts of Greenland, the Færoes and Iceland, according to papers pubHshed during the latest years, e. g. Rosenvinge's on the marine algæ of Greenland, my own paper on the Færoese marine algæ, and lastly Jonsson's on the marine algæ of Iceland, also will be found in the surrounding countries on renewed investigations. In Batter's recently published list of the marine algæ of Great Britain we thus find many of Rosenvinge's new^ Greenlandic species mentioned. Another source of error is naturally the different authors' varying opinions of species; and even if we try ever so conscientiously to make out the different synonyms, there are still, according to our present know- ledge, a great many genera — I need only mention promiscuously Lithothamnion, Sphacelaria, Mijrionema, Acrosiphonia, Spongomorpha, Cladophora, Enteromorpha, Ulothrix — of which it is almost impos- sible to make any statement with cerlainty. On the whole, it is espe- cially the classification of the Chlorophyceæ which causes the greatest difficulties ; and as the piants of this group thrive in very extreme conditions, which quality they have in common with the Cyano- phijceæ, for instance they are well fitted for standing a mixture with fresh water, it would perhaps be most correct to side with Kuckuck who says (56, p. 10): »Bei pflanzengeographischstati- stischen Zusammenstellungen scheinen mir daher die Chlorophy- ceen und Cyanophyceen eher geeignet, das Resultat zu triiben als zu klåren, und aus diesem Grunde will ich auch hier von einer Beriicksichtigung jener Pflanzengruppe absehen.« But as it cannot be denied, however. that there are several species, even among the Chlorophyceæ and Cyanophyceæ, which are of no small importance from a phyto-geographical point of view, I have used them here as far as possible for the sake of comparison. Further, it is naturally of great importance in a comparison of this nature, to exclude the species which by mistake have been in- cluded previously in the flora of a country, when closer investiga- tions have shown that it is only by erroneous determination or for some other reason, that they have been included, and thus do not belong to the flora at all. As to Greenland, so man}' species had been incorrectly admitted, according to Rosen vinge (71, p. 154) that the phyto-geographical character of the algæ -flora has been essentially altered by their omission; also, with regard to the Færoes we have been obliged to exclude several of the species 783 previously admitted in botanical works. But in spite of all this, there are still some few species of which it is doubtful, whether they have any right to belong to the Færoese algæ- flora ^ Of the 215 species (compare p. 773) found at the Færoes, 6 spe- cies besides a few new varieties and forms have been described as new; there are moreover an Ectocarpus spec. and a Hypheothrix spec, of which no statement could be given with certainty because of defective material, and which therefore by closer examination may possibly be referred to species already known, and lastly, Lyng- by e's Palmella adnata which I think may be referred to the genus Pleiirococciis , as I have examined Lyngby e's old material, but cannot determine any closer. In the comparison I leave these 3 species out of consideration^. As to the 6 new species, it is hardly probable, however, that they should be endemic at the Færoes; the 5 species: Dermocarpa Farlowii, Hijella endophytica, Myrionema færoense, and M. speciosum together with Phæostroma parasiticum are small forms, that are easily overlooked ; one of them, Myrionema færoense has besides al- ready been found at Iceland, and according to what Dr. Borne t reports (cf. my flora, 8, page 523) Dermocarpa Farlowii is found in Japan and must therefore be supposed to be widely spread. Lami- naria færoensis, on the other hånd, is a tall plant, quite the tallest found on the Færoese coast. On one side it is connected with the Greenland — North-American Laminaria longicruris, and on the other side closely with Laminaria saccharina; it has already been found at Iceland, but it is rather doubtful if it is to be found on other European coasts, as it has not already been found there; still it is of course not impossible. The bottoms of the fjords in northern Norway would most naturally be the piaces to look * Of such species I need only mention: Phijllophora membranifoUa. of which I have only seen a small fragment which had been found by Ostenfeld in a plaice-net on board the cruiser »Ingolf« in Trangisvaagfjord; as the ship had, how- ever, just come from the Sound, and as there had most likely been some fishing there, it is very probable, that this fragment had been left in the net and thus carried on. Another species which only doubtfully belongs to the Færoes is Chætopteris plumosa; in Rostrups herbarium we find good specimens sent to him by Mr. Randropp of Thorshavn, but as this alga was never found by other investigators, we have a right to doubt wiiether these specimens really belong to the Færoes. * The Chilionema spec, mentioned in my flora (8, page 427), must, I think, as mentioned above, be referred to Chilionema reptans. 784 for it. During my short visit at the Shetland Isles I sought it in vain, still it might possibly grow there. From these introductorv remarks I now pass to the proper comparison of the marine algæ flora of the Færoes with that of the adjacent countries. In order to make such a comparison, a list of the algæ -flora of the particular countries, as correct as possible and critically revised, is of course indispensable. Helgi Jonsson I and have prepared a list (12), which we hope to be service- able, and I have used it as a basis for the following compari- son, which therefore appears somewhat altered from that of the Danish issue. We may first consider the countries which are nearest to the Færoes, that is Scotland with the Ork nev s and the Shetland Isles. In Batters's recently published lists (4) of the British ma- rine algæ, where these countries are treated as a whole, about 430 species are stated as belonging to Great Britain. Of these, 163 spe- cies are also found at the Færoes. The total number will certainly be considerably increased by renewed investigations, a few species which I found during my short visit at the Shetland Isles, have already been added, namely: Fuciis inflatiis, Sphacelaria fiircigera, Prasiola crispa subsp. /na/'ina, Ulvella fiicicola and Derhesia marina^; to these must be added Sphacelaria britannica which is stated by Sau- vageau as found in Scotland; the total number is thus increased to 169 or 79 Vo of the marine algæ of the Færoes. Of the re- maining species we may certainly suppose, that the following spe- cies also occur in Scotland with the Orkneys and the Shetland Is- les: Chantransia efflorescens, found in South England; Chantransia Alariæ found, besides at the Færoes, also at North-America, at Ice- land, on the coast of Norway at Haugesund and in 1904 by myself at Christianssund, and in the same year by J. Adams at Port- rush in North Ireland (Journal of Botany 1904, p. 351), and which probably has the same distribution as Alaria esculenta; Rhodo- chorton seiriolanum which has been found in West-England; Calli- thamnion scopiilorum which according to Batters' list has not hitherto been found at the British Isles, but which must be sup- posed to be there, if Agardh's determination of its occurrence in the Mediterranean is correct; Rhododermis elegans, a variety of ^ Batters mentions D. teniiissima as occurring in Scotland; judging from a specimen kindlj' sent me bj' Mr. Batters, I think that it most likely is Der- bcsia marina. 785 which, \. pohjstromatica, has been found in North England, and the closely related species Rh. parasitica also in Scotland; Phymatoli- thon compactnm f. circiimscripta (Stromf.) Fosl. is according to Bat- ters (4, p. 97) found in West Ireland, and must therefore also be supposed to be found for instance at Shetland; Ectocarpiis dasycar- pus, found in South England; Ectocarpiis lucifiigiis, a few years ago mentioned from Heligoland, but which must be supposed to be more widely destributed as it has now been found at the Færoes; Ectocarpiis Stilophoræ found in South England; Myrionema foecunda which is perhaps closely related to M. Corunnæ; Microsyphar Polysiphoniæ found in North England; Microsyphar Zosteræ, originally mentioned from Heligoland; Sorapion Kjellmani, most likely the same as Sorapion sinmlans (see Rose uvinge, 71, p. 161, note) which has been found in South England; Petroderma maciiliforme stated from Heligoland; Sphacelaria cæspitiila which according to Batters has been found in North England, but is, however, possibly doubtful, as it has not been given by Sauvageau; Prasiola fiirfiiracea, found for instance in Germany and on the north coast of Norway; Ulothrix pseudoflacca and U. consociata which have but recently been described by Wil le as found in Kristianiafjord and of which we therefore know but little as to how far they are distributed; Pilinia maritima which is known from Greenland and the Færoes and has lately been found by Sauvegeau in the gulf of Gascony; Ulvella con fluens found in South England but which Batters supposes to be common; Codioliim gregariiim, in Batters' list only stated from South Eng- land, but which is mentioned as »not uncommon« in the same author's »List of the marine Algæ of Berwick-on-Tweed« ; Codioliim pusilliim which is possibly found in Scotland (see Batters p. 9), but which is, however, not stated from Scotland in Jonsson's and my own list, as we have felt doubtful as to the correctness of the determination because of the synonyms mentioned by Batters; Chlorogloea tuberciilosa , lately described by Wil le and probably widely distributed; Pleiirocapsa amethystea found at South England; Phormidium aiitiimnale found in North England; Microcoleiis tener- rimus found in South England. All these 26 species which are mostly small and therefore only to be discovered by thorough investigation , can with some cer- tainty be supposed to occur on the coasts of Scotland and the sur- rounding isles, and can therefore for comparison be included in the total number. The total number of species thus becomes 195. 786 Of the remainding 20 species, 6 are the above mentioned new species and 3 are forms only determined as to the genus, and of which we know nothing concerning their eventual occurrence on the coasts of Scotland. It is, however, most probable, that at any rate some of them will be found to grow there. Of the remainding 11 species, 5 belong to Acrosiphonia and Spongomorpha which should be left out of consideration because of the uncertain classification of these genera; most Hkely they will all be found, however, on the coasts of the British Isles, with the exception perhaps of A. hystrix, a northern species uncommon even in the Færoes and not found there either in its typical form. We thus have a remainder of 6 spe- cies: Rhodophyllis dichotoma, Halosaccion ramentaceum, Ptilota pecti- nata, Rhodochorion peniciUi forme, Lithothamnion læve and Alaria Pylaii. These 6 species are essentially typical to northern regions; they all belong to the coldest portion of the subarctic group and thus are species having their greatest distribution in the Polar Sea proper. Whether any of these species may be found at the British Isles is certainly very doubtful, but even if they do not grow there and some few of the above mentioned Færoese algæ, which have not yet been found in Scotland, should really be wanting there, it is, however, but a very small number of species found at the Færoes which are not found in Scotland or at any rate may not be supposed to grow there. Any further examination dealing with the great number of species found in Scotland and the surrounding isles, but not at the Færoes, has but little interest liere. It is a matter of course, that a much greater number of species must be found in a terri- tory of such an extent, than at the Færoes, and it is likewise evident that a great number of more southern forms will appear there on account of the more southern situation of Scotland. Of the species (about 250) found there, but not at the Færoes, more than one half have hitherto only been found in the southern part of Scotland, comprising the east coast about the Firth of Forth (Fife, Haddington, Berwick) and the west coast reckoned from Argyle to the English boundary; of red and brown algæ there are no less than 68 species (see Borgesen and Jonsson 12, p. XXI, group E 3). Of the remainding species (as to the red and brown algæ compare the group E 2, in the above -mentioned paper (12) p. XIX and XX) we may be sure that a great many do not grow so far north as the Shetland Isles. 787 As the Shetland I si es ave of about the same size as the Fær- oes and lie nearest to Ihe latter (the distance is almost 300 kilo- meters), a comparison between the marine algæ- flora of these two groups of islands would be particularly interesting. It is therefore to be regretted, that the marine algæ-vcgetation of the Shetland Isles has been but little investigated. The most important contribution to our knowledge of these algæ is to be found in Edmonston's Flora of the Shetland Isles (16) with a list of the marine algæ known at that period. Since then no other contribution had been published, so far as I know, until 1902 when I paid a short visit to the islands. In a small paper (9) I have published my discoveries and added the species mentioned by Edmonston. The number of species known from the Shetland Isles is about 108, as I include Enteromorpha cla- thrata and Callithamnion tetragoniim var. brachiatum, whilst Gelidium cartilagineiim must be left out, as it does not belong to the flora of the islands. Of these 108 species, about 88 are also found at the Færoes; of Rhodophijceæ and Phæophyceæ 87 have hitherto been known, of which 72 are common to both countries. But as already emphasized by me in dealing with the flora of Scot- land and adjacent islands, there is hardly any doubt, that by far the greater part of the algæ-species of the Færoes will be found there, especially on the islands that are the nearest. The only spe- cies which have less likelihood of being found there are the above mentioned 6 subarctic species, to which perhaps may be added a few more, e. g. Laminaria færoensis, Phæostroma parasiticiim etc. Of the 20 species found at the Shetland Isles but not at the Færoes we must first mention Fiicus serratiis which is very common at the Shet- land Isles but is, strangely enough, wanting at the Færoes; the other 19 are: Helminthora diuaricata , Chylocladia kaliformis, Nito- phyllum Bonnemaisonii, Spermothamnion Tiirneri, Callithamnion te- tragoniim, Ceramiiim diaphaniim, C. ciliatum, Dilsea ediilis, Aspero- coccus biillosiis, Myriotrichia clavæformis, Mesogloia vermiculata, Sac- corhiza biilbosa, Cutleria miiltifida, Fiiciis ceranoides, Acinetospora piisilla, Chætomorpha ærea, Cladophora Hutchinsiæ, Dermocarpa prasinu and Calothrix confervicola. It is not impossible, that some of these species may grow on the Færoese coasts, but most of them are forms from more southern countries, most likely having their northern limit at the Shetland Isles. It appears from this, that the Færoese algæ- flora must be regarded as a rather poor selection of the algæ of Scotland and adjacent is- 788 lands, as almost all Ihe F'æroese algæ are found on the coasts of Scotland, whereas Scotland has on the other hånd a very great num- ber of species, which are wanting at the Færoes. The greatest resem- blance is between the Færoese algæ-flora and that of the Shet- land Isles; the Færoes have a few, perhaps no more than 6, spe- cially norlhern species which will most probably not be found on the Shelland Isles; on the other band the Shetland Isles have a number (how many can only be staled when a more thorough in- vestigation of the algal vegetation of these islands has been under- taken) of forms from more southern countries which do not grow any farther north than the Shetland Isles. From the British Isles the comparison most naturally passes to the west coast of Norway. Boye (6) has given a list of the algæ found by him at a part of the coast, round Sulen north of Bergen. This part of the coast is at about the same degree of la- titude as the Færoes, and consequently of special interest; but un- fortunately Boye's list is rather defective, as is clearly proved by the faet that such species as Fucus spiralis, Desmarestia viridis, Odonthalia dentata, Enteromorpha Linza and others, are not men- tioned. I have therefore made an attempt to supplement Boye's list, partly by including the species mentioned by Areschoug in his well-known work: »Phyceæ scandinavicæ marinæ« (3), even if these algæ are not specifically stated as belonging to Bergen and its neighbourhood but only from some locality on the Norwegian west coast south of Trondhjem, partly by adding the species which Hansteen mentions in his paper (38). Further, several species from the west coast of Norway are mentioned in different papers by Foslie (e. g. 25, 27); and Mr. E. Norum of Haugesund, who has been investigating the algæ-vegetation in this neighbourhood, has kindly placed his still unpublished list of brown algæ from this part of the west coast at my disposal. Lastly, Chantransia Alariæ has been found by Rosen vinge at Haugesund and by me at Chri- stianssund. Of the species mentioned in Boye's list (according to my calculation about 165, wdiich number will certainly be much increased on further investigation), 97 are found at the Færoes, but if we include the species, which are stated by the various authors mentioned as found on the west coast of Norway, the total num- ber of common species is increased by at least 40, which makes in all about 137 species or about 64% of the Færoese algæ. If we ex- clude the green and bluish-green algæ, 190 Rhodophyceæ and Phæo- 789 phyceæ have been found in West Norway, according to our list; of these species 114 or more than 73 ^/o are found at the F'æroes. There is, however, hardly any doubt, that the greater part of the many Færoese algæ which have hitherto not been found on this part of the coast, that is about 80 species, will certainly be found there. It seems unnecessary to name all these species, but I may mention some Færoese species which may be supposed not to occur on this part of the west coast of Norway: Halosaccion ra- mentaceum, Ptilota pectinata, Rhodochorton penicilliforme , Alaria Pylaii, Laminaria færoensis, and possibly some few more. These species all belong to colder regions. Of the great number of spe- cies found at Norway and not at the Færoes (of red and brown algæ 76 species^) the greater part are of a more southerly distri- bution, but some few are more northern algæ, for instance: Tur- nerella Pennyi, which is arctic, and Litbothamnion tophiforme and Haplospora globosa which are subarctic. From West Norway we pass on to Nordland. On account of the Gulf Stream the algæ -flora here has a much more southern character than might be expected so far north ; a great many Færoese species are also found here. For our knowledge of Nordland's algæ-flora we are especially ^ According to our list they are divided in the following way into different groups : The arctic group: Turnerella Pennyi. The subarctic group: Litbothamnion tophiforme, Haplospora globosa. The cold-boreal group: Spcrmotbamnion Tiirneri, Callitbamnion Hookeri, C. roseiim. Ceraminm Dcslongcbampsii, C. circinnatum, C. diaphaniim, Dilsea ednlis, Litbothamnion intermedium , L. fornicatum, L. norvegiciim, Ectocarpns Tnrnerellæ, E. Pringsbeimii, E. terminalis, E. penicillatus, E. draparnaldioides, Myriotrichia fili- formis, Dictyosiphon Chordaria, Mesogloia vermicnlata, Spermatochniis parado.viis, Fiicus ceranoides and Fucus serratus. The warm-boreal group: Nemalion mnltifidum, Gelidium crinale, G. lati- folium, Pbyllopbora rubens, Catenella opuntia, Rbodophyllis bifida, Chylocladia kaliformis, Nitophyllum punctatum , Bonnemaisonia asparagoides , Polysipbonia spinulosa, P. simnlans, Brongniartella byssoides, Monospora pedicellata, Pleonosporinm Borreri, Callitbamnion tetragoniim, Compsotbamnion tbuyoidcs, Ceraminm flabelli- ferum, Gloiosipbonia capillaris, Halarachnion lignlatnm, Litbotbamnion membrana- ceum, L. Sonderi, L. calcareum, Litbopbylhim orbicnlatum, L. piistulatum, Melobesia farinosa, M. Lejolisii, M. minntula, Corallina rubens, Ralfsia pusilla, Myrionema intermedium, Chilionema ocellatum, Ascocyclus orbicularis, Microspongium gelati- nosum, Ectocarpus spbæricus, E. globifer, Myriotrichia repens, M. claoæformis, My- riactis Haydeni, Elacbista stellaris, o'iraudia sphacelarioides, Sphacelaria bipinnata, Asperococcus bullosus, A. compressus, Litosiphon pusillus, Striaria attennata, Myrio- cladia Zosteræ , Chordaria divaricata, Stilophora rbizodes, Saccorbiza bulbosa, Cutleria multifida, Tilopterls Mertensii, Dictyota dichotoma. 790 indebted to Kl een (51); but Kjellman (48) has fr.rnished ns with important information. We know 123 species of Rhodophijceæ and Phæophyceæ together with about 20 Chlorophyceæ from Nordland, that is very nearly 150 species. 113 of these species are also found at the Færoes, that is, 75 % of the species of Nordland and 52. 8% of the Færoese species. Nordland has 123 red and brown algæ, 95 of these are also found at the Færoes, that is, 61 °/o of the Færoese algæ of these groups. The 28 species which are found at Nordland but not at the Færoes are the following: Tiirnerella Pennyi{A)^, Brongniartella byssoides {E i), Spermothamnion Turneri{D), Callithamnion Hookeri (D), Callithamnion roseiim (D), Ceramiiim Des- longchampsii (D), C. circinimtum (D), C. diaphaniim (D), Dilsea ediilis (D), Petrocelis criienta (D), Lithothamnion intermedium (D), L. forni- catiim (D), L. tophiforme (B i), L. norvegiciim (D), Ralfsia deiista (B i), Edocarpus Turnerellæ (D), E. terminalis (D), E. horealis (D), E. ovatus (Bl), E. penicillatus (D), E. draparnaldioides (D), Myriotrichia filiformis (D), Sphacelaria racemosa (Bi), Mesogloia vermiciilaris (D), Sperma- tochnas paradoxus (D), Cutleria multifida (E i), Fuciis ceranoides (D), F. serratus (D). With the exception of some few warm-boreal spe- cies (El), the greater part of these algæ belong to the cold-boreal group (D); we are therefore justified in expecting that some of them may perhaps be found at the Færoes; the other species are arctic or subarctic, and can scarcely be expected there. On the other band, the following 60 species are found at the Færoes: Porphyra lencosticta, P. coccinea, Conchocelis rosea, Chan- transia Alariæ, Ch. efflorescens, Ch. virgatula. Choreocolax Polysipho- niæ, Harveyella mirabilis, Phyllophora Brodiæi, Actinococcus snbcu- taneus, Callophyllis laciniata, Callocolax neglecta, Sterrocolax decipiens, Lomentaria rosea, Nitophyllum laceratiim, Laurencia pinnatifida, Po- lysiphonia atrorubescens, Rhodomela subfusca, Griffithsia setacea, Cal- lithamnion scopulorum, C. granulatum, Rhodochorton seiriolanum, Rh. membranaceum, Rhododermis elegans, Phymatolithon lævigatum, Li- thophylhim incruslans, L. hapalidioides , Lithoderma fatiscens, Petro- derma maculiforme, Sorapion Kjellmani, Ralfsia clavata, Myrionema Corunnæ, M. foecundum, M. glohosnm, M. færoense, M. speciosum, Mi- crosyphar Polysiphoniæ , M. Zosteræ, Edocarpus æddioides, E. Stilo- phoræ, E. velutinus, E. ludfugus, E. tomentosoides, E. dasycarpus, E. granulosus, E. Hincksiæ, Leptonema fasciculatum, Eladiista scutulata, Sphacelaria britannica, Sph. cæspitula, Sph. furcigera, Sph. cirrosa, ^ For explanation of the letters see: Borgesen and Jonsson (12, p. XV — XXI). 791 Desniotrichiim iindnlntum , Punctaria laiifolia, Litosiphon filiforme, Phæostromd parasiticiim, Phiillitis zosterifolio , Desmarestia ligulata, Chorda tomentosa, Laminaria færoensis. It is most probable, lliat a great many of these species will be foiind in Nordland on fur- ther investigalion, the lemainding species are partly some that grow in more southern regions and most probably do not occur in Nordland, partly the species hitherlo found at the Færoes only, of whose occurrence oulside the Færoes we know nothing as yet. Of green and bluish-green species 40 have been found at the Fær- oes but not hitherto at Nordland, namely: Chlorochijtrinm inclii- siim, Codiolum gregarium, C. pusillum, Enteromorpha Linza, Pra- siola crispa '■'''marina, P. furfuracea, Ulothrix consociata, U. pseudo- flacca, U. flacca, Endoderma Wittrockii, Acrochaete repens, Pilinia maritima, Ulvella fucicola, U. confluens, Pringsheimia scutata, Uro- spora Wormskioldii , Spongomorpha vernalis, Acrosiphonia flagellata, A. incurva, A. grandis, A. hystrix^, Cladophora sericea, Gomontia po- lyrhiza, Ostreobium Queketti, Derbesia marina, Vaucheria coronata, Valonia ovalis, Chlorogloea tuberculosa, Dermocarpa violacea, D. Far- loivii, Pleurocapsa amethijstea, Hyella cæspitosa, H. endophytica, Spi- rnlina subsalsa, Phormidium autumnale, Lyngbya lutea, Microcoleus tenerrimus, Calothrix ærnginea, C. scopulorum, Rivularia atra. Most probably a great number of these species will likewise be found in Nordland. What we know of the algæ-flora of Finmark is mostly due to papers by Foslie; Kjell man, however, has also contributed important information. According to our list, 125 red and brown algæ have been found there. Of these algæ 50 red and 41 brown algæ are common to the Færoes and Finmark, that is 91 in all or 58 °/o of the Færoese algæ and 72 % of those of Finmark. The 34 species which are found at Finmark but not at the Færoes are as follows: Chantransia microscopica, Turnerella Pennyi, Polysi- phonia arctica, Spermothamnion Turneri, Ceramium Deslongchampsii, C. cincinnatum, Petrocelis Middendorffi, Peyssonnelia Rosenvingii, Phy- matolithon inuestiens, Lithothamnion flavescens, L. foecundum, L. in- termedium, L. fornicatum, L. tophiforme, L. norvegicum , Lithoderma lignicola, Ralfsia deusta, Ectocarpus terminalis, E. nanus, E. ovatus, E. penicillatus , Myriotrichia fdiformis, Sphacelaria racemosa, Phæo- ' Kl een, however, mentions both a Cladophora arcta and a Cl. uncialis, which are most probably identical with some of the species of Acrosiphonia and Spongomorpha mentioned here. 792 saccion CoUinsii, Delamarea attenuata, Coiledesme bulligera, Dictij- osiphon Chordaria, D. corymbosus , D. hispidiis, Saccorhiza derma- todea, Laminaria Agardhii, L. nigripes, Haplospora globosa, Fiicus serratus. Almost half of these species are arctic and subarctic, aild mav not be expected to grow so far south as the Færoes; the others, however, are of a more southern distribution , and some of them may therefore also be found possibly at the Færoes. The majority of the 65 Færoese species absent at Finmark are the same as were wanting at Nordland; some Færoese species not found at Nordland are, however, found here, but on the other hånd Fin- mark lacks the following 19 species: Erythrotrichia ceramicola, Phyllo- phora membranifolia , Pterosiphonia parasitica, Callithamnion arbu- scula, C. polyspermiim, C. corymbosum, Ceramium acanthonotum, Lomentaria articulata, Plocamiiim coccineum, Polysiphonia violacea, P. Brodiæi, Cruoria pellita, Criioriella Dubyi, Rhododermis elegans, Ralfsia vernicosa, Ectocarpus tomentosiis , Cladostephiis spongiosus, Litosiphon Laminariæ, Himanthalia lorea. As already pointed out with reference to Nordland, there is hardly any doubt, that some of the wanting Færoese species may also be found at Finmark though probably in lesser number. As to the green and bluish-green algæ, 37 Færoese species (I take it for granted, that Acrosiphonia hamulosa belongs to A. al- bescens) are also found at Finmark; some more species have been found here, but as the determinations in many cases greatly need to be revised, I shall not deal with them any more in this con- nection. From this comparison it is seen, that Finmark has a somewhat smaller number of species in common with the Færoes than Nord- land has, and that it has a much greater number of arctic species than the latter country. From Norway we pass on to Iceland, and here our comparison has the advantage of the up to date list worked out with much care by Jonsson, in his paper on the marine algæ-flora of this country. As already pointed out by Stromfelt, the algæ-flora of the coasts of Iceland varies greatly. He divides it into two territories: North- East Iceland with an algæ-flora of an essentially arctic character, and South -West Iceland with a north-åtlantic flora. This great difference has been confirmed by Jonsson, who, however, has kindly communicated to me, that the flora ought to be further sub- divided, as the flora of the southern part of the country is purely 793 cold-boreal, that of the eastern part arctic, whilst that of the western and northern parts is a mixed flora. As Jonsson intends to enter more fully into a comparison with the adjacent countries, in his paper on the algæ-vegetation of Iceland using this division as basis I may avoid too much detail here and simply keep to tiie two principal divisions: North-East and South -West Iceland. The total number of species mentioned by Jonsson from Ice- land is 195^, Of these 195 species, 121 Rhodophyceæ and Phæo- phyceæ and 44 Chlorophyceæ and Cyanophyceæ, in all 165, are found on the southern and south-western part of the country, that is, from LonsheiSi on the south-east coast to Låtrabjarg on the north-west coast. From Låtrabjarg to Hornbjarg on the adjacent coast of North-West Iceland with mixed flora, we further find the following 7 species: Ceratocolaæ Hartzii, Phyllophora Brodiæi, Adino- cocciis sLibciitaneiis, Piinctaria plantaginea, Monostroma groenlandicum, Åcrochæte repens and Bolhocoleon piliferiim, which have hitherto not been found in more southerly piaces in Iceland. With the latter species the total number becomes 172. 134 of these species and 5 from the adjaceilt coast, 139 in alP, are also found at the Færoes. The 76 Færoese species which are not found at South-West Iceland can be divided into the following groups : I. Species found at North-East Iceland : Laminaria færoensis, Petroderma maculiforme, Codiolum grega- riiim, C. piisilliim, Phormidinni aiihinmale. II. Species found at Greenland: Chantransia efflorescens, Harveyella mirahilis, Polysiphonia elon- gata, Rhododermis elegans, Sorapion Kjellmani, Ulolhrix conso- ciata, Pilinia maritima, Uluella con fluens, Vaucheria coronata, Calothrix scopulorum. III. Species found on the coast of North Norway: Erythrotrichia ceramicola, Chantransia Daviesii, Polysiphonia violacea, Polysiphonia Brodiæi, Rhodomela subfusca, Callitham- ' According to Mr. Foslie's latest determination, Lithothamnion circumscriptum must be excluded from the number of species, whilst on the other hånd Chantransia microscopica and Vaucheria sphærospora must be included (compare Borgesen and Jonsson, 12). ^ Of Phyllophora Brodiæi only the subspecies interrupta has been found at Iceland, but the Færoese specimen is very much like it; as to Acrosiphonia, I consider A. flaccida as belonging to A. hystrix (8. p. 512), and Spongomorpha lanosa to be the same as Sp. vernalis (12, p. XXV). 794 nion corymbosum, C. polyspermum, Furcellaria fastigiata, Polyides rotundus, Cruoriella Diibyi, Chilionema reptans, Sphacelaria cir- rosa, Cladostephiis spongiosus, Asperococciis echinatus, Halidrys siliqiiosa, Himanthalia lorea, Acrosiphonia flagellata, A. grandis, Bryopsis phnnosa, Derbesia marina; lo Ihis group we maj' most probably also refer Sterrocolax decipiens, which has certainly not been found farther norlh than the Færoes, so far as I know, but which may be supposed to grow in the same re- gions as Ahnfeltia. VI. Southern forms, not found hitherto north of the Færoes and West Norway: Porphyra leucosticta, Callophyllis laciniata, Callocolaæ neglecta, Lomentaria articulata, NitophyUum laceratum, Laurencia pinna- tifida, Polysiphonia atrorubescens, Griffithsia setacea, Callitham- nion grannlatum, Rhodochorton seiriolaniim, Phymatolithon læ- vigatnm, Lithophyllum incrustans, L, hapalidioides, Ectocarpus dasycarpus, E. grannlosns, E. hicifugus, E. velutinns, Myrionema foecundum, Microsyphar Zosteræ, Sphacelaria cæspitiila, S. furci- gera, Elachista scutnlata, Punctaria latifoHa, Desmotrichnm un- dulatum, Litosiphon Laminariæ, Valonia ovalis, Chiorogloea tuberculosa, Dermocarpa violacea, Hyella cæspitosa, Lyngbya hitea , Microcoleus tenerrimus, Calothrix ærnginea and Choreo- colax Polysiphoniæ. Lastly, we have the 6 new Færoese species, and the 3 which are only determined as to the genus. Of these, Myrionema færoense has been found at South -West Iceland and is thus aheady inckided in the total number, and Laminaria færoensis has been taken at North-East Iceland and is included in the first group; thus there are only 7 species left, of which nothing^ is known as to their oc- currence outside the Færoes, but of which some at least may prob- ably also be found at Iceland. As to group I, it seems to me, that the 5 species it includes must also be supposed to occur at South -West Iceland. I likewise think it most likely, that the 10 species belonging to group II, may be met with at Iceland. As to the 21 species of group III I think there is some probability at any rate, that some of them may be found at South -West Iceland, as they have been able to grow so far north as the coast of Norway. ^ Dermocarpa Farlowii excepted. 795 Laslly there are Ihe more southern forms belonging lo group IV. With the exception of some few species, especially lliose ol" which we know very little as to their distribution, we can in general take it for granted , that tliey are most likely not found at Iceland; thus their north western boundary is at the Færoes: in Norway, however, several of them grow much farther north. As to the 33 species found at South -West Iceland, but not at the Færoes, they can be divided into the following groups; I retain the designations used in the preceding section. I. Arctic species: Ceratocolax Hartzii, Polysiphonia arctica, Criioria arctica, Monostroma groenlandiciim; 4 species in all. II. Subarctic species: Subdivision 1 : Peyssonnelia Rosenvingii, Lithothamnion tophi- phorme, Coilodesme buUigera, Ralfsia ovata, R. deusta, Myrionema Laminariæ , Saccorhiza dermatodea, Acrochæte parasitica, Uro- spora Hartzii and Chiorochytrium Schmitzii; 10 species in all. Subdivision 2: Phaeostroma piistulosum, Chiorochytrium derma- tocolax, C. Cohnii and Cladophora hirta; 4 species in all. III. Cold-boreal species: Dilsea edulis, Rhodochorton miniitum, Petrocelis Hennedyi, Rhododermis parasitica, Ectocarpus penicillatiis, Dictyosiphon Chordaria, Fiiciis serratus, Sphacelaria radicans, S. olivacea, Codiohim Petrocelidis , Cladophora glaacescens and Vaiicheria sphærospora; 12 species in all. IV. Wa r m - b o r e a 1 species: Bonnemaisonia asparagoides ; 1 species. Lastly we have the 2 new species: Rhodochorton repens and Acrosiphonia flabelliformis , the distribution of which is unknown outside of Iceland. Firstly as to Bonnemaisonia asparagoides, it must be pointed out, that it is not perfectly certain that this species really grows on the coasts of Iceland (compare Jonsson, 41 p. 141). Of the species mentioned in group III it is probable , that some of them at any rate, especially the smaller species, will also be found to grow at the Færoese coasts. With the exception of some few species, there is, however, only a slight possibility that some of the species belonging to group II should be found on the Færoese coasts, and as to the 4 arctic species there is certainly no doubt that they are absent at the Færoes. Thus we see that even if there is an important, though not very great, difTerence between the flora of South -West Iceland and Botany of the Færoes. 51 796 that of the Færoes, in so far that 4 species belonging to the arciic group, which is quite wanting at the Færoes, have been foiind at Iceland, as also that many more subarctic species occur there than at the Færoes, the difference is much reduced, if we make the comparison between the algæ-flora of South Iceland proper and that of the Færoes. At South Iceland 82 red and brown species and 26 green and bluish-green have been found, 108 species in all. The following 12 of these species are wanting al the Færoes: Subarctic species: Lithothamnion topbiphorme, Ralfsia deiista, Urospor a Hartzii, Cladophora hirta; Cold-boreal species: Rhodochorton repens (endem.), Petrocelis Hennedyi, Rhododermis parasitica, Sphacelaria radicans, S. olivacea, Fiicus serratiis, Acrosiphonia flabelliformis (endem.); Warm-boreal species: Bounemaisonia asparagoides, which does not belong perhaps to the flora. As mentioned above, there is hardly any doubt, that at any rate some of these species will also be found at the Færoes, just as, on the other hånd, several of the species wanting at the Færoes may most likely be found at this part of Iceland, especially perhaps at the Vestmanoer. Some of the latter however can hardly be con- sidered to occur there, as for instance some of the most important: Porphyra leiicosticta, CallophyUis laciniata with Callocolax neglectiis, Lomentaria articiilata, NitophyUiim laceratiim, Polysiphonia violacea and P. Brodiæi, Griffithsia setacea, Cladosteplms spongiosiis, Punctaria latifolia, Himanthalia lorea, Halidrys siliqiiosa, Bryopsis plumosa, Valonia ovalis etc. The algæ-flora of South-Iceland must be regarded as a poorer selection of the Færoese algæ-flora, particularly as some of the warm-boreal species, found at the Færoes, are wanting there. When we turn to North-East Iceland, we see that the dif- ference between the algæ-flora there and that at the Færoes is much greater, as might well be expected. From North- and East-Iceland proper, that is, if we count from the head of Huna-Floi to Lons- heiSi on the south-east coast of Iceland, we have 92 red and brown species and 34 green and bluish-green; east of Hornbjarg along the adjacent north-western coast, we have moreover 8 brown and red algæ, and 9 green and bluish-green species, namely: Chantransia microscopica , Cystoclonium purpiirascens , Polysiphonia nigrescens, ■^'Ralfsia ovata, Ascocyclus islandicus, ^Leptonema fasciculatum, "Spha- celaria hritannica, Leathesia difformis, '''Chlorochytrium dermatocolax, Codiohim pusillum, Enteromorpha aureola, "Ulva Lactuca, "Ochlo- 797 chæte ferox, Cladophora nipestris, '-'C. sericea, Pledonema norvegiciim and "Spiriilina subsalsa. If we add these to the above-mentioned number\ ^Ye have 143 species. Of Ihese, 111 are also found at the Færoes. Thiis North-East Iceland has 23 species less in com- mon with the Færoes than South -West Iceland. Firstly, as to the 32 species found at North- East Iceland and not at the Færoes, they may be divided into the following groups: I. The A ret i c gro up: Turner ella Pennyi, Delesseria Baerii, Po- hjsiphonia arctica, Lithothamnion flavescens, L. foeciindiim, Om- phalophylliim ulvaceiim, Dictyosiphon corymbosiim, Laminaria nigripes and Monostroma groenlandiciim; 9 species in all. II. The Su baret i c gro up: Subdivision 1: Peyssonnelia Rosenvingii, Lithothamnion tophi- forme, Ralfsia densta, R. ovata, Coiledesme hulligera, Saccorhiza derniatodea, Ulothrix snbflaccida , Pseiidendocloninm marinum, Urospora Hartzii, Acrosiphonia penicilliformis ; 10 species in all. Subdivision 2: Phæostroma pustalosnm, Chlorochytrinm derma- tocolax; 2 species in all. III. The Boreal-arctic group: Chantransia microscopica. IV. The Cold-boreal group: Petrocelis Hennedyi, Lilhothamnion norvegicnm, Ascocyclns islandicns, Ectocarpus penicillatus, Spha- celaria radicans, Dictyosiphon Mesogloia, D. Chordaria, Entero- morpha aureola, Ochlochæte ferox, Pledonema norvegicnm; 10 species in all. If we compare these species with those that are not common to South -West Iceland and the Færoes, we at once perceive a very distinct difference in that the arctic group is much increased^. It is not likely that any of these 9 species will be found at the Færoes: and as to the subarctic species of the subdivision 1, it is likewise improbable that they will be found there, a few species perhaps excepted. On the other hånd the species of subdivision 2 may probably also be found on the Færoese coasts, and this is cer- tainly also the case with the greater part of the cold-boreal species. ^ It is, however, doubtful, whether they may all be found at North and East Iceland proper; the 8 species marked with asterics and which have all been found much farther north, for instance at Greenland, must, however, be supposed to occur at North Iceland, and especially at East Iceland. ' This group will most probably be still more increased on renewed inve- stigations, as we may certainly suppose, that the two arctic species at an3' rate, Cruoria arctica and Ceratocolax Hartzii, found at the other side of Hornbjarg, also grow there. 51* 798 The 104 Færoese species which have not been found at North- East Iceland are divided into the foUowing groups: The sul)arctic group: Chantransia efftorescens, Ch. uirgatiila, Hfirneijella mirahilis, Sordpion Kjellmani, Perciiisaria perciirsa, Pilinia marilima. Vivella con fluens, Pringsheimia scutata, Spongomorpha ver- nalia. All these 9 species will most likely be lound at North-East Iceland. The boreal-arctic group: Polysiphonia elongata and Rivularia atra, both of which will certainly also be found at North- East Iceland. The cold-boreal group includes the great majority, namely, 66 species, but I need not mention themhere; some of them may perhaps be found at North East Iceland, but most of them pro- bably not. Lastly, the warm-boreal group includes 27 species, none of which probably are to be found on the coasts of North-East Iceland. If we now pass on to Greenland, we find, according to Ro- senvinge (71, p. 166 — 172), that the total number of species is 167, when the algæ-flora of both West and East Greenland is included. To this number must be added Actinococcus snbculanens which Rose uvinge did not accept as a species, according to Darbi- shire's view at that time. According to Jonsson (42), we must further add 7 species, firstly Microsyphar Polysiphoniæ; tlien Ro- se n vin ge's Spongomorpha arcta includes Acrosiphonia inciirua, A. hystriæ and A. penicilli formis; also 3 species of Ulothrix must be added : U. psendoflacca, U. scutata and U. subflaccida, whilst U. consociata is partly the same as Rosen vin ge's U. implexa. Lastly, Rosenvinge's Sphacelaria olivacea includes Sphacelaria radicans and Sph. britannica. Thus the number of species we know from Greenland becomes 176. According to my judgment of the species, Enteromorpha prolifera must, however, be withdrawn, and accor- ding to Foslie's latest views the 12 species of Lithothamnion men- tioned by Rosen vinge must be reduced to 6^; the total number of species thus becomes 169. At the Færoes we meet with 104 of these species. The Færoes have 108 species (-+- 3 only determined as to the genus) that are not found at Greenland, but on the other hånd, we find 65 species there that do not grow at the Færoes; thus 49*^/0 of the 212 Færoese algæ are common to both countries. ^ Lithothamnion botrijoides, L. ftahellatum, L. colliciilosiini and L. varians bejng rcferred to L. glaciale, and L. circuniscriptum to L. compactnni. L. innestiens is doubtful at Greenland and L. fri.ticiilosiim is the same as L. intermedium. 799 As to the 111 Færoese species wanting at Greenland, they are mostly species also wanting at Iceland (namely the 54 species men- tioned as belonging to group III and group IV p. 793 — 94), and in addition the following 57 species: Porphijra coccinen, Chantransia Alariæ, Chondrns crispiis, Gigartina mamillosa, PhijUophora mcmbra- nifolia, Ahnfeltia plicata, Cystocloniiim purpur ascens, Lomentaria rosea, L. ckwellosa, Plocamiiim coccineiim, Delesseria alata, D. sangiiinea, Polysiphonia fastigiata, P. nigrescens, Pterosiphonia parasitica, Odon- thalia dentata, Callithamnion arbnscula, C. scopulorum, Plumaria ele- gans, Ptilota phimosa, Ceramiiim acanthonotiim, Diimontia fdiformis, Cnioria pellita, Phijmatolithon polijmorpimm, Lithothamnion Lenor- mandi, Lithophyllnm Croiiani, L. macrocarpiim, Corallina officinalis, Petroderma maculiforine, Mijrionema vulgare, M. Corunnæ, M. færoense, M. speciosum, Ectocarpiis tomentosns , E. fasciculatiis , E. Hincksiæ, Phæostroma parasiticum, Dictyosiphon Ekmani, Desmarestia ligulata, Leathesia difformis, Laminaria færoensis, L. hyperborea, Fiicus spiralis, Peluetia canalicidata, Codiohim pnsilhim, Enteromorpha Linza, E. cla- ihrata, Prasiola fiirfiiracea, P. stipitata, Acrochæte repens, A. albescens, Dermocarpa Farlowii, Hyella endophytica, Phormidiiim antnmnale, besides the 3 referred only to the genus. These 111 species are almost all of a more southern distribution and belong to the warm- boreal and cold-boreal groups. As to the 65 species found at Greenland and not at the Færoes, they can be referred to the following groups: The arctic group: Ceratocolax Hartzii, Callymenia sanguinea, TurnereUa Pennyi, Delesseria Baerii, D. Montagnei, Polysiphonia arctica, Dilsea integra, Petrocelis polygyna, Cruoria arctica, Lithothamnion foecundum, Ectocarpus pycnocarpus, Omphcdophyllum ulvaceum, Kjell- mania subcontinua, Coelocladia arctica, Dictyosiphon corymbosum, Myriocladia callitricha, Laminaria solidungula, L. cuneifolia, L. groen- landica, L. nigripes, Monostroma groenlandicum, M. leptodermum, Ulo- thrix scutata, Arthrochæte penetrans, Chætobohis gibbus, Urospora crassa; 26 species in all. The subarctic group: Subdivision 1 : Peyssonnelia Rosenvingii , Lithothamnion tophi- forme, Ralfsia ovata, R. deusta, Myrionema Laminariæ, Ectocarpus helophorus, E. ovatus, Sphacelaria racemosa, Phæosaccion Collinsii, Symphyocarpus strangulans, Delamarea attenuata, Coilodesme bulligera, Dictyosiphon hispidus, Saccorhiza dermatodea, Laminaria longicruris. 800 Agariim Tiirneri, Chlorochijtriiim Schmitzii, Ulothrix siibftaccida, Acrochæte parasitica, Urospora Hartzii, Rhizocloniiim pachydermiim'^, Acrosiphonia penicilliforme ; 22 species in all. Subdivision 2: Phæostroma piishilosiim , Haplospora globosa, Chlorochijtriiim Cohnii, Ch. dermatocoiax , Cladophora hirta: 5 spe- cies in all. The boreal-arctic gro up: Chantransia microscopiai, Oscilla- toria amphihia, O. tenuis^; 3 species in all. The cold-boreal group: Lithothamnion intermedium, Ecto- carpiis Pringsheimii, E. penicillatiis, Sphacelaria radicans, Dictyosiphon Chordaria, Epicladia Fliistræ, Ochlochæte ferox, Vaiicheria intermedia, V. sphærospora ; 9 species in all. Thus we see, that by far the greater part of these species be- longs to the arctic and the coldest subdivision of the subarctic group; it is not probable that any of these species will be found at the Færoes. On the olher band, some of the few species referred to the other groups will no doubt be found to grow there. This comparison thus leads to the conclusion that the Færoese algæ-tlora includes a great number of southern forms wanting at Greenland, whilst Greenland has a great number of arctic species not found at the Færoes. Lastly we may make a comparison with the algæ-Ilora of North- America. It is a well-known faet that in spite of the great distance separating the algæ- flora of North -Europe from that of the North-American coast of the North-Atlantic Ocean, they still have a great many species in common. Our knowledge of the North-American algæ -flora we owe firstly to Farlow's well- known paper (20) which is now somewhat antiquated, however, as it is 25 years since it was published, secondly to later works of Farlow% Collins and others. The latter author published in 1900 a list (13) of the North-American algæ, which has been used toge- ther with Farlow's above-mentioned paper as a basis for this comparison. Of the species mentioned in this list about 130 are also found at the Færoes, that is, more than one half (60 %) of all the Færoese species. Fa ri o w has already cailed our attention to the great resemblance between the algæ-llora of both sides of the North-Atlantic Ocean; he likewise mentions (20, p. 4) some of the commonest species that grow both east and west of the Ocean. ^ Found in freshwater at the Færoes. 801 However, many widely distributed North-European species are, as is well-known, wanting at North-America, e. g. CallophijUis la- ciniata, Lomentaria clavellosa and L. articnlata, Delesseria samjiiinea, Laiirencia pinnatifida, Polysiphonia Brodiæi, Griffithsia setacea, Calli- thamnion scopulorum and C. arbnscula, Diimontia filiformis, Ptilota phimosa, etc. On the other hånd, the coast of North-America has some species that are wanting at the European coast; of these I may mention: Grinellia americana, Polysiphonia vestita, P. Olnei and P. Harueyi, Callithamnion Baileyi, Ceramiiim Hooperi, Ectocarpiis Chordariæ, E. liitosiis. Stictyosiphon subsimpleæ, Dictyosiphon Macouni, Scaphospora Kingii; to these must also be added Aganim Tiirneri and Laminaria longicruris which are both very common on the coasts of North-America and Greenland, but have not been found further to the east. It is true, that Laminaria longicruris is stated to have been washed asliore at certain piaces of the European coasts, but it has never been found growing on any substratum, and it is indeed very probable that these hollow stalks of Laminaria are fragments of the very similar Laminaria færoensis, and have come from the Færoes or Iceland. To what result does this comparison lead? As our know- ledge of the floræ of several of the countries we have dealt with is deficient, the result can only be somewhat hypothetical. First of all it must be emphasized that the algæ-flora of the Færoes is to be regarded as a scanty selection of that of North-Scotland, with the addition of some few, more northern species not found at the British Isles. If we had a more thorough knowledge of the algæ-flora of the Shetland Isles, the flora of these islands and that of the Færoes would certainly be found to agree well, just as the algæ-vegetation of the Shetland Isles, at any rate the littoral vege- tation, agrees well, with that of the Færoese coasts (compare 9). The Færoese algæ-flora moreover much resembles that of South- West Iceland; the algæ of this part of Iceland must be regarded as a somewhat poorer selection of the algæ of the Færoes, still "with a few more northern forms Ihan at the Færoes; lastly there seems to be considerable resemblance between the algæ-flora of the Færoes and that of Nordland. When the Danish issue of this paper was published, I did not have the revised list of algæ which I have prepared later together with Mr. Jonsson, and without which a thorough exami nation of 802 the relation belween the diffcrent algæ -flora is impossible; on ac- count of this, my comparison of Ihat lime led to the result, that »the Færoese algæ-flora mostly resembles that of the nearest parts of the Brilish Isles, less that of West Norway, South Iceland and Arctic Norway, still less that of North -East Iceland, least of all that of Greenland, and a little more that of the North -American coast of the North-Atlantic Ocean«. This result has been criticized by Si m mon s. Founding his opinion on his list of algæ published in »Botaniska Notitser«, 1904 (compare Porsild and Simmons, 66, p. 203) he comes to the conclusion that the countries, judged by the degree of resemblance between their algæ-floræ and that of the Færoes should be named as follows: South -West Iceland, North Norway, West Norway, Scot- land, North Iceland, America, Greenland. The difference however is not very great, when we consider that Simmons takes all Scotland as a wdiole, which tends to reduce the resemblance with the Færoes considerably, and that I expressly deal with the British Islands closest to the Færoes, that is, the Shetland Isles. Moreover, whilst Simmons only deals with the numbers (that his list is wanting in accuracy, and his num- bers therefore not to be relied upon, has already been mentioned above), I have proceeded more hypothetically, as we know very little of the algæ-flora of many of the particular countries, and have therefore tried to point out which species we may still expect to find in the flora of the different countries; thus my result ne- cessarily becomes less precisely formulated. I do not wish naturally to deny all importance to these num- bers, on the contrarj^ I think that Sim mons's method might be rather instructive, for instance his table 1. c. p. 219, and I therefore add one of the same kind (see next page), calculated from the list prepared by Jonsson and myself. Sim mons's method is this: he adds up the number of the species of two floræ, calculates the per- centage of those species common to both and of those peculiar to each flora. Like Simmons I have in the following table only dealt with Rhodophyceæ and Phæophyceæ , and likewise only with the species that we know with certainty to be found in the parti- cular countries. The relative order is, as will be seen from the table, therefore as follows: South-West Iceland, Nordland, West Nor- way, Finmark, North-East Iceland, the Shetland Isles, Scot- land, North- America, Greenland. t?' 803 The Færoes Conimon species South -West Iceland 52 103 22 30 0/0 58 0/0 120/0 The Færoes Common species Nordland 60 95 28 330/0 52 0/0 150/0 The Færoes Common species West Norway 14 114 76 18% 490/0 320/0 The Færoes Common species Finmark 64 91 34 34 0/0 48 0/0 18 0/0 The Færoes Common species North-East Iceland 78 77 22 44.5 0/0 430/0 130/0 The Færoes Common species The Shetland Isles 83 72 15 500/0 420/0 8 0/0 The Færoes Common species Scotland 25 130 182 7.50/0 38.5 0/0 540/0 The Færoes Common species North-America 57 98 104 22 0/0 38 0/0 400/0 The Færoes Common species Greenland 87 68 43 440/0 340/0 22 0/0 As mentioned before, I do not think it justifiable to attach any grcat validity to this table, until the flora of the particular countries have been thoroughly investigated; it may serve as a hint, perhaps a correct one, but no more. In agreement with my first thesis (ep. 10, p. 123), the algæ- flora of the Færoes may thus for the present be regarded as most closely related to that of the northern coasts of the British Isles (the Shetland Isles), also to that of We s t Norway (especially Nordland), and to that of South-West Iceland. This likewise agrees well with the result attained in the preceding chapter, namely, that the Færoes, Nordland, South- West Iceland, and the Shetland Isles are especially the habitats 804 of the cold-boreal algæ. It is certainly very probable that the greatest resemblance will be foiind to exist between the F'æroese algæ -flora and that of South -West Iceland, more specially with that ol" South Iceland, as the numbers of the table above tend to show, but until the Shetland Isles have been more thoroughly in- vestigated nothing can be said with certainty. 3. Floristic Differences in the Marine Algæ-Flora of the Færoes. It is evident that no great floristic difTerences can be met with within a terrilory so small as the Færoes, which in their greatest extent from north to south only extend over a little more than one latitude, and where the hydrographic conditions are so very uniform. Thus the vegetation is in similar localities everywhere almost perfectly homogenous in its composition, and it is onlj'^ on closer observation that a few disagreements between the northern and southern parts of the islands become discernible. These disagree- ments are probably caused by the East-Icelandic Polar current which is naturally of great influence on the north coast of the islands. Nevertheless, the Polar current has at the Færoes already, partly been heated by mixture with water from the Atlantic Ocean, partly become more saline, to such a degree, that, according to Ostenfeld (65, p. 610) no arctic plankton has been found at the Færoes; on account of this, the influence of the Icelandic Polar current must likewise be supposed to be relatively small on the fixed algæ vegetation. According to the division made by Jonsson and myself, no arctic species are to be found at the Færoes, as already mentioned. Among the subarctic species there is one, however, belonging to the coldest division of this group, which is somewhat more in- teresting; it is Halosaccion ramentaceiim a circumpolar species vei-y common in the Polar Sea. It is usually found on the south and west coasts of Iceland, is frequent at the northern coasts of Norway, but does not grow south of Nordland where it has only been found at a few piaces. At the Færoes, where the southern limit of this alga is most likely found, a supposition already men- tioned by Rostrup (72, p. 16), it has been found at 3 piaces; namely, in addition to the well-known habitat in Klaksvig, also at »Havnen« on the west coast of Svino, and in Vestmanhavn. Thus the two habitats are found on the Nordreoer, which likewise shelter most of the arctic inland piants, the third on Nordstromo. 805 In Klaksvig and Vestmanhavn it grows under exactly similar outer conditions; in both piaces, especially in Vestmanhavn, much fresh and cold water streams down the rocks into the sea, and the temperature of the sea can be fairly low in winter, especially in the upper water layers. At Svino it grows on an exposed coast but at a high level in a large water basin which is only inundated by the sea at high-tide and by rough seas; here also the tempera- ture of the water can of course be somewhat reduced in the winter. Whilst this species has thus only been found in the northern territories of the islands, there are also some few others which have only been found at the southern and western coast. Griffithsia setacea, for instance, has only been found in Trangisvaagfjord on Sj'dero, likewise Polysiphonia violacea; Laurencia pinnatifida has only been gathered in Trangisvaagfjord and at Hojvig on Stromo; very stunted specimens of Nitophyllum laceratiim on the west coast of Stromo, south of Kvivig and, according to Simmons, Edocarpiis veliitinns in Kvalbofjord on Sydero, but it is certainly not excluded, that these and a few other forms, hitherto only found at the southern and western half of the islands, may also be found farther north. In this connection it should be pointed out, that the Færoes form the southern limit ^ of several species of northern distribution, as well as the northern limit of several species of southern distribution. Among the former should be mentioned, besides the above mentioned Halosaccion ramentaceiim , also Ptilota pectinata, Rhodo- chorton penicilliforme , Rhodophyllis dichotoma, which is, however, reported as found at Bergen in Norway, that is, almost as far south- w^ard, and a few more. Of species having their north-western limit here (the real northern limit of several of these species is on the west coast of Norway) the following should be mentioned: Porphijra leucosticta, Callophyllis laciniata, Callocolax neglecta, Nitophyllum laceratiim, Griffithsia setacea, Callithamnion grannlatiim, Rhodochorton seiriolaniim, Lithothamnion lævigatiim, Lithophylhim incrustans, Ec- tocarpus liicifiigus, E. velutinus, Himanthalia lorea, Valonia ovalis, and several others. 4. The Origin of the Marine Algæ-flora of the Færoes. Kjellman (48, p. 58—77) and Reinke (68, p. 96—99) are cer- tainly right in supposing that both the European and American ^ Several of these species are, however. found still farther south on the east coast of America on the Atlantic Ocean. 806 algal flora of the North Atlantic have originated from a mixture of Atlantic and Arctic species. I shall briefly explain the main rea- sons for this conclusion. There is hardly any doubt, that in tertiary times there has been a land-connection, reaching from Europe by means of the Færoes and Iceland to America. This explains the great resemblance that exists nowadays between the algal floræ on both sides of the ocean. The flora north oflhe land-conneclion differed much from that Nvhich grew south of il. The Arctic flora has gradually developed from that north of the land-connection, and as pointed out by Kjellman, it is an old flora, which has developed in the seas about the Pole and has been very rich in endemic species. But when the land-connection was broken, which probably happened in the later Tertiary period, a comingling of the species from the two formerly separated territories began and continued into the Glacial Period. During the latter, when the Polar Sea and the northern part of the Atlantic Ocean were covered by great masses of ice the algal flora was forced to go southwards, so that a flora of Arctic character probably oc- curred as far down as the coasts of South England and North France. On its way south, however, this Arctic flora met and be- came intermingled with the species of the Atlantic flora, which had been able to resist the climatic changes. When the ice again receded after the Glacial Period, this algal flora, now composed of species from two different territories, again wandered towards the North, yet a few Arctic forms, which were able to adapt themselves to the higher temperature, remained on the coasts of England and France, whilst others withdrew to the Polar Sea proper. On the other hånd, several more southerly Atlantic forms likewise followed northw^ards, and in such quantities, that they still occur so numerously as to stamp the character of the algal flora in piaces where circumstances were favourable to their growth, for instance, on the northern coast of Norway. In some such manner as this we may suppose, that the algal flora of the northern part of the Atlantic Ocean has originated. At the end of the Glacial Period its general appearance must have been much the same as it is to-day: a comingling of old Atlantic and Arctic elements. The flora of the Færoes and with it the oceanic algal flora were probably completely destroj^ed during the Glacial Period. How have the Islands again recovered their flora, and in what manner especially has the immigration of the marine algæ taken place? 807 In reference lo these questions opinion is divided into two sharply opposing canips; the one finds the explanation in a post- glacial land-bridge, which makes the immigration of the flora ap- parently easy to understand. The other maintains, that this land- bridge is very problematic and not necessary to explain the immi- gration of the flora of the islands. I shall not further discuss what has already been said con- cerning this question, but merely call attention to Warming's treatise on this matter^ and to his paper (83) published (p. 660) in this work. In the latter paper Warming again takes up the discussion of this question, and on the one band emphasizes the improbability of the existence of a postglacial land-bridge, on the other declares that the plant-immigration may have taken place by aid of the following factors: (1) hirds; (2) the wind; (3) ocean currents; (4) floating ice, floating timber, etc. and finally (5) man's agency. In reference to great distances, thus also to the Færoes, Warming thinks that the birds, the ocean currents, and the float- ing timber, etc. are of small importance or none at all. On the other band he assigns great importance to the influence of the wind and the agency of man. Shortly after this, in the spring of 1904, I published in Danish my treatise on the marine alga-vegetation of the Færoes (10), in which I likewise tried to explain the occurrence of the marine algæ on the coasts of the Færoes by means of factors at work to this day, without having recourse to the theory of the postglacial land-bridge. This hypothesis was, as already mentioned in the in- troduction, attacked by Porsild and Si m mons, who both adopt Ostenfeld's view (64) and assume, that there has been a post- glacial land-bridge by means of which the flora immigrated to the islands. They maintain moreover, that as natural conditions are now, the algal flora could not possibly traverse the sea; Porsild for ex- ample thinks it a »physical impossibility«, that the algæ should be able to cross the Gulf Stream. I have already replied (11) to this attack, quoting amongst other arguments, several hydrographers e. g. Komm. Holm, Doc. M. Knud- sen, Prof. M o h n , Prof. F r. Nansen, Kapt. Ryder, and Admiral Wandel, well-acquainted with the course of the currents in the North Atlantic, to prove that there is nothing to prevent floating ar- ^ Warming, Eug. , Om Grønlands Vegetation (Meddelelser om Grønland, København, 1888). 808 ticles from being transported to the Færoes from all sides under favourable winds. I have likewise quoted what has been published since W ar- mi ng's paper against the theory of a postglacial land-bridge by A. C. Johansen^, Helgi Pjetursson^ and also a discussion by Fridtjof Nansen who refers especially to his own works^. But the most important article concerning this matter is of recent date and due to Prof. Th. Thor od d sen. In a paper published in February 1905* he sets forth argu- ments against a postglacial land-bridge from a geological point of view. At my request, Prof. Thorodd sen has very kindly given me a short resumé of his paper, which I give here : »A land-bridge of basalt existed across the Atlantic from Scotland to the Færoes, Iceland and Greenland at the beginning of the Miocene period, but the tectonic features of Iceland, the displacements of the basalt-sheets and the relation of the beds of »Surtarbrand« to the lines of faults show, that the country was not much larger at the end of the Miocene period than it is now. The land-bridge had sunk into the sea. During the Pliocene period Iceland's systems of valleys and rivers were formed in the basaltic regions. These systems conformed closely to the present form of the country. Thereafter doleritic, later striated lava-streams tlowed down into the valleys, and at the time these doleritic lava-beds were laid down, the basaltic regions had pretty much the same essential contours as at present. The dole- ritic, striated lava-streams are either preglacial or glacial, that is, they date from an earlier period than that during which the land was last covered by ice. The pliocene valleys and fjords are con- tinued by deep channels out to the edge of the submarine plateau, and at east Iceland even on to the submarine ridge. These channels are no doubt older than the last ice period in Iceland and younger than the submarine ridge. The Red-Crag-beds in North-Iceland show also, that there was no connection between Iceland and Greenland ^ A.C.Johansen: »Om den fossile kviirtære Molluskfauna i Danmark og dens Relationer til Forandringer i Klimaet«, København 1904, p. 42. * Helgi Pjetursson: vOm nogle glaciale og interglaciale Vulkaner paa Island«, (Oversigt over d. kgl. danske Videnskabernes Selskabs Forhandlinger 1904, n:o 4), p. 266. ^ Fr. Nansen: The Oceanography of the North Polar Basin (The Norwegian N. P. Expedition, vol. III, Nr. 9 pp. 419—420). Fr. Nansen: »The bathymetrical fea- tures of the North Polar Seas with discussions of the continental shelves and pre- vious oscillations of the shore line (Ibid. Vol. IV. 1904). * Th. Thoroddsen: Hypotesen om en postglacial Landbro over Island og Færøerne set fra et geologisk Synspunkt (»Ymer«. 1904, H.4, Stockholm). 809 at the end of the Pliocene period, that the sea-level was 80 — ^100 meters above the present high water-mark and that the contours of the coast were ahnost the same as now. At the close of the glacial period, when almost the whole country was covered by inland ice, the level of the sea was at the coast 80 — 130 meters above the present high water- mark and gradually sank down to 30 — 40 meters, when the animal life at the coast had become the same as it is now. Raised beaches are to be seen all round the coast in every inlet. It seems to me for these reasons, that Iceland cannot possibly have been connected with other countries during a postglacial period.« With regard to the immigration of the flora on the whole, re- ference may be made toWarming's above-mentioned, detailed work, and I shall now pass on to discuss the possibilities of immigration as regards the algæ, especially the marine algæ. The latter, which live in the sea, are of course widely different from the land flora, so far as the present question is concerned. In my paper on the fresh-water algæ of the Færoes (7) I have pointed out, as already mentioned, that the supposition is at least tenable, that the fresh- water algæ have been transported to the islands by hirds or the wind. As to several marine algæ, especially littoral algæ, it also seems to me most probable, that birds have contributed to their distribution. It is now said, however, that migratory birds not only journey with stomachs empty but are on the whole clean (comp. Ostenfeld p. 116); yet small crusts of mud and similar substances have frequently been found on the beaks and feet of the birds. Spores and resting cells can very well be hidden in these crusts (comp. Winge's observations in Warming's paper (83, p. 676)). The birds which might be supposed to carry the algæ, for in- stance, from Shetland to the Færoes need not, however, be migratory birds only. The distance between these two groups of islands is only about 300 kilometres, and numerous birds journey everywhere across the sea between them. When fogs and tempests suddenly arise, the birds may very easily be blown off from one coast to another, and it is even very probable, that some of the many sea birds, perhaps parti- cularly the wading birds which live on the coasts of Shelland and the Færoes sometimes visit the one coast, sometimes the other. Even halfway between the islands large flocks of gulls, fulmars and gan- nets may be met with, and it cannot be supposed that these birds always return to the same coast. In order to have an opinion from a competent authority on 810 this matter I applied to Mr. Winge, Vice-Inspector at the Zoological Museum, who has very kindly supplied me with the following re- port: »Small crusts of mud, clay, elc, may he found on the beaks and feet of the hirds ^, that are killed at the Danish light- ships during migration and I consider such hirds as Charadrins pliwialis, Hæmatopiis ostreologns, Nnmenins phæopus, N. arqiiatiis which are frequent bolh on the shore and inland, particularly fitled for car- rying about spores of algæ etc. They fly excellently; they may be able to travel the distance from Shetland to the Færoes in about four hours when the weather is favourable^; and they wander 1 Mr. Winge has afterwards sent me the feet of two larks which had fallen down on the light -ship of Gedser in the autumn of 1904. 1 scraped 120 milli- grammes of mud from these. Further a wing covert of a female blackhird, fallen at the light-house of Skagen on the second of Nov. 1904, on which two things had stuck, one of tliem a Betida-fruit, which must probably have been carried from Norwaj'. On March 3rd, 1905, I at length received a sample of earth from Inspector Winge, with the following information: »Earth and fragments of plants from the feet of a lapM'ing, fallen on the light-ship of Horns Reef, March 5th, 1905. The eartii most likely is from England or Holland.« When I received the sample its weight was 360 mgr. Shortly after, I placed it in a Petri-glass under a hand-glass on filter-paper moistened with well-hoiled water. During the summer, some algæ were developed, and Professor Wille, to whom I sent the sample, has kindly stated them to be the following species: Nostoc sp. (most likel}' A'^. Linckiæ (Roth) Bom.). Anabæna sp. (most likely A. variabilis Kiitz). Ulolhrix sp. (belonging to the /?acc/c/a-group). Cijsiococciis hnmicola Nægl. Navicula sp. Moss-protonema (large and well developed) and hyphae of fungi. As it is of the greatest importance to know whether the hirds had been soiled during the transport from the light-ship to the Zoologicai Museum of Copen- hagen, I applied to Mr. Winge for information, and he kindly reported: »From the letters of conveyance I learn that all hirds sent from the light-ship of Horns Reef during the spring arrived in »boxes in paper«, that is small, tight, square, nailed wooden boxes, made on board the ship and wrapped up in cap-paper. The boxes were packed on the light-ship, lying in the North Sea about 21 Eng. miles west of the coast of Jutland, and were not opened until they had arrived in my study at the Zoologicai Museum. With the end of a clean knife I here knocked off the earth that stuck to the lapwing's feet, and wrapped it up in a paper which was left unopened till 3'ou opened it yourself. Mould and the like might possihly oc- cur in the Museum but certainl}' nothing else.« I likewise think, that the species which developed in the earth cannot possibly have been conveyed to the sample in my study at the Botanical Museum. Mould and the like might also have been convej'ed to the sample here, but such forms were not developed at all. The species found were all earth-plants, and we can certainlj'^ take it for granted therefore that thej' were present in the sample, when the bird fell on the light-ship. * Compare Wille: »Færøernes Ferskvandsalgcr« p. 18. 811 about a great deal, the whimbrel perhaps most of all.« Mr. Winge further states that »a great many niigratory hirds really go to the Færoes, not only those that are breeding there but also many on their passage north- or southward.« It seems to me very probable, that several, especially littoral algæ have been carried across the sea by the agency of these mi- gratory hirds. How easily may not the filament of a Rhizocloniiini or a Bangia stick to a bird's feathers, feet or beak. These filaments are, as is well known, so elastic and roll up so easily, that they may readily become entangled in the feathers of a bird. A small por- tion of an alga, e. g. Porphyra, Enteromorpha , Monostroma , Uha, Ulothrix, Urospora or Prasiola may also easily stick to the beak or the foot of a bird, and thus be carried off. It is of course only smal- ler algæ as a rule, or species of which a fragment is sufficient to develop new individuals, that we can suppose have been carried across the sea in this way. As to the wind, it is certainly not incredible that small frag- ments of algæ may be swept up and carried across large distances, as has been shown for both organic and inorganic substances (see Warming, 83, p. 678— 679^); but I think, that no great importance can be assigned to the wind as a means of transporting marine algæ. As to the freshwater algæ it is quite different. They often live in piaces that dry up, from which they may easily be carried a long way with the dust as resting cells or spores. But the factor of greatest importance in the distribution of the marine algæ is naturally the sea currents, and in the following pages we must examine which of the currents are the most impor- tant in this connection. We should naturally suppose that the greatest supply of fragments of algæ, resting cells or spores is furnished by the countries which are nearest to the Færoes; namely 1) the coasts of Ireland and Scotland, 2) the west coast of Norway and 3) Iceland. Each of these countries ma^^ be examined from this point of view. The coasts from which first of all the immigration would be ex- pected, are those of Ireland and Scotland, as they are the nearest to the Færoes. It cannot, however, be denied that the first view of a map of the currents of the Northern Atlantic is somewhat disappointing, if information is expected regarding the currents of these regions, as *) Cp. also: Warmiiig, Eug., Den danske Planteverdens Historie efter Istiden. Kjabenhavn 1904. Botanj' of the Færoes. 52 812 Ihey do nol seem to flow in any direction favourable lo Ihe immi- gration. As poinled out by Ostenfeld (64, p. 115) a broad arm of the Gulf Stream flows south east round the istands and seems to prevent all immigration from Europe. As it comes from the open Atlantic and has nol touched any country after the West Indies and tropical North America, it can, as Ostenfeld says, only carry tropical fruits and seeds^. These circumslances are seemingly unfavourable, and Ostenfeld also declares, thai in his opinion the current can never carry species thai will thrive in the Færoes. This is however scarcely quite right; even if the circumslances are far from favourable, il seems to me it cannot be denied Ihat marine algæ, which will thrive at the Færoes, can be carried there by cur- renls. If we look al one of the lalesl of the currenl maps of the northern part of the Allantic, as for instance Ryder's (73, lable 11) which represents the seas between Norway, Scotland and Green- land, and of which fig. 162 represents a photographic reproduction, it must be ackno wiedged, that as lo the parts of the British Isles which are nearest lo the Færoes, the direction of the current is anything but favourable, as it flows from the Færoes in an almosl straighl line towards the Shelland Isles. Circumslances are how- ever totally different as lo Ihe wesl coast of Ireland. In order lo understand this, an ordinary, Iheoretical map of Ihe currents is not sufficient; one must study a delailed chart, for instance, the excellent »Monalskarte fur den Nordatlantichen Ozean« prepared by Ihe »Deutsche Seewarte«, on which are indicated all the directions of the currents observed during one month^. Also in a paper on Rockali (Notes on Rockall Island and Bank elc. Transaclions of the Royal Irish Academy, vol. 31, Dublin 1896—1901) the observed directions of the currents (all possible directions!) and ^ Such fruits and seeds are frequentlj' washed ashore on the Færoes. Lyngbye even mentions this in »Tentamen Hj^drophytologiæ« p. 60. I myself have also found and received from inhabitants of the Færoes fruits and seeds of Cocos, Entacla giga- lobiiim and Guilandina, all well knovvn from the beach woods of the West Indies. * In order to prevent any misunderstanding, I may give a report of what Commodore G. Holm, the director of the chart-archives, has kindly communicated to me, concerning the meaning of the arrows (compare figs. 163 — 4). »The arrows indicate the observed directions of the currents. They are most likely compiled from the journals of ships, and are the mean of the directions caused bj- a cur- rent. Most likely these directions are immediately due to the wind, but in this thej' do not differ from other currents, for most currents are originally produced by the agency of the wind. The arrows must indicate surface currents extending down more than several feet. several fathoms at least. 52* 814 their rapidity (sometimes almost nil, sometimes rapid) maj^ be seen in Ihe two charts (PI. XIII and XIV) accompanying the paper. Thus it is evident, that even far out in the Gulf Stream, where the current might be expected to flow to a certain degree regularly, it may go, so to speak, in all directions. This is of course due to the faet, that the Gulf Stream flows very smoothly here, so that strong winds may give rise to currents in all directions, even across Fig. 163. Chart showing the observed directions of currents (1^'rom Deutsche Seewartes Monatsliarle,) April 1903. it. In the accompanying figs. 163 and 164 I have given portions of two charts from the year 1903; the arrows indicating the observed directions of the currents. Even if it is far from usual, it cannot, however, be denied, that algæ, floating timber, etc. may be carried as far as the Færoes by strong winds from the East or the South, which are very frequent, at any rate in the spring, and by the above-mentioned changes in the direction of the currents. They may even be carried as far as Iceland, which happened to the bottle No. 12 thrown ovérboard north-west of Rockall by the »Thorvaldsen« (see Ryder 1. c. p. XLII). Another bottle, No. 11, which was thrown overboard by the »Skålholt« was washed ashore on the Færoes. The latter bottle (see also »Ceres« bottle No. 6) pro ves 815 the irregularity of the setting of the current. The same can be said, for instance, of the bottles No. 32, 33 and 26 of the »Tejo« 1899. These bottles were thrown overboard to the south of Iceland; under ordinary circumstances they would have been carried westwards, but on account of the heavy winds from the west they were hur- ried eastwards lo the west coast of Norway. How very irregular the direction of otherwise constant currents can be on the whole Fig. 164. Chart showing fhe observed directions of currents in July 1903. (Krom Deutsche Seewartes Monatskarte.) (comp. Boguslawski and Kriimmel: Handbuch der Ozeano- graphie, vol. II, for instance p. 374 — 375) is also proved by the faet, that the current of the North Sea went in an opposite direction from the usual in December 1896 and January 1897^. I cannot but believe therefore that floating algæ, floating timber to which algæ are attached, etc. can be carried to the Færoes some- times when the wind is favourable, probably not only from the coasts of Ireland, but also from the Hebrides and the west coasts of Scotland. But it is not only from these parts of the British Islands that 1 Ful ton, T. W. The Currents of the North Sea, and their Relation to Fishe- ries. (Fifteenth Annual Report of the Fisher}' Board for Scotland for the year 1896. Edinburgh. 1897.) 816 things may be carried to the Færoes, but also from the west coast of Norway. According to Ryder \\e must suppose that there is a large circulation north of the Færoes between Iceland and Nor- way. This circulation is formed by the East Icelandic Polar Current to the wesl and the current along the coast of Norway to the east. Ryder thinks, that the northern part of the circulation is formed by a current which goes from the northern part of Norway at about Lofoten in a north-westerly and westerly direction towards Jan Mayen, and then southward along Ihe east coast of Iceland (the Easl-Icelandic Polar Current), and (inally soulh westwards to the Færoes. From here the current again runs in a norlh-caslerly direction to the west coast of Norway, after having touched the Shet- land Isles. It is also very probable, that the great distance which the algæ have to tloat, if they are carried round the whole circle, can be much shorlened, when the winds are favourable. If we look at Ryder's map of bottles thrown overboard from the »Antarctic«, we find, that a bottle (No. 15) thrown out at about 250 miles N. E. of Langanes is believed by Ryder to have been carried first in a south-westerly, then in a south-easterly direction, in a large curve close past the Færoes, and to have been washed ashore on the north coast of Norway. If algæ can now be carried from the west coast of Norway out into the North Sea, some of them may have a chance of being car- ried to the Færoes. Fortunately Prof. H. H. Gran has kindly com- municated to me an observation made by him during his expedi- lion in the North Sea (from May to June 1904). He writes that he found floating seaweed with epiphyles all over the North Sea. He has sent me extracts from his journal regarding some of the piaces between 67" 44'— 62« 57' N. Lat. and 11" 10' W. L.— 6« 22' E. L., that is to say, midway between Iceland and Norway, and on both sides of the route which Ryder supposes the above mentioned bottle to have taken. There can hardly be any doubt, that such algæ floating far out in the North Sea can be carried to the Færoes, if the wind is favourable. As to the third possibility for the immigration of algæ, namel}' from Iceland, the currents are even highly favourable. As will be seen on Ryder's map, the current from East Iceland goes straight down to the Færoes, and algæ can thus easily be carried from East Iceland to the Færoes. But as the hydrographic conditions difTer greatly on this part of the coast of Iceland from those at the Fær- 817 oes, the immigration of algæ from there is probably of but small importance. It is probable, bowever, tbat the Færoese species of a more nortbern distribution e. g. Halosaccion ramentacenm and Ptilota pectinata, wbicb are distributed more specially in tbe nortbernmost part of tbe islands bave been carried this way to tbe Færoes. It is my opinion Iberefore tbat 1) circumstances are some- wbat favourable to marine algæ being carried across and over tbe sea to tbe Færoes from tbe west and north coasts of Ireland, from the west coast of Scotland, from tbe Hebrides, etc, 2) tbat algæ can possibly be carried from tbe west coast of Norway, and 3) tbat algæ can very well be introduced from East Iceland (probably also from other parts of Ibis island). Having tbus discussed tbe possibility, tbat objecls floating in tbe sea can be carried to tbe Færoes from tbe coasts of tbe ad- jacent countries, wben circumstances are favourable, we may en- quire into what enables tbe algæ to float for a long time in tbe sea. One reason is, tbat tbey are in themselves able to float for a long while; anotber, tbat tbey are attached to floating timber, etc. As to tbe first reason, there are not a few algæ, especially lit- toral algæ, with air-bladders or with a thallus partly inflated witb air^, which keep them on tbe surface of the water for some lengtb of time. Almost everywhere in the sea one can meet with floating seaweed (comp. tbe above-mentioned observation made by Gran) often far from their habitats. Kjellman says (48, p. 73), tbat he bas found Ascophijlhim nodosiim and Fucus vesiciilosiis floating in tbe sea at different latitudes between Norway and Spitzbergen; and on the south coast of Spitzbergen he has collected Ascophyllnm witb numerous Pohjsiphonia fasfigiata. He bas not found these algæ fixed to any substratum, but others bave stated tbat tbey grow there; at any rate tbey must bave been carried there by the Gulf Stream. It ^ Of the algæ from the Færoes there are in the first place nine brown algæ which are either always or at any rate more or less frequently supplied with air- bladders: Ascophiilhim nodosnm, Fnciis vesiciilosiis, F. inflatus, F. spiralis, Himan- Ihalia lorea, Halidrys siliqiwsa, Laminaria færoensis, Scijtosiphon lomentariiis and Chorda filiim. Moreover, there are the species of Enteromorpha and perhaps a few more brown and red algæ. Here the Zostera must also be mentioned; it is well known that its thallus floats excellentl3\ These species, which are almost all rich as to numbers, float easily, and on these different algæ and Zostera there often, indeed almost always, grow a great many different epiphytes and endophytes, which are likewise carried about in the sea. 818 is siich a well-known fad, and il has so often been proved, that sea-algæ are capable of floating far out lo sea, thai furlher (hscus- sion of the malter is scarcely needed*; neverlheless I niay iUuslrale this by some examples. A great many algæ are washed ashore on the sandy and there- fore in many piaces barren west coast of Jiitland. Dr. L. Kolde- rup Rosen vinge has recently staled at the »Botaniske F'orening«, that about 40 species have hitherto been washed ashore there; se- veral of these species do not grow on the west coast of Jutland, some of them not in all Denmark. Among these algæ some per- fectly well-preserved, fruit-bearing specimens are found, e. g. of Hi- manihalia lorea, which must be supposed lo have been carried probably from the English coast. I do not know for certain, whether there are algæ which are not air-inflated and yet able to drift or perhaps only hang suspended in the water for some lenglh of time; but I am inclined to believe that such algæ do exist, some however sink down quickly, some even very quickly e. g. Fiicus serratns. Il seems, however, very probable, that several fdiform and much ramified algæ must be able to float for a long time in the sea. Dr. Jobs. Schmidt for instance informs me, that he has mel with species of Ectocarpiis and other higher forms of algæ floating near the surface of the sea east of Iceland. In this connection I may also mention, that Hessel man (Bol. No- liser, 1897) found, in addition to some larger algæ (Chorda fihim, Fiiciis vesiculosiis and Enteromorpha intestinalis) also some smaller forms, which unfortunately are not named, in Nortålge skårgård amongst drift. And in »Bolanische Untersuchungen der Pomme- rania-Expedition« vom 3. bis 24. August 1871, p. 77, Magnus de- clares that at a distance of 4 miles from the east coast of Gotland ^ I may fiirther mention a few examples: GcUdiiim cartilagineitm from the Southern part of the Atlantic has been washed ashore on the coast of Scot- land), it is therefore also mentioned in earlj^ botanical works c. g. Edmonston: Flora of Shetland), and on the coast of Norway, whcre it was found for instance by Giinnerus (comp. Se man der: Den skandinaviske vegetationens spridnings- biologi, Upsala 1901, p. 120). The American Laminaria longicniris has an inflated stalk, by aid of which it can be carried a long distance; it has been found for instance on the west coast of Jutland, at Hohuslån and Finmark (comp. Sernander 1. c). Still it is not excluded. that some of these inflated stipes of Laminaria, if not all, which have been found, came from the Færoes or Iceland, as Laminaria færoensis which grows there has similar inflated stipes. though without mucous can- als. Fossibly it might, however, be the American species which has been carried across the Atlantic. If that is so, the immigration of algæ from America to the Færoes is certainiv not excluded. 819 he has met wilh Rhodomela siibfnsca, Ceramium dinphaniim, Sphnce- larin cirrhosa and other algæ floating al)out in the sea. But even if the algæ cannot of themselves float about in the sea for a long time, they can nevertheless be carried about by adhering to floating timber or other objects tloating on the surface of the sea. Between Shelland and the Færoes I have several times seen floating timber complelely covered by a dense vegetation of algæ. In Nolsofjord near Thorshavn, I once saw a large beam quite overgrown by algæ. Amongst these were a great many well-de- veloped Alnria esculenta, carrying numerous Litosiphon Laminariæ on their lamina, several species of Ectocarpus , small forms of En- leronwrpha. Diatoms, etc. Another time I found a piece of cork with a dense, velvet-like vegetation, consisting oi Callithamnion scopiiloriim. Finally, algæ destitute of air-bladders can be carried a long distance by the sea-currents, eilher by growing on species provided with bladders or only entangled among such algæ. To the factors contributing to the distribution of the algæ one must be added which is certainly of no small importance now- adays, that is: man's agency. There can hardly be any doubt, that a great many species of algæ are carried far by the aid of navigation. It is a well-know^n faet that ships which lie near to land very quickly become green below the water-line. The reason is, that the part of the ship which is under the surface of the sea is cov- ered after a short time by a great many animals and piants, which thrive excellently on account of the movement of the ship, and the ensuing constant renewal of water. On a ship one can meet with a vigorous algæ-vegetation, rich in different species^. ^ The foUowing species were for instance collected by me on the 13th of June 1900 from the »Guldborgsund«, which had been scraped and scrubbed on the 15th of May of the same year, according to the journal of the ship. The species were: large, vigorous Alaria esculenta, several specimens more than one foot long; they preferred the most »exposed« parts of the ship, some specimens for instance grow on the screw, where however the lamina were somewhat torn and split. Of Laminaria. there were Laminaria saccharina and L. digitata, ca. half a foot long. Further fairly large specimens of Porphyra miniata, smaller specimens of Chordaria flagelliformis. Of green algæ there were Monostroma fusciim , small Enleromorpha and a great many quite small specimens of different species, Diatoms, etc. which I did not get an opportunity of examining. This comparatively luxuriant vegetation had thus been produced during about one month. It must of course not be left out of consideration, that if the bottom of a ship is ever so thoroughly cleaned, it will never become perfectly clean, unless it is taken into dock. In the above mentioned case, however, one may take it for granted, that only small spe- cimens or fragments of algæ had been left. As to the transport of algæ by ships see also Kjell man (48, pag. 59). 820 When we see, that a ship can become overgrown to such a degree d uring so short a time, as is shown by the case mentioned in the foot-note, the algæ must often have a chance of being trans- ported by ship and thus of being carried to other countries, where they can fix themselves, but of course only where there are suitable conditions of Hfe^ If we may now assume that the algæ can float far across the seas, another question presents itself, viz. if they can also stand the changes of temperature and salinity which a long pas- sage involves. As to the hrst, it may at once be said, that the changes of temperature and salinity in the seas we speak of are relatively small, and we may therefore also suppose that as a rule they have hardly any injurious influence on the algæ. It must not be for- gotten, that the changing of temperature proceeds very slowly during the passage, and that the algæ are very well fitted to these changes within certain limits 2. On the other band, the algæ are usually very sensible to sudden changes of temperature and sali- nity, as previously mentioned by me, still by no means so very sensible as was formerly assumed. In a paper (Notizen iiber die Cultur und Lebensbedingungen der Meeresalgen, Flora 1895) Olt- manns has already pointed out that freshly gathered marine algæ were much better fitted to resist changes of salinity, than algæ which 1 That tilis supposition is right is confirmed by the faet, that we frequently meet with algæ, belonging to very remote territories, on harbour moles, just as is the ease with the inland tlora found in harbours and on wharfs. In »Bulletin de la soc. hot. de F'rance« vol. 35, 1888, p. 364, Bornet for instance mentions 3 La- minariaceæ which have been met with a few times in and near harbours of the Mediterranean, but which were otherwise never found there. The Fiicus infkttiis which I found on Lerwick mole (8) and the two isolated habitats of Fnciis ser- raliis found in Iceland (J éns son, 41) can probabh' also serve as examples. F'inally, Dr. Bornet has klndly informed me b}' letter of some interesting discoveries. Clwrda filiim has been found at the entrance to the harbour of Nice; the species was never observed at other piaces in the Mediterranean. Helminlho- cladia pnrpiirea was found at the harbour of La Nouvelle, likewise in no other place in the Mediterranean. Bonnemuisonia hamifera which is distributed on the coasts of Japan and California has been found on the coasts of England and later also of F'rance (Cherbourg). The plant is dioecious, and only the female plant has been imported; it propagates by aid of buds which are found at the top of the branches like hooks. Hypnca miisciformis has been found in the Channel; its true habitat is more towards the south. ' Cp. for instance: Porter, H. C, Abhångigkeit der Breitling- und Unter- warnow-Flora vom Wechsel des Salzgehaltes. (Arch. Ver. Nat. Meckl., 1894). 821 had been cultivaled for a long time. And in a paper (Verhalten einiger mariner Algen bei Ånderung des Salzgehaltes. Oesterr. bol. Zeitschrift, 1904) Karl Techet in his »Ubersicht« comes to Ihe follow'ing conclusion: »Die individuelle Anpassungsfåhigkeit — um diese handelte es sich bei den angefiihrten Versuchen — an An- derungen des Salzgehaltes ist bei marinen Algen eine ziemlich weit- gehende und zwar sowohl bei spontaner Erhohung als sponlaner Verminderung der Salzintensitåt.« With regard lo change of salinily, I may also call attention to Ihe difTerent o])servations mentioned in my descrij)lion of Ihe vegetation. I have met with algæ, even Florideæ, growing al Ihe mouths of rivers, where Ihey were completely covered by fresh water at Iow tide, and by salt water at high tide^ These are sudden changes. With regard to the temperature, the algæ seem specifically until for resisting changes from cold to warm waler. They can prob- ably stand the reverse change much belter, within certain limits of course^. Algæ carried about by a ship are naturally much more exposed to the danger of sudden changes in the temperature and salinily, and if such changes occur the algæ simply perish^. But for short distances, as for instance belween the Shelland Isles and the Fær- oes, the difference is too small lo hurt the algæ. It is very probable, thai the white light on the surface may hurt some of the algæ, which usually grow deepest down, when they happen to be carried near the surface of the sea, but \ve know nothing for certain. We may however suppose, that they are some- times prolecled against Ihe white light; for instance by growing on the under side of wrecks or by being enlangled belween larger algæ. The motion of Ihe waves also perhaps serves to protect them in some way, as it constantly makes them lurn anolher side to Ihe light. We may also suppose that sublitloral algæ which have be- come altached as spores lo floaling timber, elc. are perhaps belter ^ See also Gomoiit, Sur la vegetation de quelques sources d'eau douce sousmarines. (Bull. soc. hot. Fr., t. 51, 1904). ^ Cf. Kjellman, Norra Ishafvets Algflora, p. 73. ^ I may brieth' illustrate this by an example. In the summer of 1899. the »Guldborgsund« was more than usuallj' foul, and was therefore ordered to go in dock at Copenhagen. But when it arrived there, the bottom of the ship was per- fectly clean, all the algæ had perished and dropped from the ship, as soon as it had entered the warm and less salt water of the »Sound«. 822 fitted to stand a strong light, than an alga which is suddenly torn from its habitat in deep water. But even if the alga is destroyed by unfavourable biologicai conditions, the reproductive cells (tetraspores, carpospores etc.) which it possibly carries may be able to survive the transport. If an alga has finally surmounted these various difficulties and been washed ashore, it is still far from being settled among the flora of the island. In order to become so, it must be able to fix itself. A great many algæ are certainly able to do so; I may just mention such species as Ceramium, Polysiphonia, Callithamnion, Ec- tocarpns, Sphacelaria, Cladophora, Acrosiphonin etc. There is hardly any doubt, that all species belonging to these very rich genera are normally supplied with rhizoids, or at any rate are very apl to form them. If therefore one of these algæ when it is washed ashore, sticks to something or other, for instance to an alga growing there, it will perhaps only be a shorl time before it forms rhizoids, which can fasten the plant to its subtratum. Even very small fragments of several algæ seem to be able to produce new individuals. Thus Oltmanns mentions (1. c), that he has cut shoots of Polysiphonia and Ceramium into very small pieces which quickly formed rhizoids and gradually developed into small piants. In different papers Tobler^ also tells us that even one cell of some Rhodomelaceæ can develop into new individuals. By cultivating a Dasya elegans he observed, that the alga divided after a short time into many small pieces, which gradually developed into small piants in the culture-glass. He therefore supposes that the algæ must some- times have a similar reproductive power in the sea. On the other band there certainly are many algæ which can- not take root again , when they have once been torn from their subtratum. That it may be possible for such algæ to immigrate into another region, they must necessarily carry reproductive cells within them. When an alga with more or less ripe tetraspores, carpospores, etc. is torn from its habitat, nothing prevents us from supposing, that after having floated for a shorter or longer time it can succeed in carrying the spores to another habitat. In the new region the spores can thrive, if the external conditions are suit- able to the species. * Tobler. F. , Zerfall und Reproduktionsvermogen des Thailus einer Rhodo- melacee. (Rerichte d. deut. bot. Ges. Rd. 20, 1902.) Ueber Eigenwachsthum der Zelle und Pflaiizeiiform. V^eisuche und Studien an Meeresalgen. (Pringsh. Jalirb. Rd. 39, 1904.) 823 After having thus taken into consideration what serves to sup- port the theory of the transport of algæ across the sea, we must, however, admit that on the whole we know very little with certainty about the matter — and so far, our conclusions must necessarily be hypothetical. Taking everything into consideration, however, it seems probable to me, and at least not incompatible with the real conditions, that perhaps all the algæ of the Færoes have been transported across the sea by the agency of the currents. They can either have floated by themselves or have been fixed to various objects floating in the sea. The agency of birds must perhaps also be taken into account. Floating ice has also perhaps contributed to the distribution of the algæ towards the end of the glacial period and shortly after. Nowadays the floating ice is of no importance to the Færoes. There is one further evidence, which indirectly but strongly confirms the hypothesis concerning the immigration by sea — viz. the faet, that hardly any island has not got its algæ-vegetation, even those at the greatest distance from the coast and of volcanic origin or built by corals. Jan Mayen serves to prove this by its algal vegetation, which is certainly very poor, so far as we yet know, but still has species not only from shallow, but also from deep water. It is naturally more difficult for algæ to pass across the sea than to travel along a coast. This serves to explain the faet, that several species common on the coasts of the adjacent countries, are wanting on the coasts of the Færoes, it likewise seems to me to very clearly refute the hypothesis of a postglacial land-bridge. Fiiciis serratus for instance is totally wanting on the coasts of the Færoes, and this is the more worth mentioning, because this alga is very common along all the coasts of North Europe. On the Shetland Isles for example, which are nearest to the Færoes, it grows abundantly. That this species has not spread so far as to the Færoes is certainly solely due to its want of air-bladders, which prevents it from floating by itself on the surface of the sea. The species of Fiiciis found in the Færoes are on the other band al- ways or at least sometimes provided with air-bladders and are thus able to float. It cannot surely be denied, that the Færoes offer an excellent locality for Fiiciis serratus, if it should succeed in reaching so far. It is therefore probable, that it would have grown 824 there abundantly, if a postglacial land-bridge had existed, by aid of which it might easily have made the passage. Several sublittoral algæ e. g. Chætopteris plumosa, Phyllophora rubens, Dilsea ediilis, Brongniartella byssoides, etc. are also wanting al the Færoes, though they are more or less frequent in the sur- rounding seas. Wlien everything is taken into account, the algæ-flora from the deeper sea has perhaps had the grealest difficulty in reaching the Islands. This perhaps explains, why some sublittoral algæ which might reasonably be expected to grow at the Færoes are not found there. The sublittoral algæ which grow deep down are probably the least fitted for floating a long time, and as to their adhering to algæ from the littoral region or to floating timber as a means of transport, this likewise seems to occur only rarely. On the other band we know but little about the first stages of development of many of these algæ. We especially want to know whelher the difTerent reproductive organs are able to Hoat for a long time in the sea apart from the mother-plant. They may possibly be carried far about as plankton ^ Finally if the algæ spores cannot float far either by themselves or by the agency of the sea currents, it is hov^'ever probable, that they can stick to the mucilage of larger algæ, or to tlie mucus of fishes, and thus be carried off. It is hardly probable, that any of the sublittoral algæ have been able to survive the glacial period, according to Geikie^. He ^ In connection with this it may be pointed out, that the late Prof. Cleve, the well-known Swedish plankton-investigator, thinks it probable that algæ-spores can be carried about bj^ tlie agencj^ of marine currents. He has even expressed the opi- nion , that the stunted forms of algæ from the Polar Sea met with in the Baltic can have been carried there together with Arctic plankton (see Botaniska Notiser, 1898, p. 269). If it is so, the short distance from the English coast to the Færoes is nothing by comparison. It should however be observed that these forms of algæ in the Baltic must rather be considered as »relicts« as pointed out by Sve- deli us (Studier ofver Ostersjons Hafsalgtlora, p. 68— 69). A remark by Oltmanns should also be mentioned here. In a paper »Notizen uber die Cultur- und Lebensbedingungen der Meeresalgen" (Flora 1895), after having emphasized the necessity or at least the desirability of having sterilised sea water for culture-experiments, he writes : »Die Steriliserung ist aber auch besonders dann unerlåssig, wenn man kleine Ectocarpeen einigermas.sen rein cultiviren will, weil das See w asser fast zu jeder Jahreszeit und an jedem Ort eine recht e r h e b 1 i c h e A n z a h 1 von S c h vv ii r m s p o r e n der v e r s c h i e d e n s t e n Species enthålt, welche alle neben den ausgesiiten Formen keimen wiirden.« (The emphasis is mine.) ^ Geikie, .lames, Prchistoric Europe. A Gcological Sketch. London 1881, p. 663. 825 writes: »bul thc ice was so thick that it filled up all the fjords and sounds between the various Islands of the archipelago, thus forniing one compact mer de glace which flowed outwards in all directions from the dominant points, and discharged ils icebergs into the surrounding ocean«. But as the Færoes lie so far out in the Atlantic Ocean , they cannot have been far from the open sea, and possibly there may then as nowadays have been strong cur- rents at the coasts, and for this reason the ice may not have reached everywhere to the bottom of the sea in the sublittoral region, some of the sublittoral algæ thus being spared. In accordance with what I have previously said about the possibilities of immigration of the fresh-water algæ (7), it seems to me that there cannot be any doubt, that the marine algæ -flora has likewise been able to immigrate to the Islands across the sea from the coasts of North Europe, and by the agency of factors which are at work to this day. Thus, the possibility of immigration always exists, and a postglacial land-bridge is not at all necessary to explain the occurrence of the algæ-flora. In perfect agreement with Warming's view regarding the land- llora, I may therefore condude by saying that the marine algæ of the islands have also been able to immigrate after the glacial period across the sea from the nearest countries, especially from the British Islands, but also from Norway and from Ice land. V. SOME BIOLOGICAL OBSERVATIONS. It is a well-known faet, that at different seasons of the year, no small difference is discernible in the appearance and luxuriance of the algal vegetation in the northern seas. This has been men- tioned by Kjellman (49) with reference to Bohusliin, and other au- thors e. g. Gran (37, p. 15) and Rosenvinge (71) have expressed the same view for other regions of the sea. On the Færoese coasts the development and luxuriance of the algal vegetation like- wise vary according to the seasons. Unfortunately, I myself have only spent some months of the spring and summer on the Færoes and have therefore only been able to make an incomplete study of the development of the vegetation throughout the year; but this want has been greatly helped out by the collection of algæ from 826 the autumn months of 1897, brought home by Mr. Helgi Jonsson, as well as by Ibe in formalion he has also given me. With these as basis we may safely conelude, that the algal vegetation of the Færoes is mnch richer and more luxuriant in summer than in winter; the greater development probably begins early in spring and continues into July or August, when the greatest luxuriance is reached, then it decreases again. But on account of the great uni- formity of the Færoese climate, the variability of which, parti- cularly as to temperature, is regulated in the sea itself, this varying development according to the seasons is hardly so great here as in many other piaces with a cold temperate climate. The algal vegetation of the Færoes agrees well, as regards the richer development in summer, with what is known even in Arctic seas, where the period for the growth of the algæ is at the brightest time of the year, for instance, on the coasts of Greenland, whilst their resting period begins at the beginning of winter (see Rosen- vinge 71, p. 239). On the other band the algal vegetation of more Southern countries is much less developed in summer. This may be remarked even on our own coasts, where the vegetation in shal- low water is more luxuriant during the months of spring and au- tumn, but less so during the warmest summer days; and in the Mediterranean Sea the littoral algæ-vegetation is, according to Ber- thold (5, p. 426), very poor in summer, but rich and well developed during the months of winter and spring. DURATION OF LIFE. PERENNIAL, ANNUAL AND SHORT-LIVED SPECIES. As on the coasts of Greenland, the algæ growing on the Fær- oese coasts are for the most part certainly perennial. It is impos- sible for the time being, however, to make any reliable comparison between the perennial and the annual species, as our knowledge of the developmental history of a great many species is still very imperfect. On the other band, several species are certainly or most prob- ably annual. To these belong: Pnrphijra umbilicalis and probably other species of Porphyra, Chantransia (9) , Erythrolrichia (9); of brown algæ may be mentioned: Chorda filiim and C. tomentosa, Castagnea virescens, Scytosiphon lomentariiis , Dictyosiphon Ekmaiii, Punctaria plantayinea, Leathesia difj'ormis, Phyllitis fascia and Ph, 827 zosterifoUa, hthmoplea sphærophora, Hinmnthalia lorea, which is, however, most likely hapaxanthic (see 8, p. 480), several species of Ectocarpus and others; of green algæ may be mentioned several species of Entcromorpha , Monostroma Greuillei and M. undiilatiim, (whereas M. fiisciim, as pointed out by Ro sen vinge (p. 238) can most likely live for more than one year), the species of Ulothrix, Acrochæte repens, Bolbocoleon piliferum, Pringsheimia scutata, Uro- spora mirabilis and U. Wormskioldii , Codiolum , Cladophora sericea and Cl. gracilis, Acrosiphonia (-?) and others. Several of these species are certainly capable of producing more than one generation in a year; as pointed out by Gran (37, p. 9), those short-lived species (e. g. Porphyra umbilicalis, forms of Entcro- morpha intestinalis, species of Ulothrix, possibly Monostroma Greuillei etc.) only belong to the littoral region, the reason probably being that only algæ growing in full daylight are able to go through all stages of development in so short a time; we may likevsåse sup- pose with some reason that the rapid development of the short-lived algæ, as pointed out by Gran, is an adaptation to the often very unfavourable biological conditions under which the littoral algæ live, especially those growing at the highest level, just where the short- lived algæ are met with. With regard to Greenland, Rosen vinge has remarked that the low temperature of the sea prolongs the lifetime of many an- nual algæ, and just in the same way the low summer temperature of the Færoes, together probably with the small amount of direct sunshine, has a preserving and prolonging influence on the annual algæ of spring and summer. For example, a great many species, which are only found towards the end of winter and during the months of spring in the Danish and adjacent seas, grow luxuriantly during the whole summer in the Færoes. Thus Porphyra umbilicalis forms well developed associations dur- ing the whole summer, whilst, according to Kjellman, it only forms associations in winter on the coasts of Rohuslån. Vigorous specimens of Diimontia fdiformis, which in the Danish seas grows in spring and beginning of summer, are found during the whole summer on the Fær- oese coasts. Fuciis inflatns f. disticha is found in the summer at Hauge- sund according to a report from Mr. Norum; but the specimens are only badly developed at that season, though vigorous when found during the months of December to March. At the Færoes, however, it Botany of the Færoes. 53 828 is well (leveloped in the summer. According to Reinke (68, p. 61)^, Phyllitis fascia in Kiel Bay is an alga of the winter and spring, and only rare in summer, whilst it is luxuriant and very common during the whole summer on the coasts of the Færoes^. Monostroma Gre- villei is likewise well developed at the Færoes during the greater part of the summer, whereas, according to Rosenvinge (71, p. 238), it only lives from the end of the winter to April or May in the Danish seas. According to Batters (4, p. 57) Porphijra miniata grows in spring and early summer on the coasts of England; at the Færoes it is found during the whole sunmier and far into the autumn. Gran calls our attention to the faet (37, p. 15), that the annual, littoral algæ usually have a shorter or longer resting period, but whilst this period is at the warmest time of the year (July-August) in Kristianiafjord, it most likely occurs at the Færoese coasts only in the autumn, and probably lasts for the greater part of the winter, as these algæ do not come to life again until early in spring. THE PERIOD AND CONDITIONS OF GROWTH AND TIME OF FRUCTIFICATION. A great many of the perennial algæ of the Færoes are certainly capable of growth all the year round even if their vital functions are somewhat reduced during the darkest time of the year. This is most probably the case, for inslance, in the Fiicaceæ and the Laminariaceæ , but, on the other band, we must suppose, that the last months of the year, as also January and perhaps February, form a resting period for several Færoese algæ, and that their proper vegetative period is the months of spring and summer. Judging from the material I have at band, I believe that many algæ begin to develop new shoots even towards the end of winter. The material collected by Jonsson during the months of autumn mainly con- sists of badly developed specimens, whose apices are often dead or dropped; only a few species, e. g. PhyUophora Brodiæi and some specimens of Fiicus had begun to form some few new shoots. In April, however, I found a great many species which already had large, vigorous shoots, e. g. Odonthalia dentata, Delesseria sanguinea ' See also Si m mons: Algologiske Notiser II. Einige Algenfunde bei Drobak (Bot. Notiser, 1898, p. 118). ^ In the Danish seas it is, however, also common in the autumn; it may be found well developed even in August. 829 and D. siniiosa, Callithamnion orbnscnht, Fiircellaria fastigiata and many otliers. In the course of winter, especially towards its end, and during the first months of spring, the new lamina appears in the periodically lamina-changing species of Laminariaceæ, e. g. La- minaria digitata, L. hyperborea and L. saccharina, and the old laminæ are Ihrown olT during the months of spring, at which time huge masses of tlie latter are washed ashore on the beach. As the vegetative period for several species is much prolonged in Greenland, the same may be said though to a less extent of several species on the Færoese coasts, compared with what is the case on our Danish coasts. Besides the above-mentioned annual algæ, which live somewhat longer, as just stated, we may find, e. g. Desmarestia aciileata with new shoots and densely covered with assimilating hairs as late as June and July, whilst in Denmark it is often hairless al- ready at this time of the year. Rhodomela lycopodioides likewise grows for the greater part of summer on the Færoese coasts, whilst Rhodo- mela siibfiisca, which is closely related to the latter, stops developing at the beginning of the summer on the Danish coasts. On the whole, several algæ which at a certain moment stop developing in our seas, seem to be able to throw out shoots much later in the year on the Færoese coasts; but whether their vegetative period is really pro- longed or whether it is due to the faet, that this period begins later and thus likewise ends later than on the Danish coasts, I cannot say as yet. Tlie latter seems at any rate rather probable. As to the time of fructification , it sets in at very different pe- riods of the year for the different species; in my paper on the Ma- rine Algæ of the Færoes I have stated the time of fructification of each species, so far as it is known. As a general rule it may be stated, that the greater part of the Færoese algæ seem to fructify during the summer season. Some species, however, have only been found with reproductive organs at other times of tlie year; e. g. Delesseria sangiiinea, found with cystocarps and tetraspores during the last months of the year; Halidrys siliquosa which is sterile in summer, but which had young receptacles in October; Furcellaria fastigiata of which I have seen specimens witli tetraspores in great quantities in November-Decem- ber, whilst specimens with antheridia or cystocarps did not occur in the material, as they are most probably developed during tlie winter montlis, as on our own coasts. Some species begin to develop their organs of fructification in 53* 830 winter, and fructify the next summer. AscophijUiim nodosiim, for instance, has young receptacles in December: it has been found with ripe receptables in April, and fructifies constantly during the sum- mer; just the same seems to be the case with Fiicus inflatus and Fiiciis vesiculosus; Himanthalia lorea has been found with young re- ceptacles in April, and they become perfectly ripe in July-September. Other species e. g. Ectocarpus lucifugus, Sphacelaria cirrhosa, Elachista scutnlata . Desniarestia aculeala etc. have only been found with ripe sporangia during the months of autumn, others, e. g. Ectocarpus tomentosoides in the end of the winter and more rarely in spring; species of Urospora, Monostroma Grevillei etc. during the months of spring and early in summer. Lastly, it seems as if a great many algæ are able to fructify all the year round; several of the above mentioned short-lived spe- cies, e. g. Porphijva umhilicalis, also Chantransia secnndata, Callo- phijllis laciniata, Euthora cristata, Delesseria sinuosa, Ectocarpus lito- ralis, Elachista fucicola, and probably many more may serve as examples. Just as the period of life and growth of several algæ is prolonged under the influence of the Færoese climate, the fruit- bearing period is often likewise of longer duration liere than in more southern countries; this is similar to what Rose uvinge has observed in Greenland. As illustrations of this may be men- tioned: Rhodochorton Rothii which in Denmark only fructifies during the winter months, but which has been found with tetraspores even in June and July on the Færoes; Ascophyllum nodosum which here in Denmark only fructifies during the winter, but which has ripe reproductive organs all the summer on the Færoes as in Green- land, and Delesseria sinuosa of which the same may be said. On the whole, the fruit-bearing period seems, as regards several species, to occur at other seasons than in more southern countries, and the Færoes seem also in this to occupy an intermediate posi- tion between Greenland, where still greater differences are observed according to Rose uvinge, and more southern countries. Before closing this treatise I would here express my grateful thanks to Professor War min g and Dr. L. Kolderup Rosen vinge, my »Opponenter« ex officio, who by their thorough criticism of the Danish issue of my work have greatly contributed to improve the present edition. 831 indp:x to literature. 1. Agardh, J G.. Species genera et ordines algarum. Vol. 1 — 3. Lundæ et Lipsiæ 1848—1876. 2. — Novitiæ Floræ Sueciæ ex Algarum familia. Lundæ 1836. 3. Areschoug, J. E., Plnceæ Scandinavicæ marinæ. Upsaliæ 1850. 4. Batters, E. A. L., A Catalogue of the British marine Algæ being a List of all the Species of Seaweeds known to occur on the Shores of the British Islands, with the Localities where they are found. (Supplement to the Journal of Bo- tany, 1902.) London. 5. Berthold. G., Uber die Vertheilung der Algen im Golf von Neapel nebst einem Verzeichnis der bisher daselbst beobachteten Arten. (Mittheil. aus der zool. Station zu Neapel. HL Band). 6. Boye, P., Bidrag til kundskaben om algevegetationen ved Norges vestkyst. (Bergens Museums Aarbog 1894—95. Nr. XVL) 7. Børgesen, F., Freshwater Algæ (Botany of the Færoes. Part. I.. Copenhagen 1901). 8. — Marine Algæ (Botany of the Færoes. Part. IL, Copenhagen 1902—03). 9. — The marine Algæ of the Shetlands (Journal of Botan3^ 1903). 10. — Om Algevegetationen ved Færøernes Kyster, en plantegeografisk Undersøgelse. Kjøbenhavn og Kristiania 1904. 11. — Om Færøernes Algevegetation. Et Gensvar. 1 og 2. Botaniska Notiser. 1904—05. 12. — and Helgi Jonsson, The distribution of the marine Algæ of the arctic sea and of the northernmost part of the Atlantic. 13. Collins, Frank S., Preliminary Lists of New Englands Piants. — V. Marine Algæ (Bhodora, Vol. 2, Boston 1900). 14. Deutsche Seewarte. Atlantischer Ozean. Ein Atlas von 39 Karten, die physi- kalischen Verhåltnissen und die Verkehrs-Strassen darstellend etc. Zweite Auflage. Hamburg 1902. 15. Dickson, H. N., The mean temperature of the surface waters of the sea round the British coasts etc. Quarterly Journal of the Royal Meteorological Society. Vol. XXV. 1899. 16. Edmonston, Th., A Flora of Shetland; comprehending a List of the flow- ering and cryptogamic Piants of the Shetland Islands, with remarks on their Topography, Geology, and Climate. &c, &c. Aberdeen 1845. 17. Ekman, Fr. L., Bidrag till kiinnedomen af Skandinaviens hafsalger. Akad. Afhand. Stockholm 1857. 18. Falkenberg, P., Die Meeres-Algen des Golfes von Neapel. (Mittheil. aus der zool. Station zu Neapel, I. Band, 1878.) 19. Farlow, W. G., List of Marine Algæ of the United States, with notes of new and imperfectly known species. (Proceed. of the Amer. Academy of Arts and Sciences 1875.) 20. Farlow, W. G., Marine Algæ of New England and adjacent coast. Reprint from report of U. S. fish. Comm. Washington 1881. 21. — Notes on Arctic Algæ, based principally on Collections made at Ungava Bay by Mr. L. M. Turner. Proceed. of the Am. Acad. of Arts and Sciences. 1886. 22. — On some new or imperfectly known Algæ of the U. S. I. Reprint from the BuII. of the Torr. Bot. Club, vol. XVI, Nr. 1. 1889. 832 23. Flahault, Ch., La flore et la vegetation de la France avec une carte de la distribution des végétaux en France. Tirage å part de lintroduction de la flore déscr. et illustr. de la France. Par H. Coste. Paris 1901. 24. Foslie, M., Contribution to knowledge of the Marine Algæ of Norway. I, East- F'inmarken. Tromso Mus. Aarshefter 13, 1891. 25. — Contribution to knowledge of the Marine Algæ of Norway. II. Species from different tracts. Tromsø Mus Aarshefter 14, 1892. 26. — Kritisk fortegnelse over Norges havsalger efter ældre botaniske arbejder. Tromsø Mus. Aarshefter. IX. 1886. 27. — The Norwegian forms of Lithothamnion. (Det kgl. norske Videnskabernes Selskabs Skrifter 1894. Trondhjem 1895.) 28. — Nye havsalger. Tromsø Mus. Aarshefter 10, 1887. 29. — The Norwegian forms of Ceramium. Kgl. norske Vidensk. Selsk. Skrifter. Trondhjem 1893. 30. — New or critical Norwegian Algæ. Kgl. norske Vidensk. Selsk. Skrifter. Trondhjem 1894. 31. — Ueber die Laminarien Norwegcns. Kristiania Vidensk. Selsk. Forhandlinger. Kristiania 1884. 32. — Algological notises. Det kongl. norske Videnskabers Selskabs Skrifter. 1891. Trondhjem 1893. 33. — The productive Organs in Turnerella septentrionalis, F^etocarpus (Streblonema) Turnerellae, a new Alga. Kgl. norske Vidensk. Selsk. Skrifter. Trondhjem 1896. 34. — Om nogle nye arctiske havalger. Kristiania Vidensk. Selsk. F"orhandlinger 1881. 35. — Gaidukow. N., Die Farbe der Algen und des Wassers. Hedwigia. Bd. XLIII. H. 2, 1904. 36. Gran, H. H., Algevegetationen i Tonsbergfjorden. (Christiania Videnskabs-Sel- skabs Forhandlinger for 1893.) Kristiania 1893. 37. — Kristianiafjordens algeflora. I. Rhodophyceæ og Phæophyceæ. (Videnskabs- selskabets Skrifter. I. Mathem. -naturvid. Klasse. 1896.) Kristiania 1897 38. Hansteen, Barthold, Algeregioner og algeformationer ved den norske vest- kyst. (Nyt Magazin for Naturvidenskaberne. 32. Christiania 1892.) 39. Hjort, Johan, O. Nordgaard and H. H. Gran, Report on Norwegian marine investigations 1895—97. Bergen 1899. 40. Holm, P. A., Skildringer af Naturen paa Færoerne (Tidsskrift for populære Fremstillinger af Naturvidenskaben. 2det Bind. Kjobenhavn 1855). 41. Jonsson, Hel gi, The Marine Algæ of Iceland, I— IV. Bot Tidsskr., 24.-25. Bind. Kjobenhavn 1902—1908. 42. — The Marine Algæ of East Greenland. Meddelelser om Grønland. XXX. Copenhagen 1904. 43. — A contribution to knowledge of the Marine Algæ of Jan Mayen. Bot. Tids- skr., 26. Bind. Kjobenhavn 1904. 44. Kjellman, F'. R., Ueber die Algenvegetation des Murmanschen Meeres an der Westkiiste von Nowaja Semlja und Wajgatsch. (Nova Acta Reg. Soc. Ups. Ser. III., Upsala 1877.) 45. — Uber Algenregionen und Algenformationen im ostllchen Skager Rack. Nebst einigen Bemerkungen iiber das Verhåltniss der Bohuslan'schen Meeres-Algen- vegetation zu der Norwegischen. (Bihang till K. Svenska Vet. Akad. Hand- lingar. Band 5. Stockholm 1878.) 46 — Om en Ceramium-form från Gotland. Ett bidrag till hafsalgernas biologi. (Ofverslgt af Kongl. Vetenskaps-Akademiens Forhandlingar 1897. Stockholm.) 47. — Handbok i Skandinaviens hafsalgflora. I. Fucoideæ. Stockholm 1890. 833 48. Kjellman, F. R., Norra Ishafvets algflora. (Vega-Expeditioncns vetenskapliga iakttagelser. Bd. III. Stockholm 1883.) 49. — Våxtlifvet under vintern i hafvet vid Sveriges vestra kust. (Hotaniska No- tiser 1886. p. 111.) 50. ~ Studier ofver Chlorophyccslagtet Acrosiphonia J. G. Ag. (Bihang till k. svenska Vet.-Akad. Handlingar. Band 18. Afd. III. N:o 5. Stockholm 1893.) 51. Kleen, E. A. G.. Om Nordlandens hogre hafsalger. (Ofversigt af k. Vetensk. -Aka- demiens Forhandlingar. N:o 9. Stockholm 1874.) 52. Knudsen, iMartin, Ingolf-Expeditionens hydrografiske Undersøgelser. (Geo- grafisk Tidsskrift. 14. Bind, 1897—98, p. 151.) 53. — Havets Naturlære. Hydrografi, med særligt Hensyn til de danske Farvande. Skrifter udgivne af Kommissionen for Havundersøgelser. Nr. 2. København 1905. 54. Kuckuck, Paul, Ueber marine Vegetationsbilder. (Bericht. d. deutsch. bot. Gesellsch. Band XV. 1897.) 55. — Uber die Algenvegetation von Helgoland. (Verhandl. d. k. k. zool. bot. Ge- sellsch. in Wien. 1897.) 56. — Meeresalgen von Sermitdlet- und kleinen Karajakfjord (Bot. Ergebn. der v. d. Gesellsch. f. Erdkunde zu Berlin unter Leitung Dr. v. Drygalski's ausgesandten Gronlandsexpcdition nach Dr. Vanhoffens Sammlungen bearbeitet. A. Krypto- gamen. Stuttgart. 1897). 57. Landt, Jørgen, Forsøg til en Beskrivelse over Færoerne. Kjobenhavn 1800. 58. Le Jolls, Auguste, Liste des algues marines de Cherbourg. Paris et Cher- bourg. 1863. 59. Lj^ngbye, H. C, Tentamen Hydrophj'tologiæ Danicæ. Hafniæ 1819. 60. — Rariora Codana (Videnskab. Medd. fra d. naturh. Foren, i Kjobenhavn 1879 — 80. 61. Mohn, H., Den Norske Nordhavs Expedition. 2. Bind. Kristiania 1883. 62. Oltmanns, Friedrich, Ueber die Cultur und Lebensbedingungen der Meeres- algen. (Pringsheim's Jahrb. Band 23, Berlin 1892.) 63. — Notizen iiber die Cultur- und Lebensbedingungen der Meeresalgen (Flora, 1897). 64. Ostenfeld, CH., Phyto-geographical studies based upon observation of »Pha- nerogamæ and Pteridophyta« (Botany of the Færoes, Part I). 65. — Phytoplankton from the sea around the F"æroes (Botany of the Færoes, Part II). 66. Porsild, M. P., og H.G. Simmons, om Færoernes Havalgevegetation og dens Oprindelse. En Kritik. Bot. Notiser. Lund 1904. 67. Rattray, John, The distribution of the marine Algæ of the Firth of Forth. (Transactions of the bot soc.. Vol. 16. Edinburgh 1886. p. 420.) 68. Reinke, J., Algenflora der westl. Ostsee deutschen Antheils. (VI. Bericht der Kommission zur Untersuchung der deutschen Meere. Kiel 1889.) 69. Rosenvinge, L. Kolderup, Grønlands Havalger. (Meddelelser om Grøn- land III.) 70. — Deuxiéme mémoire sur les Algues marines du Groenland. (Meddelelser om Grønland, 20.) 71. — Om Algevegetationen ved Grønlands Kj^ster. (Meddelelser om Grønland, 20.) (Sur la vegetation d'Algues marines sur les cotes du Grønland (ibid. p. 339)). 72. Rostrup, E., Færoernes Flora. (Botanisk Tidsskrift. Bind 4. Kjobenhavn 1870—71.) 73. Ryder, C, Some investigations relating to the ocean currents in the sea be- tween Norway, Scotland an Greenland. (Særtryk af det danske meteorologiske Instituts nautisk-meteorologiske Aarbog 1901.) 834 74. H3'dcr, C, Continuation of the abovc-mentioned treatise. (Nautisk-meteoro- logisk Aarbog 1904.) 75. Sauvageau. C, Note préliminaire sur les Algues marines du Golfe de Gas- cogne. Journal de Botanique, Tome XI. 1897. 76. — Sur quelques Myrionemacées. Annales des sciences naturelles. 8 serie. Bo- tanique, tome V. Paris 1897. 77. — Remarques sur les Sphacelariacées. Journal de Botanique. tome XIV, 1900 and tome XV, 1901. 78. Si m mons, H. G., Zur Kenntnis der Meeresalgen-Flora der Fåroer. Hedwigia, Band 36, 1897. 79. Svedelius, Nils, Studier ofver Ostersjons hafsalgflora. Akademisk Afhand- ling. Upsala 1901. 80. Stromfelt, H. F. G., Om algvegetationens vid Islands kuster. (Akademisk Afhandling. Goteborg 1886.) 81. Traill, G. W. , The marine Algæ of the Orkney Islands. (Transactions of the bot. Soc. of Edinburg. Vol. 18, 1891.) 82. Warming, Eug.. Plantesamfund. Grundtræk af den økologiske Plantegeografi. Kjøbenhavn 1895. 83. — The history of the flora of the Fæi'oes. (Botany of the Færoes. Part. II, p. 660— 681.) Copenhagen 1903. 84. Willaume-Jantzen: Færøernes Klima. (Geografisk Tidsskrift, 15. Bind). 85. Wille, N., Bidrag til Algernes physiologiske Anatomi. (K. Svenska Vetenskaps- Akademiens Handlingar, Band 21. Stockholm 1885.) 86. — Om Færoernes Ferskvandsalger og om F'erskvandsalgernes Spredningsmaader. (Bot. Notiser, Lund 1897.) 87. Orsted, A. S., De regionibus marinis. Elementa topographiæ historico-naturalis freti Øresund. (Diss. Hauniæ 1844.) 88. — Om Lovene for Farvefordelingen hos Dyrene i Havets forskjellige Dybder. (_Vidensk. Medd. fra d. naturh. Forening. 1849.) ADDITIONS AND CORRECTIONS TO THE LIST OF THE PHANEROGAMAE AND PTERIDO- PHYTA OF THE FÆROES c. H. OSTENFELD. SiNCE the publication in 1901 of my list of the flowering piants and ferns found in the Færoes^ several investigators have con- tributed to the flora. The result is not a few additions to the list together with new localities of already known species. Further, a reexamination of the material of several critical species has caused alterations in my views of their value, and besides that some spe- cialists have examined Færoese specimens of several genera, e. g. Mr. H. Dahlstedt the genus Taraxaciim and Mr. W. Becker (of Hedersleben) the genus Viola. For all these reasons I feel it ne- cessary to publish a corrected list of the vascular piants, containing the additions and corrections to the former list. For convenience I follow the same arrangement of the piants as in the main list, i. e. the alphabetical arrangement of the species within a genus, of the genera within a family (order), and of the families within the large groups: Sympetalae, Choripetalae , Monocotyledones. Gymno- spermae and Pteridophyta. With regard to the nomenclature I have used the same names as in the main list, with very few exceptions. 1 do not find it convenient to have different names to the same plant in the diff"erent parts of one work, so the changes are ex- tremely few. I have in the following list repeated all the species names, consequently also the names of piants to which I have nothing to add or correct. The list will therefore contain the names o f a 1 1 the flowering p 1 a n t s and ferns h i t h e r t o f o u n d in the Færoes. I have to mention the main contributors to this list. An Ame- rican lady Miss Elisabeth Taylor has for several years lived 1 See Vol. I pp. 41— 99. Botany of the Færoes. 54 836 in the Færoes; she has collecled many planls in Ihe different is- lands, especially in the Nordreoer. A member of the teaching staff of the Highschool at Fagralid on Bordo, Mr. R. Rasmussen is an eager and successful botanical explorer; he has investigated some parts of Bordo, Kuno and Kalso very thoroughly. Further I may men- tion one inhabitant more, Mr. Gazet Patursson, who has found many interesting piants especially in the southern part of Stromo. In 1903 I visited the islands again and studied the vegetation in the neighbourhood of Thorshavn (Slromo), but the floristic result was very poor, because this place is most often visited. In the literature after 1900 I have found only one book, in which there are remarks on the Faeroese flora, viz. W. Bisiker: Across Iceland, With illustrations and maps and an appendix by A.W. Hill on the piants collected. London (E. Arnold), 1902. The Appendix II (pp. 226 — 231) to this book is »a list of piants collected in the Faroes and Iceland in June and July 1900«. The piaces in the Færoes vi- sited by the authors are Trangisvaag (Sydero), Thorshavn (Stromo) and Klaksvig (Bordo), just the ordinary stopping-places on the way to Iceland, and all visited and investigated by many botanists be- fore. The list gives most of the plant-names without indication of which of the three piaces is the locality, only »Faroes« is added to the name, and it contains both Icelandic and Færoese species together. All this makes it probable, that the records of Draba verna, Viola ericetorum (= V. canina), Dryas octopetala, Geiim rivale and Galiiim uliginosum are incorrect, as none of these species have been found before in any of the three visited piaces, and Draba verna and Viola canina are not elsewhere recorded from the Færoes. All these spe- cies are not rare in Iceland. I therefore consider the records as incorrect and do not take them up in my list. It remains only to refer to the paper by Mr. H. Dahl stedt on the Færoese Hieracia^. He describes herein all the species found in the Færoes as new. A good many of them I refound in 1903 and was able to determine correctly, as shown by a later revision of my material by Mr. Dahlstedt. I take here the opportunity of thanking Miss Elisabeth Tay- lor, Mr. G. Patursson and Mr. R. Rasmussen for their valuable contributions to the flora, and Mr. H. Dahlstedt and Mr. W. Becker for assistance with regard to the Taraxaca, Hieracia and Violæ. I also wish to give my hest thanks to Mr. W. H. Beeby for his kind 1 See Vol. II, pp. 625— (559, pi. XI— XII. 837 help in coirecling llie English in this paper and for his friendly leaving at my disposal his large collection of Shetland piants. The abbreviations used are explained in the following manner: Str. — Stromo. Ost. = Ostero. Syd. = Sydero. Vid. = Videro. G. P. = Gazet Patursson. R. R. — R. Rasmussen. E. T. = Miss Elisabeth Taylor. I found by myself. I have examined all the specimens recorded in the list, unless there is added »according to«. The species of the following list to which a number is prefixed are truly wild-growing or quite naturalized, while a f means that the species is introduced lately and at present not completely naturalized. I. Dicotyledones. A. Sympetalae. Fani. I. BORRAGINACEAE. t Anchusa arvensis (L.) M. Bieb. 1. Mertensia maritima (L.) D. C. 2. Myosotis arvensis (L.) Roth. 3. M. palustris (L.) Roth., var. strigiilosa Rehb. Tlie specimens collected by H. G. Simmons at Ejde (Ost.) in 1895 C-^/s) bear ripe fruits. 4. M. repens Don. 5. M. versicolor (Pers.) Sm. Fam. II. CAMPANULACEAE. 6. Campanula rotundifolia L. Str.: Vestnianliavn (according to G. P.). Fam. III. COMPOSITAE. 7. Achillea millefolium L. 8. A. ptarmica L. 9. Bellis perennis L. t Chpysanthemum leucanthemum L. Str.: sparsely in a cultivated grass-field at Kirkebo (G. P., 1904). t Cirsium arvense (L.) Scop. 10. C. palustre (L.) Scop. Also found in the Nordreoer: Kai so: Husum and otlier piaces (ac- cording to R. R.y, Videro: Viderejde, in enclosed fieids (E. T.). 54* 888 11. Gnaphalium supinum L. 12. Hieracium ardisodon Dahlst. Sir.: Sandej*ær(ie near Thorshavn (!). 13. H. ciliolatum Dahlst. 14. H. constrictiforme Dahlst. 15. H. cordifrons Dahlst. Str. : Thorshavn, at the rivulet; Havnedal, at the rivulet; Syderdal(!). t H. danicum Dahlst., K. Svenska Vetensk. Akad. Handl., Bd. 25, No. 3, 1893, pp. 118 & 120. H. viilgatum Hntegrifoliiim Joh. Lange, Haandbog i den danske Flora, ed. 1, 1851, p. 456; H. integrifolium Lange, ibidem, ed. 2, 1856—59, p. 520: Flora Danica, tab. 2661 ; non H. integrifolium Fries, Herb. Normale, fase. XII, No. 23, 1846, nec Symbolæ Hierac, 1848, p. 117. In a little plantation in the enclosed fields around Thorshavn I found in 1903 a Hieracium, which I could not identify with any of the Færoese species described by Mr. Dahlstedt. I therefore sent some specimens of it to Mr. Dahlstedt, who determined them as H. dani- cum, adding, that the species probably had been accidentally in- troduced to the Færoes. His supposition is without doubt right, as will appear from the following: In 1898 the direction of the Botanical Garden of Copenhagen sent different young shrubs to Torshavn for planting and some of them were forwarded to Mr. Jone Isaksen, to whom the small plantation in question belongs. As H. danicum occurs in the shrubberies and carpets in the Bota- nical Garden of Copenhagen, it is but natural to suppose, that some fruits have been carried with the sending of shrubs to Thorshavn and have grown here. — It will be interesting to follow this intro- duction, of which we know the date so exactely. 16. H. epileucoides Dahlst. 17. H. epileucum Dahlst. 18. H. faeroense Dahlst. Str.: On grassy slopes at Thorshavn and Sandegærde (!); specimens from tilis locality have been issued in I. Dorfler, Herbarium nor- male, No. 4550. 19. H. Hartzianum Dahlst. 20. H. heterophyllum Dahlst., and var. pinnatifrons Dahlst, 21. H. kalsoense Dahlst. 22. H. leucograptum Dahlst. Str.: On grassy slopes al Hojvig near Thorshavn (!). 23. H. melanochrotum Dahlst. 839 24. H. Ostenfeldii Dahlsl. 25. H. pepamplifopme Dahlst. 26. H. peramplum Dahlst. Str. : Thorshavn, at the rivulet (J). 27. H. perintegrum Dahlst. Str.: Thorshavn, at the rivulet (!). 28. H. sarcophylloides Dahlst. 29. H. scoticifopme Dahlst. Str.: On grassy slojie at Hojvig near Thorshavn (!); specimens deter- mined by Mr. H. Dahlstedt. 30. H. Simmonsianum Dahlst. 31. H. subpubicundum Dahlst. Str.: Thorshavn, at the rivulet (!). 32. H. vetepascens Dahlst. Str.: Thorshavn (Th. Mikkelsen); Hojvig (!); Syderdal (!). var. eidense Dahlst. 33. Leontodon autumnale L., and var. Taraxaci (L.). 34. Matpicapia inodopa L., var. phaeocephala Rupr. 35. Senecio vulgapis L. t Sonchus apvensis L. 36. Tanacetum vulgape L. 37. Tapaxacum spectabile Dahlst. in Bolaniska Notiser 1905, p. 159. var. faeroense Dahlst., nov. var.; Syn. T. pahislre Rostrup, Fær- oernes Flora, 1870, p. 51, non D. C; T. laevigatum Ostenfeld, Botan. Tids., Bd. 22, 1898, p 139, non D. C; T. croceum Ostenfeld, Bot. Fær- oes, vol. I, 1901, p. 45, non Dahlstedt. A formå primaria diversiim foliis viilgo (semper in specim. ciil- tisj solo arcte (idpressis, pedicellis vulgo foliis brevioribiis + deciim- bentibiis, involiicris angustioribus basi ovato-descendente, s quo mi s exterioribus angustioribus nec non ligulis angustioribus calathioque flosculis paucioribus magis radiante. Mr. H. Dahlstedt of Stockholm has lately examined all the Færoese specimens of Taraxacum preserved in the Botanical Mu- seum of Copenhagen and has given me the results of his studies. The species, which I in my list had taken as H. croceum Dahlst., is according to Mr. Dahlstedt a special Færoese variety of his species H. spectabile. He has sent me the description just quoted and also all the following remarks concerning the Taraxaca. The T. spectabile, var. faeroense Dahlst. has been collected in the following piaces: 840 Bordo: Klaksvig (A.Poulsen, 188G); Str.: Kirkebo. Thorshavn and Sandegærde (C. A Feilberg & E. Rostrup, 1867); Sj^d.: Tværaa and Trangis- vaag, rock-ledges at Kvanhaugen (J. Hartz, C. H. Ostenfeld & E. Warming, 1897); Videro: Malinsfjæld, 300 M. (J.Hartz & C.H.Ostenfeld, 1897). Mr. Dahlstedt remarks: »The characters given above are on ly re- lative, and they are such as might possibly be induced by the place of growth, but that is not the case. I have had specimens raised from fruits from Sydero under cultivation through several genera- tions, and they keep themselves distinct from cultivated specimens of the main form (from Sverige, Jemtland)«. var. maculiferum Dahlst., nov. var. A formå primaria et a var. præcedente satis diuersiim foliis in pagina superiore + pnrpureo-macalatis squamisque paullo laxias adpressis. Syd.: Tværaa and rock-ledges at Kvanhauge (J. Hartz & C. H. Osten- feld, 1897); Str.: Havnedal near Thorshavn (C.H.Ostenfeld, 1903). Col- lected in the F"ær6es as early as 1821 (by Trevelyan'?), as there are specimens in the Riksmuseum of Stockholm, labelled: »Færoe, 1821, about 2000 feet above the Sea.« [Also coUccted on St. Kilda, Outer Hebrides by O. Paulsen, 1905]. »This variety differs only from the foregoing form by the cha- racters given in the diagnosis, but as it seems that both characters always are combined, it is probable, that they are persistent. — The variety in question is of special interest, as it in many re- spects reminds us of the nearly related species T. naevosum Dahlst.« 38. T. naevosum Dahlst., nov. spec. Folia dense et tate lobata — pinnatifido-lobata, tobis latis — an- gnstis deltoideis, iitrinqiie v. præcipne in luargine superiore ± dentata inferne angnstius lobata, inter lobos in ferne + irregntariter dentata, lobo terminali satis brevi, lato ovato-triangntari — hastato, snpra pur- pureo- V. atropurpnreo-macnlata (v. in umbrosis fere emacnlata) et in pagina superiore vulgo pitis crassis articulatis sparsis — densiusculis obtecta, petiolis + intense purpureis nervoque mediano inferne v. pro max. parte purpnreo. Inuolucrum + obscure oleraceum, magnum satis longum, squa- mis exterioribus longis lanceolatis breve acuminatis, supra medium inv. attingentibus, anguste v. inconspicue marginatis, apice + purpureis V. fuscopurpureis et in pag. interiore superne + fuscoviolaceis, laxe ad- pressis V. erecto-patentibus, apicibus vulgo + recurvato-patentibus, inte- rioribus sub apice + purpureo leviter callosis. Calathium obscure luteum, multiftorum, radians. 841 Ligulæ longæ, marginales latiiisculæ, extiis stria lata rnbro-piir- purea vittatæ, dentibiis in lig. omnibus + rnbris. Antheræ polline + repletæ. Stylus et stigmata livescentes. Achenium fusco-stramineum apice muricato-spinulosum, cæterum fere læve v. minute tuberculatum, c. 4 mm. longum, 1 mm. latum, py- ramide c. 0,9 mm. longo, rostro 8 — .9 mm. longo et pappo albo. Syd.: Ravine near Kvanhauge (J.Hartz & C.H.Ostenfeld, 1897). Geogr.area: Regio alpina and subalpina of the Scandinavian mountains. Mr. Dahlstedt writes: »The form found in a ravine near Kvanhauge seems to me to be the same as T. naevosum Dahlst., a hitherto undescribed species, which is rather comnion in the al- pine and subalpine parts of the Scandinavian mountains. The in- volucres and the fruits of the Færoese form are quite like those of the Scandinavian one, but the leaves differ in lacking the purpie- spots and the coarse hairiness on the upper side. In Scandinavia similar forms have been found here and there and ahvays growing in shadow; they want the spots, but very seldom the coarse hairs, although sometimes very sparsely occurring. T. naevosum differs from T. spectabile, var. maculiferum by longer, more patent outer phyllaries, shorter achenes, shorter beak (pyramid), but longer rostrum and developed, more or less abundant pollen. var. (?) bipinnatifidum (Rostr.) Dahlst.; T. obliquum Fr., var. bi- pinnatifidum Rostrup, Færoernes Flora, 1870, p. 51. A formå primaria foliis valde et prof unde laciniatis laciniis del- toideis valde laciniato-dentatis omnibus in apicem elongatum attenuatis. Sando: At the dune at Sandsbugt (C. A. Feilberg & E. Rostrup, 1867). »Of this very peculiar form only three fruiting individuals have been collected, and it is therefore difficult to examine the characters of the involucre and the phyllaries. It seems to agree with the main species as regards the phyllaries. The achenes also are simi- lar to those of the main species, but they are somewhat longer and more coarsely denticulate towards the top. The leaves bear the same characteristic hairiness of the upper side as the Scandinavian species of the main form. Of this last specimens from open and sunny piaces have deeply and narrowly laciniate leaves, which come rather near to the leaves of the Faeroese form , but are easily di- stinguable by their wanting the laciniation of the second order (»bipinnati fidus«), highly characterizing the Færoese form. This diffe- 842 rence is too great to justify us in regarding the form as a merely growing-place-form. I suppose that it is a race adapted to living in the sandy soils of the dune. Nevertheless closer examination of new material and also cultivation is highly desirahle. Flowering specimens are necessary to settle the question of the true (correct) systematic place of it.« 39. Tussilago farfarus L. Fam. CONVOLVULACEAE. t Convolvulus sepium L. Fam. IV. DIPSACACEAE. 40. Succisa pratensis Moench. t Trichera arvensis (L.) Schrad. Fam.V. ERICACEAE. 41. Calluna vulgaris Salisb. 42. Erica cinerea L. 43. Loiseleuria procumbens (L.) Desv. Str.: Kirkebo Rejn from c. 250— 300 M. (Lyngbye, Rostrup, G. P., !). Fam. VI. GENTIANACEAE. 44. Gentiana campestris L., subsp. islandica Murb., mostly »ad subsp. germanicam Murb. accedens«. 45. Menyanthes trifollata L. Fam. VII. LABIATAE. 46. Brunella vulgaris L. 47. Galeopsis tetrahit L. None of the collected specimens seems to belong to G. bi/ida Boenn. f Lamium dissectum With. t L. intermedium Fr. t L. purpureum L. 48. Mentha aquatica L. Sando: Sandslid (accord. to G. P.). The few specimens existing in our herbarium (from Vaago and Stromo) are quite sterile, without any flowers at all. 49. Thymus serpyllum L. Resides the main form with purpie corollas a form with rose-coloured (pink) flowers has been found on Fuglo, at about 550 M. (J. Hartz & C.H.Ostenfeld, 1897). 843 Fam.VIII. LENTIBULARIACEAE. 50. Pinguicula vulgaris L. 51. Utriculapia vulgaris L. Miss E. Taylor has discovered this interesting addition to the Færoese flora. She found it in 1904 and reinspected the locality in 1906, and in both summers she only saw specimens without any flowerstalks at all. Vaago: plenty in pools and ditches about half way from Midvaag to Sorvaagsvatn, in the outfields. Fam. IX. LOBELIACEAE. 52. Lobelia dortmanna L. Fam. X. PLANTAGINACEAE. 53. Litorella lacustris L. 54. Plantago coronopus L. Syd.: Cliffs on the west coast opposite Lopra (Ove Paulsen). 55. P. lanceolata L., and f. depressa Rostr. t P. major L. 56. P. maritima L. Fam. XI. PLUMBAGINACEAE. 57. Armeria elongata (HofFm.) Koch., mostly var. maritima (Mill.) Willd. Fam. XII. PRIMULACEAE. 58. Anagallis tenella L. 59. Lysimachia nemorum L. 60. Primula acaulis (L.) Jacq. Fam. XIII. PYROLACEAE. 61. Pyrola minor L. Kuno: Skard, both at the top of the mountain and in its lower parts, flowering and fruiting sparingly (R. R.); Vid.: Malinsfjæld, c. 400— 500 M.; Mornefjæld, c. 300 M.; and eastern ridge of Villingedalsfjæld; in all piaces with flowers (E. T.). Fam. XIV. RUBIACEAE. t Galium aparine L. 62. G. palustre L. 63. G. saxatile L. 844 Fani. XV. SCROPHULARIACEAE. 64. Alectorolophus groenlandicus (Chab.) Ostf. emend. 65. A. minor (Ehrh.) Wimm. c^ Grab. I hope to be able in another place to treat tbe northern species of Alectorolophus minor, sens. lal. 66. Bartschia alpina L. J. Landt in liis book on tbe Færoes (1801) mentions, tbat Bartschia occurs in tbe Nortb-Stromo, but as no later investigator had found it, and as tbe statements of tbe Rev. Landt are not al- ways correct, I bad omitted it in my bst. Now I bave got evi- dence of tbe correctness of its occurring in tbe Færoes, as botb Miss Taylor and Mr. R. Rasmussen bave sent me specimens. Yaago: near Sorvaag (sent to R. R. from a relative, Mr. Niclas Ras- mussen, but without more exact indication of the locality); Vid.: rock- ledge on Mornefjæld, ab. 300—400 M. (E. T.). In my list (I. c. pp. 55 — 56) I bave put down tbe result of my examination of tbe Færoese Euphrasias, but I expressed some doubt witb regard to tbe correctness of my determinations, as I must admit, tbat tbe bmits betv^^een tbe small-flowered forms were arbitrary. I bave taken up again tbe study of tbese interesting piants and bave had a large material at my disposal, consisting of tbe collections made by Mr. J. Hartz and myself in tbe Færoes in 1895, 1896, 1897 and 1903 togetber witb smaller collections from other investi- gators. As some of the species in question have been described upon specimens from Sbetland or Scotland, it was necessary to have material from tbese countries for comparison. We have in the Botanical Museum of Copenhagen a good many Eupbrasias from Sbetland and Scotland sent me from my friend Mr. W. H. Beeby and from the English monograpber of the genus the late F. Towns- end, and furtbermore Mr. Beeby bad sent me all bis Sbetland- Euphrasias on loan. I have therefore been able to decide several points of interest, baving in my bands specimens from tbe original locality and original collecting-date (»co-types« of the American bo- tanists) of E. foulacnsis Towns., E. paludosa Towns. (= E. scotica Wettst.) and E. horealis (Towns.) W^ettst.; besides tbe type speci- mens of E. arctica Lange and E. gracilis Fr., f. atropurpiirea Rostr. are in our herbarium. The following treatment will show, tbat the result of my ex- amination is a reduction of the number of species. I think, tbat it is because the authors bave had too scanty material at their di- 845 sposal, that they have described so many species — this is a com- mon fault by monographers working in a country far away, a fault which now-a-days often is committed with regard to northern spe- cies belonging to genera, monographed by Central-European botanists. 67. Euphrasia borealis (Towns.) Wettst. This is the only Færoese species with larger corollas, and it is therefore easily distinguished from the other species. On the other hånd it is very near E. brevipila Burnat et GremH, and is perhaps, as suggested by Mr. Townsend, only an eglandular from of it. 68. E. curta Fr., Wettst. Small specimens of this species are not rare beyond the en- closed fields in the lowlands; they are rather broadleaved and tend to approach E. latifolia Pursh, without giving place for any doubt with regard to their naming. 69. E. minima Jacquin apud Schleich.; R. Wettstein, Monogr. der Gatt. Euphrasia, 1896, p. 151; E. scotica Wettst., 1. c. p. 170; Townsend, Journ. Botany, 1897, p. 425; Ostenfeld, Bot. Fær., I, p. 56; E. pahidosa Townsend, Journ. Botany, 1891, p. 161, non R. Br. ; E.gracilis Ostenfeld, Bot. Færoes I, p. 56, non Fr.; E. latifolia Osten- feld, ibid. p. 56, non Pursh ; E. arctica Lange apud Rostrup, Færoernes Flora, Bot. Tids. IV, 1870, p. 47, ex parte (quoad plantæ færoenses). In 1891 Mr. F. Townsend published a note in the Journal of Botany on a new form of Euphrasia, which he named E. pahidosa. He had found it in boggy ground around Braemar in Scotland. In some respects it was like E. gracihs, but differed, among other cha- racters, in its small, whitish flowers, its short and rather broad capsule and its wet habitat. As R. Brown had applied the name pahidosa to another species (from Australia), R. Wettstein in his monograph (1896, p. 151) altered the name of Townsend's plant to E. scotHca Wettst. (more correctly spelt scotica), and under this name the plant has gone since (e. g. by Townsend, Monograph of the Bri- tish species of Euphrasia, Journ. of Botany, 1897, p. 425, and Eu- phrasia scotica, Journ. of Botany, 1903, p. 57). Wettstein considers E. scotica to be very near E. minima and says (1. c. p. 171), that the only difference of importance lies in the length of the capsule in proportion to the calyx, but adds that he does not know if this difference is constant. Shortly after Town- send (Monograph, p. 426) declares, that it is not constant, as he has found specimens of E. scotica with capsules exceeding the calyx; he says that »a marked distinction seems to lie in the form of the 846 upper leaves and bracts of E. scotica which are narrower thaii those of E. minima and have a cuneate base.« I have examined many hundred specimens of E. scotica from Scotland , Shetland and the Færoes and have compared them with many specimens of£. mznznja both from the Alps and from Scandinavia, and I can not find any distincHon which holds good. I feel pretty sure, that the Scottish elc. plant is identical with true E. minima. Townsend who has seen a good deal of my Færoese material, has determined many speci- mens with capsules exceeding the calyx as E. scotica, specimens which are quite like the typical E. minima from the Alps. As pointed out by R. Wettstein (1. c. p. 159) E. minima varies much with regard to the colour of the corolla; the true E. scotica represents a form with pale or whitish flowers (f. pallida Gremli), but from this we find all possible variations of colour until a form with dark purpie corolla (f. purpurascens Wettst. 1. c. p. 159). The main form is very common in the Færoes, and also the purple- flowered form occurs frequently in the Færoese heaths; it is the same form which has been described as E. foulaensis Towns, apud Wettstein (1. c. p. 139). I have examined Mr. W. H. Beeby's speci- mens from Hamnafeld on Foula, Shetland, upon which F. Town- send has made his description, and they are, after my opinion, only rather coarse, unbranched E. minima with dark purpie corol- las and long capsules ; the specimens were found among heather and this explains their somewhat flexuose stem. Both Wettstein (1. c. p. 140) and Townsend (Monograph, p. 423) compare it with E. lati- folia Pursh; but it is easily distinguised from it by its nearly glabrous leaves; common to both forms are the obtuse teeth of the leaves and bracts. The same form has been described in 1870 by E. Rostrup (Færoernes Flora, p. 48) as E. gracilis, f. atropurpnrea Rostr., which consequently is the name to be used. I have seen Rostrup's speci- mens (from Hesto) and found them almost identical with Beeby's specimens of E. fonlaensis. The synonymy of tlie form is then as tbllows: E. minima Jacq., f. atropurpnrea (Rostr. sub E. gracili); E. minima, f. purpurascens Welisi., Monographie, p. 159; E. Foulaensis Towns. ap. Wettstein, 1. c. p. 139; Townsend, Journ. of Botany, 1897, p. 422; E. atropurpnrea (Rostr.) Ostenfeld, Bot. Færoes, I, p. 55; E. gracilis Fr., f. atropurpnrea Rostrup, Færoernes Flora, Bot. Tids., IV, 1870, p. 48. E. gracilis Fr. is distinguished from it by its slender, erect, often 847 branched stem, longer inlernodes, small and quite glabrous leaves, rather short and small capsule and dark-coloured stem and leaves. 70. Pedicularis palustris L. 71. Veronica alpina L. Kuno: not rare at higli levels. f. i. Nakken (R. R.). 72. V. beccabunga L. f V. hederifolia L. 73. V. fruticans Crantz; V. saxatilis Scop. A very interesting discovery for the flora. Kai so: Mountain nortli of Husum, ab. 500 M. (E. T.). 74. V. offlcinalis L., and f. glabrata Fristedt. 75. V. serpyllifolia L. As mentioned in my list (1. c. p. 57) there occurs in the hills a variety: var. Borealis Læstad., which is easily distinguised from the main species by its denser clothing of the short raceme, lower growth and bright-blue corollas. The clothing consists of a pubescense of short crisped hairs, among which longer patent hairs, partly glan- dular, especially in the flower-stalks, occur. L. M. Neuman (Sveriges Flora, 1901, p. 132) considers this form as a separate species, viz. V. borealis (Læstad.) Neuman, and I should think he is right in so doing. It is probably identical with V. hnmifiisa Dickson. Besides from the Færoes, where this form is rather common in the hills, I have seen specimens from Lapland (collected by Læ- stadius), from the Alps, the Apennines and the higher mountains of North America (especially in British Columbia). It seems therefore to have a wide range of distribution. Fam. XVL VACCINIACEAE. 76. Vacclnium myrtillus L., and f. pijgmaea Ostf. 77. V. uliginosum L., and f. microphylla (Lge.). 78. V. vitis idaea L. S t r. : Between Leinum and Kvalvig, fruiting in Aug. 1 900 (accord. to E. T.). B. Choripetalae. Fam. XVIL CALLITRICHACEAE. 79. Callitriche autumnalis L. 80. C. hamulata Kutz. In my preliminary list (Botan. Tids., vol. 22, 1898, p. 140) I have reported C. pedunciilata D. C. from the Færoes, while in my list in Bot. Færoes, vol. I (p. 59) I have altered it to C. hamulata. I now 848 think that both opinions are correct, as I take C. pedunciilata as merely a variety of C. hamiilata, distinguished by its stalked friiits. Such a form: C. hamiilata Kiilz., L pedunciilata (D. C. pro sp.) has been found in S trom 6 (Miavevatn). 81. C. stagnalis Scop. Fam. XVIII. CARYOPHYLLACEAE. t Agrostemma githago L. 82. Alsine verna Bart., var. hirta (Wormskj.) Lge. 83. Cerastium Edmondstonii (Watson) Mm'b. & Ostf. Cerastiiim alpiniim has been recorded from Fugl 6 in my list (1. c. p. 60), but on closer examination I find it to be only a male form of C. Edmondstonii, which elsewhere ahvays occurs as a her- maphrodite. The male specimens from Fuglo differ from the type in the more branched inflorescence and the whole shape of the tufts, but the broad sepals and the narrowness of the white mem- branous margins of the bracts point decidely to C. Edmondstonii. With regard to the name of this species S. Murbeck (Bota- niska Notiser, 1898, p. 246) and I myself (1. c. p. 60) have pointed out that the oldest name must be C. latifoliiim, />, Edmondstonii Watson, London Botan. Soc. Catalogue of British Piants, 1844. The species name C. arcticiim given by Joh. Lange, Flora Danica, fase. 50, p. 7, 1880, to forms of this neighbourhood, can not be retained for Iwo reasons: P it is much later than Watson's name. 2^ Lange himself has mixed two distinct species together, as his description and the drawings in the plate are based upon partly C. Edmond- stonii from East-Iceland, partly C. alpiniim from Greenland (a con- densed dwarf form). From this it will be evident, that the name C. arcticiim in all cases should be omitted. 84. C. glomeratum Thuill. 85. C. tetrandrum Curtis, and var. zetlandiciim Murb. 86. C. trigynum Vill. 87. C. vulgare Hartm., mostly var. alpestre (Lindbl.). 88. Honckenya peploides (L.) Ehrh., and var. major Rostr. 89. Lychnis flos cuculi L. 90. Melandrium rubrum (Weig.) Garcke. Myggenæs (accord. to G. P.); Vaago: Midvaag, rock-ledge on the fowling clifTs (accord. to G. P.). 91. Sagina nivalis (Lindbl.) Fr. 92. S. procumbens L. 849 S. procumbens L. >( subulata (Sw.) Prsl. Str. : bare, gravelh', damp piaces between Thorshavn and Hojvig (!, 1903); Syd.: Frodeby (E. Rostrup & C. A. Feilberg. 1867). Probably not rare. 93. S. subulata (Sw.) Prsl. 94. Silene acaulis L. 95. Spepgula arvensis L. An examinalion of the Færoese specimens with mature frnits showed, that they all belong to S. sativa Boenn. The localities of the examined individuals are: Str.: Thorsvig (J. Hartz & C. H. Ostenfeld, 1897\ Thorshavn (E. Rostrup & Feilberg, 1867, CH. Ostenfeld, 1903;; Ost.: Ejde (E. T.). 96. Stellaria media (L.) Cyril. As an unusual locality the following is worthy of notice: Vid.: Vil- lingedalsfjæld, in a hole under a hammer, ab. 400 M. (E. T.). 97. S. uliginosa Murr. Fam. XIX. GHENOPODIACEAE. 98. Atriplex Babingtonii Woods. Mr. O. Paulsen has examined all the specimens of Atriplex collected in the Færoes, and he supposes, that all belong to A. Babingtonii. As his determinations are deciding with regard to all fruiting individuals, I think it better to omit the two other species from the flora, viz. A. hastata and A. patula, at least until further researches alter the question. The species is common on the coasts; specimens with fruits have been collected as follows: Bordo: Bordovlg (J. Hartz & C. H. Ostenfeld, 1897); Str.: Sande- gærde (E. Rostrup & C. A. Feilberg, 1867), Thorshavn (J. Hartz & C. H. Osten- feld, 1897; O.Paulsen, 1903). Thorsvig (J. Hartz & C. H. Ostenfeld, 1897); Syd.: Trangisvaag {E. Rostrup & C. A. Feilberg, 1867), between Trangis- vaag and Tværaa (C.H.Ostenfeld, 1895); Vaago: Sorvaag (E.Rostrup & C. A. Feilberg), Midvaag (J.Hartz & C.H.Ostenfeld, 1897). t Chenopodium album L. Fam. XX. CORNACEAE. 99. Cornus suecica L. Fam. XXI. CRASSULACEAE. 100. Sedum rhodiola D. C. 101. S. villosum L. 1^ 850 Fam. XXII. CRUCIFERAE. Arabis alpina L. As stated in my list (1. c. p. 66) Trevelyan records it from the hills of Kuno, but as this island lately has been much investigated by Miss E.Taylor and Mr. R. Rasmussen without refinding the Arahis, I doubt the statement of Trevelyan. 102. Arabis petraea (L.) Lam. t Brassica campestris L. t B, napus L. t B. nigra (L.) Koch. 103. Cakile maritima Scop., var. latifolia (Poir.) 104. Capsella bursa pastoris (L.) Moench. 105. Cardamine hirsuta L., both var. campestris Fr. and var. silvatica (Lk.). 106. C. ppatensis L. 107. Cochlearia officinalis L., coll. As pointed out in my list (I. c. p. 67) the Cochlearias of the Fær- oes are very varying, but I had not succeeded in discovering dis- cernible limits. During 2 — 3 years I have had several Cochlearias under cultivation in the Botanical Garden of Copenhagen and among them some raised from seeds collected in the Færoes. The results were, that many distinct forms occur, which keep their special characters from generation to generation, but it seems very difficult, if possible, to express these distinctions verbally. Consequently my opinion is, that at present we are not able to distinguish the elementary spe- cies of Cochlearia, but must retain the old collective name C. offici- nalis. I do not doubt, that many forms will be segregated in fu- ture, and I think the English and Scottish botanists are on the right way, when separating C. alpina H. C. Watson, C. micacea Marshall and C. groenlandica »L.« from the common C. officinalis »L.« I have compared authentic specimens of C. micacea Marshall (see Journ. of Botany, vol. XXXII, 1894, p. 289, pi. 345, 346; cfr. Journ. of Botany, XXX, 1892, p. 225, pi. 326 A) with Færoese specimens grown in bare gravelly piaces in the hills, and I have found them quite like; but on the other hånd there are specimens which do not match any of the above mentioned »species« in their typical shape, but are intermediate-looking. 108. Draba hirta L., f. riipestris (R. Br.). 109. D. incana L. t Raphanus raphanistrum L. 851 f Sinapis alba L. f S. arvensis L. 110. Subularia aquatica L. As the localities in my list (1. c. p. 67) through a fault in the printing are incomplete, I give them all here again: Sando: lake between Skopen and Sand; Str. : Leinumvatn and Miavevatn, lake on Vardebakken, at ab. 300 M.; Havnedal, a little pool at ab. 200 M.; Ost.: Kornvatn near Næs, at ab. 100 M.; Toftevatn. Fani. XXIII. DROSERACEAE. 111. Drosera rotundifolia L. Pam. XXIV. EMPETRACEAE. 112. Empetpum nigrum L. Pam. XXV. GERANIACEAE. t Geranium molle L. 113. G. silvaticum L. Pam. XXVI. HALORAGIDACEAE. 114. Myriophyllum alterniflorum D. C. Pam. XXVII. HYPERICACEAE. 115. Hypericum puichrum L., f. prociimbens Rostr. 116. H. quadrangulum L. Str.: near Hojvig (Th. Mikkelsen); Vaago: refound on the old lo- cality at Sorvaag (E. T., 1906). Pam. XXVIII. LINACEAE. 117. Linum catharticum L. Pam. XXIX. OENOTHERACEAE. 118. Chamaenerium angustifolium (L.) Scop. Bordo: Fagralid, rock-ledge, sparingly and not producing any flower during the last five years (accord. to R. R., 1901—05); Kai so: Husum, ravine, flowering (accord. to R. R.). 119. Epilobium alsinifolium Vill. 120. E. anagallidifolium Lam. E. anagallidifolium Lam. X palustre L. 121. E. iactiflorum Hausskn. 122. E. montanum L. 123. E. palustre L. Botany of the Færoes. 55 852 Fam. XXX. OXALIDACEAE. 124. Oxalis acetosella L. Recorded from the Færoes by Trevelyan, but not refound until Mr. R. Rasmussen in 1904 discovered it on Kuno. He has laler found it in several piaces with flowers and fruits; he writes to me: »Oxahs occurs everywhere in the ravines and talus of débris from Nakken (the northern cape) to Skardsgjov, but only in the eastern half of the island«. Fam. XXXI. PAPAVERACEAE. 125. Papaver radicatum Rottb. Kuno: on the hill-plateau at the southern end of the island (J. Hartz & C.H.Ostenfeld); everywhere in the hills of the eastern half of the is- land (accord. to R. R.). Fam. XXXII. PAPILIONACEAE. 126. Lathyrus pratensis L. Myggenæs, and Vaago: Midvaag (accord. to G. P.). 127. Lotus corniculatus L., f. carnosa (Pers.). Kuno: Skard, and Kalso: Husum (accord. to R. R.). t Pisum sativum L. t Trifolium hybridum t T. pratense L. t T. ppocumbens L. 128. T. repens L. 129. Vicia cracca L. Rordo: F'agralid, in a ravine (accord. to R. R.); Str. : Kirkebo, on two spots in the outfields (accord. to G. P.); Vid.: Viderejde, in the en- closed field (E. T.). Fam. XXXIII. POLYGALACEAE. 130. Polygala serpyllacea Weihe. 131. P. vulgaris L., var. Ballii (Nyman) Ostf. Fam. XXXIV. POLYGONACEAE. 132. Koenigia islandica L. 133. Oxyria digyna (L.) Campd. 134. Polygonum amphibium L. 135. P. aviculare L. t P. convolvuius L. 136. P. viviparum L. 137. Rumex acetosa L. 853 f R. acetosella L. t R. crispus L. 138. R. domesticus Hartm. R. domesticus Hartm. X obtusifolius L. 139. R. obtusifolius L., f. agrestis Fr. Fam. XXXV. PORTULACACEAE. 140. Montia lampposperma Cham.; Syn. M.rivularis auctt.,nonGmel. In my list (I. c. p. 73) I have mentioned, that all the Montias from the Færoes belong to the form with finely netted and shiny seeds, and that this form is more northern than M. minor Gmel. I have foUowed most of the botanists in naming this form »Montia rivularis Gmel.«, but this name is not correct according to the opi- nion of H. Lindberg. He has pointed out (Medd. af Soc. pro F'auna et Flora Fenn., vol. 27, 1901, pp. 18—21), that the differences between the tw^o species described by G. G. G mel in in his Flora Badensis (1806) are based only upon the vegetative parts of the piants, and the characters taken from the testa of the seeds are not at all mentioned. Upon this last character Ad. de Chamisso has first laid emphasis, when he in Linnæa (1831, p. 565) described his new species M. lamprospenna; this author says, that the seeds of M. minor and M. rwiilaris have just the like structure, and piaces his new species as a contrast to them. This view is after H. Lind- berg — with the opinion of whom I agree — the correct one, and consequently the Montia which occurs in Greenland, Iceland, the Færoes, Scandinavia etc. must bear the name M. lamprosperma Cham. It consists of a smaller (annual?) form — the typical — and a larger perennial water-form : var. horeo-rivularis Lindb. fd., both found in the Færoes^. Fam. XXXVL RANUNCULACEAE. 141. Caltha palustris L., and var. radicans (Forst.). 142. Ranunculus acer L., and f. pumila (Whbg.). This species varies very much after its much varying habitats. In luxuriant rock-ledges a tall and robust form with the stem pa- tently stiff-hairy beneath (f. velutina Lindbl.) is met with. In bare gravelly piaces in the hills the f. pumila Whbg. is the substitute. This last form is — after the description — probably the same as ^ The more southern species, M. minor Gmel., with smaller, tuberculate-netted, somewhat opaque seed-testa has a parallel development, the water form of which must be named M. minor Gmel., var. rivularis (Gmel.) Lindb. fil. 55* 854 R. icelandicus (should be R. islandiciis) puhlished by Davis (Minne- sota Botan. Studies, IV, 1900 p. 472, Nalive and cultivated Ranun- culi of North America and segregated genera) upon specimens col- lected by Miss E. Taylor in Seydisfjord in Easl-Iceland (which does not belong to North America!). 143. R. auricomus L. Recorded from the Færoes by Trevelyan and rediscovered by Mr. G. Patursson in 1904. Str.: Kirkebo, the lowest ledge of the hill above Kirkebo. t R. ficapja L. The Rev. J. Landt mentions this species from Kirkebo (Str.), and Mr. G. Patursson has seent me pretty and large specimens taken April 10., 1904 in full flower. It occurs on the churchyard and around the ruins of the cathedral; without doubt it has been introduced in former times. 144. R. flammula L. My form speciosa (1. c. p. 74) resembles in many respects the Scottish species R.scoticiis Marshall (= R. peliolaris (Lange) Marshall, Journ. Botany, XXX, 1892, p. 289, pi. 328), but ditTers in the more numerous, larger and broader petals a. o. 145. R. glacialis L. 146. R. repens L. 147. R. reptans L. I should prefer to take this as a distinct species and not as a form of R. flamnuila, as I have done in my list (1. c. p. 74). 148. Thalictrum alpinum L. Fam. XXXVII. ROSACEAE. 149. Alchimilla acutidens Buser; Syn. A. Wichiirae Buser. 150. A. alpina L. 151. A. faeroénsis (Lange) Buser. In J.Dorfler's Herbarium normale, Cent. XLVII, Nr. 4654, Ice- landic specimens of this inleresting species have just (Dec. 1906) been distributed, and the well-known authority in Alchimilla Mr. R. Buser of Geneva has thereof taken the opportunity of giving a rather exhaustive list of citations concerning the form in question. He is of the opinion, that it is a subspecies o( A. splendens Christ, to which it is very near-allied, but its quite peculiar geographical distribution removes it so much from the alpine A. splendens and its different races, that it deserves — after my opinion — specific rank. 855 152. A. filicaulis Buser, wilh var. dcmidnta Buser and var. ve- slila Buser. The var. denmlala has becn found on Vid.: Villingedalsfjækl, ab. 500 M. (E. T.). 153. Dryas octopetala L. Kuno: Gjovadal ncar Skard (U. R.); Vid.: Mornefjæld (accord. to E.T.). 154. Geum rivale L. Str. : Ravnabjorginisoiilh of Kirlccbo and rock-ledge near Kirkeb6(G.P.). 155. Potentilla anserina L. 156. P. erecta (L.) Dalla Torre. 157. P. palustris (L.) Scop. Rediscovcrcd in ils old locality given by Mohr on Vaago: Gaase- dal (E. T.). 158. P. verna L.; P. maculata Pourr. 159. Rosa mollis Sm. 160. Rubus saxatilis L. 161. Sibbaldia procumbens L. 162. Spiraea ulmaria L. Fam. XXXVIII. SALICACEAE. 163. Salix glauca L. Kuno: Skard (accord. to R. R.; the locality is perhaps the same as »Hills on the south side from about 250 m.«); Vid.: Ormedalen (E. T.). 164. S. herbacea L. 165. S. phylioifolia L. Fam. XXXIX. SAXIFRAGACEAE. 166. Saxifraga caespitosa L.; S. decipiens Ehrh. 167. S. hypnoides L. 168. S. nivalis L., and var. tenuis Whbg. 169. S. oppositifolia L. 170. S. rivuiaris L. 171. S. stellaris L. Fam. XL. UMBELLIFERAE. t Aegopodium podagraria L. Str.: The churchyard and gardens in Kirkebo (G. P.); gardens in Velbestad (accord. to G. P.); a plantation in Thorshavn (!). 172. Angelica silvestris L. 173. Archangelica officinalis Hoffm. 174. Haloscias scoticum (L.) Fr. S a n d 6 : Troldhoved (accord. to G. P.). 856 Fam. XLI. URTICACEAE. 175. Urtica dioica L. f U. urens L. Fam. XLII. VIOLACEAE. Mr. W. Becker of Hedersleben (Germany) has examined Ihe material of Violæ from the Færoes and has kindly sent me Ihe following notes. 176. Viola palustris L. 177. V. silvestris (Lam.) Rchb., var. nov. rotiindato-crenata Becker; Syn. V. Riviniana Ostenfeld, Bot. Færoes, I, p. 80, non Rchb. Folia plerumqiie rotiindato-cordata,subacuminata, rotiindato-crenata. Mr. Becker means that the Færoese form of the Silvatica-group does not exactely match V. silvestris, var. Riviniana, but represents an intermediate stage between V. silvestris, typica and var. Riviniana. From his letter I quote: »it is especially noteworth with regard to the inferior systematical rank of the V. Riviniana, that there occur in islands forms which come more or less ncar to V. Riviniana or may be identical with it and at the same time show the characters of the true V. silvestris.« 178. V. tricolor L., subsp. nov. faeroensis Becker. Folia ovato-rotnndata, plane crenata, ad apicem obtnsissima, par- tim snbemarginata , ad basim in petiohim breviorem abrnpte vel sub- abrnpte angnstata. Stipnlae hjrato-incisae, lacinia terminalis lata ob- tnsa, laciniae laterales breves oblongae, obtusiusculae, introrsnm i — 2, extrorsum 3 — A. Sepala late lanceolata. To this description Mr. Becker add the following nole: »The form in question is characlerized by the broad-ovate leaves and by the stipules pinnatifid to a smaller degree than the main form (coming near to the stipules of Viola cornnta)«. In my list (1. c. p. 80) I had named the form »F. tricolor L. subsp. gemiina Wittr., formå« and had mentioned, that it is perennial. I should now like to name it: V. tricolor L., subsp. gemiina Wittr., var. faeroensis (Becker), as it without doubt belongs to the subspecies gemiina in Wittrock's sense. All the specimens from the Færoes are of this form. It is found in the southern part of Str. (Kirkebo and Velbestad) and on Sando (Sand, and other piaces according to Dr. Knud Poulsen), and it occurs always around the houses and in the enclosed fields. 857 II. Monocotyledones. Fam. XLIII. COLCHICACEAE. 179. Narthecium ossifragum (L.) Huds. 180. Tofieldia palustris Huds. New to the færoese flora. Bordo: Mountain between Kvannesund and Klaksvig (E. T., 1903). Fam. XLIV. CYPERACEAE. 181. Carex atrata L. 182. C, binervis L. C. caespitosa L. X Goodenoughii Gay should be ornitted, see under C. Goodenoughii. 183. C. dioica L. 184. C. echinata Murr.; C. stelliilata Good. 185. C. flacca Schreb.; C. glaiica Scop. C. flacca Schreb. X Goodenoughii Gay. The Rev. G. Kiikenthal has examined the specimens, which Mr. C. Raunkiær determined as the just mentioned new hybrid, and he does not believe in the hybrid nature of them. He writes: Nil nisi formå spicuHs abbreviatis Caricis Goodenoughii Gay. 186. C. flava L. C. flava L. X fulva Good. 187. C. fulva Good.; C. Hornschuchiana Hoppe. 188. C. Goodenoughii Gay. The specimens which in my list (1. c. p. 81) are named C. cae- spitosa X Goodenoughii do not belong to this hybrid, It would a priori also have been curious, if a hybrid should occur in a country where one of the parent species, at least at present, has not been found. The form in question must be referred to var. juncella F'ries, characterized by its densely tufted growth and the very narrow long leaves. I have in 1903 reexamined its onlj^ known locality in the Færoes, viz. : Syd.: near Tværaa, boggy place in the enclosed field (!). C. Goodenoughii Gay X rigida Good, 189. C, incurva Lightf. 190. C. Jeporina L. 191. C. Lyngbyei Hornem.; Syn. C. cryptocarpa C. A. Mey. 858 The name C. Liinohijei is older Ihan C. crijplncarpa, and as it has been pubHshed by Hornemann just on specimens from the Færoes (collectcd by H. C. Lyngbye), I feel it necessary in a Hst of Færoese piants lo prefer it. The type-specimens, which have been pictured in Flora Danica, tab. 1888, are in the herbarium of the Botanical Museum of Copenhagen. C. Lyngbye! Hornem. X rigida Good., f. siib-Lijnghijei G. Kliken Ihal. In a salt-marsh at the head of Trangisvaagfjord on Syd. I found in 1897 a curious robust Carex-form growing among C. Lyng- byei, C. Goodenoiighii and C. salina, subsp. kattegalensis. I took it at first as a hybrid between the two lirst named species, but could later not find a satisfying determination of it. Now the Rev. G. Kliken thai has examined it and given il the name placed above, remarking al foUows: »multo magis ad C. Lyngbyei acce- dens quam C. haematolepis Drej.^ sed praeler spiculis erectis squa- mae latae basi involventes uninervae culmusque rigidus C. rigidam indicant.« The specimens are tall and robust, of about the same shape as C. Lyngbyei; the leafy shoots bear long and broad leaves, higher than the flovs'ering shoots. 1 — 2 male spikes with brown, obtuse scales, 2—3 shorl-cylindrical or ovoid, erect female spikes; the lower one short-stalked, with a leafy bracl, as long as or longer Ihan the short upper ones, which often are male at the top. The female scales are dark-brown or blackish with pale midvein ovate, subacute, about as long and broad as the perigynium. 192. C. panicea L. pilulifera L. pulicaris L. pulla Good.; C. saxatilis auctt. above Kuno village, c. 400 M. (R. R.). rigida Good. salina Whbg., subsp. kattegalensis Fr. Str.: Saltmarsh at Selletræ (J.Hartz & C.H.Ostenfeld), a somewhat divergent form, which nevertheless also by the Rev. G. Kiikenthal is referred to the form here given. 198. Eriophorum poiystachyum L. 199. E. vaginatum L. 200. Heleocharis multicaulis Sm. ^ In my Flora Arctica, I, 1902 (p. 76) I have showed, that C. haematolopis Drej. according to the authentic specimens is a hybrid between C. Lyngbyei and C. rigida. 193. C. 194. C. 195. C. Kuno: 196. c. 197. c. 859 201. H. palustris (L.) R. Br. Mr. II. Lindberg ol" Helsingfors has exainined the specimcns from the Færoes and has testilied, Ihat all are: Scirpns palustris L., subsp. eiipaluslris Lindb. fil. 202. H. uniglumis Lk. 203. Scirpus caespitosus L. All the specimens from the Færoes belong to var. germanica (Palla), but are somewhat merging into var. anslriacn (Palla). Mr. E. Broddesson of Lund (Sweden), \vho has examined our speci- mens has labelled most of them »Scirpus germanicus Palla, ad auslriacum Palla vergens.« 204. S. pauciflorus Lightf Fam. XLV. GRAMINPZAE. 205. Agropypum junceum (L.) P. B. A. junceum (L.) P. B. X repens (L.) P. B. 20(3. A. repens (L.) P. B. 207. Agrostis canina L. A. canina L. X vulgaris With. 208. A. stolonifera L.; A. alba L. 209. A. vulgaris With. 210. Alra alpina L., vivipara. I think it is better to keep this form distincl from A. caespitosa and not take it as merely a viviparous form of it. 211. A. caespitosa L., var. hrevifolia Hartm. 212. A. flexuosa L. 213. Airopsls praecox (L.) Fr. Myggenæs (accord. to G. P.); Nol so (E.Rostrup & C. A. Feilberg); Str.: Kirkebo (G. P.J; Syderdal (!). 214. Alopecurus geniculatus L. f A. pratensis L. 215. Anthoxanthum odoratum L. t Apera spica venti (L.) P. B. t Avena sativa L. t A. strigosa Schreb. t Briza media L. f Bromus mollis L. 216. Catabrosa aquatica (L.) P. B. t Dactylis glomerata L. 217. Digraphis arundinacea (L.) Trin. 860 218. Elymus arenarius L. 219. Festuca ovina L., vivipara. t F. pratensis Huds. 220. F. pubra L., and f. arenaria Osbeck. The large, broad-Ieaved form which I in my list have named var. planifolia Trautv., is according to Mr. R. Rasmussen common in Ihe cliffs inhabited by sea-fowls. I do not know whether the variety name is correct or not, but the form is a very remarkable one. 221. Glyceria distans (L.) Whbg. 222. G. fluitans (L.) R. Br. 223. G. maritima (Huds.) Whbg. Syd.: On the west-coast opposite Lopra, ab. 125 M. supra mare (Ove Paulsen). 224. Holcus lanatus L. 225. H. mollis L. t Hordeum vulgare L. f Lolium multiflorum Lam. t L. perenne L. Ost.: Ejde (»sown with seeds from England, bul nearly superseded by other grasses«. 22. Aug., 1817, H. C. Lyngbye). 226. Molinia coerulea (L.) Moench. 227. IMardus stricta L. t Phieum pratense L. 228. Phragmites communis Trin. 229. Poa alpina L., vivipara. 230. P. annua L. 231. P. glauca M. Vahl. 232. P. nemoralis L., f. ghtiicanlha Gaud. 233. P. pratensis L., f. hninilis Khrh. 234. P. trivialis L. In my list (1. c. p. 90) I have described a f. paUida, but this name must fall, as f. pallescens Stebler & Volkart, 1895 is earlier. 235. Psamma arenaria (L.) R. & S. 236. Sieglingia decumbens (L.) Bernh. t Triticum vulgare Vill. Fam. XLVI. IRIDACEAE. 237. Iris pseudacorus L. Fam. XLVH. JUNCACEAE. 238. Juncus balticus Willd. 861 239. J. biglumis L. Str.: Kirkcborcjn, ab. 350 M. (!). 240. J. bufonius L. 241. J. conglomeratus L. 242. J. effusus L. 243. J, lampocarpus Ehrh. 244. J. obtusiflopus Ehrh. 245. J. squarrosus L. 246. J. supinus Moench. 247. J. trifidus L. 248. J. tpiglumis L. 249. Luzula arcuata (Whbg.) Sw. 250. L. campestris (L.) D. C. 251. L multiflora (Ehrh.) Lej. 252. L. silvatica (Huds.) Gaud. 253. L. spicata (L.) D. C. Fam. XLVIII. JUNCAGINACEAE. 254. Triglochin palustre L. Fam. XLIX. LILIACEAE. 255. Scilla verna Huds. Mr. G. Patursson writes, that according to a communication from Mr. Skaalum of Thorshavn this pretty httle plant grows in several piaces on Nol so and Sando. Fam. L. ORCHIDACEAE. 256. Habenaria albida (L.) R. Br. Str.: Ravnabjorgini near Kirkebo (G. P.); Vid.: rock-ledge on Morne- fjæld, ab. 400 M. (E. T.). 257. H. viridis (L.) R. Br. ; Coeloglossiim Hartm. Kai so: iMouiitain near Husum (E.T.); Vid: Malinsfjæld, ab.400M.(E.T.). 258. Listera cordata (L.) R. Br. Bord 6: Fagralid (B. R.), flowering in May— June, t'ruiting in August; Vid.: Malinsfjæld, slope ab. 2—300 M., flowering (E. T.). 259. Malaxis paludosa (L.) Sw. 260. Orchis latifolius L. Myggenæs: »Lundelandet (accord. to E. T.); Str.: Velbestad (ac- cord. to G. P.). 261. O. maculatus L. 262. O. masculus L. Str.: Between Velbestad and Kirkebo, and Ravnabjorgini near Kirkebo (G. P.). 862 Fnm. LI. POTAMOGETONACEAE. 2(k}. Potamogeton alpinus Balb. 264. P. filiformis Pers. 265. P. gramineus L. P. gramineus L. X perfoliatus L. 266. P. natans L. 267. P. perfoliatus L. 268. P. polygonifolius Pourr. 269. P. praelongus Wulf. 270. P. pusillus L. 271. Ruppia maritima L., 1". roslellata (Koch). 272. Zostera marina L. Pam. LII. SPARGANIACEAE. 273. Sparganium affine Schnizl. III. Gymnospermae. Pam. LUI. PINACEAE. 274. Juniperus communis L., f. nana (Willd.). IV. Pteridophyta. Fam. LIV. EQUISETACEAE. 275. Equisetum arvense L. Specimens referablc to f. pseudosilualica Milde have been found on Str.: Slope at Sandegærde ncar Thorshavn (!). 276. E. heleocharis Ehrli., f. limosa (L.). Str.: Kirkebo (G. P.); Ost.: Lervig, in dilchcs (R. R.). 277. E. palustre L. 278. E. pratense Ehrh. Kuno: near the village (accord. to R. R.); Str.: Kirkebo (G. P.). 279. E. silvaticum L. Fam. LV. HYMENOPHYLLACEAE. 280. Hymenophyllum peltatum (Poir.) Desv. Fam. LVI. ISOÉTACEAE. 281. Isoétes echinosporum Dur. 282. I. lacustre L. 863 Fam. LVII. LYCOPODTACEAE. 283. Lycopodium alpinum L. 284. L. annotinum L. 285. L. selago L. Fam. LVIII. OPHIOGLOSSACEAE. 286. Botrychium lunaria (L.) S\v. Fam. LIX. POLYPODIACEAE. 287. Aspidium dryopteris (L.) Baumg. Str. : Kirkeborcjn, rock-ledge sloping towards Arge (!). 288. A. filix mas (L.) Sw. 289. A. lonchitis (L.) Sw. Vid.: Rock-ledge on Mornefjæld, ab. 300 M. (accord. to E. T.). 290. A. phegopteris (L.) Baumg. 291. A. spinulosum (Miill.) Sw., subsp. dilutatiim (Hoflm.) Sw. 292. Asplenum adiantum nigrum L. 293. A. trichomanes L. 294. Athyrium filix foemina (L.) Rolh. 295. Blechnum spicant (L.) With. 296. Cystopteris fragilis (L.) Bernh. 297. Polypodium vulgare L. F^am. LX. SELAGINELLACEAE. 298. Selaginella selaginoides (L.) Link. LIST OF POPULAR PLANT NAMES FROM THE FÆROES COMPILATEU BY GAZET PATURSSON. At the request of Mr. C. H. Ostenfeld the author of the present list l\ has tried to put together the popular nanies of the færoese piants. The names originate parti}' from pvipers published by other authors, partly from my own knowledge supplemented by that of fellow-coun- trymen. During several years I have been interested in collecting the plant names used on my native istands, and I have always noted the names I heard, when travelling from village to village, from istand to island. I am fully aware that the list is not at all complete, but I hope it may be of some value as an attempt to refer the popular names of the færoese piants to the species fixed by their Latin names. Aarmosi: Astrophyllum sp. Adam oy Eva^ : Orchis maculatus. Akeleia: Lychnis Ilos cuculi, Orchis maculatus (Sv.). Akiiriill: Eriophorum polystachyum. Apaldiir: Pyrus malus. Arvi: Stellaria media, Montia lam- prosperma. Båla see Korki. Baldiirsbrå : Matricaria inodora, var. phaeocephala (BaldursblaS, Sv.). Baraldin: Juniperus communis. Bdtsmans hattiir: Brunella vulgaris. Bergs/^ra: Oxyria digyna. Ber(ja)lyiigur: Empetrum nigrum. Bervisa: The flower of Berlyngur. Biåiskinn see Mjolkasolja. Birki: Betula sp. Bjollika: the rhizome of Equisetum arvense. Bjollikustrå : the sterile stems of Equisetum arvense. Bjargahvonn see Hvonn. Bldber: Vaccinium myrtillus (the fruit). Bldberlyngiir: Vaccinium myrtillus. Bjolgagras : Equisetum silvaticum. Bladtari: Laminaria saccharina. Bldhattiii'^ : Viola silvestris (Sv.). Bldkolla^: Succisa pratensis; Viola silvestris. Bludgras: Blechnum spicant. Bjargablomstiir: Melandrium ru- brum. Blik see Korki. Blod riitari : Fucus vesiculosus, Asco- phyllum nodosum. Bék: Fagus europæa. Bolatari: Ascophyllum nodosum. Bord(a)gras: Juncus squarrosus, Ca- rex salina Whbg. BorfJiir see Borågras. Borkubondi'^ : Orchis maculatus. Breiåskora: Tussilago farfarus. Brobber^: Thymus serpyllum. Bukkablaå : Menyanthes trifoliata. Borkiwisa: Potentilla erecta; the rhi- zome: Borka. Bygg : Hordeum vulgare. Bokkiir: Silene acaulis. Ddi: Galeopsis tetrahit. Diinniiber^ : Vaccinium uliginosum. ' The inhabitants of Thorshavn call the white-flowercd form Eva , the red- flowered form Adam. ^ According to E: Bhihattur :=- Blåkolla =^ Succisa prat.; Blåkolla ^ Viola silvestris. ^ Sv. says: it is tlie root, which is called Borkubéndi. * After E: Bro aber, Broåberg. ^ Atter Sv. : Dunnuber =^ Blåber. 865 Dyngjiimosi: Grimmia hypnoides. Eik: Quercus robur. Einir: Juniperus communis. Eyrisgras: Cochlearia officinalis. Eyrissyra: Oxyria digyna. Eldiirt: Geranium silvaticum. Eyingras: Euphrasia sp. Flindiir see Korlvi. Fransagras: Lathyrus pralensis, Pri- mula acaulis. Fisibjolgiir: Lycoperdon sp. Fiira: Pi nus. Glanssolja: Ranunculus acer and R. repens. Giilbeita : Equisetum silvaticum. Goliibrå: Plantago major, Brunella vulgaris, Gentiana campestris, Po- lygala serpyllacea, Epilobium {'!). Hagasolja: Taraxacum, Leontodon autumnale, and Hieracium. Hårtari: Desmarestia aculeata. Havri: Avena sativa. Heimahvonn see Ilvonn. Hjålpiråt: Sedum rhodiola. Hogvgras: Tussilago farfarus. Hostagras: Lycopodium selago. Hoyloppa: Anthoxanthum odoratum. Hoyliis: Myosotis arvensis. Hoytari: Desmarestia aculeata. Hiisagras: Poa trivialis and P. pra- tensis. Hundaland: Agaricaceæ. Hveiti: Triticum sp. Hvonn: Archangelica officinalis; cul- tivated: Heimahvonn ; wild: Bjar- gahvonn; the stem of the fruiting specimens: Hvonnjuliir. Homilia: Rumex sp. Iglagras: Potamogeton polygoni- folius. Jardarsipa: Peltigera canina. Javni: L3'Copodium alpinum. Joansokugras : Plantago lanceolata; the leaves: Langskora. Jomfru Marin grås: Orchis macu- latus. Jordepli: Solanum tuberosum. Kaggaber: Vaccinium uliginosum. Kannubjolla see Kannubjolluvisa. Kannubjollustrå : Equisetum arvense. Kanniibjolhwisa : Equisetum arvense; the tubers of the rhizome: kannu- bjolla. Karsi: Cardamine hirsuta. Kaltaklogv : Narthesium ossifragum. Korki: Lecanora tartarea; the apo- thecia: Bdla, the thailus: Blik, Flindur and Tel. Kay ril (Sv.), Koyring (¥.): Lamina- ria saccharina. Krdka: the sporophylls of Tangur. Kråkiiber: Empctrum nigrum (espe- cially the immature fruit). Krokja, Krokjiigras : Vicia cracca. Krusemynta : Mentha sp. Langskora see Joansokugras. Linbinda: Brassica campestris (also Sinapis and RaphanusX Llragras: Succisa pratensis. Lodnaskora : Tussilago farfarus. Loppiigras: Gentiana campestris. Liisagras: Lycopodium selago. Lyngiir: Calluna vulgaris. Lundasina see Sina. Marleggiir: Isoetes lacustre; marine alga in Sundalag. Mekja : Iris pseudacorus. Mekjiigestiir: flower of Mekja. Meistarnrt: Haloscias scoticum. Miåkjalskora: Alchimilla alpina. Mirkjallur: the midrib of Tangur. Mjaåariirt: Spiræa ulmaria. Mjolber: Galium saxatile (F.). Mjolkasolja: Taraxacum; Ullakona, when the flower has fadened; Bidiskinn, when the fruits are dropt. Mogras: Eriophorum and Juncus sp. Mi'ira, Miirgras: Potentilla anserina, the root: Miirild. Myrimosi: Sphagnum sp. Myrisnipugras : Orchis maculatus (and latifolius). Myrifipa: Eriophorum polystachyum. Myrisolja: Caltha palustris; the le- aves: Sdljariidiir, Soljukappar. Ndtil: Galeopsis tetrahit (F.). Nohir: Urtica dioica. Ovrokjå : Pedicularis palustris. Pålmi: Salix phylicifolia. 866 Perikum : Hypericumquadrangulum. Pina: Chorda fdum. Pninella: Brunella vulgaris. Punyagras: Alectorolophus minor. Piintalastrå : Aira caespitosa. Ravnaleikiir: Orchis niaculalus. Reyåhlomstiir: Melandrium rubrum. Reinfann: Tanacetum vulgare. Riski: Luzula silvatica. Rivugras: Vicia cracca. Roygras: Anthoxanthum odoratuni. Royggj : Digraphis arundinacea. Reipatari: Himanthalia lorea. Royuber: Cornus suecica. Rossaber: Rubus saxatilis. Rallik: Achillea millefolium. Rugiir: Secale cereale. Sand(s)arvi: Elymus arenarius (Sv.). Sev: Scirpus pauciflorus. Seyåasmæra see Smæra. Sina, Sinugras: Festuca rubra; in thc fowling Cliffs: Lundasina. Siftiisokugrcis (Svitunswokugras) : Alo- pecurus pratensis, Phleum pra- tense (L.). Sif: Juncus conglomeratus. Sjeyskura: Alchimilla vulgaris. Sjogvgras: Armeria elongata. Skrikkjugras: Vicia cracca; Poten- tilla anserina (Sv.). Skiiagras: Scirpus cæspitosus. Skræpurut: Silene acaulis. Sliggj : Porphyra umbilicalis. Slokja : Angelica silvestris. Smæra : Trifolium repens (the flower oftenest: Smæra, the leaves Sey- åasmæra). Snjallibjalli : Alectorolophus minor. Soleya, Solja: Ranunculus; Caltha palustris (Sv.). Soljurudiir, Soljiikappar see Myri- solja. Sortugras: Geranium silvaticum, Spi- ræa ulmaria. Spågvagras : Callitriche. Spungras: the leaves of Orchis maculatus. Svarlber: Empetrum nigrum. Stargras: Carex. Steinamosi: Parnielia saxatilis. Sleinbrå : Galium saxatile. Sidrur: Carex, Elymus arenarius. Summardåi: Bellis perennis. Svinahvånmir : Ranunculus acer. Svinasolja : Ranunculus repens. Svovngras: Hypericum pulchrum. Svovnurt: Haloscias scoticum. Sijra : Rumex acetosa. Sotagras: Vicia cracca. Siiir: Parmelia saxatilis. Sol: Rhodymenia palmata. Tålgagras: Galium saxatile. Tarablod ra : the basal part of Rei- patari. Tari: Marine Algae. Tangur: Alaria esculenta ; the stem: Tangbjolla. Tel see Korki. Tirilshattur: Taraxacum and Leon- todon. Tirilstunga: Lotus corniculatus. Tirilsgras: Equisetum. llstil: Cirsium palustre. Trollakambur : Aspidium lilix mas. Trollablomstur : Brunella vulgaris. Trollkonu/ingur: Cystoptcris fragilis. Trælamosi: Polytrichum commune. Tiigvugras: Silene acaulis. Tussatongul: Chorda filum. Tussingur: Laminaria hyperborea; the fringed and torn leafof Tangur. Tongul: the stem of Laminaria. Ullakona se Mjolkasolja. Undirloyugras: Pinguicula vulgaris. Valnlilja: Cardamine pratensis. Veikur: Juncus conglomeratus. Viriksgras: Hypericum pulchrum. Vælaks: Hordeum hexastichum. Note: The letters E., F., L. and Sv. are abbreviations for A. C. E vensen, C. A. Feilberg, J. Landt and Svabo, wbo all have collected popular names of the Færoese piants. THE LAND-VEGETATION OF THE FÆROES, WITH SPECIAL REFERENCE TO THE HIGHER PLANTS, BY C. H. OSTENFELD. INTRODUCTION. AN attempt is made in these pages to give an account of the vege- . tation of the Færoes, including its dependence on, and adapt- ation to external conditions; the whole together forms a contribution to the plant ecology of the Færoes. It is not my intention to discuss questions relating to the Færoese flora and the history of its immi- gration, because the floristic and phyto-geographical aspects of the land-vegetation have already formed the subject of contributions in »The Botany of the Færoes«, Vol. I (Ostenfeld 1901 b, 1901 c). In a later supplementary list (Ostenfeld 1907), I also revised the vascular piants found on the islands, adding those species foand since the issue of the earlier memoirs, and in other ways bringing the floris- tic list up to date. My observations on the vegetation of the islands were made chiefly during the summer of 1897, the period referred to in »Bot. of the Færoes«, Vol. I. Previously I had made short visits in 1895—1896 with the Ingolf-Expedition, and in 1903, in addition, I spent three weeks on Stromo (Thorshavn) and on Sydero. The experience thus gained led to the publication of the memoirs referred to and also provided material for the present contribution, which was first published in Danish in December 1906 (Ostenfeld 1906). That this paper has its shortcomings and defects I know too well. The principal reason for this is, that all my observations were made during summer (between May 7th and Sept. lOth), hence I have no personal acquaintance with conditions during the stormy, sleety Færoese winter, and am obliged to supply this lack in my information by purely meteorologicai statistics relating to the Færoese climate (Willaume-Jantzen, 1899 and 1905). Fortunately however, I have obtained a certain amount of supplementary infor- mation of great value from a botanical point of view. This consists Botany of the Fa-roes. 5(5 868 of a number of notes from Dr. Knud Poulsen on the nature and duration of the snow-covering during the winters 1901 — 02 and 1902 — 03, and also relating to the time when certain piants began to flower in the spring of 1902 and 1903. I take here the opportunity of tendering my sincere thanks to Dr. Poulsen, a friend of many years duration, for the great care and regularity with which he pre- pared these notes, especially so since the vocation of physician allowed him only a scanty leisure. In more ways than one, these notes have been of great benefit in enabling me to grasp the peculiarities of the Færoese vegetation. I have also been permitted to use some notes made by Mr. R. Rasmussen, Fagralid on Bordo; these include records of the com- mencement of tlowering periods, and also lists of piants in different habitats on the Nordreoer (Northern Islands). Acknowledgements are also due to Mr. C. Jensen and Mr. J. Hartz for valuable assistance with regard to the mosses and li- chens mentioned in this paper; Mr. Jensen especially has aided me greatly in identifying the numerous examples of mosses which I collected. For most of the photographs reproduced here I am indebted to Professor E. Warming and Dr. F. Borge s en. The former has also placed at my disposal a number of notes, some taken on the Færoes in 1897, others being annotations on the Danish edition of this memoir. The material for this memoir was put together in 1897, but its completion has been delayed for many reasons which need not be detailed. To delay was easy, when one had a certain reluctance in attempting a task which seemed unsatisfactory on account of the insufficient preliminary study possible, in what was really a sojourn of a single summer. As the work progressed this reluctance disappeared almost en- tirely, and in recalling my stay on the islands I could not forget the hearty hospitality and helpfulness extended to me everywhere on these charming crag-girt isles. If in the smallest degree this paper will aid any efforts towards the welfare and success of the islands, then I feel that in this way it may in some measure repay my debt to their inhabitants. The plan of this paper is to give in the first part a brief histo- rical survey of our knowledge of the Færoese vegetation, especially as regards its ecology. The influence of external conditions on the vegetation is next considered along with different biological features of importance with regard to the physiognomy of the vegetation and its composition. The main part of the paper then foUows as a description of the plant-formations, with a review of the adapta- tions of the various piants to the conditions under which they live. The English edition difl'ers in several respects from the Danish one, as I have altered the treatment of some parts (especially Chap. III.) and have put in many corrections and amendments. The translation, made by a Danish translator, has been corrected and revised by Dr. W. G. Smith of the University of Leeds, and I wish here to thank him sincerely for his assistance in the neces- sary, but not always easy task of finding Enghsh equivalents for Danish ecological terms. The full text of this paper has thus been rendered available for English readers. I. HISTORICAL REVIEW OF LITERATURE RELATING TO THE VEGETATION OF THE FÆROES. Prior to the detailed investigations carried out during the last 10 years, the principal source of our knowledge of the vegetation of these islands was E. Rostrup's treatise: »Færoernes Flora« (1870). Some of the still earlier works refer to vegetation in more or less casual m anner. Nowhere, however, is there any special considera- tion of that aspect which is our particular objective, namely: a de- scription of the plant associations of the Færoes, and their conditions of li fe. It is only regarding the associations of cultivation (the Farm- land with its cornfields and grass-meadows) that the earlier authors give any information. Jorgen Landt in 1800 published a description of the Færoes with a long list of piants, but says nothing about the vegetation in the uncultivated part of the country. There is a detailed chapter on the condition of agriculture (1. c, p. p. 292^ — 320) and on haymaking (p. p. 320—328), and we are told how the inhabitants treat the soil brought under cultivation. The methods then in use, more than a hundred years ago, were much the same in their principal features as those now employed and described later in this paper. The ;70 nips; the cultivation of potatoes, which is now rather extensive, was then only beginning, while oats and other crop-plants were then as now of no great importance. Nature was left, as it is to- day, to clothe the cultivated fieids with grass after the bariey-har- vest, and so in the course of a few years to form a meadow. It is noteworthy to what degree conservatism has ruled in agricultural afTairs during the past century, and how only in recent years there has been any progress. Pastor H. C. Lyngbye, the author of the well-known, impor- tant work on x-Xlgae: »Tentamen Hydrophytologiae danicae«, visited the Færoes in 1817. From his pen we have (Lyngbye 1822) some: »Anmærkninger til kort Efterretning om Færoerne efter Sir Macken- zie«. This paper includes a somewhat abbreviated translation ot Mackenzie's paper, which however contains little of much interest to us, merely the usual description of methods of cultivation. Lyng- bye's comments, on the other hånd, contain some valuable botani- cal information. His description of the ascent of Skællingfjæld on Stromo (1. c, p. p. 125 — ^126) is specially worthy of notice on account of the detailed observations on the vegetation of the summit-plateau of the mountain. — He states that the summit is »in piaces covered with moss, especially Trichostemum [Grimmia hypnoides and G. eri- coidesj^, and in other piaces it is bare sand and gravel with Koeni- gia islandica appearing sparsely here and there.« Lyngbye has in these words defined the two highland plant-formations, distinguished in this paper as Grimmia-heath and Rocky-flats. He gives a list for the latter which, in addition to a number of lichens and mosses, includes the following species of flowering piants: Festnca vivipara [F. ovina vivipara], Aira montana [A. flexiiosa montana] , Koenigia islandica, Polygoniim vivipariim, Rumex digynns [Oxyria digyna], Saxifraga palmata [S. caespitosa], S. stellaris, Cerastiiim alpiniim [C. Edmondstonii] , Arabis hispida [A.petraea], Statice Armeria and Salix herhacea. This collection of species coincides exactly with my own conception of a rocky plateau as seen in the Færoes. Lyngbye also describes (1. c, p. p. 130) how a cornfield, left uncultivated, be- ^ Byg, Bigg or Bere are names applied to several hardy varieties of Hordeiim viilgare with short six-rowed ears; the crop is grown onlj' in the upper part of the Oat-zone (e. g. in Scotland) or on inferior soils, and is much less valuahle than tlie finer quaHties of brewing Barleys grown in the Wheat-zone. (Note by W. G. Smith). ■ The names enclosed in square Ijrackets are those now used. 871 comes covered with weeds (my Melamorphic Vegetation), and grad- iially arrives at the condition of a grassy meadow with Holcus Uinatiis and mollis, Antho.vanlhiim, Festuca pratensis^ and Agrostis vulgaris; in other words the iisual »B6«-formation. A record is also made of the pecuharity that the houses are all thatched with green turf, on which the grass thrives so well, that »one can mow hay on the housetop« (1. c, p. 149), and a list of the piants growing on the roof of Sorvaag church on Vaago is given. A few years after Lyngbye, the danish geologist Forchham- mer visited the Færoes accompanied by a Scotsman, W. C. Tre- velyan, who published a short paper on the vegetation and climate of the Færoes (Trevelyan 1835 — 37). What he has to say regarding. the vegetation is mainly quoted from the memoirs of Landt and Lyngbye, but some original observations are given in his note on- an ascent of Malinsfjæld on Videro, which Trevelyan and Forch- hammer made on July 18th 1821. The list of species growing on« the summit-plateaux corresponds exactly with Lyngbye's list from Skællingfjæld, and Trevelyan adds the attitudes at which the dif- ferent alpine species were first observed during the ascent. The next contribution to our knowledge of the vegetation of the Færoes is that of Ch. Martins, who came to Thorshavn with the French corvette »La Recherche«, and during his stav there made excursions to Stromo and Nolso, His observations formed the basis of a comparative study of the flora of the Færoes, Shet- land and Iceland with regard to the routes and agents concerned in the immigration of these floras (Martins 1848). It is beyond the province of this paper either to review or to criticise this part of Martins' contribution. There is the oft-repeated description of agri- cultural conditions and the methods of cultivating the soil. Short sketches are given, however, of the vegetation in the immediate neighbourhood of Thorshavn, and on Nolso from the coast to the sumniit of the mountain on that island. The characteristic rounded rocks north of Thorshavn with their varying vegetation are accurately described, and he distinguishes be- tween the barren and exposed tops, where Armeria is so prominent, and the boggy hollows with Eriophorum, etc. On Nolso he men- tions the vegetation of the sandy strand with Honckenya, Cochlearia and Potentilla anserina, and attention is draw'n to Nardus stricta as ^ The identification is not correct; Festuca riibra is probablj' what is nieant. 872 the dominant plant of the mountain slopes; »it had, so to speak, excluded all other vegetation« (1. c, p. 369). Martins' lists are very faulty, and a nurnber of species reported from the vicinity of Thors- havn have not been confirmed by more recent observers, although undoubtedly this district has been more thoroughly examined than any other part of the Færoes. A picturesque description of the Færoes was written by P. A. Holm (1855) in a somewhat general and popular style. The vegetation of the thatched roofs, the grassy »Bo's«, and the cultivation of corn, potatoes and turnips all come under no- tice. Regarding the vegetation outside of the area of cultivation, we are told that »Tue-Kogleaks« (Scirpiis caespitosiis) and »Tue-Star« (presumably Jiinciis squarrosus is meant) form pretty green tussocks on the moors, with the intervening spaces occupied by many kinds of »Siv« {Jnnciis lampocarpus and Carex-species) and »Kæruld« (EriophonimJ, the prevailing sombre tone being relieved by the co- loured flowers of »Benbræk« (Nartheciiim) and »Fjæld- Vibefedt« (must be Pingiiiciila viilgaris, not P. alpina). What he refers to here is the Grass-moor and the Sedge-moor as they are found on the »Hauge« i. e. the uncultivated parts of the lower zone. Holm also mentions the Calluna Heath (Lynghede) with its »Rævlinger« (Empetnim) and Vaccinia, also the rock-vegetation, and he adds a few notes on the mountain vegetation. Our review of the literature now brings us to the memoir of Rostrup, which is based on a voyage made in 1867 along with Mr. C. Feilberg round most of the islands. An effort was made to complete the list of piants as far as possible and to extend it over all groups of piants. Earlier, often incorrect, statements were revised and verified, and many new discoveries by the authors themselves were added. Thus for the first time a complete and reliable floristic list was obtained for the Færoes. Rostrup also gives a general account of the vegetation, and this up till now has been the main source of information on this subject. At the time Rostrup wrote this paper (1870), systematised studies on vegetation in relation to its environment were hardly known. Hult's »Forsok til analytisk behandling af våxtformationerna« (1881) did not appear till 10 years later, and still later Warming published his »Plante- samfund« (1895), the flrst systematised work on the vegetation of the earth from an ecological point of view. Rostrup therefore does not attempt a consecutive account of the plant-associations, but 873 merely describes the vegetation of difTerent localities: e. g. »Bo«, moor, shore, lake etc. For each locality the common and domi- nant piants are recorded with remarks on features of biological in- terest. A few examples will illustrate his method. Describing a »Bo«, Rostrup enumerates the principal characteristic grasses, and he points out that the grasses of the Færoes are particularly liable to appear as »viviparous« forms, a condition which »may probably be attri- buted to the moistness of the atmosphere«. The grasses of the thatched roofs are included in this list. Then follows a list of those »flowering« herbs which occur most frequently among the grasses. The »Bo« at Sand on Sando, a well-known Færoese locality for flowering piants is described at some length. He con- cludes with a list of species, which although most at home in »the higher regions«, may appear occasionally in the »Bo«. Other plant- associations and their habitats are dealt with in like manner (1. c, p. 12—20). The introduction also contains some interesting notes on the duration of species. Rostrup points out the peculiar absence of trees and shrubs, and that most of the piants are perennial herbs, while the comparatively few annual and biennial piants which are found, occur with few exceptions as weeds in the arable land, in enclosed fields, or near the sea. He estimates that »about half a score of true mountain piants« are annuals or biennials, and gives the names of six species. Only one of these however, is an annual, another is a parasitic plant, and the remainder are peren- nials. The annual referred to is still the one annual species on the Færoese mountains, viz. Koenigia. During the next 20 years no papers were published on the vegetation of the Færoes, but early in the nineties a fresh start was made. The first to be mentioned is a short contribution by two Eng- lish ladies, Miss Copland and Miss Birley (1891), who visited the Færoes in 1889: this contains few observations which had not been made already. In describing an ascent of Odnedalstind on Stromo a few common mountain piants are recorded: Raniincuhis glacialis, R. acer piimila etc. In 1895 the Swedish botanist, Dr. H. G. Simmons made a voyage to the Færoes, principally to study marine algae, but he also made collections on land, and gives brief floristic notes on these. (Simmons 1896). 874 Dr. F. Borge sen paid his first visit to the islands in the same year. My first observations were also made about this time when I landed at Trangisvaag on Sydero on the outward and homeward voyages of the Danish Ingolf-Expedition. A short note on our com- bined observations on the vegetation with some hsts of piants was pubhshed later (Borgesen and Ostenfeld-Hansen 1896). The systematised investigation of the vegetation of the islands, projected by Prof E. Warming, was begun during the foUowing year (1896), when Mr. C. Jensen investigated the moss-flora of most of the islands, and Dr. Borgesen studied the algal flora of several of them. C. Jensen's excellent account of his voyage (Jen- sen 1897) contains many extracts from his diary which are of eco- logical interest. There are numerous observations on the vegetation and on the habitats of particular species, as well as reflections on the influence of environment. This memoir has been of much service to me, and will be referred to again and again in my chap- ter on plant-associations. Mr. Jensen has not limited his observa- tions to mosses only, but has recorded many noteworthy facts regard- ing the vegetation as a whole, his extensive knowledge of the flowering piants enabling him to give descriptions of plant-associa- tions superior to those resulting from a one-sided study of either flowering piants or mosses alone. The plant-associations of the Færoes are also dealt with in P. Feilberg's »Fra Lier og Fjælde« (From slopes and mountains) which was printed privatelj'^ as a manuscript (1900). The author spent the summer of 1899 on the Færoes and Shetland, principally to study vegetation from an economic point of view, and he gives a delightful account of his travels. The paper deals mainly with the grassland of the cultivated area, the cultivation of corn and potatoes, 'and the lower zone of uncultivated land (»Haugen«). Although the purely botanical side is only of minor importance from the author's point of view, yet he gives some excellent descriptions of certain plant-associations. The grass-vegetation of the »Bo« and the cultivated fields is first dealt with at considerable length; in a less degree the grass- slopes and grass-moors, boggymoors, heather-moors and Calluna- heaths. Special attention is directed to the constitution of the soil and its influence on the composition of the vegetation, and the exception- 875 ally wide distribution of humous and peaty soils in the Færoes is referred to again and again. The result of my own travels in 1897 were published in 1901, C. Jensen's method of treatment being followed. This may be re- garded as a preliminary study to the present paper, hence need not be formally summarised, nor is it necessary to consider further the short account of the vegetation which I wrote for the article »Færoerne« in the Danish periodical »Atlanten« (1905 — 06, p. 216 — 221). With regard to the present and most recent contribution it may be stated that a Danish edition appeared in 1906. In addition to these more strictly botanical papers, one may find many references to the vegetation scattered throughout the pages of memoirs on tlie Færoes, many of which have appeared in recent years. The titles of some of these may be given without attempting a formal summary of observations generally of minor botanical importance: J. Lomholts descriptive article on the Fæ- roes in the periodical »Nord og Syd« (1898); Pastor I. F. Ronne's popular pamphlet on the Færoes (1900); James Currie's descrip- tions of the islands in »The Scottish Geographical Magazine (1906); and the article in »Atlanten«, already mentioned, which includes a description of the agriculture by L. Berg (1906). Bibliographical lists of works relating to geography and natural history are given in »Atlanten« (1905—06); under »Færoerne« (by J. J.) in S al mon- se n's Lexicon; and in »Færo Amt« (The district of the Færoes) in Trap's large topographical and statistical work »Danmark« (3rd edition). This summary I trust, has now dealt with all the more im- portant contributions on the vegetation (plant-associations) of the Færoes. ^ It may be stated, however, that only a brief reference has been made to agricultural literature, because it is regarded as lying some- what beyond the scope of this paper which is purely scientific. For the same reason horticulture has been passed over. Both de- partments are worthy of special consideration by experts on the respective subjects. 1 Professor Eug. Warming in his introduction to »Tlie Botany of the Færoes«, vol. I (1901) has given a short description of botanical investigations in the islands; this agrees essentially with the snmmary given here. 876 II. THE INFLUENCE OF EXTERNAL FACTORS ON THE VEGETATION. The vegetation of the Færoes bears the impress of the peculiar climatic conditions which result from the geographical position of the islands. The decidedly insular chmate \vith cool summers and absence of extreme cold in winter, with an all-prevailing humidit}^ and abundant rainfall, all these together produce moist conditions in the soil al most everywhere and favour the rapid production of humus. Excellent accounts of the Færoese climate are given by Wil- laume-Jantzen in »Geografisk Tidsskrift« (1899) and »Atlanten« (1905). In »Bot. of Fær.«, Vol. I. (Ostenfeld 1901 a) I have given a short abstract of the first of these papers. These accounts show at once a climate with characteristics so marked that they must of necessity have a determining influence on the vegetation. It ^\i\\ therefore be advantageous to consider in some detail these climatic conditions in so far as they influence plant life. Vegetation is affected not by climate alone, but also by edaphic factors, nor can the influence of man and animals be left out of account. There are thus three main groups of factors to be inves- tigated vv'hen considering the relation between external conditions and the vegetation. 1. Climatic factors. a. Temperature. Willaume-Jantzen s sketch of the cli- mate makes it quite evident that there is an unusually slight variation in the temperature of the air in the different seasons of the year. The climate is decidedly insular. The coldest season extends into the middle of March, and the highest temperatures begin to be registered in July and August. Observations made at Thorshavn over a period of 30 years^ give the following averages (in Centigrade) for the months of the year : ' In mj' abstract (Ostenfeld 1901 a) I have only been able to use Willaume- Jantzen's first paper which is based upon observations during 25 years; therefore the figures given there do not exactly correspond with those pnblished here. 877 January Febiiiaiy March April May June July August 3« 2 S«^ 2 3M 5^ 5 7^ 2 9'^ 7 10'^ 8 10« 8 September October November December. 9*^ 4 6« 7 5° O 3*^ 5 The mean annual temperature is 6*^ 5 C. These show, that winter and early spring (December-March) have nearly the same temperature, about 3*^3— 3° 2 C. Consequently there is no winter to speak of with prolonged periods of tempera- tures below zero. So far as vegetation is concerned, however, the mean readings for any period are of less value than the real reading, because it is more especially the extreme ranges of temperature, which are of importance in this connection. The lowest tempera- ture in Thorshavn during 30 years is — 11° 6 C. It is furthermore of interest to note, that the annual number of »ice-days« (i. e. 24- hour periods during which the temperature does not exceed zero) is only 8, and there are 70 »frost-days« (i. e. 24 hours during which the temperature falls for some time below zero). These are com- paratively small numbers. Even on »frost-days« the temperature is seldom much below zero, as is proved by the faet that in Thors- havn during a period of 30 years only 9 days have yielded a tem- perature below — 10*^ C. It must also be considered whether the changes of temperature occur often and suddenly, or whether the periods are long, and the transitions slow and uniform. It is not easy to obtain good data to answer this question, but all things considered the changes seem to be rapid. An abrupt change from frost to thaw, especially in spring, will invariably have an injurious effect upon the vegetation, and on the Færoes one frequently hears of damage from this cause. These sudden changes are confirmed by the observations on the snow-covering in the winters 1901 — 02 and 1902 — 03, made by my friend Dr. Knud Poulsen. These I have given in full on a later page, but one part of his notes is particularly applicable here: De- cember 30th and 31st, 1902, heavy fall of snow, and a fairly deep snow-covering; January Ist, 1903, rain; Jan. 2nd, snow disappeared in the lowlands; Jan. 4th, black-frost; Jan. oth and 6th, continuous fall of snow; Jan. 7th, snow with wind; Jan. 8th, rain, the snow melting; Jan. 9th, black-frost: etc. Thus within ten days the di- strict has been twice covered with snow, twice cleared of snow by rain, and twice in the grip of black-frost. One can easily conceive that changes of this kind must have some influence on the growth 878 of piants. Even with the relatively high winter temperature, piants must be very hardy to endure changes so rapid and frequent as these. On the other hånd, compare the summer temperatures and their influence. The two hottest months are July and August with a mean temperature of 10'' 8 C. In other words, there is no real summer, as understood in Europe. If however the actual tem- peratures are taken into account, the conditions are not so un- favourable. The highest actual temperature recorded in thirty years is 21° 2 C. But even in summer considerable variations in tem- perature from day to day are frequent, mainly as a result of the extremely changeable winds. Our knowledge of atmospheric temperature may be summed up in a general statement: The temperature of the air in winter is compara- tively high, as a rule above 0° C, occasionally a rapid fall below zero may occur for a brief period, yet ex- treme low readings nev er occur. The coldest period ex- tends into the middle of the spring, and it is April be- fore the temperature begins to increase. July andAugust are the hottest months, but even then the temperature is rather low, seldom rising above 20*^ C; rapid changes also occur du ring this season. Piants capable of thriving under these conditions, must there- fore be adapted to comparatively low temperatures, but there is no need for any special adaptability to extreme and continuous cold. b. Rainfall and atmospheric moistiire. The rainfall is considerable, the annual amount being 1570 mm. At all times of the year there is an abundant downpour; it is least during the spring and summer months, viz. April (93 mm.), May (88), June (77), July (87), whereas December and January have 185 and 184 mm. respectively. The rain generally falls as a fine drizzle, the total amount for any given period being small. This is indicated by the faet that in spring and summer there are no less than 20 rain-days per month, or two thirds of the entire month; yet the actual amount recorded in this period is only 86 mm. or an average of 4.3 mm. for each day with rain. In winter the number of rain-days is 28 per month. The mean annual number of rainy days is about 280, which leaves only 85 days without rain in the year (23 p. cent). 879 Fogs are also frequent, and take the place of rain, especially during the summer months. The annual number of fog-days is 54, of which 30 occur during the months of June, July and August, whereas the months from December to March have only one foggy day each on the average. It must be noted, however, that all these meteorological observations have been taken in Thorshavn, con- sequently in the lowlands. The number of foggy days amongst the mountains is certainly much greater, but unfortunately no record of these is available. The relative humidity of the atmosphere is great, namely 82 p. cent., p. annum. On comparing the figures for each month it is found, that there is no great variation during the different sea- sons. The lowest mean humidity, 78 — 80 p. cent., is during the spring months, February to May; the greatest is in the summer, July to September (85 p. cent.). The variation in the relative humidity is not, unfortunately, indicated by these figures, but, through the courtesy of the »Da- nish Meteorological Institute«, I have been able to consult a table showing the lowest records of relative humidity observed. This table States, that periods may occur, during which the humidity falls to 30—50 p. cent., but this is exceptional (e. g. 30 p. cent. has been observed only once in 25 years, and 31 to 39 only seven times). The lower readings occur without relation to season, but on the whole are more frequent in early summer. According to M. Knudsen (1900), the humidity and the temperature are both de- pendent upon the direction of the winds, or rather, whether the wind comes from the »East-lcelandic polar stream«, or from the »Gulf stream». When the wind blows from the polar stream, the average relative humidity is 72 — 81 p. cent. ; if from the Gulf stream, then 81 — 90 p. cent. is attained. Winds from the polar stream also lower the temperature over the Islands to a considerable extent. Thus though the atmospheric humidity may become very low on rare occasions, it is open to us to regard it as almost always comparatively high. Under these conditions, and with the prevailing moistness of the soil, there is no necessity for the development of structures, such as dense hair-coverings, etc, which protect piants from the effects of excessive evaporation. It is actually the case, as we shall see later, that woolly or tomentose species are the exceptions amongst Færoese piants ; in faet, xerophiious characters are exhibited by com- 880 paratively few species, and these are almost all piants of the moun- tain or swamp. c. The snow-covering. In arctic regions, where the winter is severe and enduiing, the presence or absence of snow is of the upmost importance for vegetation. The mild winters of the Færoes render a snow-covering less essential, but even here the plant co- vered by snow possesses advantages over the unprotected one. The snow not only serves to mitigate the alternations of frost and thaw, Fig. 165. Snow-clad landscape, Trang'isvaagfjord on Sydero. (From photo. by O. Effersoe.) but it also prolects the piants from direct insolation during frost, and from the dessicating effect of severe storms. As no information respecting the duration of the snow-covering is to be found amongst the meteorologicai statistics, I prevailed on Dr. Knud Poulsen to make as many observations on snow-cove- ring as possible, during his sojourn of two winters in Thorshavn. His notes are extensive, but I have no hesitation in quoting them at full length, because they are valuable both from the purely me- teorologicai and from the botanical point of view. The Winter 1901 — 1902. November 12th: the first snow fell, covering the mountains and lowlands with a thin coating. 16th: Iowlands free from snow, the south- sides of the mountains practically in the same condition. 17th: snow again, covering every place witli a thin and equal coating. 20th: snow gone everywhere. December 9th: a thin coat of snow every where. lOlh: a thaw setting in. Uth and 12th: thaw; the lowlands snowless. 13th and 14th: a heavy fall of snow, the lowlands covered. 16th: the lowlands almost free from snow. 17th: snow and wind. 18th: fall of 881 snow. 19th: snow commencing to disappcar in the lowlands. 21st: rain. 22nd: the snow has disappeared in the lowhinds, and practically on the mountains. 26th: snowless nearly everywhere. 27th Dec. to 9th January: essentially unchanged; no precipitation of any consequence; apparently very little evaporation, because the few patches of snow in sheltered piaces on the mountain-slopes and on the summit-plateaux do not seem to decrease in size to any considerable extent. lOth: snow- storm. Uth: snow everywhere, cahn. frosty. 12th: snow-stomi. 14th: rain. 15th: snow has ahiiost disappeared. 25th: snow-storm, snow everywhere. 30th : thaw. 31st: snow has disappeared ahnost every- where. February 4th: snow everywhere. 4th— 13th: frequent falls of snow. 13th: the snow lies in great drifts, especially on the gentle northern and western slopes; on the other hånd, on the summit-plate- aux and in the lowlands, the snow-covering varies from a few ctm. to 30— 60 ctm. 14th: thaw has commenced. 15th: the snow has already partially disappeared in the lowlands. 16th: the snow has disappeared alniost everywhere; rain. March lOth: fall of snow, the mountains covered with a thin coat of snow; thaw in the lowlands. 19th and 20th: once more a little snow. 21st: snow-storm during the night, a thin coat of snow everywhere. 23th: the snow lies, as usual, mostly on the northern and western sides; snow also in the lowlands. 27th: much has evaporated and nielted; the patches of snow become fewer and more scattered, as one approaches the sea. We w'ere told, there is still much snow on the northern istands, extending down to the sca. 29th: fall of snow. 31st: thaw, storm from the east. April Ist: almost free from snow. 3rd: fall of snow. 4th: thin coat of snow on the moun- tains. 6th : coating of snow on the summit-plateaux, scattered patches on the mountain-sides. 6th— 9th: thaw. lOth— 12th: rain. 12th: only a few patches of snow remain on the summit-plateaux. May 2nd: the patches have nearl}^ disappeared. 3rd— 4th: slight fall of snow. 4th: thin cover of snow on the mountains. According to report, much snow on the northern islands all the wa\' down to the sea. lOth: much snow on the mountains of the northern islands personally observed); south- wards completely free from snow. The following remarks by Dr. Poulsen may be added to these notes: »The parts of the islands which I have been able to observe to some extent, include the southern parts of Stromo and Ostero, as well as Sando and Nolso. These are all comparatively low, and they are^ presumably mainly on this account, snow-covered to a much less extent than the great northern part. Nevertheless there is a marked difference between Nolso, for example, and the immediate neighbourhood of Thorshavn on the one hånd, and, on the other, the district about 25 km. further north on Ostero. Nolso frequently becomes snowless in a few hours, presumably because it is small, low and dome-shaped, sloping on almost every side steeply towards the sea. Snow is never present on it in large quantity, nor does it form drifts. When the expression »snow-covered« is used in these notes it is only an approximation, because extensive bare piaces may always be 882 found, whilc outstanding boulders, rocky ledges (»HanireO etc. are never, striclly speaking, buried in snow. A real snow-field suitable for snow- skating is thus a rare occurrence. In the early part of February, when we had the largest fall of snow, nuich more snow than usual, w^e were told, I was obliged to walk on several occasions from Thorshavn to Velbestad on the west- side of Stromo. The snow was distributed as follows: A number of large and deep snow-drifts occurred on the gradualh' ascending slopes from Thorshavn to the summit-plateaux (c. 200 M.); alternating with these were areas entirely free from snow, and other stretches covered with a smooth uniform coating of snow from 30 to 60 ctm. deep; the plateau itself w^as covered by a layer of snow with a depth from 15 to 30 ctm., so that the larger stones and boulders were not covered. The steeper slope from the plateau downwards to Velbestad was prac- tically without a single snowless spot, even the »Hamre« (rock-ledges) being buried. The snow everywhere reached almost to my waist, and had gathered in many piaces into enormous snow-drifts. From what I saw myself at this and other piaces, and from what I have been told, I believe that this example will convey a tolerabl}^ good idea of the conditions at this period. The snow disappears very rapidly as a rule, as shown by the notes. Indeed, it is often surprising how a heavy rain may cause, in the course of a couple of hours, the rapid disappearance of a comparatively large amount of snow, and this quite as much on the mountains as in the lowlands. I find, after ever3'thing is taken into consideration, that there is much less difference between the snow on the mountains and in the lowlands than I had expected.« The Winter 1902-03. November 15th: a thin layer of snow on the summits of the northern Islands. December 25th : snow, 2—3 ctm., forming a uni- form covering in the lowlands and on the mountains. 26th: heavy and general fall of snow. 27th: heavy rain, the snow^ melts. 28th: the snow entirely gonc in the lowlands, but still isolated wreaths and patches on the mountains. 30th and 31st: heavy fall of snow, forming everywhere a uniform coat, 2—8 ctm. January Ist 1903: rain. 2nd: snow gone in the lowlands. 4th: black-frost. 5th and 6th: general fall of snow. 7th: drifting snow. 8th: rain, the snow melts. 9th: black-frost. lOth: snow-storm. Uth: black-frost. 12th: snow, 15—60 ctm., lying on the mountain-plateaux, the larger stones and boulders exposed; much snow with snow-wreaths, 1 — 1.25 metres deep, lying on the western side of Stromo (the mountain slope towards Hesto-fjord); only a little snow on the eastern side of Stromo and on Sando. 13th and 14th: thaw and rain. 15th: snow gone. 18th: a little snow yesterday and during the night, now thaw and snow gone. February Ist: snow. 2nd: snow, 2—2.5 ctm., everywhere in the lowlands and on the mountains. 3rd: rain. 8th : heavy fall of snow, almost 15 ctm., Iowdands and mountains uniformly covered. lOth: the snow completely gone in the lowlands and partially so on the mountains. 20th: hardly any change, snow in patches on the mountains. 22nd: slight fall of snow, chiefly on the 883 mountains. 23rd: uniform layer of snow on the mountains, about 5 ctni.; no snow in the lowlands. 24th: the snow on the mountains partly gone. 26th: a thin layer of snow still lying on the higher mountains. March 6th: patches of snow here and there only. lOth: almost en- tirely gone. April 5th: snow. 7th : snow gone again. lOth: fall of snow. 12th: fall of snow with some wind. The layer of snow about 5 ctm. deep, except in occasional deeper drifts. 14lh: fall of snow, lying about 15 ctm. deep. 15th: slight fall of snow. 17th and 18th: the snow evaporating rapidly. 19th: rain. 20th: lowlands and moun- tains almost snowless; scattered wreaths on the tops of the mountains. 21st: slight fall of snow in the evening. 22nd: fall of snow, with thaw. 23rd: snow gone in the lowlands, scattered spots and wreaths on the mountains. 26th: almost all the snow gone. May 8th and 9lh: fall of snow; the mountains snow-covered, but not south of Thorshavn, the lowlands snowless. Dr. Poulsen further states: »These observations as a rule hold good only for the southern part of Stromo, with Nolso, Hesto and Sando. These istands present somewhat similar conditions as regards snow-covering. Nolso, however, always loses its snow first of all, and it is under snow to a less extent than the others, particularly Stromo; this is probably due to its situation and form. The same is true to some extent with regard to Hesto. Much more snow falls on the nor- thern part of Stromo and on the northern istands, and the mountains there have been more or less snow-covered through the enlire winter; this has certainly been the case on the higher mountains. This is due probably to the much greater altitude of the mountains, though the dilference in latitude and situation must also be of some importance. So far as my information goes, snow and the snow-covering on Sydero have been rather less than in Thorshavn district, although part of that istand has a high elevation. On the whole, less snow has fallen during this winter than in the one before, there being very little previous to Christmastide.« These notes on the two winters present an original and in all essentials a perfect picture of the snowfall and the snow-covering on the Færoes. The prominent feature is the frequent and rapid changes between snow, thaw and black-frost. The speedy disappearance of snow and frost is due doubtless to the proximity of a relatively warm ocean; the mean temperature of the ocean at Thorshavn from January to March being 5^ 5 C. The snow is never allowed to stay Ion g, and a con- stant snow covering throughout the entire winter never occurs. Hence on the Færoes the snow-covering plays quite a subordinate part in protecting the vegetation against the CO Id of winter; on the other band the coId itself is not severe enough to be of prime importance. The snow probably exer t s its greatest influence as a form of precipitation, Botanv of the Færoes. a;7 884 since it is soon converted into waler. Only on rare occasions does Ihe snow-covering become a protection against the iiniled forces of frost and wind; this was the case in the period F'ebruary 4th to Fig. 166. Malinsfjæld (750 ni. altitucle) on Videro. Upper part snow-clad, the lower slopes being entirely clear. (Krom photo. by K, Rimestad.) 13th of 1902, the greatest and most lasting snow-covering of these two Winters. The observations apply, as emphasized by Dr. Poulsen, only to the lower districts. A more continuous snow-covering seems to occur, however, on the mountains of the northern islands. The climate there is presumably colder, since sub-arctic piants are known to be located there in greater numbers, e. g. Dryas, Veronica al- pina, Saxifraga riuiilaris, Papaver radicahim, Salix glanca, etc. Un- fortunately we have no exact information on the climatic conditions in the mountains, and must be content with what may be gathered from the scanty records available. Observations of this kind car- ried on for several years on the summit-plateau of one of the 885 higher mountains would form an extremely valiiable contribution to meteorology and to plant-ecology. d. Movements of the air (ivind). The Færoes have earned a well-deserved reputation for their stormy chmate. The mean records show that at Thorshavn only 12 p. cent p. ann. of the days are calm, the remaining 88 p. cent have some wind, frequently a stiff breeze. During the winter-season one in four or five of the soutliwesterly winds becomes a gale or storm, but in summer-time only one in thirty does so. If the strength of the wind be expressed by a scaleofO — 6, then the average figures are 2,o — 2,2 in winter, and 1,3 — 1,5 in summer. The winds blow in nearly equal proportions from all points of the compass; though there is some difference in the frequency, in that they are oftenest S. W. and W., N. and N. E. coming next. There is, however, this great difference between the two groups: the former bring heat and moisture, whereas the northerly winds bring cold and little moisture (cfr. M. Knudsen 1900). The great influence of the wind on vegetation (mainly as a desiccating agent, but also as a mechanical one), is a well known faet emphasized in the case of northern countries by many bota- nists, particularly E. Warming, N. Hartz (1895), O. Kihlman and more recently by Adolf Hansen. The elfects of the wind are very marked in many piaces on the Færoes. One characteristic type of vegetation, primarily determined by the wind, occurs in its most typical form on the so-called »Eider« (see Fig. 167). These valley- like depressions extending across the istands are swept by winds of such strength, that at times it is impossible to maintain an up- right position. All piants there are dwarfed to such a degree, that they form a closely shorn green sward in which the flowers open amongst the leaves instead of rising some distance above, as under ordinary conditions. As an instance there is in a sward of this kind at Vaag on Sydero an abundance of Lijchnis flos ciiculi with flowering stems not exceeding 2 — 5 ctm., instead of the usual height of 20—50 ctm. The force of the wind is also indicated by the vege- tation of the summit-plateaux of the mountains. Here the sward of dwarfed piants may be found wilh portions torn up and carried away (Fig. 170). It is even possible by examination to determine for any given place the direction of the strengest wind, from the faet that the vegetation forms strips or elongated patches, the long axis of which is at right angles to the wind. In the case of sum- 57* 886 mils with only a meagre rocky-flat vegetation, Ihe niajority of the piants shelter around and on the lee-side of stones. Another example of the power of the wind in the mountains may be given : One frequently comes across small patches of poM^dery soil quite devoid of any vegetation whatever; they are generally situated in a shallow depression liable to be flooded with water, but when this evaporates, the soil dries up and forms cracks arranged in a kind Fig 167 An »Eide« at Kvalbo on Sydero. The soil over large patches has been carried ofT by the wind ; the portions where the plant-covering still remains form slightly elevated tables or pedestals with scarped or overhanging hoUowed out margins. (From photo. by E. Warming.) of polygonal honeycomb (»Rudemark«). In summer the gaping cracks are seen to be filled with stones varying in size from a nut to a closed fist (see Fig. 168); on the surface, however, there are hardly any stones. This we regard as evidence that the wind tumbles these stones about, and causes them to lodge in the cracks. The few trees to be found, chiefly in and about Thorshavn, are all planted in piaces sheltered from the wind, and cannot grow any higher than their shelter allows; they have the same stunted growth as the trees on the western coast of Jutland. e. Light. There are no forests on the Færoes, and the piants receive the fuU benefit of any light which may reach the surface 887 of any given place, according lo the geographical position of the islands and tlie local lopographical conditions. The frequency of fog has ah'eady heen referred to, and the amount of Hght available on foggy days is of course comparatively limited. In addition, the atmosphere is ahiiost constantly more or less overcast. The mean amount of cloud is about 7,4 (scale 0-10; O = clear; 10 = com- pletely overclouded). In other words three-fourths of the sky on an average are covered with clouds. There are annually only 18 clear days (O — 2 of the scale), but 184 »dark« days with the sky entirely cloudy are recorded. The full effect of the sun is lost from IS Mountain pliiteau on Kirkeborejn on Stronio. Bare »Rudemark« showing large and small slones blown togethcr into the cracks. (From photo. by Author). this cause, and an additional loss results from the high northerly latitude of the islands. The angle of incidence of the sun's rays to the earth's surface at noon in winter is only 5", and this only holds good if the surface is flat, a rare circumstance on the Færoes where almost the whole surface is more or less sloping. According to the exposure of the slope the vegetation receives light, together with heat, to a greater or less extent than on a flat surface. A slope facing south is thus more favourably situated in this respect than one facing northwards, and a marked difference in the com- position of the vegetation is the result. The southern exposures carry a great wealth of flowering piants, while the northerly ex- posures have a larger proportion of mosses. The Calluna or heather moors of the islands, for instance, are always met with on slopes facing southwards; similarly the rock-terraces (»Hamre«) towards the south always have the greatest variety of flowers. It follows naturally that piaces of this kind offer the best climatic 888 conditions for vegetation, and that heie it will reach its fullest de- velopment when the edaphic conditions are also favourable. This distinction between the north and south slopes is a result of the sun's influence, or, in other words, the combined eftects of the varying degrees of light, heat, and the amoiint of soil-moisture resulting from the exposure. It is a fundamental difference which we shall have occasion to refer to frequently in the chapter on plant- formations. 2. Edaphic factors. a. Nature of the soil. The geological slructure of the Færoes is very uniform (see geological account Vol. I, pp. 24 — 31). The islands consist entirely of basalt laid down in almost horizontal beds separated by thin strata of tuff and clay; on Sydero and Myggenæs the clay includes thin layers of coal. Both tuff and clay have originated from the basalt, hence this rock alone need be considered when dealing with the chemical composition of the Færoese soil. Mr. O. B. Boggild has drawn my attention to an analysis of Færoese basalt^, which states that it should contain 10, i6 p. cent CaO. Although this is the only definite statement available, Mr. Boggild still regards it as a fair estimate of the probable percentage all over the Færoes, since it is the amount found in most basalts, and they are fairly uniform. This is a high percentage of lime compared with the 1 — 2 p. cent in common granite, and piants need have no lack on this account. On the whole, basalt must be regarded as producing a fair soil for piants, partly because of its chemical com- position, and partly because of its comparatively rapid decay. The basalt on weathering forms a reddish-brown soil, of fine texture. The process of disintegration is most noticeable on the mountains, where the soil is more or less free from vegetation, and, if one may judge by personal observations, the weathering proceeds always very rapidly there. The numerous mountain-plateaux are covered as a rule with loose shattered flakes of basalt, with here and there a block falling into fragments, and one rarely sees large exposures of solid rock. The finer disintegrated particles soon settle down, among the coarser material, or they are transported by wind and water, to collect again in sheltered nooks, in cracks and crevices, etc. The frequent and abrupt changes between thaw ^ Durocher in Annales des Mines, 19, 1841, p. 559. 889 and frost in winter, togelher with the plenleous inin, which lodges in every depression, will considerably iiaslen the work of destruc- tion. A part will also be taken by the lichens and mosses, which thrive on the solid rock (e. g. Lec/c/ea-species, Placodwm and An- rfreaea-species). Thus it is that the products of weathered basalt form the pri- mary soils, in which piants settle and by their growth carry the transformation to further stages. The calcareons nature of the soil is still further increased on many parts of the coast by broken shells of Molluscs, fragments Fig. 169. Sandsbugt on Sando. Flat expanse of sand traversed \iy the stream in the foreground which bends and is seen again to the right ; the open sea in the distance, with a steamship at anchor. (From photo. by O. Eflfersoe). of Corallinaceæ, etc, which have drifted ashore. These fragments may be present in such quantity, that they give the soil a whitish tint; thus Hvidenæs (White Naze) a short distance north of Thors- havn derives its name from the abundance of shells there. The soil in piaces of this sort is naturally much more calcareous than elsewhere. On the coast where the finer particles of the soil are washed out, there remains a rather dark coarse-grained coast sand, which is presumably less fitted for the nutrition of piants than the un- washed soil; it never has, however, that white barren colour such as one finds, say on the washed out quartz-sand on the west coast of Denmark. Sand-strands are not common on the Færoes, yet may be found over small areas in almost any fjord. The greatest accumulation 890 of sand on the Færoes is found, as indicaled by the place-names, on Sando at the head of Sandsbugl. Here (Fig. 169) a wide flat stretch of sand has formed around the outlet of the river, and on the land- ward side the drifted sand forms a veritable dune. A dune-forma- tion in its earlier stages occurs also at Midvaag on Vaago. De- posits of coarse sand or gravel are also formed along the margins of numerous small lakes and streams which have washed away Fig. 170. Mountain plateau on Nolso. The wind has torn up mo.st of the plant-carpet and carried it off along with the fine-soil, leaving a bare gravelly plain with isolatcd plant-covered patches. (From photo. by E. Warniing). the finer particles of soil. The vegetation on these piaces is gene- rally sparse, but its meagre development ought not to be put down to any quality of the materials themselves, it is rather due to the unstable condition and the recent formation of the substratum. Similar conditions exist on the rocky floors of the upland plateaux, except that here wind more than water has carried away the finer soil (Fig. 170). The fine-soil appears to be an excellent medium for the growth of piants. Wherever it is allowed to lie at rest, it becomes covered by a closecarpetof piants, and these through their death and decay bring about great changes in difTerent ways according to the con- dition of moistness. 891 b. The moistness of the soil. Tlie abiindaiil rainfall gives rise to innumeiable watercourses of small size and ot'ten aclive for a short time only. One may safely say of the Færoes that they are irrigated by fresh-water. The water bears with it dissolved materials and the fine-soil, which accumulate when the course of the water is blocked. If the impediment is such as to prevent the water from going further, then it gathers to form a pool or a lake, according to circumstances. If the lake is fairly large, it never be- comes completely overgrown by vegetation, because growth pro- ceeds much too slowly for Ihis. If, however, the accumulalion of water is small, then gradually a swamp of Eriophornm is formed, followed by Carices, etc, and (inally Ihere remains a tirm bed of peaty matter. Peaty soil plays an important part on the Færoes. The ex- cessive moisture in the soil and air, along with the low temperature, result in that incomplete decomposilion of vegetable matter, which is characteristic of peat. The condition may be still further assisted by a close covering of piants with roots interwoven so as to hin- dering the access of air. In this way most soils with a fairly large supply of organic matter become humous^ Every kind of humous soil may be found from the fairly dry conditions in the smaller lowland tracts of Calluna-heath and the expanses of Grimmia-healh on the mountain-plateaux, onwards to the moist spongy swamp-peat. Much the greater part of the area under vegetation has peaty soil. Hence the Færoeses in bringing soil under cultivation encourage the outflow of water by draining and other aerating operations. They strive to change a hydrophilous sedge-moor into a mesophilous grass-meadow, and the soil from peat lo mould or loam. Mould. The mould-earth is confined as a rule to the cultivated areas, although beyond these it also occurs to a limited extent. When the drainage conditions are good and the situation favour- able, as on slopes and mountain-terraces (Hamre), one finds an abundant vegetation of grasses and herbaceous piants (Græsli) with a substratum of mould. As might be expected, every gradation occurs from mould to peat, and on quite small areas these changes may be so frequent as to give the vegetation a varied and motley appearance. In the lower zone of the hill pasture one may fre- ' For the formalion of humus in arctic countries see H. Hesselman: Om mykorrhiza-hildningar hos arktiska våxter; Bili. Sv. Vet. Akad. Handl., vol. 26, III. 2, 1900. 892 quently find monld and grass on Ihe more elevated parts, with a sedge-vegetalion on Ihc pealy soil of the depressions; Ihe iormer as it were on the crest ol" a wave, the latter in the tiough^ The soil of the Færoes in relation to the vegetation may be classified as follow^s: game Mineral soils Soils with much or 1. soHd rock "^^tt^^ 2. gravel and sand (sea-sand, 4. peaty soil calcareous sand) 5. mould. 3. fine soil. 3. The influence of man and animals on the vegetation. a. Man. The Færoes have been inhabited for about ten cen- turies by a race engaged mainly in the rearing of domestic animals and to a less extent in cultivating the soil. It is therefore only natural, tiiat this occupation by man has influenced the vegetation by altering or modifying its composition. It has already been stated by E. Warming (1903, p. 680) and myself (Ostenfeld 1901 c, p. 118), that a number of the piants of the Færoese flora have immigratcd through the aid of man 2. This has certainly played, and still plays an important part in the vegetation of northern countries, where in former times the Norsemen settled as colonists, for example in Iceland and the southern part of Greenland. It is also probable that the presence of several European species (e. g. Calliina) in Newfoundland and adjoining lands, is due to the roving propensi- ties of these ancient Norsemen. It is not my intention to enter further into this matter here, but to confine myself to the influence of man upon the vegetation as it exists at the present time. On the Færoes the only vegetation entirely produced and main- tained by human action is that limitcd to the small areas on which barley and potatoes are cultivated and to numerous small gardens ^ Earthworms, so characteristic of mould-formation. also occur on the P"æ- roes. Several of the lesser species of the genus Lambriciis appear to he common. According to the Identification of the Zoologicai Museum of Copenhagen, my col- lections include no less than 4 species (L. tnrgidus, piirpiireus, Boeckii and siib- nibiciindus). ■^ Professor W. says (1. c, p. 680): Ostenfeld mentions it, but very briefly (p. 117), as follows: »Further. . . man has doubtless introduced and keeps on introducing new species« ; but he appears to have overlooked my further reference to the question on the page which follows. * 893 laid out especially in recent years and principally at Thorshavn. The vegetation here consists partly of the few species under culti- vation, and partly of weeds, most of which arc piants native to the Færoes. A similar vegetation of weeds or ruderal species may be found round the houses and along paths and roads in the rural districts (e. g. Montia lamprosperma and Cardamine hirsuta). One must also include as products of cultivation the »Bo«, fenced and drained enclosures, with a vegetation which will be dealt with in detail later. Outside the enclosed area, traces of human influence are not evident to any noteworthy extent, except the change brought about by the removal of peat and turf, which produces other conditions in the amount of moisture and thereby affects the original vegetation. b. Domestic animals. Sheep play far and away the greatest part among the domesticated animals. There are on the Færoes rather more than 15,000 inhabitants and about 100,000 sheep, in other words about 7 sheep per head of population. These sheep are allowed to roam at large all the year round, and their influence upon the vegetation is undoubledly enormous. Outside the enclosed land, the vegetation is almost everywhere closely clipped by the grazing of the sheep, and the taller piants have little chance of flowering and fruiting. F'ull development is only possible in piaces inaccessible to sheep, such as mountain-terraces and small Islands in lakes^ A typical example of this was found on a little isle (Holm) in Vatnsdals-lake on Sydero between Trangisvaag and Kvalbo. When I visited the valley in the summer of 1897, I was impressed with the uncommon appearance of this isle when seen from a dis- tance. The green vegetation looked as if flecked with white, and on closer inspection this was seen to be due to numerous inflore- scenses of Eriophoriim polystachyiim. This plant was also quite common in the swampy vegetation on the shores of the lake, but one might search long and not find a single inflorescence here, where the sheep had free access. Numerous stalks could be found which ought to have borne inflorescences, but these were all bitten ^ Mr. W. H. Beeby informs me that this is also tlie case in Shetland; lie saj's, one must search for the less common species on the small Islands (Holms) in the lakes, and there also the best developed specimens of the commoner species are found; several rare species have hitherto only been recorded from the >holms« (cfr. W. H. Beeby, Ann. Scott. Nat. Hist., 1907, p. 236). 894 off. On wading lo Ihe isle, I could see that it was covered by a vegetation, consisting maiiily of the same Eriophorum, Liiziila sil- vatica and Cnrex binernis, all the piants with an average height of 75 ctm. Indeed, one wonders liow the Fæiocse vegetation would look if there were no sheep ! The inflnence of the sheep is more perceptible on the vegeta- tion of lower levels, less so in the mountains, where only a sparse vegetation occurs (the rock-flat formation and the Grimmia-heath). Cows, horses and geese, the other domesticated animals on the Færoes, are of much less importance, as they are fewer in number, and, so far as the cattle are concerned, they obtain their food partly from the enclosures. I believe one may say without exaggeration, that the character and featnres of the vegetation outside the enclosures in the lower zones o f the Færoes are in a h i g h d e g r e e due to the grazing of sheep. c. Birds. The cliffs of the Færoes are well known as the nesting piaces of thousands of sea-birds, — Guillemots (Urin grylle and U. troile), Puffins (Fraterciila arctica), Gulls (Lams tridactijhis, etc.^, Fulmar-Petrels (Fulmarm glacialis) and Cormorants (Phalacro- corax). As a natural result, the vegetation on these cliffs has a character of its own. The bird-manure containing a large propor- tion of nitrogenous matter (uric acid, etc.) furnishes, more or less directly, good nutriment for the piants. The sea-fowl cliffs can almosl always be distinguished, even from a distance, by the vege- tation, which consists of tall, vigorous piants. Some piants, such as »Kvan« (Archangelica officinalis), are nearly only to be found growing wild on the sea-fowl cliffs, and others occur there in specially luxuriant forms. On the other hånd a number of piants, especially mosses, are entirely cleared out by the activity and the excrements of the birds. During an excursion to Nolso in August 1897 I had occasion to observe the vegetation on a large talus of débris (Ur) situated on the east side. Here the puffins (»Lunder«) nested at some piaces, but not at others, and there was a very noticeable difference in the appearance and composition of the vegetation. Piaces fre- quented by the puffins were dominated by a luxuriant blue-green form of Festuca riibra (»I^undasina»); and on one of the smaller patches only 8 phanerogams and 7 mosses were noted. Much richer however was the vegetation on the parts of the »Ur« not 895 distiirbed by puffins; we noted liere 27 phanerogams and 33 mos- ses; the latter being predominaiit and giving the vegetation its character, while the phanerogams taken altogether could not be compared in quantity and luxuriance with the mass of Festuca of the puffm-Ur. This example will illustrate how birds may exert considerable influence in selecling the piaces where they breed, and in altering the vegetation on them. Something of the same kind happens on Myggenæsholm , the nesting place of a large colony of Gannets (Siila bassana), and on Kirkeboholm, where the Eiderducks (Somateria mollissima) nest. The other wild animals are of secondary importance so far as vegetation is concerned. The rabbit (Lepiis timidiis) is not nume- rons enough to play an important part. Invertebrates may also be left out of account^. III. SOME BIOLOGICAL FEATURES. Before proceeding to the description of the plant-formations, it may be well to notice some biologicai features of the species ap- pearing therein, confining ourselves, however, to the higher piants. The features dealt with in this chapter are: 1 '^ Duration of life. 2^ Biological types, in Raunkiær 's acceptation. 3^ Vegetative propagation and structure of the shoots. 4" Time of flowering. 5*^ Maturation of fruit. 6*^ Distribution in attitude. It seems advisable that I should give an alphabetical list of the vascular piants which, in my opinion, are to be regarded as native or as quite naturalized in the Færoes. The species listed are those indicated by numbers in my supplementary floristic list (Ostenfeld 1907), but I have omitted the species which are not completely naturalized. ^ As alreach' stated. cartlnvorms and other animals, including bacteria, take part in the formation of monld, and hence may be regarded as indirectly impor- tant for vegetation. 896 Alphabetical list of the vascular piants (phanerogams and pterid oph y tes), growing wild or quite n a I u r a 1 i z e (1 in the Færoes. Tlic species with names printed in ilalics are of comnion occurrence. KEY TO THE ABBREVIATIONS AFTER THE PLANT-NAMES. Biologicai type (R a u n k i æ r). Ch. -= Chamæphyte. G. = Geophyte. HH. = Helo- or Hydrophyte, Her. = Hemicryptopliyte. Nph. = Nanophanerophyte. Th. = Therophyte. Duration of li fe. I annual (summer-annual). II ^ hapaxanthic (i e. flowerlng onl}' once), but not annual. d. = perennial without power of migration (se- dentary or spot-bound). perennial with subterrean wandering shoots. perennial with epiterrean wandernig shoots. ;. sub. epi. Distribution in Altitude. suljalp. str. ---= Species found only in the lowland and the lower parts of the mountain slopes. sulKiip. lat. - species found in the lower regions, and also exccptionally on the mountain-piatcaux. su!)a!p.-alp. species found both in the lower regions and on the mountain-plateaux. alp. lat. — = species found on tlie mountain-plateaux, and also exceptionally in the lower regions, alp. str. = species found only on the mountain-plateaux. 1 Achillca millefolhim Her. 2 — ptarmica Her. 3 Agropyrum junceum G. 4 -- repens G. 5 Agroslis canina Her. G — stolonifera Her. 7 — valgavis Her. 8 Aiva alpina Her. 9 — caespitosa Her. 10 — flexnosa Her. 1 1 Airopsis praeeox Th. 12 Alchimilla aciilidens Her. 13 — alpina Her. 14 — faeroensis Her. 15 — filicaiilis Her. 16 Alectorolophus groenlandieus . . . Th. 17 — minor Th. 18 Alopecnrns geniculatiis Her. 19 Alsine verna, hirta Her. 20 Anagallis tenella Ch. 21 Angelica silvestris Her. 22 Anthoxanthnm odoratnm Her. 23 Arahis petraea Ch. 24 Airhangelica officinalis Her. sub. subalp. str. sub. subalp. str. sub. subalp. sir. sub. subalp. str. epi. subal]).-alp. epi. subalp. lat. sub. subalp. lat. sed. subalp.-alp. sed. subalp.-alp. sub. subalp.-alp. I. subalp. str. sed. alp. lat. sed. alp. lat. sed. alp. lat. sed. subalp.-alp. I. subalp. lat. I. subalp. str. epi. subalp. str. sed. alp. str. epi. subalp. str. sed. subalp. lat. sed. subalp. lat. sed. alp. lat. sed. subalp.-alp. 897 25 Armeria elongata Her. 26 Aspidium dryopteris G. 27 — filix mas Her. 28 — lonchitis Her. 29 — phegopteris G. 30 — spinalosiim, dilatatum . Her. 31 Asplenum adiantuni nigruni .... Her. 32 — triehomanes Her. 33 Alhyriiim filix focmina Her. 34 Atriplex Babinglonii Th. 35 Bartschia alpina Her. 36 Bellis pereniiis Her. 37 Blechmim spicant Her. 38 Botrychium hinaria G. 39 Brunella valgaris Her. 40 C ak ile maritima Th. 41 Callitriehe autumnalis HH. 42 — hamulala HH. 43 — stagnalis HH. 44 Calliina viilgaris Ch. 45 Caltha pahistris Her. 46 Campanula rotundifolia Her. 47 Capsella bursa pas toris Th. 48 Cardamine hirsiila Th.andHcr. 49 — pratensis Her. 50 Carex atrata Her. 51 — bineruis Her. 52 — dioica G. 53 — echinala Her. 54 — flacca G. 55 — flava Her. 56 — fulva Her. 57 — Goodenoughii G. 58 — ineurva G. 59 — leporina Her. 60 — Lyngbyei G. 61 — panicea G. 62 - pilulifera Her. 63 — piilicaris Her. 64 - pulla G. 65 — rigida G. 66 — salina, kattegalensis G. 67 Catabrosa aquatiea Her. 68 Cerastiiim Edmondstonii Ch. sed. 69 — glomeratum Th. 70 — tetrandum Th. 71 — trigN^num Ch. 72 — valgåre Ch 73 Chamaenerium angustifolium . . . Her. 74 Cirsium palustre Her. scd. subalp.-alp. sul). subalp. sir. sed. subalp. str. sed. alp. str. suh. subalp. lat. sed. subalp. str. scd. subalp. sir. sed. subalp. str. sed. subalp. str. I. subalp. str. sed. alp. str. CJ)!. subalp. str. sed. subalp.-alp. sed. subalp.-alp. epi. subalp. str. I. subalp. str. epi. subalp. str. cpi. subalp. str. epi. subalp. str. sed. subalp. lat. sed. subalp. lat. sed. subalp.-alp. I. subalp. str. II. subalp. lal. sed. subalp.-alp. sed. alp. str. scd. subalp. str. Slib. subalp. str. sed. subalp. lat. sul). subalp. str. scd. subalp. lat. sed. subalp. str. sub. subalp.-alp. sub. subalp. str. sed. subalp. str. sub. subalp. str. sub. subalp. lat. sed. subalp. sir. sed. subalp. str. sub. alp. str. sub. alp. str. sub. subalp. sir. epi. subalp. str. fand epi. ) alp. lal. I. subalp. str. I. subalp. str. epi. alp. str. sed. subalp.-alp. sub. subalp. str. II. subalp. str. 898 75 Cochleario officinalis Her. 76 Cornus snecica Her. 77 Cystopieris fragilis Her. 78 Digraphis arundinacea HH. 79 Draba hirta Ch. 80 — incana Her. 81 Drosera rotundifolia Her. 82 Dryas oetopetala Ch. 83 Elymns arenariiis G. 84 Empelrnin nigriim Ch. 85 Epilolnnm alsinifoliiim Her. 86 — anagallidifolium .... Her. 87 — lacliflonim Her. 88 — montanum Her. 89 — paliistre Her. 90 Eqnisetiim arvense Her. 91 — heleocharis HH. 92 — paliistre Her. 93 — pratense Her. 94 — silvaticiim Her. 95 Erica cinerea Ch. 96 Eriophoriim polystachyiim G. 97 — vaginatum Her. 98 Eiiphrasia borealis Th. 99 — curta Th. 100 — minima Th. 101 Festnca ovina Her. 102 — riibra G. 103 Galeopsis ktrahil Th. 104 Galiiim paliistre Her. 105 — saxalile Ch. 106 Gentiana eampestri.s Th. 107 Geranium silvaticiim Her. 108 (ieum rivale Her. 109 Glyceria distans Her. 110 — fluitans HH. 111 — maritima Her. 111 Gnaphalium supinum Her. 113 Habenaria albida G. 114 — viridis G. 115 Haloseias seotieum Her. 116 Heleoeharis niultieaulis Her. 117 — paluslris G. 118 — unigluniis G. 119 Hieraeium ardisodon Her. 120 — ciliolatum Her. 121 — eonstrietiforme Her. 122 — cordifrons Her. 123 — epileueoides Her. 124 — epileucuni Her. n. subalp.-alp. sub. subalp. lat. sed. subalp.-alp. sub. subalp. str. .sed. alp. lat. sed. subalp.-alp. .sed. subalp. sir. epi. alp. str. sub. subalp. str. epi. (and sed.) subalp.-alp. sub. subalp.-alp. epi. alp. lat. sub. subalp.-alp. sed. subalp. str. sub. subalp. str. sub. subalp. str. sub. subalp. str. sub. subalp. str. sub. subalp. str. sub. subalp. str. sed. subalp. lat. sub. subalp.-alp. sed. subalp.-alp. I. subalp. str. I. subalp. lat. I. subalp.-alp. sed. subalp.-alp. sub. subalp.-alp. I. subalp. str. sub. subalp. str. epi. subalp. str. II. subalp. lat. sed. subalp. str. sed. subalp. str. sed. subalp. str. epi. subalp. str. epi. subalp. str. epi. alp. str. sed. subalp. lat. sed. subalp. lat. sed. subalp. str. sed. subalp. str. sub. subalp. str. sub. subalp. str. sed. subalp. str. sed. subalp. str. sed. subalp. str. sed. subalp. str. sed. subalp. str. sed. subalp. str. 899 125 Hieracium faerocnse Her. sed. 126 — Hartzianum Her. sed. 127 -- heterophyllum Her. sed. 128 — kalsocnse Her. sed. 129 — leucograptum Her. sed. 130 — melanoehrotum Her. sed. 131 — Ostenf'eldii Her. sed. 132 — perampliforme Her. sed. 133 — peramplum Her. sed. 134 — perintegrum Her. sed. 135 — sareophylloides Her. sed. 136 — seotieifornie Her. sed. 137 — Simmonsianum Her. sed. 138 — siibnibiciindiim Her. sed. 139 — vcterascens Her. sed. 140 Holciis lanatiis Her. sed. 141 — nwUis G. sub. 142 Honckenya peploides Her. sub. 143 Hymenophyllum pellatnm Ch. epi. 144 Hypericnm piilchnim Her. sed. 145 — quadrangulum Her. sed. 146 Iris pseudaeorus HH. sed. 147 Isuctes echinosporiini HH. sed. 148 — laciistre HH. sed. 149 Juneus baltieus G. sub. 150 — biglumis Her. sed. 151 — biifonius Th. I. 152 — cong lorne ratiis Her. sed. 153 — effusiis Her. sed. 154 — lampocarpns Her. sed. (and sub.) subalp. str. 155 — obtusitlorus G. sub. subalp. str. 156 — sqiicurosus Her. sed. subalp.-alp. 157 — siipiniis Her. epi. (and sed.) subalp. str. 158 — trifidns Her. sed. 159 — triglumls Her. sed. 160 Juniperus communis, nana .... Npli.(andCh.) sed. 161 Koenigia islandica Th. I. 162 Lathyrus pratensis Her. sub. 163 Leontodon antiimnale Her. sed. 164 Linum catharticiiin Th. II. 165 Listera eordata G. sub. 166 Litorella lacastris HH. epi. 167 Lobelia dortmanna HH. sed. 168 Loiseleuria procumbens Ch. sed. (and epi) 169 Lotus eornieulatus, earnosus . . Her. sed. 170 Luzula areuata Her. sub. 171 — campestris Her. sub. 172 — multiflora Her. sed. 173 — silvatica Her. sed. 1 74 — spicata Her. sed. Botany of the Færoes. 58 subalp. str. subalp. str. subalp. str. subalp. str. subalp. str. (?) subalp. lat. subalp. str. subalp. str. subalp. str. subalp. lat. (?) subalp. lat. subalp. str. subalp. str. subalp. str. subalp. str. subalp. str. subalp. str. subalp. str. subalp. str. subalp. lat. subalp. str. subalp. str. subalp. lat. subalp. lat. subalp. str. alp. str. subalp. str. subalp. str. subalp. str. subalp.-alp. alp. lat. subalp. lat. subalp.-alp. subalp. str. subalp. lat. subalp. str. subalp. str. subalp. lat. subalp. str. alp. str. subalp. str. alp. str. subalp. str. subalp. lat. subalp. lat. subalp.-alp. 900 175 Lychnis flos ciiciili Her. sed. subalp. str. 176 Lycopodium alpimim Ch. epi. subalp.-alp. 177 — annotinuni Ch. epi. subalp. str. 178 — selago Ch. sed. subalp.-alp. 179 Lysimachia nemorum Ch. epi. subalp. str. 180 Malaxis paludosa Her. sed. subalp. str. 181 Matricaria inodora, phaeocephala Her.(andTh.) sed. subalp. str. 182 Melandrium rubrum Her. sed. subalp. lat. 183 Mentha aquatica Her. sub.(andepi) subalp. str. 184 xMenyanthes trifoliata HH. sub. subalp. str. 185 Mertensia maritima Her. sed. subalp. str. 186 Molinia coenilea Her. sed. subalp. lat. 187 Montia lamprosperma Th.andHH. sed.(and epi) subalp. lat. 188 Myosotis aruensis Her. H. subalp. str. 189 — palustris, strigulosa . . . Her. epi. subalp. str. 190 — repens HH. epi. subalp. str. 191 — versicolor Th. I. subalp. str. 192 Myriophyllnm allerniflonim .... HH. sub. subalp. lat. 193 Nardus stricta Her. sed. subalp.-alp. 194 Narthccium ossifragum Her. sub. subalp. lat. 195 Orehis latifolius G. sed. subalp. str. 196 — nmciilatus G. sed. subalp. lat. 197 — masculus G. sed. subalp. str. 198 Oxalis aeetosella Her. sub. subalp. str. 199 Oxyria digyna Her. sed. subalp.-alp. 200 Papaver radicatum Her. sed. alp. str. 201 Pedieularis palustris Her. H. subalp. str. 202 Phragmites eommunis G. sub. subalp. str. 203 Pinguiciila vulgaris Her. sed. subalp.-alp. 204 Plantage eoronopus Th. II. subalp. str. 205 — lanceolata Her. sed. subalp. str. 206 - maritima Her. sed. subalp.-alp. 207 Poa alpina Her. sed. alp. lat. 208 — onmia Th. I. subalp. str. 209 - glaiica Her. sed. subalp.-alp. 210 — nemoralis Her. sed. subalp. lat. 211 — pratensis Her. sub. subalp. str. 212 — triuialis Her. sed. subalp. str. 213 Polygala serpyllacea Ch. sed. subalp. lal. 214 — vulgaris, Ballii Her. sed. subalp. str. 215 Polygonum aniphibiuni Her.(andHH.) sub. subalp. str. 216 — auiculare Th. I. subalp. str. 217 — viviparum G. sed. subalp.-alp. 218 Polypodium vulgare Her. epi. subalp. str. 219 Potamogeton alpinus HH. sub. subalp. str. 220 — filiformis HH. sub. subalp. str. 221 — gramineus HH. sub. subalp. str. 222 — natans HH. sub. subalp. str. 223 — perfoliatus HH. sub. subalp. str. 224 — polygonifolius HH. sub. subalp. lat. dOl 225 Potamogeton praelongus HH. 226 — pusillus HH. 227 Potentilla anserina Her. 228 — erecla Her. 220 — palustris HH. 230 - verna Her. 231 Primula acaulis Her. 232 Psamma arenaria G. 233 Pyrola minor Her. 234 Ranuncnhis acer Her. 235 — aurieomus Her. 236 - /lammiila Her. 237 — glacialis Her. 238 — repens Her. 239 - reptans HH. 240 Rosa mollis Nph. 241 Riibiis saxatilis Her. 242 Riimex acelosa Her. 243 — domcsticiis Her. 244 — obtiisi/olhis Her. 245 Ruppia maritima HH. 246 Sagina nivalis Her. 247 — prociimbens Ch. 248 — siibiilata Her. 249 Salix glauca Ch.(and 250 — herbacea Ch. 251 — phylieifolia Nph. 252 Saxifraga caespitosa Ch. 253 — hypnoides Ch. 254 — nivalis Her. 255 — oppositifolia Ch. 256 — rivularis Her. 257 — stellaris Her: 258 Seilla verna G. 259 Scirpiis caespitosus Her. 260 — paueiflorus Her. 261 Sediim rhodiola Her. 262 — villosnm Her. 263 Selaginella selaginoides Ch. 264 Senecio viilgaris Th. 265 Sibbaldia prociimbens Ch. 266 Sieglingia deciiinbens Her. 267 Silene acaulis Ch. 268 Sparganium af fine HH. 269 Spergiila arvensis Th. 270 Spiraea iilmaria Her. 271 Stellaria media Th. 272 — iiliginosa Her. 273 Subularia aquatiea Th. 274 Siiccisa pratensis Her. sub. subalp. str sed. subalp. str. epi. subalp. str. sed. subalp.-alp. sub. subalp. str. sed. alp. str. sed. subalp. str. sub. subalp. str. sub. alp. lat. sed. subalp.-alp. sed. subalp. str. epi. subalp. lat. sed. alp. str. epi. subalp. str. epi. subalp. str. sed. subalp. str. epi. subalp. lat. sed. subalp.-alp. sed. subalp. str. sed. subalp. str. sub. subalp. str. sed. alp. str. epi.(andsed.) subalp.-alp. sed. subalp. lat. ^^ph.) sed. alp. str. sub. subalp.-alp. sed. subalp. str. sed. subalp.-alp. epi. subalp.-alp. sed. alp. lat. sed. (and epi.) alp. lat. sub. alp. str. sed.(and sub.) subalp.-alp. sed. subalp. str. sed. subalp.-alp. sub. subalp. str. sed. subalp.-alp. sed. subalp.-alp. sed. subalp.-alp. I. subalp. str. epi. alp. str. sed. subalp. str. sed. subalp.-alp. sub. subalp. lat. I. subalp. str. sed. subalp. str. I. subalp. str. epi. subalp. str. I. subalp. lat. sed. subalp. str. 902 275 Tanacetiim vnlqare Her. 276 Taraxacum naevosiim Her. 277 — spcclabile, var Her. 278 Thalidnim alpimim Her. 279 Thymiis serpylhim Ch. 280 Tofieldia palustris Her. 281 TrifoUiim repens Her. 282 Triglochin palustre Her. 283 Tussilago farfarus G. 284 Urtica dioica Her. 285 Utricularia vulgaris HH. 286 Vaccinium myrtillns . Ch.fandNph 287 — uliginosum Ch.(andNph 288 — vitis idaea Ch. 289 Veronica alpina Her. 290 — beccabunga Her. 291 — frutieans Ch. 292 — officinalis .. Ch. 293 -- serpyllifolia Her. 294 Vicia eraeea Her. 295 Viola palustris Her. 296 — silvestris, rotundato-crenala Her. 297 — trieolor, faeroensis Her. 298 Zostera marina HH. sub. subalp. str. sed. subalp.-alp. sed. subalp.-alp. sub. subalp.-alp. epi. subalp.-alp. sed. alp. str. epi. subalp. str. sub. subalp. str. sub. subalp. str. sub. subalp. str. sed. subalp. str. sub. subalp.-alp. sub. subalp.-alp. sub. subalp. lat. epi. alp. str. epi. subalp. str. sed. alp. str. epi. subalp.-alp. epi. subalp.-alp. sub. subalp. str. sub. subalp. lat. sed. subalp.-alp sed. subalp. str. sub. subalp. str. 1. Duration of life. Annuals and other hapaxanthic species which oniy flower once, although not necessarily in the first year of growlh, are not strongly represented in the natural plant-formations of the Færoes. Even in the culture-formations, perennial species are dominant except in cornfields. It is natural to expect that the insular climate with mild winters, cold summers and high rainfall Ihroughout the year, must inlluence in a favourable way the dominance of perennial species. The list of piants shows that there are 298 wild species of vas- cular piants in the Færoes. The preponderance of perennial species will be seen at a glance from the following tabulated summary; I gives the number of summer-annuals, II the olher hapaxanth- species, and III the perennial species. Actual numbers. Percentages. I II III Total I II III Total Vascular Cryptogams . Gymnosperms .... . I 24 24 .: 1 1 1 .1 92 i 95 152 1 178 i 100 100 100 100 Monocotyledons Dicotyledons 3 18 "s" 3,2 96,8 10,1 ! 4,5 85,4 100 100 Vascular piants. . . 21 8 1 269 1 298 7,0 2,7 1 90,3 100 903 Certain species merit special attention, which in other countries may be hapaxanlhic, but in the Færoes, in my opinion, they must be reckoned among the perennials: Montia lamprosperma occurs both as an annual and as a perennial, because it is so in the natural formations, while it is only in cornfields and gardens that its annual form is found. The case is much the same with Matricaria inodora, var. phaeocephala, which is perennial in the sand- strand formation, but perhaps becomes annual, when it occurs as a rare plant in potato-fields near the coast. The condition is less clearly defined with regard to Cardamine hirsnta; in shady piaces among cliffs, it is certainly not annual, but still, as a rule hapaxan- lhic, hence I have placed it in the second column; as a weed in gardens it may sometimes be an annual. In an earlier paper (Bor- gesen and Ostenfeld Hansen 1896, p. 146), I pointed out that Cochlearia officinalis in the Færoes sometimes behaves as a peren- nial plant, since the same individual is able to produce flowers more than once (see Fig. 171); this, however, is exceptional. This tendency to })ass from hapaxanthic to perennial has gone farther in Viola tricolor and in Alopecurns geniculatns, both of which are true perennials in the Færoes. The table shows that of the 298 vascular piants not less than 269 (c. 90 p.ct.) are perennial, and only 29 (c. 10 p.ct.) are hapaxanthic. (I). Of the latter 21 (c. 7 p.ct.) are sum nier-annuals viz: Airopsis praecox (c) Galeopsis tetrahit (c) Alectorolophus groenlandicus (p) Juncus bufonius (c) — minor (p) Koenigia islandica (n) Atriplex Babingtonii (h) Myosotis versicolor (c) Cakile maritima (h) Poa annua (c) Capsella bursa pasloris (c) Polygonum aviculare (h) Cerastium glomeratum (c) Senecio vulgaris (c) — tetrandrum (h) Spergula arvensis (c) Euphrasia borealis (p) Stellaria media (c) — curta (p) Subularia aquatica (w) — minima (p) The habitats of these annual piants are indicated after the names: c, meaning that the species occurs in the culture-formations, /?, in the halophile fermations, n, in the natural inland-formations, IV, that it is a water-plant, and p, that is a parasite. We find that 904 as m a n y as 10 o f t h e 21 a n n u a 1 species have t h e i r home in the culture-formations , 4 in the halophile formations (sand- strand formation), and 5 are parasites (all belonging to the Rhin- antheæ). Of the remaining 2 species, one is a water-plant, so that there remains only one single annual species — Koenigia Kig. 171. Perennial specimens of CochlearUi officinalis from a wet rock-ledge at Trangisvaag on Sydero, 9th May 1895. /. Stem-rudiments from tlie first period of flovvering, // from the second period, ///. stem witli flowering scapes of the third period. i. long internodes. a. specimen at least 4 years old, probably 5 years. b. specimen at least 3 years old, probably 4 years. islandica — as an inhabitant ofthe nat u ral land- vegetation of the Færoese interior (the parasite Rhinantheæ of course ex- cepted). The annual species confined to the cnltivated soil are weeds in the fieids and gardens, along roadsides and round houses, and it is open to suppose that they owe their existence on the Færoes to mankind, but as they are so common I have included them as quite naturalized species. We are therefore justilied in describing the F'æroese vegetation as relatively deficient in annual herbs. The reason for Ibis may be traced, as suggested above, to the insular climate and the occurrence 905 of rain throughout the whole year. Perennial species are thus en- abled to conlinue their growth nearly all the year, the winter rarely causing a complete cessation; yet growth as a whole is a slow^ process because of the constant low temperature, and the fre- quent interruptions. (II). The other hapaxanthic species (column II in the table) in- clude: (a) wintering-annual herbs — i. e. piants which ger- minate in autumn, but do not flower until next summer; (hj true biennial or pleiocyclic piants. The list comprises 8 species (all dicotyledons): Cardamine hirsuta (n) Linum catharticum (n) Cirsium palustre (n) Myosotis arvensis (c) Cochlearia officinalis (n, h) Pedicularis palustris (p) Gentiana campestris (n) Plantago coronopus (h) The letters in brackets indicate the habitat as in the case of annuals. The difference betw^een the habitat of the two categories of hapaxanthic herbs is a striking one. Whereas the majority of the summer-annuals (I) inhabit cultivated land and the sea-shore, 5 of the species (II) given above have their home in natural inland forma- tions of the lowlands, such as mountain-crags with a good exposure, rock-ledges and slopes; only one species, Myosotis arvensis, occurs in the »Bo« (the cultivated meadow); one is a parasite, and one a true shore-plant; Cochlearia officinalis is both a shore-plant and a mountain-plant, being most frequent on bare gravelly piaces on the hill-plateaux. Thus, most of the small number of Færoese hapaxanthic species which are not summer-annuals, live in the more luxuriant habitats of the lowland. It looks as if they were not quite happy in their surroundings, and must resort to these piaces which receive most sunshine and w^here the soil is a well-drained mould. (III). The perennial pian