THE BOTANY OF ICELAND EDITED BY L. KOLDERUP ROSENVINGE PH. D. AND EUG. WARMING PH. D., SC. D. VOL.1 WITH ONE PLATE AND 82 FIGURES IN THE TEXT COPENHAGEN J. FRIMODT LONDON JOHN WHELDON & CO. 1912-1918 PUBLISHED BY THE AID OF THE CARLSBERG FUND H. H.THIKLKS I'.OCTKYKKKRI. KOBKNHAVN CONTENTS PART I. Page PREFACE Ill 1. HELGI JONSSON: The Marine Algal Vegetation of Iceland. 1912 1 2. TH. TIIORODDSEN: An Account of the physical Geography of Iceland. 1914 187 PART II. 1918. 3. ERNST 0STRUP: Marine Diatoms from the Coasts of Iceland (with one Plate). (Printed in 1916) 345 4. AUG. HESSELBO: The Bryophyta of Iceland 395 THE BOTANY OF ICELAND EDITED BY L. KOLDERUP ROSENVINGE PH. D. AND EUG. WARMING PH. D.. SC. D. PART I 1. THE MARINE ALGAL VEGETATION BY HELGI JONSSON, PH. I). (PUBLISHED BY THE AID OF THE CARLSBEHG FUND) COPENHAGEN J. FRIMODT LONDON JOHN W HELD ON & CO 1912 PRINTED BY H. H. THIELE PREFACE. IT was mentioned in the preface lo the "Botany of the Faeroes" (Co- penhagen & London, 1901 — 1908) that, on the completion of that work, Iceland would be the one island among the dependencies of the Danish kingdom in the Atlantic which was in most need of a thorough and systematic investigation as regards its botany, and the hope was expressed that this would be commenced as early as the year 1909. This hope has been so far realized that we, the undersigned, are now able to publish the first paper on the subject, viz. "The Marine Algal Vegetation" by Dr. HELGI JONSSON of Reykjavik. Iceland, however, is so large compared with the Faeroes that the investigation will not only be far more difficult to carry out, but \vill also extend over a far longer period. We hope that specialists in botany may be able, at short intervals, to visit the island and make collections and notes. Thus, even in 1910 a young bryologist, A. HESSELBO, studied the moss -vegetation when he took the photographs, published in the present paper - and this summer he is again paying a visit to the island, after which he will prepare an account of the moss-flora and the moss- vegetation of Iceland. Next summer we hope that a young lichenologist will be able to set to work in a similar manner, and will be followed by others, until the work can be completed with a general account of the vegetation and the plant-geographical position of the island. L. KOLDERUP ROSENVINGE. EUG. WARMING. COPENHAGEN, AUG. 1912. 1. THE MARINE ALGAL VEGETATION OF ICELAND BY HELGI JONSSON PH. D. WITH 7 FIGURES IN THE TEXT CONTENTS. Patfe Introduction 1 I List of the Marine Alga? 5 II Life-conditions of the Marine Algal Vegetation 25 1. The Nature of the coast 25 2. The Ocean 27 A. The Movements of the Ocean 27 B. The Temperature of the Water 32 C. The Salinity 40 3. The Air 42 A. The Temperature 42 B. The Humidity 44 C. Precipitation, Amount of Cloud, Foggy days, Wet days 44 D. Winds ' 46 4. The Light 47 III The Horizontal Distribution of the Species and the Components of the Algal Flora 48 The Components of the Algal Flora 58 A. The Arctic Group 58 B. The Subarctic Group 59 C. The Boreal-Arctic Group 60 D. The Cold-Boreal Group 61 E. The Warm-Boreal Group 62 Floristic Boundaries 66 IV Comparison with neighbouring Floral Districts 68 V The Vertical Distribution of the Species 80 A. Upper Littoral Zone 85 B. The lower Littoral Zone and the Belt below down to a depth of about 10 metres 88 C. The Sublittoral Species ... 91 Lower Limits of Growth . 92 Page VI Marine Algal Vegetation and Sea-grass Vegetation 95 Account of the Marine Algal Vegetation and the Sea-grass Vegetation ... 98 A. The Marine Algal Vegetation 99 a. The Littoral Zone 99 aa. The Photophilous or strictly Littoral Communities 99 1. The Prasiola stipitata-association 100 2. The Community of Filiform Algae 102 3. The Community of Fucacea? 106 4. The Enteromorpha-association 120 5. The Acrosiphonia-association 121 bb. The Shade-vegetation 123 6. The Hildenbrandia-association 123 7. The Rhodochorton-association, the Sphacelarietum bri- tannici and the Polysiphonietum urceolatse 123 cc. The Vegetation of Tide-Pools 125 b. The Semi-littoral Zone 128 8. The Monostroma-association 129 9. The Chorda-association 130 10. The Community of Rhodymenia 132 11. The Polysiphonia urceolata-association 133 12. The Community of Corallina 135 13. The Crustaceous alga-association 138 c. The Sublittoral Zone 139 14. The Community of Laminariacea? 140 15. The Desmarestia-association 151 16. The Deep-water Community of Floridea? 152 17. The Lithothamnion-association 154 18. The Community of Crustaceous Alga? 155 B. The Sea-grass Vegetation 157 The Zostera-association 157 VII Differences in the Vegetation in East and South Iceland 159 VIII Some Notes on the Biology of the Alga? along the coast of Iceland 166 1 . Duration of Life . 166 • 2. Periodical Changes 169 3. Littoral Winter-vegetation of Reykjavfk 180 Bibliography 183 INTRODUCTION. IT is far from being a fact that the Marine Algal Flora and Marine Algal Vegetation of Iceland can be regarded as sufficiently known; this does not apply in the same degree, however, to all parts of the coast. East Iceland, South Iceland and South-west Iceland are better known in this respect than North-wrest Iceland and North Iceland. Also, as is very natural, the littoral vegetation is better known than the sublittoral, as it is easier of access and may be investigated directly on the spot, while, as regards the sub- littoral vegetation , one has to be content with what is obtained from dredgings. Very little has previously been written with regard to the marine algal vegetation of Iceland. Strom felt, who travelled in Iceland in the summer of 1883, has treated the algal flora exhaustively (see Jonsson, 31) in his valuable work "Om Algevegetationen vid Is- lands kuster" (70) and has given a critical review^ of the older lite- rature of the marine algal flora of Iceland; but, on the other hand, he has dealt very briefly with the marine algal vegetation. He fol- lows Kj ell man in dividing the vegetation in question into a litto- ral and a sublittoral vegetation. Stromfelt found the littoral vege- tation poorly developed in many places he records, however, a luxuriant littoral vegetation from Reykjavik, Eyrarbakki and Eskifjor5ur. The sublittoral vegetation is mentioned even more briefly, and is emphasized as being more uniform than the littoral as regards its distribution and the species which compose it. Strom- felt does not make any definite statement regarding the elittoral vegetation, owing to his not having dredged in sufficiently deep \vater, but he considers it improbable that any vegetation worthy of notice occurs at that depth, as he did not find any rich vegeta- tion at a greater depth than 10 — 12 fathoms. The reason why Stromfelt found the littoral vegetation on the north coast so poorly developed may be two-fold: it may result The Botany of Iceland. I. \ 2 H. JONSSON from the drift-ice having remained at the coast during the whole summer of the previous year (1882), but it may also be due to the fact that, in this case, Stromfelt went by steamer from port to port, and could scarcely have obtained a thorough knowledge of the coast in as much as the steamer usually stops only a short time at each port. Stromfelt mentions the following algal formations: the Fucacece- formation which is reported from Holmanes and Seley in E. Ice- land; the Laminaria- formation, under which a subvegetation of red alga3 is mentioned. Further, a Monosfroma-vegetation is recorded as occurring near Ekifjor5ur at a depth of two fathoms on a sandy bottom, and a //a/osacczo/7-formation at extreme low-water mark on Holmanes. Stromfelt expresses the opinion, moreover, that a Cora llina - formation formed by Lithothamnion- species must exist, but he does not say anything definite regarding this point, as he received almost all the Lithothamnion- species from the fishermen (70, pp. 10, 11). The description of the vegetation is evidently based on observations made in places where Stromfelt stayed for a longer time, viz. Eskifjor5ur and Rey5arfjor5ur in E. Iceland, and Eyrarbakki in S. Iceland. I have incorporated Strom felt's obser- vations with my own in my description of the vegetation. Stromfelt, on the other hand, treats exhaustively of the dis- tribution of the species along the coasts. Thus he is the first to substantiate the existence of two floral districts in the sea on the coasts of Iceland: a cold-water flora in NE. Iceland and a warm- water flora in SW. Iceland. In a table he gives a summary of the distribution of the species along the coast of Iceland, and states whether they are found in the Norwegian Polar Sea, the North Atlantic and the Greenland Sea. He records 33 species as common to NE. Iceland and SW. Iceland, 33 species as growing in NE. Ice- land and absent from SW. Iceland, and 28 species as growing in the latter district and absent from the former. Thus 66 species in all are recorded from NE. Iceland and 61 species from SW. Iceland. Of the species given by Stromfelt as being found in or absent from NE. Iceland and SW. Iceland respectively, later investigations have proved that far the greater number are common to both places, but then, again, other species have been found which are characteristic of the different districts. My description of the algal vegetation along the coast of Ice- land is based mainly on my own observations, and further on MARINE ALGAL VEGETATION 3 Stromfelt's work, as well as on Ostenf eld's observations. The latter mainly concern the littoral zone, and originate from E. Iceland (Holmanes), SW. Iceland (Reykjavik, NjarSvfk) and S. Iceland (Sta5ur on the south side of Reykjanes). Ostenfeld, moreover, has given information of the sublittoral vegetation of Myrakollur in NW. Iceland. My own observations are drawn from various places encircling the whole of Iceland. In E. Iceland I have especially investigated Berufjor5ur, Rey5arfjor5ur and Sey5isfjor5ur, and everything which is narrated of the algal vegetation from E. Iceland originates from these fjords. In N. Iceland I have examined Eyjafjor5ur fairly ac- curately, from its innermost part to the submarine ridge off Hrfsey, and I have, moreover, in the course of my journey, investigated the head of Hunafloi. When travelling by the mail steamer "Laura" round NW. Iceland I visited all the fjords from Skutulsfjor5ur to Patreksfjordur. I stayed only a short time in each fjord, as I ac- companied the boat from port to port, and was only able to dredge and investigate the littoral zone in the vicinity of the towns. In SW. Ice- land I have been at the south side of Brei5ifjor5ur, and have dredged along the stretch of coast from Rost in Hvammsfjor5ur to Hjallasand- ur, and have also examined the littoral zone over a far larger area, not only round Sna3fellsnes but also in Dalasysla. Round Reykjavik I have dredged and investigated the littoral zone many times. In S. Iceland I have investigated the Vestmannaeyjar most thoroughly and have, in addition, dredged and examined the littoral zone at Eyrarbakki. All remarks concerning the algal vegetation of S. Iceland are based on observations drawn from the western part of the south coast. The eastern part of the south coast from about Stokkseyri eastward is, as far as I know, a sandy coast, a "desert" devoid of algal vege- tation. I have not dredged further east than round the Vestmanna- eyjar, but on my trip through S. Iceland in 1901 I saw very few algaB cast ashore, which may be regarded as a sure sign that a desert lies beyond, because, where algal vegetation exists, it is quite common, with a landward wind, for large, often astonishingly large quantities of alga? to be thrown up on the shore. What might not be found then, on the south coast, where the swell of the Atlantic rolls up onto the flat shore, if any algal vegetation existed further out! Nor can it be expected that anything but a desert exists off this coast, as the bottom consists of sand, and the coast lies exposed to the sea, like the west coast of Jutland. Where, on the other hand, there H. JONSSON: MARINE ALGAL VEGETATION are rocks, there vegetation is sure to occur. At Vik in Myrdal near the southern point of Iceland there was, for instance, a poor vege- tation on the rocks. When one considers how great is the extent of Iceland's coast- line, one cannot expect this to be sufficiently elucidated as re- gards the distribution of marine algae by the few and scattered investigations which have been undertaken. For a long time, then, I nourished the hope of being able to undertake further investiga- tions, and therefore constantly deferred publishing a comprehensive description of the algal vegetation. Now, however, I have decided to delay no longer and hope in the future to be able to make a more extensive contribution in several respects towards the eluci- dation of the algal vegetation. I. LIST OF THE MARINE ALG.E. THE following List of the Marine Algae of Iceland is extracted from my earlier publications (Jonsson, 31, Borgesen and Jonsson, 14) and from the paper by Henning Petersen (57) on the species of Ceramium. It gives only the names of the species with synonyms, their distribution in the different coastal districts and some new habitats. The limitation of species is unchanged except in the case of the genus Ceramium and in Clathromorphum circumscriptum (Stromf.) which is included in Clathromorphum com- pactum (Kjellm.) as proposed by Foslie. One species, Vaucheria sphcerospora Nordst. (Borgesen and Jonsson, 14), is omitted from the list as it can scarcely be called a Marine Alga. Of Ceramium 5 species are added. Thus the number of species is: 76 Rhodophycese 67 Phaeophyceae 51 Chlorophycese 6 Cyanophyceae Total. . . 200 species. The coastal districts are the following (see the map, p. 7):- East Iceland (E. Icel.), from L6nshei5i (Eystra horn) to Langanes. North Iceland (N. Icel.), from Langanes to Hornbjarg (Kap Nord). Northwest Iceland (NW. Icel.), from Hornbjarg to Latrabjarg. Southwest Iceland (SW. Icel.), Brei5ifjor5ur and Faxafloi from Latra- bjarg to Reykjanes. South Iceland (S. Icel.) from Reykjanes to Vestmannaeyjar and east- wards to L6nshei5i (Eystra horn). In "The Marine Algae of Iceland" (Jonsson, 31) the district NW. Icel. is larger; it reaches from the inner end of Hiinafloi to Latra- bjarg instead of as now from Hornbjarg to Latrabjarg. Localities from the part of the coast which stretches from Hiinafloi to Horn- H. JONSSON bjarg are referred to NW. Icel., in the paper mentioned above, but in the present work (cf. Jonsson, 33, p. 11) to North Iceland. These localities are: Hrutafjor5ur, Prestsbakki, Kolbeinsa, Skalholtsvik, KollafjarSarnes, Broddanes and Grimsey in Hunafloi. RHODOPHYCE^. BANGIOIDE^E. Fam. Bangiaceae. Bangia fuscopurpupea (Dillw.) Lyngb., K. Rosenv., 61, p. 831. E. Icel., N. Icel., SW. Icel., S. Icel. Porphyra umbilicalis (L.) J. Ag., K. Rosenv., 61, p. 830; P. laciniata Stromf., 70, p. 34. Common in all parts of the coast of Iceland. Porphyra miniata (Ag.) Ag., K. Rosenv., 61, p. 826; Diploderma m., D. tenuissimum, D. amplissimum Stromf., 70, p. 33. Found in all parts of the coast. Porphyropsis coccinea (J. Ag.) K. Rosenv., 65, p. 69; Porphyra coccinea Jonsson, 31. • SW. Icel.: Reykjavik, S. Icel. Conchocelis rosea Batters. Found in all parts of the coast. FLORIDE.E. Fam. Helminthocladiaceae. Chantransia virgatula (Harv.) Thur., K. Rosenv., 61, p. 824. NW. Icel., SW. Icel. Chantransia secundata (Lyngb.) Thur., K. Rosenv., 61, p. 824. Found in all parts of the coast. Chantransia Alariae H. Jonsson, 31. SW. Icel., S. Icel. : Eyrarbakki. Chantransia microscopica (Naeg.) Fosl. On Cladophora gracilis in the littoral zone. Thallus has long' hairs. Published in Borgesen and Jonsson, 14. N. Icel.: Kolbeinsa. The specimens mentioned under this name belong most probably to another species of Chantransia with a unicellular base. MARINE ALGAL VEGETATION 7 Fam. Gigartinacese. Chondrus crispus (L.) Stackh., Stromf., 70, p. 31. NW. Icel. (cast ashore), SW. Icel., S. Icel. Gigartina mamillosa (Good, et Wood.) J. Ag., Stromf., 70, p. 31. Found in all parts of the coast, common in SW. Icel. and S. Icel. Ahnfeltia plicata (Huds.) Fries, Stromf., 70, p. 31. Cast ashore on N. Icel. and NW. Icel.. common in SW. Icel. and S. Icel. Nor d -I s 1 a, n d S AND S v d - L s 1 Fig. 1. Phyllophora Brodifci (Turn.) J. Ag. *interrupta (Grev.) K. Rosenv., 61, p. 821. E. Icel., NW. Icel. Phyllophora membranifolia (Good, et Wood.) J. Ag.5 Stromf., 70, p. 30. SW. Icel., S. Icel. Actinococcus subcutaneus (Lyngb.) K. Rosenv., 61, p. 822. E. Icel., NW. Icel. Ceratocolax Hartzii K. Rosenv., 62, p. 34. NWT. Icel. Fam. Rhodophyllidaceae. Cystoclonium purpurascens (Huds.) Kiitz., Stromf., 70, p. 30. N. Icel., N\V. Icel. (cast ashore), common in SWT. Icel. and S. Icel. H. JONSSON Turnerella Penny! (Harv.) Schmitz, K. Rosenv., 62, p. 29. E. Icel., N. Icel. Euthora cristata (L.) J. Ag., Stromf., 70, p. 27. Common around the coast of Iceland. Rhodophyllis dichotoma (Lepech.) Gobi, Stromf., 70, p. 26. Common around the coast of Iceland. Fam. Rhodymeniaceae. Rhodymenia palmata (L.) Grev., Kjellman, 36, p. 147; Stromfelt, 70, p. 27 ; R. pertusa Stromf., 70, p. 28. Very common around the coast of Iceland. Lomentaria clavellosa (Turn.) Gaill.; Le Jol., Liste des Algues mar. de Cherb., p. 132, var. sedifolia Ag. S. Icel. Lomentaria rosea (Harv.) Thur., Le Jol., Liste des Alg. mar. de Cherb., p. 131, Fig. Harv. Phyc. Brit., T. 358 and 301. S. Icel. Plocamium coccineum (Huds.) Lyngb., Stromf:, 70, p. 27. S. Icel. Halosaccion ramentaceum (L.) J. Ag., Kjellm., 36, p. 153; Stromf., 70, p. 29; H. scopula Stromf., 70. Common around the coast of Iceland. Fam. Delesseriaceae. Delesseria alata (Huds.) Lam., Stromf., 70, p. 24. SW. Icel., S. Icel. Delesseria Baerii (Post, et Rupr.) J. Ag. ; *corymbosa (J. Ag.) K. Rosenv., 61, p. 806. There is a specimen of this plant in the herbarium of the Botanical Museum in Copenhagen; it is labelled "'Islandia d. Morck." Delesseria sinuosa (Good, et Wood.) Lam., Stromf., 70, p. 24. Common around Iceland. Delesseria sangvinea (L.) Lam. ; Hydrolapathum s. Stromf., 70, p. 26. E. Icel.; rather common in SW. Icel. and S. Icel. Fam. Bonnemaisoniaceae. Bonnemaisonia asparagoides (Wood.) C. Ag. In the herbarium of the Botanical Museum in Copenhagen there are three specimens of this species, said to have been collected in Ice- MARINE ALGAL VEGETATION 9 land. On one of the labels is written "misit Faber." In Flora Danica T. 2579 a specimen of this plant is figured, regarding which Liebmann writes: "ad littora Islandiae pr. Reykjavik legit beatus Faber, cujus spe- cimina mecum communicavit cl. Hofman-Bang.'1 Fam. Rhodomelaceae. Pterosiphonia parasitica (Huds.) Falkenberg, Die Rhodomelaceen des Golfes von Neapel. Polysiphonia p. Kjellman, 36, p. 117; H. Junsson, 31, p. 142. S. Icel. Polysiphonia urceolata (Lightf.) Grev., Stromf., 70, p. 24. Common around Iceland. Polysiphonia fastigiata (Roth) Grev., Stromf, 70, p. 24. NW. Iceland.; common in SW. Icel. and S. Icel. Polysiphonia arctica J. Ag., K. Rosenv., 61, p. 800. E. Icel., N. Icel. and NW. Icel. common ; SW. Icel. Polysiphonia nigrescens (Huds.) Harv., Kjellman, 36, p. 126. N. Icel., SW. Icel. Rhodomela lycopodioides (L.) Ag., Stromf., 70, p. 23. Common around Iceland. Odonthalia dentata (L.) Lyngb., Stromf., 70, p. 23. Common around Iceland. Fam. Ceramiaceae. Callithamnion Arbuscula (Dillw.) Lyngb., Stromf., 70, p. 32. Rather common in SW. Icel. and S. Icel. Callithamnion scopulorum C. Ag., Spec. Alg. (2), p. 176. SW. Icel., S. Icel. Plumaria elegans (Bonnem.) Schmitz, Syst. Uebersicht der bisher bekannten Gattungen der Florideen, Flora oder allgem. bot. Zeit., 1889; Ptilota e. Kjellman, 36, p. 172. SW. Icel.. S. Icel. Ptilota plumosa (L.) Ag., Stromf., 70, p. 32. N. Icel. ; common in NW. Icel., SW. Icel. and S. Icel. Ptilota pectinata (Gunn.) Kjellm., Stromf., 70, p. 32. Common in E. Icel., N. Icel. and NW. Icel ; rather rare in SW. Icel. Antithamnion Plumula (Ellis) Thur. /? boreale Gobi, K. Rosenv., 61, p. 787. A. boreale Stromf., 70, p. 32. E. Icel., N. Icel., NW. Icel., SW. Icel. 10 H. JONSSON Antithamnion floccosum (Mull.) Kleen, Stromf., 70, p. 32. E. Icel., SW. Icel., S. Icel. Ceramium acanthonotum Carm., Kjellm., 36, p. 171. SW. Icel., S. Icel. Ceramium Deslongchampii Chauv., Petersen, 57, p. 108; Ceramium rubrum ex pte. Jonsson, 31. SW. Icel.: Reykjavik (L. Kolderup Rosenvinge, 5/e 1886). Ceramium fruticulosum Kutz., Petersen, 57, p. 108. SW. Icel. : Seltjarnarnes (Helgi Jonsson, 28/8 1907). Ceramium circinnatum Ag., Petersen, 57, p. Ill ; Ceramium rubrum ex pte. Jonsson, 31. SW. Icel.: Stykkisholmur (Helgi Jonsson, 16/e 1897), Skerjafjordur (Helgi Jonsson, 10/7 1905). Ceramium arborescens J. Agardh, Petersen, 57, p. 112; Ceramium rubrum ex pte. Jonsson, 31. N. Icel.: Hrisey (Helgi Jonsson, 2/7 1898); NW. Icel.: Latravik in Adal- vik (C. H. Ostenfeld, 8/7 1896); SW.Icel.: Reykjavik (L. Kolderup Rosen- vinge, 5/6 1886). Ceramium atlanticum Petersen, 57, p. 112; Ceramium rubrum ex pte Jonsson, 31. SW.Icel.: Grotta (Helgi Jonsson, 17/e 1908), Hafnarfjordur (Hjalmar Jensen, 7/5 1890); S. Icel.: Sta6ur (C. H. Ostenfeld, 12/e 1896), Eyrarbakki (Helgi Jonsson, 31/5 1897), Vestmannaeyjar (Helgi Jonsson, 14/5 1897). Ceramium rubrum (Huds.) Agardh, Petersen, 57, p. 113; Jonsson, 31, ex pte. N. Icel., NW. Icel., SW. Icel., S. Icel. Rhodochorton Rothii (Turt.) Naeg., K. Rosenv., 61, p. 791. E. Icel., N. Icel., NW. Icel., SW. Icel. (common), S. Icel. Rhodochorton repens H. Jonsson, 31. S. Icel. Rhodochorton minutum Suhr. Descr. in Reinke's Atlas (59), Fig. Reinke's Atlas T. 40. SW. Icel. Rhodochorton penicilliforme (Kjellm.) K. Rosenv., Les Algues ma- rines du Greenland in Ann. Sc. nat., 7e Ser., XIX. E. Icel., N. Icel., NW. Icel., SW. Icel. MARINE ALGAL VEGETATION7 11 Rhodochorton membranaceum Magnus, K. Rosenv., 61, p. 794; P. Kuckuck, Beitrage zur Kenntniss der Meeresalgen, 1897. E. Icel., N. Icel., NW. Icel., SW. Icel. Fam. Dumontiaceae. Dumontia filiformis (Fl. Dan.) Grev., Stromf., 70, p. 30. E. Icel., SW. Icel. (rather common), S. Icel. Dilsea edulis Stackh., Sarcophyllis edulis Kjellm., 36, p. 152. SW. Icel. Fam. Squamariaceae. Petrocelis Hennedyi (Harv.) Batters, A list of the Marine Algae of Berwick-on-Tweed. N. Icel.: Hraunakrokur (O, Davidsson), NW. Icel, SW. Icel, S. Icel. Cruoria arctica Schmitz, K. Rosenv., 61, p. 784. SW. Icel. Cruoria pellita (Lyngb.) Fries, Kjellm, 36, p. 142. SW. Icel, S. Icel.: Eyrarbakki. Peyssonellia Rosenvingii Schmitz, K. Rosenv., 61, p. 782; Haema- tostagon balanicola Stromf., 70, p. 25? E. Icel, N. Icel, NW. Icel, SW. Icel. Rhododermis parasitica Batters, A list of the Marine Algae of Ber\vick-on-Tweed. NW.Icel, SW. Icel, S. Icel. Fam. Corallinaceae. Lithothamnion glaciale Kjellm., Stromf., 70, p. 18. E. Icel, N. Icel, SW. Icel. Lithothamnion Ungeri Kjellm., 36, p. 91 excl. syn.; L. intermedium Stromf., 70, p. 19. E. Icel, N. Icel, NW. Icel. Lithothamnion tophiforme linger, Foslie, The Norwegian forms of Lithothamnion, 1895, p. 119; L. soriferum Stromf., 70, p. 18. E. Icel, N. Icel, S\V. Icel.: HvalfjorcUir (Horring), S. Icel. Lithothamnion flavescens Kjellm., 36, p. 98. E. Icel. Lithothamnion foecundum Kjellm., 36, p. 99. E. Icel, N. Icel. 12 H. JONSSON Lithothamnion laeve (Stromf.) Foslie, List of species of the Litho- thamnia p. 7; Lithophyllum laeve Stromf., 70, p. 21. E. Icel., N. Icel., NW. Icel., SW. Icel., S. Icel. Lithothamnion Lenormandi (Aresch.) Foslie, The Norwegian forms of Lithothamnion, 1895, p. 150. SW. Icel. Phymatolithon polymorphum (L.) Foslie, List of species of the Lithothamnia p. 8; Lithothamnion polymorphum Stromf., 70, p. 19. S. Icel. Clathromorphum compactum (Kjellm.) Foslie, Lithothamnion com- pactum Kjellm., 36, p. 101 ; Clathromorphum circumscriptum (Stromf.) Fosl., Lithothamnion circumscriptum Stromf., 70, p. 20. In all parts of the coast. Lithophyllum Crouani Fosl., List of species of the Lithothamnia, p. 10. N. Icel., NW. Icel., S. Icel. : Eyrarbakki. Dermatolithon macrocarpum (Ros.) Fosl., Revised systematical survey of the Melobesiese, p. 21; Melobesia macrocarpa Stromf., 70, p. 23. SW. Icel., S. Icel. Corallina officinalis L., Stromf., 70, p. 18. N. Icel., NW. Icel., SW. Icel., S. Icel. Hildenbrandia rosea Kiitz, Stromf., 70, p. 24. Common around the coast of Iceland. Fam. Myrionemaceae. Lithoderma fatiscens Aresch., emend. Kuck., Bemerk. I (47), p. 238. E. Icel., N. Icel., NW. Icel., SW. Icel. Petroderma maculiforme (Wollny) Kuck., Bemerk. II (47), p. 382. N. Icel. Ralfsia ovata K. Rosenv., 61, p. 900; 62, p. 94. N. Icel. : Husavik (Ove Paulsen), Prestsbakki; SW. Icel. Ralfsia clavata (Carm.) Farl., Mar. Alg., p. 88; Reinke (59) Atlas T. 5 and 6, figs. 14 — 20; Stragularia adhserens Stromf., 70, p. 49, T. II, figs. 13—15. E. Icel., N. Icel., NW. Icel., SW. Icel. MARINE ALGAL VEGETATION 13 Ralfsia verrucosa (Aresch.) J. Ag., Reinke (59), Atlas T. 5 and 6, figs. 1—13. E. Icel, N. Icel.. S\Y. Icel. Ralfsia deusta (Ag.) J. Ag., K. Rosenv., 61, p. 898. E. Icel., N. Icel., SW. Icel., S. Icel. Myrionema vulgare Thuret, Sauvageau, 66, p. 185. N. Icel., SW. Icel. Myrionema Corunna3 Sauvag., 66, p. 237. S. Icel. Myrionema globosum (Rke) Fosl., New or critical Norw. Algae, p. 17; Ascocyclus globosus Rke, 58, p. 46; Atlas (59) T. 17; Phycocelis globosus K. Rosenv., 62, p. 86, figs. 19—20. E. Icel., N. Icel., NW. Icel., SW. Icel. Myrionema fasroense Borgs., 13, p. 424. SW. Icel. Myrionema Laminar!® (K. Rosenv.), Dermatocelis LaminariaB K. Rosenv., 62, p. 89, fig. 21. SW Icel. Ascocyclus islandicus H. Jonsson, 31, p. 149. N. Icel. Probably this species will prove to be identical with A. sphcerop horns Sauv., cf. Jonsson, 31, p. 151 and Kylin.1 The last-named author writes that A. islandicus without doubt is identical with A. sphcerop ho rus, but he gives no particulars as to the chromatophores of the last-named species, I therefore must still regard the Icelandic plant as a distinct species. Fam. Ectocarpaceae. Microsyphar Polysiphonia3 Kuck., Beitrage (48) p. 29. NW. Icel., SW. Icel. Streblonema acidioides K. Rosenv., 61, p. 894; 62, p. 80; Phyco- celis secidioides Kuck., Bemerk. I (57), p. 234. E. Icel., N. Icel., SW. Icel. Streblonema StilophoraB Cr. var. ccespitosa K. Rosenv., 61, p. 892. Found in all parts of the coast. Pylaiella littoralis (L.) Kjellm.; Ectocarpus littoralis, Kuck., 48, p. 7; Rosenv., 61, p. 881 ; Pylaiella littoralis, Pylaiella varia Kjellm., 35, p. 83. Common around the coast of Iceland. 1 Harald Kylin, Zur Kenntnis der Algenflora der Norwegischen Westkiiste, Arkiv for Botanik. Bd. 10, No. 1, 1910. 14 H. JONSSON Ectocarpus tomentosoides Farl., New or imperfectly known Algae of U. S., reprint from Bull. Torr. Bot. Club, Vol. XVI, 1889, p. 11, T. 87, fig. 4; K. Rosenv., 61, p. 180; Gran, En norsk form af Ectoc. tomentosoides Farl., Christiania Vidensk. Selsk. Forhandl. for 1883, No. 17; Kuckuck, Ueber Polymorphic bei einigen Phseosporeen in Festschrift fur Schwendener, p. 370, figs. 5 — 7. E. Icel., N. Icel., NW. Icel.; common in SW Icel. and S. Icel. Ectocarpus tomentosus (Huds.) Lyngb., Hydr. Dan. (51) p. 132; Kjellman, 35, p. 73. SW. Icel., S. Icel. Ectocarpus confervoides (Roth) Le JoL, Kuck., 48, p. 19; Kjell- man, 35, p. 77, ex pte. ; K. Rosenv., 61, p. 883, ex pte. Found in all parts of the coast. Ectocarpus siliculosus (Huds.) Lyngb., Hydr. Dan. (51) p. 131; Kjellman, 35, p. 78; Kuck. 48, p. 15. N. Icel., SW. Icel. Ectocarpus penicillatus (Ag.) Kjellm., 35, p. 76; E. confervoides f. penicillata Kjellman, 39, p. 79. E. Icel., N. Icel., SW. Icel. Ectocarpus fasciculatus (Griff.) Harv., Kjellm. 35, p. 76. SW. Icel., S. Icel. Ectocarpus Hinksiae Harv., Manual, p. 59; Phyc. Brit., T. 22; Sauvageau, Observations relatives a la sexualite des Pheosporees (Journal de Botanique, 1896). S. Icel. Fain. Elachistaceae. Leptonema fasciculatum Rke, 58, p. 50; var. subcylindrica K. Rosenv., 61, p. 879. N. Icel., NW. Icel., SW. Icel. Elachista fucicola (Veil.) Aresch., emend. K. Rosenv., 61, p. 878; E. fucicola Stromf., 70, p. 49. a typica is the most common, /? lu- brica (Rupr.) K. Rosenv. is rather common. In all parts of the coast. Fam. Sphacelariaceae. Sphacelaria britannica Sauvag., 67, p. 50. N. Icel., SW. Icel., S. Icel. MARINE ALGAL VEGETATION 15 Sphacelaria radicans Harv., Sauvag., 67, p. 27, fig. 14; Reinke, 60, T. Ill, fig. 1; Kuck., Bemerk. I (47), p. 229, fig. 4. E. Icel.. N. Icel,, SW. Icel., S. Icel. Sphacelaria olivacea Pringsh., emend. Sauvag., 67, p. 54. NW. Icel., SW. Icel., S. Icel. Chaetopteris plumosa (Lyngb.) Kiitz., Sauvag., 67, p. 44; Stromf., 70, p. 52; K. Rosenv., 61, p. 903; Reinke, 59, Atlas T. 49— 50. E. Icel., N. Icel., NW. Icel., SW. Icel. Fam. Punctariaceae. Omphalophyllum ulvaceum K. Rosenv., 61, p. 872, fig. 19. E. Icel. Punctaria plantaginea (Roth) Grev., K. Rosenv., 61, p. 871; 62, p. 71; Stromf., 70, p. 50. E. Icel., N. Icel., NW. Icel. Litosiphon filiformis (Rke), Pogotrichum filiforme Rke (59), Atlas, p. 62, T. 41, figs. 13— 25; K. Rosenv., 61, p. 869; Kuck., Ueber Poly- morphie bei einigen Phaeosporeen, Festschrift fur Schwendener, p. 360. E. Icel., N. Icel., SW. Icel., S. Icel. Isthmoplea sphaerophora (Harv.) Kjellm., 36, p. 276; Reinke (59), Atlas T. 30; Pylaiella curta Foslie, Nye havsalger, in Tromso Mu- seums Aarshefter, X, 1887, p. 181; Kjellman, 35, p. 85; Fosliea curta Rke, Atlas, p. 45. E. Icel., NW. Icel., SW. Icel., S. Icel. Stictyosiphon tortilis (Rupr.) Rke, Atlas, T. 31—32; K. Rosenv., 61, p. 868; Phloeospora tortilis Stromf., 70, p. 51; Phloeospora sub- articulata Kjellman, 39, p. 78. E. Icel, N. Icel., NW. Icel, SW. Icel. Phaeostroma pustulosum Kuckuck, Ueber einige neue Phseosporeen d. westl. Ostsee, Bot. Zeit. 1895, p. 182, T. VII; K. Rosenv., 62, p. 68, fig. 15. E. Icel, NW. Icel, SW. Icel. Scytosiphon Lomentaria (Lyngb.) J. Ag., K. Rosenv., 61, p. 863; 62, p. 62; Stromf., 70, p. 50. In all parts of the coast. Phyllitis zosterifolia Rke, 58, p. 61; K. Rosenv., 61, p. 862. E. Icel, NW. Icel, SW. Icel, S. Icel. 16 H. JONSSON Phyllitis fascia (O. F. Mull.) Kiitz., K. Rosenv., 61, p. 862. Gathered in all parts of the coast. Fam. Dictyosiphonaceae. Coilodesme bulligera Stromf., 70, p. 48, T. II, figs. 9— 12; K. Ro- senv., 61, p. 862; 62, p. 61, fig. 13. E. Icel., NW. Icel, SW. Icel. Dictyosiphon Ekmani Aresch., Obs. phyc. 3 (7), p. 33. SW. Icel. Dictyosiphon Mesogloia Aresch., Obs. phyc. 3 (7); Reinke, 58, p. 64. N. Icel. Dictyosiphon Chordaria Aresch., Obs. phyc. 3 (7); Reinke, 58, p. 63; K. Rosenv., 61, p. 861; Coilonema Chordaria Stromf., 70, p. 51. E. Icel., SW. Icel. Dictyosiphon corymbosus Kjellm., 36, p. 267; Stromfelt, 70, p. 51. N. Icel. Dictyosiphon hippuroides (Lyngb.) Kiitz.; Kjellm., 36, p. 268; Strom- felt, 70, p. 51. N. Icel., NW. Icel., SW. Icel., S. Icel. Dictyosiphon foeniculaceus (Huds.) Grev., Kjellman, 36, p. 269; K. Rosenv., 61, p. 859; Stromf., 70, p. 52. E. Icel., N. Icel., NW. Icel., SW. Icel. Fam. Desmarestiaceae. Desmarestia viridis (Mull.) Lam., K. Rosenv., 61, p. 859; Dichloria viridis Stromf., 70, p. 51. Common around the coast of Iceland. Desmarestia aculeata (L.) Lam., Stromf., 70, p. 51; K. Rosenv., 61, p. 857. Common everywhere along the coast. Desmarestia ligulata (Lightf.) Lam. S. Icel. : Vestmannaeyjar (Ove Paulsen). Fam. Chordariacese. Castagnea virescens (Carm.) Thur., K. Rosenv., 62, p. 58; Eudesme virescens Stromf., 70, p. 47. E. Icel., N. Icel., SW. Icel. MARINE ALGAL VEGETATION 17 Leathesia difformis (L.) Aresch., Kjellm., 36, p. 252. N. Icel., SW. Icel. Chordaria flagelliformis (Mull.) Ag., Stromf., 70, p. 47; K. Rosenv., 61, p. 854. Common everywhere along the coast. Fam. Chordacese. Chorda tomentosa Lyngb., Hydrophytologia Danica, p. 74; K. Rosenv., 61, p. 854. E. Icel., N. Icel., SW. Icel. Chorda Filum (L.) Stackh., K. Rosenv., 61, p. 853; Stromf., 70, p. 47. E. Icel., N. Icel., NW. Icel., SW. Icel. Fam. Laminariaceae. Saccorrhiza dermatodea (De la Pyl.) J. Ag. ; K. Rosenv., 61, p. 850; Phyllaria lorea Stromf., 70, p. 42. E. Icel., N. Icel., NW. Icel., SW. Icel. Laminaria saccharina (L.) Lam., Kjellman, 36, p. 229; 35, p. 24; Stromf., 70, p. 42. f. typica; f. linearis J. Ag., Kjellman 36, p. 229; 35, p. 25; Stromf., 70, p. 42; Borgesen, 13, p. 451, fig. 85; f. latifolia Kjellm., 35, p. 26 ; Laminaria saccharina f. latis- sima Kjellm., 36, p. 230; Stromf., 70, p. 43? This species is common everywhere along the coast, especially the principal form ; f. linearis is rarer and f. latifolia is only met with in E. Icel. and NW. Icel. where it occurs gregariously. Laminaria fasroensis Borges., 13, p. 454. E. Icel., N. Icel. Laminaria nigripes J. Ag., emend. K. Rosenv., 61, p. 842. ft atrofulva (J. Ag.) K. Rosenv. (1. c.); Laminaria discolor, La- minaria nigripes f. oblonga Stromf., 70, pp. 43 — 44. E. Icel. Laminaria digitata (L.) Lam., Kjellman, 36, p. 240; 35, p. 22; Stromf., 70, p. 45. f. genuina Le Jol. 49; Kjellman, 35, p. 23; f. stenophylla Harv. Phyc. Brit., T. 338; Laminaria stenophylla Stromf., 70, p. 45; J. Ag. De Lam., p. 18; Kjellm., 35, p. 24; f. cucullata Le Jol., 49. F. genuina is common everywhere; f. stenophylla: E. Icel., SW. Icel., S. Icel.; f. cucullata: E. Icel., NW. Icel. The Botany of Iceland. I. 2 18 H. JONSSON Laminar i a hyperborea (Gunn.) Foslie, 20, p. 42; Stromf., 70, p. 44; Laminaria Cloustoni Le JoL, 49, p. 577; fig., Fosl., 20, T.I. E. Icel., N. Icel., NW. Icel.; common in SW. Icel. and S. Icel. Alaria Pylaii (Bory) J. Ag., emend. K. Rosenv., 61, p. 838; Alaria Pylaii and Alaria membranifolia Stromf., 70, p. 39. f. typica K. Rosenv.; f. membranacea (J. Ag.) K. Rosenv. Common everywhere along the coast. Alaria esculenta (L.) Grev., Kjellm., 36, p. 212; 35, p. 19; Alaria esculenta, Alaria linearis and Alaria flagellaris Stromf., 70, pp. 38— 41 ; A. flagellaris K. Rosenv., 62, p. 49. f. aastralis Kjellm.; f. fasciculata Stromf.; f. pinnata (Gunn.) Kjellm. This species is exceedingly common everywhere along the coast. Fam. Fucaceae. Fucus spiralis L., Kjellm., 36, p. 202; Stromf., 70, p. 35; Borge- sen, 13, p. 472; Fucus Areschougii Kjellm., 35, p. 11. f. typica; f. borealis Kjellm. E. Icel., N. Icel., SW. Icel., S. Icel. Fucus inflatus L., M. Vahl, Fl. Danica (30), T. 1127; Foslie, Krit. Fortegnelse, Tromso Mus. Aarshefter, IX, p. 109; Kjellm., 35, p. 11; K. Rosenv., 61, p. 834; Borgesen, 13, p. 465 ; Fucus evanescens Stromf., 70, p. 35; F. edentatus, F. furcatus and F. evanescens J. Ag., 3, p. 40; F. furcatus Kleen, 43, p. 29; F. edentatus De la Pyl., 15, p. 84. f. typica. F. furcatus Kleen ex pte. ; F. evanescens auct. ex pte. ; F. edentatus De la Pyl.; fig. Flora Danica (30) T. 1127; Borgesen, 13, figs. 90 and 91. f. evanescens (C. Ag.) F. evanescens C. Ag., Sp. p. 92 et auct. partim. f. linearis (Huds.) K. Rosenv., 61, p. 834; F. linearis Hudson Flora anglica London 176?, Oeder Flora Danica (30) T. 351. f. exposita. F. distichus Lyngb. Hydr. Dan. (51) p. 6, exclus. syn.; F. distichus a, robustior J. Ag. 3, p. 37, Kjellman 36, p. 210; F. inflatus f. disticha Borgesen, 13, p. 465. This species is common everywhere along the coast. MARINE ALGAL VEGETATION 19 Fucus serratus L., Kjellm., 36, p. 196. SW. Icel., S. Icel. Fucus vesiculosus L., Kjellm., 36, p. 198; Stromf., 70, p. 34. f. typica, fig. Harv. Phyc. Brit. T. 204. f. turgida Kjellm. f. sphcerocarpa J. Ag. This species is common everywhere. Pelvetia canaliculata (L.) Dec. et Thur., Stromf., 70, p. 38. SW. Icel., S. Icel. Ascophyllum nodosum (L.) Le Jol., K. Rosenv. , 61, p. 832; Ozo- thallia nodosa Stromf., 70, p. 34. Common along the coast. CHLOROPHYCE.E. Fam. Protococcaceae. Chlorochytrium Cohnii Wright, K. Rosenv., 61, p. 963. SW. Icel. Chlorochytrium inclusum Kjellm., 36, p. 320, T. 31, figs. 8—17; K. Rosenv., 61, p. 963; 62, p. 119. E. Icel., N. Icel., NW. Icel., SW. Icel. Chlorochytrium dermatocolax Rke, 58, p. 88; K. Rosenv., 61, p. 964; Svedelius, 71, p. 72. N. Icel., SW. Icel. Chlorochytrium Schmitzii K. Rosenv., 61, p. 964; 62, p. 119. SW. Icel. Codiolum Petrocelidis Kuck., Bemerk. (47), p. 259, fig. 27. SW. Icel. Codiolum gregarium Al. Braun, Algarum unicellularum genera nova et minus cognita, Lipsiae, 1855, p. 19; Borgesen, 13, p. 517. E. Icel. Codiolum pusillum (Lyngb.) Kjellm., Borgesen, 13, p. 518; Vau- cheria pusilla Lyngb. Hydr. Dan. 51, p. 72, T. 22. N. Icel. Fam. Ulvaceae. Percursaria percursa (Ag.) K. Rosenv., 61, p. 963. SW. Icel. 2* 20 H. JONSSON Enteromorpha aureola (Ag.) Kiitz., Tab. phyc., Vol. VI, T. 40, III; Ulva aureola Ag. Ic. alg. europ. (1), T. 29; Capsosiphon aureolus Gobi; Ilea fulvescens J. Ag., Ulvacese p. 114; ? Solenia fulvescens Ag., 2, p. 420; Enteromorpha quaternaria Ahlner in Wittr. et Nordstedt Alg. exsicc., Nos. 138 and 139. N. Icel. Enteromorpha Linza (L.) J. Ag., Ulva enteromorpha a, lanceolata Le Jol., 50, p. 42. SW. Icel., S. Icel. Enteromorpha intestinalis (L.) Link., emend. K. Rosenv., 61, p. 957; Borgesen, 13, p. 487. f. genuina K. Rosenv. 1. c. p. 957 ; Enteromorpha intestinalis Stromf., 70, p. 52. f. micrococca (Kiitz.) K. Rosenv. 1. c. p. 957. f. compressa (L.) K. Rosenv. 1. c. p. 958; Enteromorpha compressa f. typica and E. complanata f. subsimplex Stromf., 70, p. 53. f. minima (Naeg.) K. Rosenv., I.e. p. 959; Enteromorpha minima Stromf., 70, p. 53. f. prolifera (O. F. M Ciller) Borgesen, 13, p. 490. Enteromorpha prolifera K. Rosenv., 61, p. 960. This species is common everywhere along the coast. Enteromorpha clathrata (Roth) Grev., Kjellm., 36, p. 287; Ulva clathrata Le Jol., 50, p. 48 (partim); Enteromorpha compressa f. race- mosa Stromf., 70, p. 53. E. Icel., N. Icel., SW. Icel., S. Icel. Monostroma groenlandicum J. Ag., K. Rosenv., 61, p. 954, fig. 53. E. Icel., N. Icel., NW. Icel. Monostroma Grevillei (Thur.) Wittr., emend. K. Rosenv., 61, p. 946. var. typica K. Rosenv. 1. c. Monostroma Grevillei Wittr., 76, p. 57; Stromf. 70, p. 54 partim (e specim.). var. arctica (Wittr.) K. Rosenv. 1. c. Monostroma arcticum Wittr., 76, p. 44; Monostroma latis- simum Stromf., 70, p. 54. var. intestiniformis K. Rosenv. 1. c. Enteromorpha intestinalis Stromf., 70, p. 58 partim (e spec.). Var. typica and var. arctica are common along the coast; var. inte- stiniformis: E. Icel., SW. Icel. MARINE ALGAL VEGETATION 21 Monostroma undulatum Wittr. 76, p. 46,. T. Ill, fig. 9; K. Rosenv., 61, p. 945; Monostroma Grevillei Strom f., 70, p. 54 partim (e specim.). In all parts of the coast. Monostroma fuscum (Post, et Rupr.) Wittr., emend. K. Rosenv., 61, p. 940; M. Blyttii, Stromf. 70, p. 54. f. typica is common along the coast; f. grundis.: E. Icel., N. Icel. Ulva Lactuca L., K. Rosenv. 61, p. 839; Stromf., 70, p. 54. N. Icel., NW. Icel., SW. Icel., S. Icel. Fam. Prasiolaceae. Prasiola polyrrhiza (K. Rosenv.). Gayella polyrhiza K. Rosenv., 61, p. 936; Prasiola crispa subsp. marina Borgesen, 13, p. 482; Prasiola crispa f. submarina Wille, 73, p. 13. SW. Icel., S. Icel. Prasioia furfuracea (Mert.) Menegh. Imhauser, 29, p. 266; Foslie Contrib., I, p. 127; Borgesen, 13, p. 486. E. Icel., N. Icel., SW. Icel. Prasiola stipitata Suhr; Imhauser, 29, p. 272 ; Kjellman, 36, p. 303. E. Icel., N. Icel., SW. Icel., S. Icel. Fam. Ulothricaceae. Ulothrix consociata Wille, 73, p. 25. var. islandica H. Jonss. N. Icel. Ulothrix subflaccida Wille, 73, p. 29. E. Icel., N. Icel. Ulothrix pseudoflacca Wille, 73, p. 22, T. II, figs. 64— 81. E. Icel., SW. Icel., S. Icel. Ulothrix flacca (Dillw.) Thur., K. Rosenv., 61, p. 935, fig. 44; Wille, 73, p. 18, T. I— II, figs. 54—63. Common around the coast of Iceland. Fam. Chaetophoracese. Pseudendoclonium submarinum Wille, 73, p. 29, T. Ill, figs. 101- -134. E. Icel. Entoderma Wittrockii (Wille) Lagerh., K. Rosenv., 61, p. 934. N. Icel., SW. Icel, S. Icel. 22 H. JONSSON Acrochaete papasitica Oltm. Bot. Zeit. 1894, p. 208; K. Rosenv., 62, p. 114. SW. Icel. Acrochaete repens Pringsh., Beitrage p. 2, T. II; Huber, 28, p. 306. NW. Icel. Bolbocoleon piliferum Pringsh., Beitrage p. 2, T. II; Huber, 28, p. 308, pi. 13, figs. 8— 12. E. Icel., N. Icel., NW. Icel. Fam. Mycoideaceae. Ulvella fucicola K. Rosenv., 61, p. 926, fig. 40. Pseudopringsheimia fucicola (Rosenv.) Wille in Engler u. Prantl : Die naturlichen Pflanzen- familien, Nachtrage zu I. Theil, Abtheil 2, p. 89. E. Icel., N. Icel., SW. Icel., S. Icel. Pringsheimia scutata Rke, 58, p. 81, Atlas T. 25. NW. Icel., SW. Icel. Ochlochaete ferox Huber, 28, p. 291, T. X; K. Rosenv. 61, p. 931, fig. 41. N. Icel. Fam. Cladophoraceae. Urospora mirabilis Aresch., K. Rosenv., 61, p. 918, fig. 35; 62, p. 106. Common along the coast. Urospora Hartzii K. Rosenv., 61, p. 922, fig. 38. E. Icel., SW. Icel., S. Icel. Urospora Wormskioldii (Mert.) K. Rosenv., 61, p. 920, fig. 36. In all parts of the coast. Chaetomorpha tortuosa (Dillw.) Kleen, K. Rosenv., 61, p. 917. E. Icel., N. Icel., SW. Icel. Chastomorpha Melagonium (Web. et Mohr) Kutz., K. Rosenv., 61, p. 917; Stromf., 70, p. 55. Probably common along the coast of Iceland. Rhizoclonium riparium (Roth) Harv. , K. Rosenv., 61, p. 913; 62, p. 103. f. polyrhiza K. Rosenv., 1. c. p. 913. f. valida Fosl., K. Rosenv., 1. c. p. 915. f. implexa (Dillw.) K. Rosenv., 1. c. p. 915. E. Icel., N. Icel., SW. Icel. MARINE ALGAL VEGETATION 23 Spongomorpha vernalis (Kjellm.) Wille, Acrosiphonia vernalis Kjellm., 41, p. 82. SW. Icel. Acrosiphonia albescens Kjellm., 41, p. 55, T. IV, fig. 21; Borgesen, 13, p. 507, fig. 103; Spongomorpha arcta Stromf., 70, p. 54, ex pte. Common along the coast of Iceland. Acrosiphonia incurva Kjellm. 41, p. 61. Common along the coast. Acrosiphonia hystrix (Stromf.) H. Jonss., 31. f. typica H. Jonss., Spongomorpha hystrix Stromf., 70, p. 54, Cladophora diffusa Stromf., 70, p. 55 ex pte. (e specim.); Cladophora (Spongomorpha) arcta y hystrix K. Rosenv., 61, p. 907. f. littoralis H. Jonss. E. Icel., N. Icel.. NW. Icel., SW. Icel. Acrosiphonia flabelliformis H. Jonss., 31. S. Icel. Acrosiphonia penicilliformis (Fosl.) Kjellm., 41, p. 80 forma. E. Icel. Cladophora rupestris (L.) Kiitz., K. Rosenv., 61, p. 909; Stromf., 70, p. 55. N. Icel., NW. Icel., SW. Icel., S. Icel. Cladophora hirta Kiitz., Kjellm., in Wittr. et Nordstedt Exsicc., No. 1041. SW. Icel, S. Icel. Cladophora sericea (Huds.) Aresch., 8, p. 194, forma. N. Icel., SW. Icel., S. Icel. Cladophora glaucescens (Griff.) Harv., Phyc. Brit. T. 196; Le Jol. Alg. mar. d. Cherb. Exsicc., 66. SW. Icel. Cladophora gracilis Kiitz., Kjellm. in Wittr. et Nordstedt Exsicc., No. 1040. E. Icel., N. Icel., SW. Icel. Earn. Gomontiaceae. Gomontia polyrrhiza (Lagerh.) Born, et Flah. sur deux nouv. gen. d'Algues perfor. Journ. de Bot. Tom. II, 188, p. 163. E. Icel, N. Icel, NW. Icel, SW. Icel. 24 H. JONSSON : MARINE ALGAL VEGETATION Fam. Phyllosiphonaceae. Ostreobium Queketti Born, et Flab., Sur quelques plantes vivant dans le test calcaire des mollusques, p. 15, pi. IX, figs. 5 — 8. E. Icel., N. Icel., NW. Icel., SW. Icel. CYANOPHYCE.E. Fam. Chamaesiphonacese. Pleurocapsa amethystea K. Rosenv., 61, p. 967, var. E. Icel., N. Icel.; common in NW. Icel., SW. Icel. and S. Icel. Fam. Oscillator iaceae. Plectonema norvegicum Gomont, Bull, de la Soc. hot. de France, tome XLV1, 1899. N. Icel. Phormidium autumnale (Ag.) Gomont, emend. Jobs. Schmidt, 68, pp. 348 and 410. E. Icel. Spirulina subsalsa 0rsted, Beretning om en Excursion til Trin- delen, Kroyers Tidskrift 3. Bd., pp. 566, 1842. N. Icel., S. Icel. Fam. Rivulariaceae. Calothrix scopulorum (Web. et Mohr) Ag., emend. Jobs. Schmidt, 68, pp. 390 and 414. E. Icel., N. Icel. Rivularia atra Roth, Gatalecta botanica, III, p. 340, 1806. SW. Icel. II. LIFE-CONDITIONS OF THE MARINE ALGAL VEGETATION. 1. THE NATURE OF THE COAST. THE coast of Iceland consists partly of rock and partly of sand. The rocky coasts are rich in algal vegetation, while the sandy coast is most frequently a "desert.'' Here and there vegetation may be found, however, on the sandy coast, where this is not exposed to violent movement during any length of time. The vegetation then consists of short-lived species. The Rocky Coast. This abounds in indentations of various size: inlets, fjords and bays. The size of the fjords varies greatly; for example, the largest, Faxafloi, is ten geographical miles long and twelve geographical miles broad, and Breidifjordur is eighteen geographical miles long and ten geographical miles broad. The smaller fjords, on the other hand, are short and narrow indenta- tions. Thus, owring to the indentations on the coast, the exposure is apt to vary greatly. The extreme points and the outer portions of the fjords have, as a rule, an exposed position, while in the interior of the fjord the water is generally calm. The rocks on the coast consist of basalt; in some places, how- ever, tuff-coasts exist, and especially on the south coast. The fjord- coasts of Iceland, which comprise South-west, North-west, North and East Iceland, are mainly composed of basalt. The basalt varies considerably but, as far as I have seen, it has no significance as regards the distribution of the species, and no difference is seen, for example, in the vegetation on the dolerite and the ordinary basalt coasts. What is of prime importance to the vegetation is not the rock itself but the nature of its surface. The surface is, as a rule, very uneven, being eroded by water, weather and wind, and furrowed by numerous fissures. Its nature is, therefore, such that the algae can easily attach themselves to it. 26 H. JONSSON The rocky coast is, as a rule, of solid rock, consisting of preglacial basaltic lava. Postglacial basaltic lava is found, nevertheless, in some places, as for example on Snasfellsnes. In many places, large stretches of the coast are covered with debris (Ur5) from the mountains. Where the debris or the new lava predominates the surface is ge- nerally very uneven, and one then finds distinct elevations with large and small depressions interposed; such a coast is usually covered with an abundant and multifarious algal vegetation, if the conditions are in other respects favourable to the existence of algae. There is no range of skerries (Skaergaard) as there is, for example, on the coast of Norway. Yet a number of islands and rocks occur in the fjords, especially in Brei5ifj6r5ur. In this fjord are found indications of a range of skerries running parallel with the coast and along a considerable stretch of it, and marking the outward limit of the Zosfera-vegetation. The Sandy Coast. Almost the entire coast of S.Iceland is sandy shore or gravel shore. As a rule, such bottoms afford a mobile substratum because each wave which breaks on the beach shifts the particles backwards and forwards. A sandy coast is also met with, here and there, in other parts of the country, but is then found, as a rule, alternating with rocky parts; thus, the sandy or gravelly shore is often predominant at the head of small indentations which at the sides are bounded by projecting masses of rocks. Clayey Shore is also found fairly frequently in the interior of the fjords. The rocky coast is, as a rule, abundantly overgrown, and this is frequently the case also with the sublittoral gravel-bottom, while the sand and gravel bottoms laid bare periodically by the shifting tide are not, as a rule, overgrown, and, in any case, only with short- lived species. On clayey and muddy bottoms, on the other hand, alga3 are seldom or never found, while Zostera often covers such a bottom and forms submarine "green meadows." Despite the small udesert" areas, one may say that the coasts are covered with a zone of continuous algal vegetation - if we ex- clude the eastern portion of S. Iceland. This algal zone varies greatly in width, accommodating itself to the precipitousness of the coast. In a bay as shallow as Faxafloi the algal vegetation has a great extension seawards, while it is far more limited, for example, on the steeply descending submarine declivities in the fjords of the east coast. MARINE ALGAL VEGETATION 27 2. THE OCEAN. As regards the ocean, the chief points are its movements, tem- perature and salinity. A. The Movements of the Ocean. These are tides, waves and currents. All these movements of the ocean are of very great importance to the life of the algae. a. Tides. By the alternate rise and fall of the tide a part of the shore is laid bare, and the vegetation growing there must then be capable of maintaining life in the air for a longer or shorter period. Those plants which grow highest up in the zone thus left dry, are exposed during the greater part of the period between the one flood-tide and the next, or for about 10 — 11 hours in every 12. The plants occurring lowest down in the zone, on the other hand, are not exposed during spring-tide for more than one hour in every 12, and they are submerged the whole time during neap-tide. The upper limit of the algal vegetation is, moreover, dependent on how high the tide rises, i. e. the height of the flood-tide. The Height of the Flood-tide. The following data regarding the height of the flood-tide are taken from "Den islandske Lods" (1903) and from the alterations and additions to it which have been published. The height of the flood-tide is greatest in SW. Iceland and least in E. Iceland. The height of the flood-tide at spring-tide is recorded as being about 14 feet from Reykjavik (SW. Iceland), 10 — 11 feet from NW. Iceland, 5 — 5Ys feet from the north coast and 5 feet from E. Iceland. In many places there is a great difference as regards the height of the flood-tide during the spring and neap tides. To illustrate this more fully I give the following figures from some localities on the different parts of the coast: Spring-tide Neap-tide South Iceland, Vestmannaeyjar 8 — 10 feet 4 feet Eyrarbakki 10 6 South-west Iceland, Reykjavik cir. 14 4 Stykkisholmur 12 6 North-west Iceland, Dyrafjordur 11 5—6 North Iceland, Akureyri 5*/2 - lx/2 - East Iceland, Djiipivogur 7l/z - 2l/2 - From Elli5aey near Stykkisholmur in SW. Iceland the height of the flood-tide during spring-tide is recorded as being 14 feet and 28 H. JONSSON during neap-tide 7 feet, and it is also recorded from the latter place that the low-water at neap-tide lies about 3 feet above that at spring-tide. When extraordinary conditions prevail the height of the flood- tide may be still greater; thus, 18 — 19 feet has been recorded from Reykjavik and 18 feet from Stykkisholmur. From this it is seen that the height of the flood-tide varies greatly, which must necessarily affect the algal vegetation in several respects, especially as concerns the upper limit of its growth. Thus there is a great difference in the highest flood-mark (at spring-tide) and in the lowest flood-mark (at neap-tide). The tidal wave or the tide, moreover, shows irregularities, because neither the highest flood-mark (spring) nor the lowest flood-mark (neap-tide) is constant. The limit of the flood-tide fluctuates perpetually between a high- mark and a low-mark. The high-mark or the highest limit of the flood-tide is seen distinctly on rocky coasts from the action of the sea on the rock. On flat coasts, the high-mark can be distinguished by detached algae and various other bodies which accompany the tidal wave and remain at the highest level reached by the water. This high-mark lies considerably higher than the upper limit of the algal vegetation. The low-mark is not as easy to distinguish as the high-mark, but it will almost coincide with the upper limit of the Pelvetia- Fucus spiralis association. Above this the Bangia association (Ulo- thrix, Bangia, and others) is found, the extreme limit of which will almost coincide with an average water-level which, however, does not lie midway between low-mark and high-mark, but rather nearer the former. The upper limit of the algal vegetation thus lies some- what above the limit of the flood-tide at neap-tide. The same rule holds good, of course, on a very exposed coast. That the place is exposed means that there is a heavy swell, which causes the sea to rise higher up on the coasts, both at neap-tide and at spring-tide. By the upper limit of the marine algal vegetation is meant that boundary line above which marine algae do not occur in the form of associations; on the other hand, no account is taken of single individuals or groups of individuals being found higher up, in crevices or pools, as such an occurrence must be considered accidental, because they are carried up to this height with the high water or by far-reaching breakers. They do live, of course, but do not thrive, and have evidently gone beyond their real area of distribution. In MARINE ALGAL VEGETATION 29 this respect, however, those species should be excepted which can grow both in salt and in fresh water, as for example, Enteromorpha intestinalis (the principal form) and others. Naturally, such species cannot be taken into account when defining the upper limit of growth of marine algal vegetation. If we compare the upper limit of the marine algal vegetation in Iceland with the same limit in Greenland it appears that they agree almost completely, as Rosen vinge (63, p. 89) sets the limit in Greenland almost at flood-mark at neap-tide. On the other hand there seems to be an incongruity with the Faeroes, as Borgesen (11 and 12) sets the limit far above highest flood-mark in exposed places and almost at uppermost flood-mark in sheltered places. If we institute comparisons with more distant coasts, for instance with the west coast of Sweden, the algal vegetation of which has lately been described by Kylin (43), we find that, as regards the upper limit, the case is the same as in Iceland, that is, the upper limit is coincident with an average water-level which lies higher in exposed places than in sheltered ones. In Iceland, indeed, on a very exposed coast, marine algae can be found rather high up, and if the sea is smooth and calm they may appear to be rather far away from the water; but on returning to the same place when the sea is in motion we see that it washes over them, and \ve no longer think it strange that they grow in so high a position. While the marine algal vegetation, as mentioned before, only extends upwards to an average water-level, it frequently happens on flat coasts that the land-vegetation is sub- merged at spring-tide. This occurs both in the interior of the fjords and on the lower islands, and may generally be distinguished by the appearance and the components of the vegetation ; marine algae, however, do not occur among such vegetation. Some phanerogams, e. g. Atriplex, Mertensia, Cakile, etc., also grow below the upper limit of the littoral zone. According to Ro- sen vinge it happens in Greenland also that the land- vegetation is submerged at spring-tide. b. The Swell. Here, those inequalities in the surface of the ocean which in every -day language are called waves, and that volume of water which, with a sea-wind is forced in towards the land, are treated collectively. The volume of \vater which is driven towards the coast by a sea- wind raises the water-level. While the tidal wave rises, the sea moved by the wind works together with 30 H. JONSSON it, but when the water falls it counteracts the tidal wave; this circum- stance is of great importance on exposed coasts where the wind blows frequently, as it shortens the period of desiccation. These move- ments in the sea are naturally somewhat irregular, but the irregu- larities are quite equalized in the long run, and therefore the effect of these movements may very well be regarded as constant. The high-mark which the water leaves on the coasts is due to the tide and to these movements jointly. A frequent sea-wind has a favourable effect on the vegetation in the zone laid bare by the tide, as mentioned above, but where a frequent land-wind is blowing the effect is the reverse, as this counteracts the rise of the tidal wave and accelerates its fall, whereby the period of desiccation is prolonged. The effect of the waves beating on the algae is great. In ex- posed places, that is in places where the swell is heavy, the plants must be able to withstand the drag of the waves. The species which grow in these places therefore have a tough, leathery and narrow frond, whereas species in quiet waters have a delicate frond, often broad and brittle. The consistency of the frond thus accom- modates itself to the force of the beat of the waves, and in partially exposed places, or in places where the beat of the waves is not strong, but yet fairly considerable, we find the consistency of the frond to be about midway between what it is in exposed and in calm places. Like all other movements of the sea, the waves also are of great importance to the algal vegetation by the fact of their con- stantly providing fresh particles of water. c. Currents (Fig. 2). Along the coast of Iceland the warm water of the Atlantic Ocean meets the cold water of the Arctic Ocean. The Gulf Stream washes the south coast of the country and sends an arm northward along SW. Iceland and NW. Iceland, and along the entire north coast warm water can be traced (the eastern arm of the Irminger Current) to Langanes; and from thence the arm turns toward the south along the coast of E. Iceland (Nielsen, 52, p. 13), where it mixes with water from the East Iceland polar cur- rent, which comes from the Norwegian Sea (Helland-Hansen and Nans en, 27, p. 287, where the current is called the East Iceland Arctic Current). In this manner characteristic coast-water arises at E. Iceland. Regarding this Nielsen (52, p. 13) writes that the Irminger Current "gives the waters over the coast shoal of East MARINE ALGAL VEGETATION 31 Iceland a physiognomy different from that of the surrounding sea, the distribution of temperature being different even if the tempera- ture is not in any important degree higher than in the East Ice- landic polar current itself." The conditions of the currents in N. and E. Iceland are evidently very complicated, and only the direction of the current of the warm water has been given above in outline, but as the observations are so few it is difficult to form an opinion as to how the condi- ^ \~; ; t S\\ x \ v ' / Fig. 2. Map of Currents. (Helland-Hansen and Nansen.) tions of the currents along these coasts vary in other respects from year to year or according to the seasons. Further, other current -movements occur in the coast-water which may be deemed to be of importance to the vegetation, that is, such movements as are due to varying specific gravity. In the summer (Nielsen, 52, p. 8) the surface-water along the coast is lighter on account of its mixing with fresh water. A surface- current from the coast outwards then arises, and an under-current from the depth towards the coast. In the winter the surface-water along the coast becomes heavier owing to cooling, and sinks. Then a surface-current from the ocean towards the coast arises, and an under-current from the coast towards the depth. 32 H. JONSSON B. The Temperature of the Water. a. The temperature of the ocean around the coasts is not suf- ficiently known, and consequently the mean values cannot be given. I give, therefore, as an example, some actual measurements, as even these may be instructive in several respects. As regards NW. Iceland, N. Iceland and the northernmost part of the coast of E. Iceland I rely on the measurements carried out during the year 1904 on board the Danish Deep Sea Exploration ship "Thor" (Niel- sen. 52). East Iceland just south of Langanes. April24. St.15. f;o°99;N;lat. 14" 26 W. long. 178 m. Depth (m.) Temp. (C°) Salinity °/oo 0 0.63 34.70 25 0.74 34.72 50 0.73 34.72 100 0.71 34.72 66° 10' N. lat. 66° 16' N. lat. Aug. 12. St. 99, ! 40 29' W. long. Aug. 1 3. St. 100, 13o 36, w long 189m. 284m. Depth (m.) Temp. (CO) Salinity o/00 Depth (m.) Temp. (C°) Salinity %o 0 7.60 34.43 0 8.57 34.51 10 7.53 34.44 10 8.31 34.51 25 6.90 34.51 25 6.00 34.69 50 5.89 34.69 35 4.87 75 5.20 34.78 50 3.58 34.88 100 4.61 34.85 75 3.32 34.91 100 3.08 34.92 North Iceland east of Eyjafjor5ur. 66° 32' N. lat. 66° 33' N. lat. April 14, 17o50,wlong St. 74, 18o10,wlong 175m. 75m. Depth (m.) Temp. (C°) Salinity %o Depth (m.) Temp. (C°) Salinity °/oo 0 1.50 34.85 0 8.40 33.91 25 1.65 34.87 10 7.49 34.79 50 1.68 34.87 17 7.19 34.85 100 1.70 34.85 21 6.34 25 5.49 34.82 50 5.06 34.87 73 4.81 34.88 MARINE ALGAL VEGETATION 33 Cf .o 66° 14' N. lat. July 21. St. /3, 17o28,w>lona 197 m. Depth (m.) Temp. (C°) 0 10.5 10 7.12 17 6.52 25 6.74 50 5.95 75 5.37 100 4.84 Salinity %o 31.62 31.11 34.43 34.69 34.76 34.88 34.96 66°14'N. lat. Aug. 15. St. 104, 17o28,wlong 226 m. Depth (m.) Temp. (G°) 0 8.60 10 8.52 25 7.84 35 6.71 50 5.90 75 5.63 100 5.40 Salinity °/oo 34.27 34.27 34.47 34.60 34.82 34.86 34.91 North Iceland west of Eyjafjordur. April 23. ot. 1 O, f)(\( £i\J 220m. Depth (m.) Temp. (C°) 0 2.34 25 2.41 50 2.47 100 2.46 101 222 m. Salinity °/oo 34.94 34.96 34.97 34.97 ,_ 66° 31' N. lat. April 23. St. 12, 22° 25' W. long. Depth (m.) 0 40 60 62 m. Temp. (C°) 1.20 1.05 1.05 Depth (m.) Temp. (C°) Salinity °/oo 0 4.50 32.18 10 4.70 32.94 17.5 5.65 33.53 25 6.30 33.89 30 6.11 33.93 37.5 4.42 33.95 50 4.27 34.34 62.5 5.45 34.75 75 5.62 34.83 90 5.84 34.94 100 6.12 35.05 Salinity °/oo 34.76 34.81 34.81 66° 29' N. lat. June 2. St. 51, 22° 25' W. long. Depth (m.) 0 10 25 60 62 m. Temp. (C°) 3.52 3.49 3.49 3.49 Salinity °/oo 34.67 34.67 34.70 34.70 66° 30' N. lat. Aug. 24. St. 107, 22° 27' W. Ion* 46 m. Depth (m.) Temp. (G°) 0 8.91 10 8.96 20 8.93 30 8.91 45 8.72 Salinity °/ 34.58 34.58 34.58 34.58 34.61 The Botany of Iceland. I. 34 H. JONSSOX North-west Iceland. 66° 17'N. lat. 66° 33' N. lat. April 22. St. 10, 23°14'W.long. APril 23' Stll> 23° 37' W. long. 84 m. Depth (m.) Temp. (C°) Salinity °/00 125 m. Depth (m.) Temp. (C°) Salinity %o 0 1.42 34.69 50 2.82 34.99 120 2.94 35.05 0 82 2.40 2.97 34.97 66° 20' N. lat. June 2. St. 52, 23° 31' W. long. 66° 19' N lat Aug. 24. St. 108, 23o 27, w. long. Depth (m.) 0 10 25 50 75 100 142 m. Temp. (C°) 4.25 4.24 4.17 4.19 4.25 4.34 115 m. Salinity %o 34.92 34.92 34.98 34.99 34.99 35.01 Depth (m.) Temp. (C°) Salinity °/oo 0 9.42 34.67 10 9.52 34.66 25 9.46 34.66 35 9.39 — 50 8.16 34.85 75 7.37 34.91 110 6.76 35.01 Almost at the boundary between NW. Iceland and SW. Iceland. 65° 32' N. lat. 65 32' N lat 65 29' N lat June 26. St. 61, 24° 34' W. long. Au§" 2 6" St' 109' 24° 37'5 W. long. Depth (m.) 0 5 10 20 40 41 m. Temp. (C°) 8.17 7.93 7.77 7.66 7.61 Salinity %o 34.13 34.25 34.34 34.50 34.52 Depth (m.) 0 10 15 25 40 43 m. Temp. (C°) 10.26 10.20 10.20 10.21 10.21 Salinity %o 34.54 34.54 34.53 34.54 34.57 The measurements recorded show distinctly the range of the temperature in April and August, 1904. By taking successively the stations 11 (April 23rd), 13 (April 23rd), 14 (April 23rd) and 15 (April 24th) it is distinctly seen how the temperature of the surface- water of the ocean along the north coast of Iceland decreases from west to east, as shown by the following figures: St. 11 St. 13 St. 14 St. 15 2.40° 2.34° 1.50° 0.63° A similar decrease of warmth from west to east, but in a far lesser degree, appears to occur in the month of August. Station 106 (Aug. 23rd) shows a much lower temperature than MARINE ALGAL VEGETATION 35 was to be expected, which is unquestionably due to the water from the East Greenland polar current, as the ice was still, or had re- cently been, in the neighbourhood. South Iceland. From the ocean south of Iceland there are also measurements to hand carried out on board the "Thor" (Nielsen, 53). South coast west of Dyrholaey. JulyS. St.63,1904,^03oiw.long. Depth (m.) 0 25 50 104 106 m. Temp. (C°) 10.89 9.77 8.06 7.85 Salinity %o 35.14 35.14 35.14 35.16 63° 08' N. lat. July 9. St. i 64,1904,21c 1 30' W. long. 662 m. Depth (m.) Temp. (C°) Salinity °/oo 0 10.44 35.16 25 10.18 35.16 50 8.07 35.16 100 7.67 35.19 63°16'\ lat 63°^5'N lat July 12. St. 67, 1904, 19o17,w long Aug. 31. St. 114,1904, 20o03'W.long. Depth (m.) 0 25 50 100 765 m. Temp. (C°) 11.45 10.09 8.14 7.77 150 m. Salinity %o 35.03 35.14 35.21 35.21 Depth un.) Temp. (C°) Salinity °/oo 0 11.70 34.49 10 11.70 34.51 20 11.55 34.65 30 11.31 34.92 40 11.31 35.03 45 10.95 35.07 50 9.78 35.04 75 7.97 35.22 100 7.74 35.22 South coast east of Dyrholaey. May 23. St. 46, 1905. S ^ W/fong. 60 m. Depth (m.] 0 10 25 58 Temp. (C°) 7.51 7.14 6.87 6.90 Salinity %o 34.81 34.99 35.14 35.16 There are moreover some notes, given by Knudsen (44), on the temperature and salinity of the surface-water of the ocean south of Iceland. They are based upon the measurements carried out on 3* 36 H. JONSSON board the mail steamer "Laura" on its route from Scotland to Ice- land during the years 1897- -1904. Between longitude 17° and 18°, near the coast of Iceland, in a south-easterly direction from Dyrholaey, the mean temperature of the year (1897—1904) is stated (44) to be 8.8° and the mean salinity during the same period 35.19. The main features regarding the temperature of the ocean around Iceland then are as follows At the south coast warm, pure Atlantic water of a high (above 35 %o) and somewhat varying salinity occurs ; at SW. Iceland there is a somewhat similar sea ; at NW. Iceland and N. Iceland there is Atlantic water mixed with cold water of low salinity from the East Greenland polar current; and lastly, at E. Iceland Arctic water occurs (with a temperature of 0° to 2° and salinity from 34.6 per cent, to 34.9 per cent. [He Hand- Han- sen and Nansen, 27, p. 287]): the East Iceland polar current mixed with water from the Atlantic current. The change of temperature in the surface-layers of the water, the cooling process during winter and the heating process during summer, reaches down almost as deep as the algal vegetation, and is consequently of no slight importance to the latter. b. The Temperature in the Fjords. Respecting the tempe- rature of the surface-water of the ocean throughout the year infor- mation is given in the "Meteorologisk Aarbog" (Meteorological Year- book) regarding three stations in Iceland: Papey, Grfmsey and Stj'k- kisholmur. The following figures show the seasons' mean tempera- ture of the ocean for a period of five years (1902 --1906), chosen arbitrarily. Grfmsey is omitted, however, as the observations there have often been incomplete. Winter Spring Summer Autumn Papey (E. Iceland) 0.9° 1.7° 6.0° 4.3° Stykkisholmur (SW. Iceland). 0.4° 1.8° 9.6° 6.4° Vestmannaeyjar1 (S. Iceland). 4.1° 6.1° 10.4° 7.0° The winter in Papey is warmer than in Stykkisholmur, and the monthly mean temperatures during the winter, of the period mentioned, are there all positive; while in Stykkisholmur, February ( — 0.8) and March (---0.2) have negative numbers. The summer is much warmer 1 The figures for the Vestmannaej'jar constitute the mean of the period from July 1st, 1877 to Dec. 31st, 1906. According to J»orvaldur Thoroddsen, Lysing Islands, 2. Bd., pp. 350 — 351, Kaupmannahofn 1910. MARINE ALGAL VEGETATION 37 in Stykkisholmur. A comparison of these two places is, however, not equivalent to a comparison between E. Iceland and SW. Iceland, the situation of the stations being quite different. Papey is an island lying isolated in the ocean; Stykkisholmur, on the other hand, is a good example of the thermal conditions existing in the calm fjords. Regarding the temperature at various depths in the interior of the fjords some observations are to hand made during the summer. These are, however, too few and scattered to be given in mean values. I give, therefore, as an example, some actual measurements from different parts of the coast of Iceland. In East Iceland the measurements of temperature wrere car- ried out on board the Survey vessel "Diana" (Fisheries' Report ("Fiskeri-Beretning") for the financial year 1899—1900), and of these the following are given : Depth in fathoms Temp. (C°) Hellisfjordur (19/5) : . . . 0 1.7 10 1.6 Ladmundarfjordur (7/e) 0 5.0 cir. 33 1.5 Bakkafjordur (18/6) 0 8.0 7 2.5 Vopnafjor6ur (2%) 0 4.5 13J/2 1.8 Finnafjordur (17 e) 0 3.8 8V2 2.4 Finnafjordur (18/?) 0 6.5 5 6.3 Breiodalsvik (16/s) 0 4.6 Bottom 4.3 The low bottom-temperature in June is probably due to the East Iceland polar current. In addition to these Ssemundsson1 has published the fol- lowing measurements of the temperature and salinity of the surface of the fjords in East Iceland. Temp, of the Salinity surface (G°) °/oo ' Djupivogur (15 8; 9.0 29.47 high-water. 7.4 33.14 low-water. Faskruu-sfjoro-ur 19/s). . 10.0 33.67 at Bu6ir. 9.5 32.23 at Mjoeyri. 9.4 33.54 at Brimnes. 9.0 32.19 at Hofoi. i.. 7.0 34.71 at Kolfreyjusstadur. 1 Bjarni Ssemundsson. Fiskirannsoknir. 1898, Andvari, XXIV arg. 38 H. JONSSON Vattarnes (21/g) . . Eskifjordur (13/s) (24/8) Nor6fjor6ur (27/s) , Mjoifjordur (3%) . Sey6isfjor6ur (7/s) (8/8) (V9) (3/9) Borgarfjordur (4/9) Vopnafjor6ur (5/9) Temp, of the surface (C°) 8.5 9.3 5.0 5.0 5.0 7.5 6.5 9.5 7.5 8.0 8.5 7.8 7.5 7.0 7.0 6.5 7.1 7.5 Salinity °/oo ' 34.58 1.31 6.73 23.71 34.45 34.45 33.00 9.23 22.01 22.01 25.04 30.65 31.96 34.45 34.45 33.01 34.45 33.27 near land. in the fjord, after rain, near land, outflowing current, out in the fjord, in outflowing current, out in the fjord, in inflowing current, near land, at Brekka. head of the fjord. out in the fjord. at Skalanes. at Brimnes. at Dvergasteinn. out in the fjord, near land. From the most westerly part of the north coast and from the northern part of the north-west coast as also from Brei5ifjor5ur measurements are to hand of the temperature and salinity of the ocean at various depths in the interior of the fjords taken by Bjarni Ssemundsson (Fiskirannsoknir, 1908, Andvari, XXXIV arg.) of which the following are given: North coast, Steingrimsfjordur (31/?) . . . . Steingrimsfjordur (2/s) Hrutafjordur (7/s) North-west coast, Mj6ifj6rSur, the inner part (20/7) Depth in metres 0 15 30 55 0 15 35 70 100 0 15 30 44 0 25 50 Temp. C° 10.8 9.7 9.2 3.2 10.6 9.9 7.8 5.8 4.2 7.5 6.0 4.8 4.5 11.8 6.5 3.2 Salinity °/oo ' 27.4 34.3 34.5 34.0 34.6 34.7 34.7 34.6 34.7 35.0 30.6 34.5 34.9 MARINE ALGAL VEGETATION 39 Depth Temp. Salinity in metres C° %o Skotufjordur, the inner part (16/7) 0 11.4 30.5 15 8.9 35 7.9 34.3 70 7.4 34.7 105 4.6 34.7 Isafjar6ardjiip, l1/^ mile NW. of Ogur- nes(18/?) 0 11.8 32.2 15 11.5 35 8.5 34.0 70 6.8 34.9 100 6.6 35.5 South-west coast, Skardsstod in Brei6ifjordur (13/s) 0 10.9 34.0 (outside) 8 10.8 34.5 Skardsstod 0 11.1 24.8 (near land) 6 11.0 34.3 Bjarni Ssemundsson has kindly given me the permission, moreover, to make use of his hitherto unpublished measurements of the temperature of the ocean at SW. Iceland (BreiSifjordur and Faxafloi), in the interior of the fjords, from the summer of 1909. Of these the following are given: Breidifjordur Stykkisholmur (the port /?) Depth in metres 0 Temp. C° 10.6 Salinity °/09 " 35 1 About 1/2 a mile SE. of Vadstakksey (7/7) 0 10.1 t-J t-F • X 34.2 15 10.1 30 10.1 34.6 63 10.0 35.1 About l/2 a mile SE. of Hrappsey (10/7) 0 11.5 33.9 15 10.8 30 10.7 34.0 70 10.5 34.2 Kolgrafarfjor6ur, the interior (17A) .... 0 11.0 33.9 10 10.6 20 10.6 34.2 Kolgrafarfj6r6ur the mouth (18/?) 35 0 10.5 9.8 34.1 34.7 15 9.7 34.8 Hvammsfjordur about 2 miles SE. of Lambey (10/V) 0 11.4 32.7 15 11.0 30 11.0 33.7 45 10.9 34.1 40 H. JONSSON Faxafloi 5 miles SW. of Akranes (26/?) Knararnes About 2 miles SE. of I'or- modssker (31/?) Borgarnes (3/s) . - (5/8) - Within Seleyri BorgarfjorSur, the mouth (7/s) Hvalfjor6ur, Hvammsdjup Hvalfjordur, Galtarvikurdjup Depth in metres 0 15 30 60 0 0 15 30 45 0 0 0 5 0 10 20 0 15 30 60 0 15 30 50 85 Temp. C° 11.5 10.6 9.2 7.8 14.0 12.2 10.2 9.5 9.1 12.4 12.4 11.2 10.6 11.7 11.6 11.5 11.6 11.4 11.2 11.1 11.7 11.5 11.1 11.1 11.0 Salinity %o ' 34.4 34.7 34.9 34.0 34.6 34.6 34.7 26.9 18.9 8.8 14.7 28.9 34.4 33.4 34.0 34.2 33.9 34.0 34.4 34.3 high-water, low-water. From the measurements given above it will be seen that the temperature of the water in the interior of the small fjords is nearly the same from surface to bottom, while a regular decrease of warmth is immediately felt with the increase of depth in the more open waters. The temperature of the fjord-water is evidently dependent on the climate of the country; but regarding the temperature, during winter, for instance, we know nothing. Nor am I prepared to treat of the distribution of warmth in the coastal water in a more exact manner, as from the scattered observations made during the summer, which are at our disposal, no satisfactory results can be arrived at regarding the thermal conditions in which the algal vegetation exists all the year round. C. The Salinity. The degree of salinity is given above, together with the tem- perature. MARINE ALGAL VEGETATION 41 a. The Sea. South of Iceland the salinity varies but slightly, with the exception of the coastal water itself. During the summer, at any rate, the salinity of the coastal water must be considerably lower on account of the great amount of fresh water brought down by all the rivers of the south. On the other coasts and especially those of N. and E. Iceland the salinity varies according to whether the layers of water originate from currents which are deficient in salt or from the Irminger current, and also with the amount of fresh water streaming out from land. At station 106 (see above) the inferior salinity is evidently due to cold water from the Greenland current, and at station 73 it is probably due to fresh water. b. The Fjords. The salinity inside the fjords varies consider- ably and the variation is dependent upon the amount of fresh water which intermingles with the water in the fjord, partly in the form of river water and partly in the form of precipitated moisture. The lowest salinity in the fjords was 1.31 °/oo in Eskifjor5ur after rain. Heavy rainfalls must be capable of causing such an inferior salinity in other places also, especially in narrow fjords, but this will not last long, and as the littoral algae can endure heavy showers during low-tide, they will not suffer to any extent worth mentioning. In places where the salinity is as low as in SeyMsfjordur (9.23 °/oo), where a rather large river disembogues, the algal vegeta- tion occurs sparingly, although algae are found, especially green and brown algas. At a place like Borgarnes, where the salinity is low (18.9; 26.9) on account of fresh water from Hvita, the algal vegeta- tion occurs more abundantly than at the very head of Sey5isfjor5ur by the river, but the inferior salinity excludes certain species, for instance, Polysiphonia fastigiata, although Ascophyllnm occurs abun- dantly. Further out, where the sea is more saline, it is not absent. t/ It is especially in the surface-water of the tjords that the sali- nity varies so much. It is greater in the large open fjords, such as Faxafloi, than in the small land-locked fjords, such as HvalfjorSur, a circumstance which must certainly be due to river-water. The figures given show also that the salinity of the surface-water of the smaller fjords is less in the inner part than in the outer part. The same difference seems to appear also between the deeper layers of water of the inner and outer parts. As a rule, the salinity of the deeper layers is higher and more stable, which must be beneficial to the vegetation in the depths. 42 H. JONSSON 3. THE AIR. The climate is of special importance to that part of the algal vegetation which is exposed during low-tide. The temperature is possibly of least importance in a climate where high and very low degrees of temperature do not occur, or are, at any rate, rare. The degree of humidity of the air and the cloud-covering are, on the other hand, highly important to the algal vegetation which is left dry. The movements of the air are also of importance, especially as it produces movements in the sea. A. The Temperature. The following means (19 years)1 from a number of stations on different parts of the coast are here given for the elucidation of the thermal conditions. Papey E. Iceland Winter -1.1 Spring 0 1 Summer 6 0 Autumn 3 3 The year 2 1 )r6ur -1.4 08 7 6 3 5 2 6 N. Iceland hofn -4.0 -1.9 6 5 1 6 0 5 :V . — 2.3 — 1.5 6.1 2.9 1.3 Grimsey SW. Iceland Stykkisholmur - 2.2 0.8 8.9 3.9 2.9 S. Iceland Vestmannaeyjar 1.1 3.8 9.7 5.2 5.0 Eyrarbakki -2.0 2.1 10.2 3.5 3.5 From the figures given above it will be possible to form an opinion of the thermal conditions in the places mentioned, and these are altogether such that an algal vegetation left dry can thrive everywhere along the coast. The extremes will not have a sufficiently injurious effect on the vegetation for it to be noticeable in the long run. High degrees of temperature, about 20 °C for example, occur rarely in the summer, and will have no permanent effect. Very low degrees of temperature in the winter will not injure the vegetation left dry to any extent worth mentioning, as it is then partly pro- tected by snow (at the very top) and partly by ice. I do not consider the cold in the winter injurious to the vege- tation which is left dry, as the algae certainly endure being frozen fairly well. At least I have seen uppermost in the littoral zone, Willaume- Jantzen, Meteorologiske Middeltal og Extremer for Fseroerne, Island og Gronland, Kjobenhavn, 1899. MARINE ALGAL VEGETATION 43 early in May, algae which had been frozen hard during the night, apparently quite unharmed and alive when thawed, nor could one perceive next day that they had suffered at all; but, as I was tra- velling, I was not able to observe them more than these two days. It is also a foregone conclusion that the algae left dry must freeze in the winter when the cold is severe, but it does not appear that they suffer thereby. If the cold cannot be said to have any directly injurious effect, yet indirectly it may hurt the vegetation (though not to any great extent) by the fact that the water freezes and the beach becomes ice-covered. During severe winters a covering of ice may be found during the greater part of the winter in the smaller fjords, and especially where the fjord-water is abundantly mixed with fresh water, and even if the winters are quite mild, yet from time to time the water next the beach may freeze. In the littoral zone and \j on rocks which are laid bare during low-tide, the ice forms in ac- cordance with the substratum, and if this is uneven the ice breaks. At high- water the ice-covering is lifted up; the pieces of ice may then freeze together again, and break once more with the next ebb- tide. During spring-tides in particular these movements are rather considerable and the plant-covering may be a good deal damaged thereby: but if one regards the coasts in their entirety these distur- bances will prove to be of small importance. The drift-ice is much more dangerous to the algal vegetation as the icebergs scrape the rocks with which they come in contact. Strom felt, when travelling in Iceland in 1883, the year following one of the years notable on account of the great quantity of ice, found the littoral vegetation poorly developed in the north country. This most certainly resulted from the drift-ice having blockaded the coast during the whole summer of 1882. In the summer of 1898, I saw on the promontory between Sey5isfjor5ur and Lo5mundarfjor5ur distinct signs of the drift-ice which had been there in the spring. The injurious influence of the drift-ice consists mainly in the fact that it scrapes away the vegetation from the parts with which it comes in contact; possibly also in the fact that it reduces the tem- perature to far below normal. That the plants suddenly find them- selves in a much colder medium than thev are accustomed to must «/ produce a check upon them, particularly on the more sensitive species. The marine plants, however, are less affected by this than the land vegetation. As a rule, ice-years occur at fairly long intervals ; 44 H. JONSSON consequently the damage which the ice causes is not noticeable in the long run , it is noticed chiefly in the same year or the year following, and is remedied comparatively quickly. B. The Humidity. Very great importance must be ascribed to this as regards the algal vegetation left exposed. During the period of desiccation there is always the danger of the evaporation becoming too great, espe- cially if the air is dry. The more humid the air , the better the algae wrill be able to maintain life in it. The following figures from four places, each situated on a different part of the coast, show the mean humidity of the air as percentages (Willaume-Jantzen 1. c.). Winter Spring Summer Autumn E. Icel. Berufjordur (23 years) 77 78 81 80 N. Icel. Grimsey (21 years) 83 83 85 86 SW. Icel. Stykkisholmur (20-23 years) . 88 85 83 86 S. Icel. Vestmannaeyjar (12 years) 81 79 82 81 As a comparison with the Faeroes might be of interest, the figures showing the mean humidity as percentages at Thorshavn Willaume-Jantzen, I.e.) are appended. Winter Spring Summer Autumn Thorshavn in the Faeroes (25 years). .81 79 84 84 From these figures it appears that the humidity of the air in Berufjordur is less than in the Faeroes while the humidity of the air at Grimsey and also at Stykkisholmur is greater than in the Faeroes. The humidity of the air in the \vinter and the spring in the Vestmannaeyjar and in the Faeroes is the same, while at the latter place it is a little greater in the summer and autumn. C. Precipitation, Amount of Cloud, Foggy days, Wet days. a. Precipitation. The following figures show for purposes of comparison the mean downfall in millimetres at four places in Iceland, one on each coast, and at Thorshavn in the Faeroes (Wil- laume-Jantzen, 1. c). Winter Spring Summer Autumn The year E. Icel. Berufjordur (23 years). .. 348.0 222.7 203.7 340.3 1114.7 N. Icel. Grimsey (16-22 years). .. 83.5 64.8 85.6 139.9 373.8 SW. Icel. Stykkisholmur (1 8-22 yrs.) 191.5 115.2 113.6 203.8 624.1 S. Icel. Vestmannaeyjar (15 years). . 354.1 257.3 252.1 402.2 1265.7 Thorshavn in the Fseroes (25 yrs.). 510.9 485.2 272.4 324.6 1593.1 MARINE ALGAL VEGETATION 45 As shown by the figures there is a considerable difference in the amount of precipitated moisture. That of Thorshavn is greatest, next come the Vestmannaeyjar, followed by Berufjor5ur. The preci- pitation at Stykkisholmur is not more than half that in the Vest- mannaeyjar, and that of Grimsey is not more than a fourth part of that in the Vestmannaeyjar. b. Mean Amount of Cloud (Willaume- Jantzen, 1. c.). Scale 0—10. Winter Spring" Summer Autumn E. Icel. Berufjordur (23 years) 6.4 6.6 6.9 6.6 N. Icel. Grimsey (22 years) 8.5 8.2 7.9 8.5 S\V. Icel. Stykkisholmur (22 years) 7.1 6.4 6.0 6.9 S. Icel. Vestmannaeyjar (18 years) 6.2 6.1 6.1 6.3 Thorshavn in the Faeroes (25 years).. 7.4 7.1 7.7 7.5 The amount of cloud is greatest in Grimsey, and there is no great difference between the remaining three coast-stations in Ice- land. In Thorshavn, however, the amount of cloud is considerably greater and consequently this place approximates to Grimsey. c. Foggy and Wet days. Mean number of Foggy days (Wil- laume-Jantzen, 1. c.). Winter Spring Summer Autumn The year E. Icel. Berufjordur (23 years). 44.0 52.0 67.0 49.0 212 N. Icel. Grimsey (22 years) ... 2.4 13.0 31.0 7.0 53 SW.Icel.Stykkisli61mur(22 years) 1.0 2.6 4.1 1.3 9 S. Icel. Vestmannaeyjar (18 years) 7.0 12.0 21.0 12.0 52 Thorshavn in the Faroes (25 yrs.) 3.0 10.0 29.0 9.0 51 The number of foggy days in Berufjor5ur is remarkably high, and at Stykkisholmur is extremely low. In Grimsey the number is much lower during the winter months, and higher during the summer months than it is in the Vestmannaeyjar; there are also small differences during spring and autumn. There are only small differences between the Faeroes and the Vestmannaeyjar, except in the winter, when the Vestmannaeyjar have twice as many foggy days. Mean number of Wet days (Willaume- Jantzen, 1. c.). Winter Spring Summer Autumn The year E. Icel. Berufjordur (23 years). 52 43 34 48 177 N. Icel. Grimsey (22 years) ... 40 29 29 45 143 S\V.Icel.Stykkisholmur(22 years) 58 47 40 52 197 S.Icel.Vestmannaeyjar(18years) 64 55 47 59 225 Thorshavn in the Faeroes (25 yrs.) 81 66 58 74 279 At all seasons the number of wet days is highest in the Faeroes 46 H. JONSSON and in the Vestmannaeyjar, yet considerably higher in the former place. Grimsey has the fewest wet days, and Berufjor5ur and Styk- kisholmur have somewhat similar numbers. It is in the spring and the summer especially that the desicca- tion, during the period of exposure, may have an injurious effect in the zone laid bare along the coast. The amount of cloud is of course important, since clouds diminish the danger of desiccation, but the mean figures are not sufficiently elucidatory. Bright sunny days are not propitious to the vegetation left exposed, especially if several such days occur in succession ; and if this takes place at neap-tide, the vegetation which is found above Pelvetia-Fucus spiralis is in danger. Although the weather in Iceland varies greatly, longer periods w^hich are damp or dry often occur. Clear days are not uncommon in the spring and summer, and periods of even a week or more of bright weather are not rare. On bright sunny days in summer the temperature may rise rather high; I have measured 20° C. on such a day in a pool in the littoral zone, in the plant- covering itself, and the temperature of the air may rise even higher. The periods of bright and dry weather are certainly of impor- tance as regards the upper limit of growth of the algal vegetation during the summer. The Faeroes are probably less favoured by clear weather than Iceland, and the difference in the upper limit of growth of the algal vegetation in Iceland and in the Faeroes may possibly be partly explained by this. D. Winds. The following figures show the annual percentage (W ilia ume- Jantzen, I.e.) of the w7inds: N. . NE. . E. . . SE. . S... . SW. . W. .. NW. , Calm BerufjorcSur 6 24 4 6 8 15 4 23 10 Grimsey 8 18 20 16 4 5 12 7 10 Stykkisholmur Vestmannaeyjar 3 18 20 16 11 10 7 3 12 13 3 23 9 8 10 8 4 22 MARINE ALGAL VEGETATION 47 It happens rather frequently that the winds are stormy and, as an example, the annual percentage of storms for Stykkisholmur (from */9 1845 to 31/2 1892) may be given: — N. 32, NE. 61, E. 13, SE. 17, S. 44, SW. 31, W. 26, NW. 11. The frequency of "calm" is 10% at Grimsey and in Berufjordur and the frequency of "wind" is therefore 90 % in both places; at Stykkisholmur the frequency of "calm" is 12 % and that of "wind" 88%; in the Vestmannaeyjar the frequency of "calm" is 22% and that of "wind" 78 %. At Thorshavn, in the Faeroes, the annual "calm" is 11 % and the frequency of "wind" 89 %, somewhat the same, therefore, as at Stykkisholmur and greater than in the Vestmannaeyjar. 4. LIGHT. The influence of light on the distribution of the algal associa- tions and on their appearance is, as is well-known, exceedingly great. Without doubt most investigators assume that the main di- vision of algal vegetation into a green, a brown and a red zone is due to the quality of the light, but one cannot on that account consider the intensity of the light to be of no importance. To what extent the shades of colour in the red algae are to be regarded as an adaptation to the intensity of the light or to the quality of the light, I find rather difficult to decide. I agree with Berth old and Oltmanns in thinking that the Floridece may be characterized as shade-plants in the same sense that we speak of shade-vegetation in lava-clefts and in other places where there is a faint light. By shade-plants I understand plants which prefer feebly illuminated spots, and do not, as a rule, thrive in the full light of day. In the tidal region (Part VI) the littoral Floridece evidently prefer crevices and grottoes, i. e. feebly illuminated places, and thus prove themselves to be shade-plants. I shall not enter more fully into the question of light, as I have made no experiments in that connection and, moreover, the subject requires to be reinvestigated (Oltmanns, 54). III. THE HORIZONTAL DISTRIBUTION OF THE SPECIES AND THE COMPONENTS OF THE ALGAL FLORA. IN the following list (Table I) of the hitherto known Marine Algae of Iceland a letter (A, B1? B2 , C, D, E17 E2) is placed before each species, showing to which plant-geographical group I refer it (cf. Borgesen and Joiisson, 14). A indicates the arctic group, B! sub-division 1 of the subarctic group, B2 sub-division 2 of the subarctic group, C the boreal-arctic group, D the cold-boreal group and E the warm-boreal group. The letter c placed after the name of the species indicates that it has been found in all the five coastal districts (E. Icel., N. Icel., NW. Icel., SW. Icel. and S. IceL, see above, p. 5); a (c) placed after the name of the species indi- cates that it probably occurs in all parts of the coast. Table 1. The Distribution of the Species along the coast. E. Icel. N. Icel. NW.Icel. SW.Icel. S. Icel. c Rhodophyceae. Bangia fusco-purpurea (c) . . 4- 4- 4- 4- c Porphvra umbilicalis c . 1 4- i 4- 4- i + 4- B9 I V P. miniata c 1 -4- 4- 1 4- i 4- i 4- *-* 2 D Porphyropsis coccinea . ... 1 i l i 4- i 4- B, Conchocelis rosea c 4- 4- 4- i 4- i 4- c Chantransia microscopica (c) 1 -I- i i i D G. Alaria? i 4- 4- C C. secundata c 4- 4- 4- 4- 4- B, C. virgatula (c) .... i 4- i 4- "2 E, Chondrus crispus (+) 4- 4- D Gigartina mamillosa (c) + 4- V 1 / 4- 4- C Ahnfeltia plicata i c4-) (40 i 4- i 4- B2 D Phyllophora Brodiaei * interrupta .... P. membranifolia _L i V 1 / • • v 1 / + i 4- 4- B, Actinococcus subcutaneus 4- 4- i 2 A Ceratocolax Hartzii i i 4- D Cvstoclonium uuruurascens . 4- i 4- 4- 4- H. JONSSOX! MARINE ALGAL VEGETATION 49 Table 1. The Distribution of the Species along the coast (continued). E. Icel. N. Icel. NW.Icel. SW.Icel. S. Icel A Turnerella Pennyi B2 Euthora cristata c B, Rhodophyllis dicbotoma (c) B;, Rhodymenia palmata c Ej Lomentaria clavellosa D L. rosea Ej Plocamium coccineum Bj Halosaccion ramentaceum c -j- D Delesseria alata A D. Baerii * corymbosa ? B ., D. sinuosa c D D. sanguinea E2 Bonnemaisonia asparagoides . Ej Pterosiphonia parasitica D Polysiphonia urceolata c D P. fastigiata A P. arctica D P. nigrescens B2 Rhodomela lycopodioides c B2 Odonthalia dentata c D Callithamnion Arbuscula Ej C. scopulorum D Plumaria elegans B, Ptilota plumosa B i P. pectinata C Antithamnion Plumula v. boreale (c) . D A. floccosum (c) D Geramium acanthonotum D C. Deslongchampii D C. fruticulosum D C. circinnatum Ej C. arborescens D C. atlanticum C C. rubrum (c) G Rhodochorton Rotbii c D R. repens D R. minutum Bj R. penicilliforme (c) C R. membranaceum (c) D Dumontia filiformis (c) D Dilsea edulis D Petrocelis Hennedyi A Cruoria arctica E! G. pellita B! Peyssonellia Rosenvingii D Rhododermis parasitica The Botany of Iceland. I. •f + -f -f 9 + + 4- + -f -h • • + -f 50 H. JONSSON Table 1. The Distribution of the Species along the coast (continued). B, D B; A A B, D D B! D D D C B: D B, B2 D B, D D B, D B, D D B: D C B, D C C D D D B2 B2 B: D D B, A Lithothamnion glaciale (c) E. Icel. N. Icel. NW.Icel. SW.Icel. S. Icel. 4- 4- 4- + 4- 4- + + + 4- + + + + + + 9 + + + + + + 4- 4- _i_ + 4- + + + + + + • • + + + + + + 4- • • 4- _j_ • • 4- 4- 4- 4- • • + • • • • 4- 4- 4- + 4- • • • • • • + 4- % + 4- * • 4- + 4- 4- 4- 4- 4- 4- 4- + 4- 4- 4- • • -f + 4- I + 4- i 4- + 4- + + 4- 4- • • 4- • . 4- 4- + 4- + 4- 4- • • • • • 4- + • • • • i • + 4- 4- 4- 4- • • 4- + • • + 4- + 4- L. Un fieri (c) . L. tophiforme (c) L. flavescens L. foecundum L. laeve c . . . L. Lenormandi Phvmatolithon polvmorphum Clathromorphum compactum c Lithophyllum Crouani Dermatolithon macrocarpum Corallina officinalis Hildenbrandia rosea c Phaeophyceae. Lithoderma fatiscens (c) Petroderma maculiforme Ralfsia ovata (c) R. clavata (c) R. verrucosa (c) R. deusta (c) Myrionema vulgare M. Corunna? 1 + 4- 4- 4- 4- 4- + + t . + 4- M. glohosum (c) M. faeroense M. Laminarise (c) Ascocvclus islandicus Microsyphar Polvsiphoniae (c) Strcblonema aecidioides (c) S. Stilophorae v. csespitosa c Pvlaiella littoralis c Ectocarpus tomentosoides c E. tomentosus E. confervoides c E. siliculosus (c) E. penicillatus (c) E. fasciculatus E. Hinksiae Leptonema fasciculatum v. subcylin- drica (c) Elachista fucicola c Sphacelaria britannica (c) S. radicans (c) . S. olivacea Chaetopteris plumosa (c) Orrmhalonhvllum ulvaceum . MARINE ALGAL VEGETATION 51 Table 1. The Distribution of the Species along the coast (continued). E. Icel. N. Icel. NW.Icel. SW.Icel. S. Icel. B, Punctaria plantaginea (c) 4- 4- 4_ ~ B2 Litosiphon filiformis (c) \ 4- i 4. 4_ • • _4_ B2 Isthmoplea sphaerophora (c) i -1- i 4_ 1 4_ 1 _i_ B, Stictvosiphon tortilis (c) i 4- 4. i 4_ I _|_ 1 B2 Phaeostroma pustulosum (c) i 4- i^ i 4- 1 4. C Scvtosiphon Lomentaria c i 4- 4- i 4- i 4_ • • 4_ D Phvllitis zosterifolia (c) . i 4- i i 4- T^ _(_ 1 4. G Ph. fascia c i 4- 4. i 4- l^ 4- i 4_ BI Coilodesme bulligera. \ 4- i^ i 4- i 4_ 1 D Dictyosiphon Ekmani. i i i 4- D D. Mesogloia . 4- i D D. Chordaria (c) . 4- i 4. A I), corvmhosus i 4- i B, D. hippuroides (c) . i 4- 4- • • 4_ 4_ B, D. foeniculaceus (c) 4- i 4- i 4- i 4_ i^ B, Desmarestia viridis c i 4- i 4- i 4- i^ 4_ 4. ~ Ba D. aculeata c i 4- i 4. i 4- i 4. _|_ EI D. ligulata . i \ i i 1 4_ D Castagnea virescens M . 4- 4- J- i E: Leathcsia difformis i \^ 4- i 4- B., Chordaria ttagelliformis c 4- i -I- 4- i 4- _|_ B, Chorda tomentosa (c) i 4- i i 4- i^ B2 C. Filum (c) + i 4- 4. i 4- B! Saccorrhiza dermatodea (c) + i 4- i -1- i 4- I) Laminaria saccharina c 4- i -I- i 4- i 4- 4_ Bj L. faeroensis i 4- i 4- i i i^ A L. nigripes . i B, L. digitata c 4- 4. 4- 4. 4_ D L. hvperborea c \ 4- i 4_ i 4- i 4- i^ 4_ BJ Alaria Pvlaii c i 4- i 4- i 4- i 4- \ 4_ D A. esculenta c 4- i 4- i 4- i 4- i 4- D Fucus spiralis (c) i 4- i 4- i i 4- i 4_ B2 F. inflatus c ... i -I- i 4- 4- 4- \ 4_ D F. serratus i i i i 4. i^ C F. vesiculosus c 4- 4- _j_ i 4. 4_ D Pelvetia canaliculata i n^ i i 4- i^ 4_ C Ascophvllum nodosum c 4- 4- 4- i 4- i^ 4- B2 Chlorophyceae. Chlorochvtrium Cohnii (c) i • i i 4. i B2 C. inclusum (c) .... 4- 4- 4- i 4. B., C. dermatocolax (c) . . . i i 4- i i 4- B! C. Schmitzii (c) i i 4- D Codiolum Petrocelidis i 4- B2 C. gregarium (c) . 4- i D C. pusillum (c) . i 4- 4* H. JONSSON Table 1. The Distribution of the Species along the coast (continued). : E. Icel. N. Icel. NW.Icel. SW.Icel. S. Icel. B, Percursaria percursa (c) 4- D Enteromorpha aureola 4- i E, E. Linza t 4- 4- c E. intestinalis c 4- 4- 4- i 4- i 4- c E. clathrata (c) 4- i 4- i i 4- 4- A Monostroma groenlandicum i 4- i 4- 4- i r B2 M. Grevillei c i 4- i 4_ i 4- i 4- B, M. undulatum c i 4- i 4- i 4- 4- i 4- B, M. fuscum c i 4- i 4_ i 4- i 4- i 4- *•* i c Ulva lactuca (c) i^ 4- i 4- i 4_ i 4- D Prasiola polvrrhiza (c) . i i i 4- i 4- D P. furfuracea (c) . . . . 4- 4_ i 4_ i D P. stipitata (c) i 4. i 4. i 4_ 4- B, Ulothrix consociata v. islandica i i 4_ • i B, U. subflaccida (c) T 4_ • • B, U. pseudoflacca (c) 4- i • • 4_ 4- i B., U. flacca c i 4- • • 4_ 4_ I 4_ i 4- B2 D Pseudendoclonium submarinum (c) . . . Entoderma Wittrockii (c) i + 1 • • 4. i i i 4- B, Acrochsete parasitica (c) 1 4- i D A. repens • • 4_ \ B, Bolbocoleon piliferum (c) 4- 4_ I 4_ B9 Ulvella fucicola (c) i 4_ 1 I 4_ 4- 2 B, Pringsheimia scutata (c) i^ V I i D Ochlochcete ferox (c) . . I i \ iJ o Urospora mirabilis c 4- 1 4. 4_ i 4- B, U. Hartzii fc) . 4- I i i 4. i 4- *•• i Bo • * U. Wormskioldii c • 4- 4. 4_ i 4_ i 4- C Chaetomorpha tortuosa (c) i i i 1 i \ 4- i B, C. Melagonium c 4- 4. 4_ i^ 4_ 4- C Rhizoclonium riparium (c) i 4- 1 4_ t \ 4_ i D «j Spongomorpha vernalis (c) i 1 i 4. D Acrosiphonia albescens c 4- 4. 4_ i 4_ • 4- Bo A. incurva c . . i 4- 1 4_ i 4_ i 4_ i 4- B, A. hystrix (c) 1 i 4- 1 4_ T 4_ \ 4_ i D A. flabelliformis i 1 T^ i^ 4- Bj A. penicilliformis 4- Bo Cladophora rupestris i • • • • 4. 4- 4- B., C. hirta (c) i i 4- i 4- B2 C. sericea (c) • • 4_ 1^ 4- i 4- I) C. glaucescens (c) 1 • • i^ 4- i G C. gracilis (c) . 4- 4_ n 4- C Gomontia polyrrhiza (c) i 4- 1 4. 4_ i 4- B2 Ostreobium Queketti (c) i -U 4_ 1 i i 4. 1 1 t i^ MARINE ALGAL VEGETATION 53 Table 1. The Distribution of the Species along the coast (continued). E. Icel. N. Icel. NW.Icel. SNV.Icel. S. Icel. B, Cyanophyceae. Pleurocapsa amethvstea c 4. 4- 4- 4- 4- L»2 I) Plectonema norvegicum (c) i i 4. i i i E Phormidium autumnale (c) -u c Spirulina subsalsa (c) i 1 4- c Calothrix scopulorum (c) 4- -u i c Rivularia atra (c) \ i 4- i From Iceland (31, 14 and 57) there are published 76 species of red algae, 67 species of brown, 51 of green and 6 af blue-green 200 species in all. All of these, of course, are not equally common along the coast, and their habitats, as far as these are known, in the five districts into which the coast is divided, are given in the above table. That table shows that comparatively few species occur in all the coastal districts (in the table, such are indicated by the letter c placed after the name of the species). In all parts of the coast are found 15 species (20 %) of red algae, 18 species (26.8 °/o) of brown, 10 species (19.6%) of green and 1 species (16.6%) of blue-green. Thus, of the 200 species there are 44 species (22 %) which are common to all the coastal districts. If we take into consideration the fact that the great stretch of coast round the whole of Iceland is as yet far from accurately investi- gated, we may expect, after future investigations, not only that several more species will be found, but also that the distribution of the species in the different parts of the coast will prove to differ from what is at present stated to be the case. Therefore, in the above list of the distribution of the species along the coast I have placed a (c) after the name of those species which, as I surmise, are pro- bably to be found in all parts of the coast. I base this supposition partly on the position of the habitats already known along the coast, and partly on the occurrence of the species in the adjacent floral districts, e. g. the Faeroes and Greenland. The mark (c) is subjoined to 14 species of red algae, 29 species of brown, 32 species of green and to 5 species of blue-green. On adding to this the above-mentioned species which have the letter c subjoined we get 29 species (38 %) of red algae, 47 species (70 %) of brown, 42 species (82 %) of green and 6 species (100 %) of blue- green. Thus, of the 200 species 124 prove to be common to all parts 54 H. JONSSON Red alga; Brown alga? Green algse Blue-green alga; 20 27 19 17 38 70 82 100 of the coast. The percentage of each group is of the greatest im- portance, and for the sake of explicitness I have arranged these figures in a tabular form, both those which refer to c and those which refer to c + (c). Species common to all the coastal districts, given as percentages. All the groups collectively 22 c + (c) 38 70 82 100 60 On considering these two series of figures it becomes evident that the figures given for c Hh (c) come nearer to the real facts, while those given for c merely indicate an incomplete knowledge of the coastal distribution of the species. The fact is that, \vhere there is not a greater climatic difference between the different parts of the coast than is the case in Iceland, it may always be expected that, as regards the common species, the highest numbers will fall to the green and to the blue-green algae; and where the hydrographic differences between the different parts of the coast are as pronounced as they are in Iceland, it is natural that the smallest number will fall to the red alga3, and just as naturally the brown algae will in this re- spect be placed almost midway between the red and the green algae. Therefore, as regards the floristic difference between the different parts of the coast, particular stress is laid on the remaining 76 species. In the following table they are arranged according to their habitats. Under A, those species are given which either occur in E. or N. Iceland only, or are most common there, and thence are dis- tributed southward along the north-west coast as far as SW. Ice- land. Under B are given species which either have been found in S. or SW. Iceland only, or are most common there, and thence have a distribution northward along the north-west coast, many of them having, moreover, an eastward distribution along the north coast. Table 2. The Distribution of the 76 species not common to all the coastal districts. "3 "3 "3 "3 "3 "3 "3 73 "3 "3 A ~. •— u u u •^ - u B • Ed £ > 5; v~. •,' z ^ ^ c^i Z e/5 *^ en + 1 . Lomentaria clavellosa • • • • -)- 2. L. rosea 1. Lithothamnion flavescens 4- 4- 3. Flocamium coccineum MARINE ALGAL VEGETATION Table 2. The Distribution of the 76 species not common to all the coastal districts (continued). 55 A 13 u I fli 4) U 0 i i 13 o u Z u Cfl ;/: B W Z Z o> * w Z • 2. Omphalophyllum ulva- ceum + 9 • • 9 • • * • • < > • 9 + + [ + + • * • • • • -f + + 4. Bonnemaisonia aspara- goides 5. Pterosiphonia parasitica 6. Rhodochorton repens 7. Phymatolithon polymor- phum 8. Myrionema Corunnae 9. Ectocarpus Hinksiie 10. Desmarestia ligulata 11. Acrosiphonia flabelliformis 12. Porphyropsis coccinea 13. Chantransia Alariae 14. Phyllophora membrani- folia 15. Delesseria alata 16. Callithamnion Arbuscula 17. C. scopulorum 18. Plumaria elegans 19. Ceramium acanthonotum 20. Cruoria pellita 21. Dermatolithon macro- carpum 22. Ectocarpus tomentosus 23. E. fasciculatus 24. Fucus serratus 25. Pelvetia canaliculata 26. Enteromorpha Linza 27. Ceramium Deslongchampii 28. C. atlanticum 29. C. fruticulosum 30. C. circinnatum 31. Rhodochorton minutum 32. Dilsea edulis mandi 34. M}rrionema faeroense 35. Dictyosiphon Ekmani 36. Codiolum Petrocelidis 37. Chondrus crispus 38. Polysiphonia fastigiata 39. Rhododermis parasitica 40. Sphacelaria olivacea 41. Cladophora rupestris 42. Acrocha?te repens • • 3. Laminaria nigripes 4. Acrosiphonia penicilli- formis . . > • • • * • • * + + + + + + + + • • 5 Delesseria Baerii . 6. Turnerella Pennvi + + + + + -f • • * • 7. Lithothamnion foecun- dum 8. Laminaria fasroensis . . . 9. Petroderma maculiforme. • • • • 11. Dictyosiphon Mesogloia. . 12 D corvmhosus • • * • • • * • • • 13. Enteromorpha aureola . 14. Ulothrix consociata v. is- landica ... . • • * • 15. Monostroma groenlandi- cum + + + + • • i * • • + + 16. Phyllophora Brodisei v. interrupta 17. Actinococcus subcutaneus 56 H.JONSSON Table 2. The Distribution of the 76 species not common to all the coastal districts (continued). A 1 a "3 o NW.lcel. 1 •j-. ^ u m ttJ 13 g — • *> I "3 u V) s B 18. Ceratocolax Hartzii + • • (+) (+) + + 43. Ceramium arborescens 44. Ahnfeltia plicata 19. Polysiphonia arctica .... + + 4- + • • 4- 4- + + 45. Cystoclonium purpura- 20. Ptilota pectinata + + * • i i 4 + + + scens 46. Ptilota plumosa 47. Petrocelis Hennedyi 21. Peyssonellia Rosenvingii. 4- 4- 4- 4- . . . . + 4- + + 48. Gorallina officinalis I * .- + 49. Lithophj'llum Crouani 22. Coilodesme bulligera .... 4- . . 4- 4- . . . . 4- . . + . . 50. Polysiphonia nigrescens 23. Cruoria arctica + (+) + + + Sl.Myrionema vulgare 52. Leathesia difformis 53. Delesseria sanguinea Total . . . 15 13 8 1 5 1 10 13 40 40 Total To illustrate more distinctly llo^Y the species with a north- eastern distribution (A) and those with a south-western distribution (B) intermingle in N., NW. and SW. Iceland I subjoin the following figures taken from the preceding table: 2. Icel. N. Icel. NW. Icel. SW. Icel. S. Icel (1) 10 13 40 40 15 9 + 4 8 5 • • B Notes on the species. Of the 4 species (A, 1 — 4, Tab. 2) which have been found only in E. Iceland, Nos. 1 and 3 occurred in great abundance in several of the fjords, while Omphalophyllum was found only in ReySarfjor5ur, where it occurred abundantly, and Acrosiphonia in one place only. Delesseria Baerii, I suppose (31, p. 140), has originated from either E. or N. Iceland. Tarnerella is most common in E. Iceland, and in addition to the habitat in N. Iceland which has been published (31, p. 135) has been found by B. Saemundsson in Steingrimsfjor5ur in the most western part of the north coast; consequently it must have a wide distribution along this coast. With regard to Lithothamnion foecan- dum and Laminaria fceroensis it must be assumed that thev are / v more frequent in E. and N. Iceland than is known at present, and as the latter species occurs in the Faeroes, it may well be expected to be met with on other parts of the coast of Iceland. MARINE ALGAL VEGETATION 57 With respect to the 6 species (A, 9 — 14) which have been found only in N. Iceland, it cannot be assumed that they are confined to the north coast, and it is highly probable that they have a much wider distribution on both sides. Dictyosiphon corymbosus and Ulo- thrix consociata v. islandica must, however, be assumed to belong more closely to E. and N. Iceland. «/ As regards the 8 species (A, 15 — 22) which occur so far west or south as NW. Iceland or SW. Iceland, it must be supposed that their absence from N. Iceland (and E. Iceland [Ceratocolax, CruoriaJ) is due merely to insufficient knowledge regarding their distribution. Of these species, those which extend to SW. Iceland have not been found, however, further south than in Breidifjordur, with the ex- ception of Peyssonellia which has been found in Faxafloi near Reykjavik. Of the species given under A, Nos. 1 — 8, 12 and 14 — 23 must consequently be supposed to have an east-northward distribution along the coast of Iceland, while Nos. 9, 11 and 13 must be sup- posed to have some other principal distribution; one species (10) is endemic in the most western part of the north coast. Under B, 53 species are recorded. Of these 10 have been found only in S. Iceland , the majority of these in the Vestmannaeyjar only; to these must be added Bonnemaisonia (31, p. 141) which I believe to have been found in S.Iceland. That is, 11 species in all, one of which, however, Rhodochorton repens, is endemic. 17 species have been found only in S. and SW. Iceland (B, 12—28) and 8 species in SW. Iceland only. Thus, there are in all 36 species which are known from S. and SWT. Iceland only. 6 species (B, 37- -42) have a more northern distribution, as they have been found in NW. Ice- land. Consequently, there are 42 species which are known only from S. and W. Iceland (NW. Iceland included), but of these species there are two, Cladophora rupestris and Codiolum Petrocelidis which probably have a more northern distribution. 10 species (B, 43 — 52) which have principally a south-western distribution (the fact that some of them have not been found in NW., SW. and S. Iceland is probably due to insufficient knowledge regarding their distribution) have been found also in N. Iceland. But all these species do not reach eastward along the north coast to the same extent, Ptilota plumosa and Corallina officinalis have been found furthest east in the eastern part of the north coast; Lithophyllum Crouani and Myrionema vulgare in Eyjafjor5ur; Ahnfeltia and Peirocelis extend to 58 H. JONSSOX SkagafjorSur ; while Cystoclonium, Polysiphonia nigrescens and Leathesia do not extend further than the most western part of the north coast (the small fjords in Hunafloi). Only Delesseria sanguined now remains; strictly speaking, this appears to belong to the south and south-west coast, but has on one occasion been found in E. Iceland, cast up on the shore. It did not appear to have come from a distance, and it probably grows there, though I did not come across it in the dredgings. Thus all these 53 species have on the whole a southern and western distribution in Iceland. The above shows that there is a large neutral territory where the species with a south-western and those with a north-eastern distribution meet and intermingle. This boundary area comprises almost the whole of the north coast, the north-west fjords and, to a certain extent, the northern part of the south-west of the country. (For further details see below under the Floristic Boundaries.) THE COMPONENTS OF THE ALGAL FLORA. At present 200 species of Marine Algae are known from Iceland. In the plant-geographical groups established by Borgesen and my- self (Borgesen and Jonsson, 14) these are distributed in the following manner. The definition of the groups is here reproduced almost literally from the publication mentioned. A. The Arctic Group. The species of this group belong to the arctic area of the sea. The southern limit of this area extends from the north and east of Norway southward to the south-east point of Iceland, where the boundary is sharply defined. From E. Iceland the boundary line extends to the north of Iceland between Iceland and Greenland, and then turns considerably southward to the North Atlantic coast of America. The flora of the boreal area of the Atlantic passes without any distinct limit into the arctic algal flora on both sides of the Atlantic. In Iceland the limit is distinct only at the south- east point whereas the boundary is very indistinct on the north- east part of the coast. Some of the species of this group occur, but only rarely, south of the border-zone. MARINE ALGAL VEGETATION 59 Rh o do p hy ce ae. Ceratocolax Hartzii. Cruoria arctica. Turnerella Pennyi. Lithothamnion flavescens. Delesseria Baerii corymbosa. L. foecundum. Polysiphonia arctica. Phseophyceae. Omphalophyllum ulvaceum. Laminaria nigripes. Dictyosiphon corymbosus. C h 1 o r o p h }r c e 36. Monostroma groenlandicum. B. The Subarctic Group. Subdivision I. The species of this subdivision are common in the Arctic Sea, and are rather common in the cold-boreal area of the Atlantic Ocean as far south as the Faeroes and Nordland; some of them occur, although rarely, as far south as England. Rhodophycese. Rhodophyllis dichotoma. Peyssonellia Rosenvingii. Halosaccion ramentaceum. Lithothamnion tophiforme. Ptilota pectinata. L. laeve. Rhodochorton penicilliforme. Clathromorphum compactum. Phaeophyceae. Lithoderma fatiscens. Chaetopteris plumosa. Ralfsia ovata. Coilodesme bulligera. R. deusta. Saccorrhiza dermatodea. Myrioncma globosum. Laminaria fseroensis. M. Laminarias. L. digitata. Streblonema secidioides. Alaria Pylaii. Sphacelaria britannica. Chlorophyceae. Chlorochytrium Schmitzii. U. pseudoflacca. Monostroma undulatum. Acrochaete parasitica. M. fuscum. Urospora Hartzii. Ulothrix consociata v. islandica. Acrosiphonia hystrix. U. subflaccida. A. penicilliforme. Subdivision II. This subdivision includes species, which are either common in the Arctic Sea and the North Atlantic from western France- England northward, or which, if not common, are at least all equally frequent. 60 H. JOXSSOX R h o d o p h y c e as. Porphyra miniata. Conchocelis rosea. Chantransia virgatula. Phyllophora Brodisei * interrupta. Actinococcus subcutaneus. Euthora cristata. Rhodymenia palmata. Delesseria sinuosa. Rhodomela lycopodioides. Odonthalia dentata. Ptilota plumosa. Lithothamnion glaciale. P h se o p h }T c e ae. Ralfsia clavata. Ectocarpus tomentosoides. Leptonema fasciculatum v. subcylin- drica. Elachista fucicola. Punctaria plantaginea. Litosiphon filiformis. Isthmoplea sphserophora. Stictyosiphon tortilis. Phaeostroma pustulosum. Dictyosiphon hippuroides. D. foeniculaceus. Desmarestia viridis. D. aculeata. Chordaria flagelliformis. Chorda tomentosa. C. filum. Fucus inflatus. Chlor Chlorochytrium Cohnii. C. inclusum. C. dermatocolax. Codiolum gregarium. Percursaria percursa. Monostroma Grevillei. Ulothrix flacca. Pseudendoclonium submarinum. Bolbocoleon piliferum. Ulvella fucicola. o p h y c e se. Pringsheimia scutata. Urospora mirabilis. U. Wormskioldii. Choetomorpha Melagonium. Spongomorpha vernalis. Acrosiphonia incurva. Cladophora rupestris. C. hirta. C. sericea. Ostreobium Queketti. Pleurocapsa amethystea. Cyanophycere. C. The Boreal-Arctic Group. The species of this group are common in the Arctic Sea and the boreal area of the Atlantic at least as far south as the Atlantic coast of North Africa; probably some of them have a far greater southern distribution. Some of them might possibly be considered cosmopolitan. Bangia fuscopurpurea. Porphyra umbilicalis. Chantransia microscopica. C. secundata. Ahnfeltia plicata. Rhodophyceae. Antihamnion Plumula v. boreale. Cera mi um rubrum. Rhodochorton Rothii. R. membranaceum. Hildenbrandia rosea. MARINE ALGAL VEGETATION 01 Pylaiella littoralis. Ectocarpus confervoides. E. siliculosus. Scytosiphon Lomentaria. Enteromorpha intestinalis. E. clathrata. Ulva lactuca. Chaetomorpha tortuosa. Spirulina subsalsa. Calothrix scopulorum. P h se o p h y c e ae. Phyllitis fascia. Fucus vesiculosus. Ascophyllum nodosum. Chlorophy ceae. Rhizoclonium riparium. Cladophora gracilis. Gomontia polyrrhiza. C y a n o p h y c e ae. Rivularia atra. D. The Cold-Boreal Group. The species of this group have their area of distribution from western France — England northward to S. Iceland, the Faeroes and Nordland— Finmark. Some few species have occasionally been found in the Arctic Sea, especially in the White Sea and the Murman Sea, and some few reach as far south as the Mediterranean and North Africa. Rhodophyceae. Porphyropsis coccinea. Chantransia Alariae. Gigartina mamillosa. Phyllophora membra nifolia. Cystoclonium purpurascens. Lomentaria rosea. Delesseria alata. D. sanguinea. Polysiphonia urceolata. P. fastigiata. P. nigrescens. Callithamnion Arbuscula. Plumaria elegans. Antithamnion floccosum. Ceramium acanthonotum. C. Deslongchampii. Petroderma maculiforme. Ralfsia verrucosa. Myrionema vulgare. M. Corunnae. M. faeroense. Ceramium fruticulosum. C. circinnatum. C. altanticum. Rhodochorton repens. R. minutum. Dumontia filiformis. Dilsea edulis. Petrocelis Hennedyi. Rhododermis parasitica. Lithothamnion Ungeri. L. Lenormandi. Phymatolithon polymorphum. Lithophyllum Crouani. Dermatolithon macrocarpum. Corallina officinalis. P h se o p h y c e ae. Ascocvclus islandicus. %, Microsyphar Polysiphonia?. Streblonema Stilophorae v. caespitosa. Ectocarpus tomentosus. E. penicillatus. 62 H. JOXSSOX Ectocarpus fasciculatus. Castagnea virescens. E. Hinksiae. Laminaria saccharina. Sphacelaria radicans. L. hyperborea. S. olivacea. Fucus spiralis. Phyllitis zosterifolia. F. serratus. Dictyosiphon Ekmani. Pelvetia canaliculata. D. Mesogloia. Alaria esculenta. D. Chordaria. Chlorophyceae. Codiolum Petrocelidis. Entoderma \Yittrockii. C. pusillum. Acrochaete repens. Enteromorpha aureola. Ochlochaete ferox. Prasiola polyrrhiza. Acrosiphonia albescens. P. furfuracea A. flabelliformis. P. stipitata. Cladophora glaucescens. Cyanophy ceae. Plectonema norvegicum. E. The Warm-Boreal Group. The majority of the species referred to this group extend at least as far south as the Mediterranean and the Atlantic coast of North Africa. According to the different distribution northward the group is divided into three parts of which only the one reaches as far north as S. Iceland. 1. Species extending as far north as S. Iceland, the Faroes and Northern Norway, and at least as far south as the Mediterranean «/ > and North Africa. Rhodophyceae. Chondrus crispus. Callithamnion scopulorum. Lomentaria clavellosa. Ceramium arborescens. Plocamium coccineum. Cruoria pellita. Bonnemaisonia asparagoides. Pterosiphonia parasitica. Phseophyceae. Desmarestia ligulata. Leathesia difformis. Chlorophyceae. Enteromorpha Linza. C y a n o p h y c e ae. Phormidium autumnale. According to the above the number of species in the groups is as follows: — MARINE ALGAL VEGETATION 63 Rhodo- Phaeo- Chloro- Cyano- phycese phycese phyce:e phyceie 1. The arctic group 7 3 1 » = 11 species (5 .5%) "j 1 ^ / 2. The subarctic group: Subdivision I 8 13 10 5> ^1 31 (15 5%) \ * ** .U / \J ) 3. The subarctic group: Subdivision II 12 17 20 1 = 50 — (25 0%") \t~is • vy / \> J 4. The boreal-arctic group 10 7 7 0 27 (13 .5%) 5. The cold-boreal group . 31 25 12 1 69 34 .5%) 6. The warm-boreal group 8 2 1 1 12 (6 .0%) 76 67 51 6 If we divide the six groups into two parts, A: the first three groups, and B: the last three1 groups, we obtain the following figures : A, 92 species ' (46 %) and B, 108 species (54 °/o). The floral district must therefore be determined as boreal, be- cause more than half of the species belong to the last three groups. Of these groups the cold-boreal is the most important because its species form 64% of the total number of species (108) in all three groups. This floral district has not, however, a purely boreal cha- racter, as the subarctic group is rich in species and gives a rather high percentage (41 %). The floral district, then, is characterized to a very high degree by a boreal element, and next by a subarctic element. If we consider only the red and the brown alga?, 143 species in all, the cold-boreal character is a little more strongly pronounced than the subarctic. The figures are : Arctic 10 species (7 %), sub- arctic 50 species (35%), boreal-arctic 17 species (12%), cold-boreal 56 species (39 %) and warm-boreal 10 species (7 %). The first three groups have 60 species (42 %), the last three 78 species (58 %). If we compare the five divisions of the coast with respect to the number of species in the different groups, we obtain the figures given in Tables 3, 4. If, for instance, we select the red and the brown algae (Table 4) as a basis, then the difference which exists in the different parts of the coast is very evident. In E. Iceland the arctic group contains the greatest number of species, and this number if we follow the divisions of the coast in the order of the tables decreases 1 The boreal-arctic group is included in the boreal groups, as its species, though common in the arctic district, have a far larger area of distribution out- side this. 64 H. JONSSON Table 3. Red algae, Brown algae, Green algae, Blue-green algae collectively. E. Icel. N. Icel. NW. II Icel. SW. Icel. S. Icel. Number of species ° 1 «! = o $ % 3 a Z. «> tfi •§«-! § ° £ % Z "* Number of species * Number of species % Arctic group 8 25 38 20 20 1 7 22 34 18 18 C.I K 24 37 26 29 9 4 20 30 21 24 C.I Q 16 39 16 18 2 3 17 41 17 20 c. 2 2 27 44 23 53 6 18 28 15 34 4 12 28 16 43 9 11 26 15 39 9 Subarctic group I Subarctic group II Boreal arctic group .... Cold-boreal group Warm-boreal group .... Total . . . 112 123 94 155 108 Table 4. Red and Brown algae collectively. E. 1 eel. N. ] eel. NW. Icel. SW. Icel. S. Icel Number of species % Number of species % Number of species % Number of 1 species % | 3; |*1 °/o Z * Arctic srouu.. 7 9 4 5 2 3 2 2 » » Arctic group 7 9 ^ 5 2, 3 2 2 » » Subarctic group 1 18 23 19 22 13 18 20 17 8 10 Subarctic group II 25 31 24 28 27 37 26 23 17 20 Boreal arctic group .... Gold-boreal group 13 17 16 21 17 21 19 24 13 16 18 22 16 46 14 40 13 15 38 45 Warm-boreal group .... » > 2 2 2 2 5 4 , 8 10 i Total . 80 87 73 115 84 uniformly in the other parts of the coast, and is reduced to 0 in S. Iceland. In E., N., NW. and SW. Iceland the number of the species in the subarctic group I is practically identical in propor- tion to the number of species, taken as a whole, in these parts of the coast; S. Iceland has a distinctly smaller number, only 10 °/o. The subarctic group II is represented most abundantly in NW. Ice- land; E. and N. Iceland come next; but S. and SW. Iceland have a considerably lower percentage (about 20 %). The percentage in the boreal-arctic group is practically identical in all parts of the coast. The cold-boreal group presents almost the same percentage in E., N. and NW. Iceland, while the percentage in the group in MARINE ALGAL VEGETATION 65 SW. and S. Iceland is almost double this. The warm-boreal group is not represented in E. Iceland, and only with extreme rarity (1 species) in N. and NW. Iceland. SW. Iceland has only 4 species, while S. Iceland has 8 (10 %). There is a very great similarity between S. Iceland and SW. Iceland, if the arctic group in SW. Iceland is excluded; on the other hand, the difference is greatest between E. Iceland and S. Iceland, as is shown by the following figures: East Iceland Arctic group 9 °/o Subarctic groups 54 % Boreal-arctic group 16 % Cold-boreal group 21 % Warm-boreal group 0 °/o South Iceland 0% 30% 15% 45% 10% If we assume that the species marked (c) (Table 1) are to be considered as common to all parts of the coast (see above), the distribution of the 76 not-common species becomes decisive with reference to the floristic difference between the parts of the coast. In the following table, therefore, it is shown how these 76 species are arranged in the six plant-geographical groups of alga3. Table 5. Group-division of the 76 not-common species (see Table 2). A E. Icel. N. Icel. NW.Icel. SW.Icel. S. Icel. B . . . . • • . • Arctic group. Arctic group 8 5 3 2 • • . . • • . . Subarctic group I. Subarctic group I .... 5 4 3 3 • • • • 1 2 2 2 Subarctic group II. Subarctic group II.... 2 . . 2 . . • • . . 1 1 1 1 Boreal-arctic group. Boreal-arctic group . . . . . . • 1 6 8 31 28 Cold-boreal group. Cold-boreal group .... . . 4 . . . . • * . . 2 2 6 9 Warm-boreal group. Warm-boreal group.. . . • • • . . • • • • 1 10 13 40 40 Total number of species Total number of species 15 9-f 4 8 5 16 23 21 45 40 The figures in this table show what has been already shown by those which I have given in Tables 3 and 4; but the arctic The Botany of Iceland. I. O 66 H. JONSSON element in E. Iceland and the boreal element in S. and SW. Iceland are much more sharply defined; and this is natural, as the species assumed to be common are omitted. Even if we consider only the known distribution of the species, in its entirety (see Table 1), the distribution of the here-mentioned 76 species will still be the most essential reason for the floristic difference between the parts of the coast. The species assumed to be common are 124 (see above). Of these none are arctic, 25 belong to the subarctic group I, 46 to the subarctic group II, 26 are boreal-arctic, 26 cold-boreal, and one (Phormidium antamale) is warm-boreal. Of the 76 not -common species, 11 are arctic, 6 belong to the subarctic group I, 4 to the subarctic group II, one (Ahnfeltia plicata) is boreal-arctic, 43 are cold-boreal and 11 warm-boreal. If we add together the numbers representing the species of the corresponding groups as regards the 124 species assumed to be common and the 76 not-common species (Table 5), and compare with Table 3, we find that the species are more numerous in each district, but that the percentages are almost the same. If the groups are divided in two parts, A and B (see above) so that A includes the arctic and subarctic groups, and B the three other groups, the following figures are obtained (see Table 4) : E. Icel N. Icel. NW. Icel. SW. Icel. S. Icel. A 50(63%) 47(54%) 42(58%) 48(42%) 25(30%) B 30(37%) 40(46%) 31(42%) 67(58%) 59(70%) The arctic group is poorly represented in all the districts of the coast (see Table 4) and therefore the figures mentioned above under A apply chiefly to the subarctic group; the floral districts of E. Iceland, N. Iceland and NW. Iceland are thus subarctic. E. Ice- land is subarctic to a greater extent than N. and NW. Iceland. SW. Iceland is a boreal floral district with a very considerable subarctic element, and thus resembles the coasts of Iceland taken as a whole (see above). S. Iceland is a boreal district with a slightly subarctic element. Floristic Boundaries. The mixed character of the flora in N. and NW. Iceland has been alluded to several times in the foregoing pages, and is clearly seen from the tables given, as, for example, Tables 2 and 5. Here, MAKINE ALGAL VEGETATION <)7 neither, is anv distinct boundary found between the boreal and «. V subarctic floral districts, and the north and north-west of the country V must, strictly speaking, be considered a large boundary-area, a view which accords well, also, with the hydrographic conditions. Here, the boreal, subarctic and arctic species intermingle. At the south-eastern point of Iceland there is, on the other hand, a rather distinct hydrographic boundary along the stretch from Vestrahorn to Eystrahorn (or LonsheiSi). The greater part of this coast is sandy, and difficult of access for the investigation of the algal vegetation. From my own observations I can only say that Berufjor5ur, the most southerly point in E. Iceland which I have examined with regard to its algal vegetation, has a cold-water flora, and that the Vestmannaeyjar, the most easterly locality on the south coast which I have examined for the same purpose, have a warm-water flora. The boundary must lie between them, and I */ conclude, especially from the hydrographic conditions and the dis- tribution of the Plankton-associations, that it is situated just on the stretch of coast already mentioned. Ove Paulsen (55 and 56) has given valuable information respecting this boundary, and it is evident from his investigations that the boundary varies to a slight extent, the facts being that in May-June it has been found in the vicinity of Eystrahorn (see 55, map I), but in July-August at Vestra- horn (see 55, map II). If algae grow on this stretch of coast, one may conclude that there exists a mixed flora resembling that of N. and NW. Iceland. Whether boreal species can be carried to E. Iceland in this manner is at present not easy to say with certainty, yet it seems to me that the occurrence of Dumontia filiformis and Delesseria sanguined, both of which are absent in N. and NW. Ice- land, can be most easily explained in this way. IV. COMPARISON WITH NEIGHBOURING FLORAL DISTRICTS. IN Table 61 is given a survey of the plant-geographical distribu- tion of red and brown algae collectively, in certain subarctic and boreal floral districts. These are so arranged that those floras with the largest arctic element stand furthest to the left. The arctic and subarctic percentages decrease while the boreal percentage increases to the right. The boreal-arctic group is practically similar everywhere, which is also natural according to the geographical dis- tribution of the group. The warm-boreal group is not represented in the subarctic floras, and the arctic group is quite infinitesimal in SW. Iceland and Nordland, and is entirely absent from S. Iceland and the Faaroes. In regard to species, the cold-boreal group is ex- tremely poor in East Greenland and Spitzbergen, somewhat richer in West Greenland and considerably richer in E. Iceland. By grouping the species, as is done above (cf. Borgesen and Jonsson, 14), the character of the floral districts can be deter- mined according to those groups which are richest in species. Thus, I characterize a group as subarctic when more than half of its species are reckoned to the subarctic group. In a similar manner a district is boreal when more than half of its species belong to the boreal groups (bor. arct., cold-bor., and warm-bor.). Similarly, in an arctic district the species belonging to the arctic group must constitute more than half of the number of species be- longing to the district, and, in addition to the subarctic group, only the boreal-arctic w^ill be represented. Of the floral districts men- tioned by Borgesen and Jonsson (14) none are arctic according 1 With the exception of Iceland the numbers of the species of red and brown alg« are taken from Borgesen and Jonsson 1. c. (14, p. 22). In regard to East Greenland the numbers are corrected according to Rosen vinge (64), and to West Greenland two species have been added: Ectocarpus maritimns and Chantransia collopodd. H. JONSSON : MARINE ALGAL VEGETATION 69 to the definition here employed. The Siberian Sea, however, comes nearest to it. From here 23 species are known (14), of which 9 (39%) are arctic, 11 (48%) subarctic and 3 (13%) boreal- arctic. This district is at the boundary between arctic and sub- arctic. Regarded superficially it may appear strange that none of the districts are arctic, but on closer inspection this is easily under- stood, the reason being that some of the districts (14) are too large and consequently acquire a mixed character. In this respect I shall merely point out, for instance, that both Spitzbergen and East Greenland (and probably West Greenland) ought to be divided into two districts. As already mentioned, none of the districts recorded in Table 6 is arctic. East Greenland, Spitzbergen and West Greenland have almost the same percentage as regards the arctic species (Table 6), and as this percentage is rather high in proportion to that of the boreal species, these districts could be termed arctic-subarctic, in contra- distinction to E. Iceland where the arctic percentage is four times less than the percentage of the boreal species. The boreal districts recorded here (Table 6, p. 70) should, strictly speaking, be called cold-boreal. If we call the first three groups (in Table 6) A and the three last B the percentages will be as follows: E. Greenl. Spitzb. W.Greenl. • E. Icel Finm. SW. Icel. S. Icel. Peer. Nordl. A 81 77 72 63 46 42 30 29 27 B 19 23 28 37 54 58 70 71 73 As the table shows, SW. Iceland agrees most closely with Fin- mark, while S. Iceland and the Faeroes are nearly alike, as Borse- V sen (12, p. 804) also supposes. If we take Iceland as a whole, we get 143 species (red and brown algae collectively), 10 (7 %) arctic, 21 (15 %) subarctic (sub- division I), 29 (20 %) subarctic (subdivision II), 17 (12 %) boreal- arctic, 56 (39 %) cold-boreal and 10 (7 %) warm-boreal. These figures are almost the same as those given for Finmark (see Table 6) and differ, essentially from the figures given for SW. Iceland, only by the higher percentage of arctic and warm-boreal species. If, on the other hand, we take the first three groups collectively and the three last groups in a similar manner, we obtain the same percentages as for SW. Iceland. On combining different parts of the coast, as for instance, E. Iceland and N. Iceland, we get almost the 70 H. JONSSON r* O r, O O ~ c 2 C =T £* ~ n ~' ? a crci . re <. » I- • < C* c » * -G » "I — • rt O °S £ S' -* -• ifi 5 — - 1 I 3 ™' *" l™~ •c = ? - $* o — • =2 •§ o C/5 ss o CC o ft SJ ^ r> ta c/> co co > H 0 r* £L — f6 a SJ 2" r ? r 3 3 3 « — i w cr. 2: ir. cm rt 2 « <^ ^ 2 ^ w' m — 2". cr=. crci !^Q c & r> ~. ~, -< — — ~f **• f**, r*\ *" J-, 0 O w oc; 5 5£ S 2 £ •? r- X w w ^3 ^- w S ~ = « ^ "H t • 00 in >-» VT 1C tO •-» 4^. 1C 4^> 00 Ci C" Number of species C 3 £ i— ' Ci CC to >-i Cn 4^ W W O 00 ^0 2 Z B. <~>~ cs o co to i-» >-' w-i cc ci to *>• o Number of species Sf ^ 3 "S- i— ' Ci co to •— ' OC ^ O ^J W •-k C^ tO 1C i-1 1C Ci 4^ 00 Ci 00 e~" fp ft S V — r* 00 0 _l _l ^ IO H-l -j cc co en oo *J Number of species W NH tO i-1 m CO tO i-» ci ** t-i cc « c~ O ft • 1—4 to en • 4^ I-* tn tO IO Cn CT 00 O CO f* ~J Number of species 3 2 CO i-» *>. tO H-* 4!» Ci 4»- C CO ^3 CS ^s P i ?r* I— t VT 4^ i-i 4i- IO IO CT SS Ci C5 Ci O IO Number of species CA ^ 4-- •-* 4^. tO i-1 ^. O 4*- O CO ^1 tO ^ fT 2- OO *»- W i-* tO "-1 cc oo co ss\ ^i oo Number of species cn HH i— * i£k i— » CC IO i— ' O C^ i3 O O O Sr O — Cn •o &3 -J i_i ^ CO i-» • O Cn ^.1 Ci —t *»• Number of species •n R M i-k *». i-1 IO IO CO ^-1 i— ' ID O C5 o O: ft 75 H-t to CO i-1 CO >-* CO 1C i-* CC I-* Ci tO O tO >-» Number of species z; o "3 ^- !-» 4^- — » IO i-1 •-» — «O W Ci CS O "-1 c"~ ET ^ p 05 -. O 3 55* rt- C ft} rt- •*• S 3 § £ QTQ S 3 Si c O T3 09 3 O a •* DO 30 o 2 -! U. — 3 35 a S co £ o a ^ r. 3 CO rt- p O* W W W » <* *<" MARINE ALGAL VEGETATION 71 same figures as for E. Iceland, and on combining NW., SW. and S. Iceland we get almost the same figures as for SW. Iceland. E. Iceland and N. Iceland have 101 species of red and brown algas collectively. Of these 8 (8 %) are arctic, 20 (20 %) subarctic (sub- division I), 28 (27 %>) subarctic (subdivision II), 17 (17 ° o) boreal- arctic, 26 (26 %) cold-boreal and 2 (2 %>) warm-boreal. NW., SW. and S. Iceland have 131 species of red and brown algae collectively. Of these 3 (2%) are arctic, 20 (15%) subarctic (subdivision I), 29 (22 %) subarctic (subdivision II), 16 (12 °/o) boreal-arctic, 53 (41 °/o) cold-boreal and 10 (8 %) warm-boreal. As regards the components of the flora, both Iceland taken as a whole, and SW. Iceland resemble Finmark; S. Iceland resembles the Faeroes and Nordland, and E. Iceland resembles the White Sea.1 It is evidently not due to chance that the resemblance of the «/ floral districts happens thus. The situation of Iceland just south of and at the boundary between the arctic and the cold-boreal districts corresponds exactly with the situation of Nordland — Finmark- -White Sea in relation to this boundary. Iceland and the White Sea are at the boundary itself, and in Finmark it certainly will be possible to distinguish parts of the coast writh a similar mixed flora as in N. and NW. Iceland; the southern part of Finmark will then be something like SW. Iceland, while S. Iceland, as already mentioned, corresponds with Nordland. This comparison shows only the relation between the quantity of the species of the floral districts within the different groups, but gives no information as to how far the species are common to all those districts. Then it remains to be investigated how many species Iceland has in common with the other districts. At the present time a comparison of the floras will, however, scarcely give any satisfactory results, because all the districts in question are not equally well-known. By future investigations a greater number of species will unquestionably be found in the majority of the floral districts, and the quantity of the species will thus be altered, but the relation between the number of the species of the different groups will, howrever, undoubtedly remain unaltered. 1 According to Borgesen and Jonsson (14) 52 species are known from the White Sea, of which 3 '6 %>) are arctic, 10 19 %>) subarctic (subdivision I), 19 (37°/o) subarctic (subdivision II), 8 (15 %) boreal-arctic and 12 (23%) cold-boreal. The subarctic species constitute 56 % of the entire number of species and the character of the flora is consequently subarctic in the same degree as that of E Iceland. 72 H. JONSSON If we compare S. Iceland and the Faeroes with regard to common species, the following figures are obtained: South Iceland 84 species1 The Faeroes 157 species not common common not common 10 (6%) 74 (44%) 83 (50%) Here it should be noted that S. Iceland is so very little known »/ that one is scarcely justified in comparing it with such a well in- vestigated district as the Faeroes. Many of the species which in this respect are peculiar to the Faeroes will certainly be found in S. Iceland and, at any rate, 32 of them are known from other parts of Iceland, principally from SW. Iceland. If we choose a larger district of the coast of Iceland, for in- stance, the boreal district (S. Iceland and SW. Iceland) for compa- rison with the Faeroes, the following figures will be obtained: S. and SW. Iceland 126 species The Faeroes 157 species not common common not common 25 (14%) 101 (55%) 56 (31%) On comparing Iceland with the Faeroes we obtain the following figures : Iceland 143 species The Faeroes 157 species not common common not common 37 (19%) 106 (55%) 51 (26%) The 37 species which grow in Iceland and are absent from the Faeroes are the following: fChantransia microscopica. fPetrocelis Hennedyi. Ceratocolax Hartzii. Criteria arctica. Turnerella Pennyi. fPeyssonellia Rosenvingii. Delesseria Baerii. fRhododermis parasitica. Bonnemaisonia asparagoides. 2 Lithothamnion flavescens. Polysiphonia arctica. L. foecundum. Ceramium Deslongchampii. fL. tophiforme. C. fruticulosum. fL. Ungeri. C. circinnatum. fRalfsia ovata. C. arborescens. R. deusta. fRhodochorton minutum. fMyrionema Laminaria? fR. repens. fAscocyclus islandicus. fDilsea edulis. fEctocarpus penicillatus. These and the following figures apply to red and brown algae collectively. I attach no importance to the fact of this species having occurred in Ice- land, as it has not yet been found again. MARINE ALGAL VEGETATION 73 fSphacelaria radicans. fS. olivacea. Omphalophyllum ulvaceum. fPhaeostroma pustulosum. Coilodesme bulligera. fDictyosiphon Mesogloia. fDiclyosiphon Chordaria. D. corymbosus. Saccorrhiza derinatodea. Laminaria nigripes. fFucus serratus. Which of these species may be found in the Faeroes is not easy to decide (cf. also Borgesen, 12, p. 795). But it does not appear altogether improbable that 19 species (marked with a f before the name) could occur there. The 51 Faeroese species which are not found in Iceland are the following: fErythrotrichia ceramicola. fPorphyra leucosticta. fChantransia efilorescens. fC. Daviesii. Choreocolax Polysiphoniae. fHarveyella mirabilis. Callophyllis laciniata. Callocolax neglectus. Sterrocolax decipiens. fLomentaria articulata. Nitophyllum laceratum. Laurencia pinnatifida. Polysiphonia violacea. fP. Brodiaei. fP. elongata. P. atrorubescens. Rhodomela subfusca. Griffithsia setacea. tCallithamnion polyspermum. fC. corymbosum. C. granulatum. fCeramium Boergesenii. Rhodochorton seiriolanum. fFurcellaria fastigiata. fPolyides rotundas. fRhododermis elegans. Phymatolithon laevigatuni. Lithophyllum incrustans. L. hapalidioides. fSorapion Kjellmani. fMyrionema foecundum. fM. speciosum. fChilionema reptans. Microsyphar Zosterae. Ectocarpus velutinus. fE. lucifugus. E. dasycarpus. E. granulosus. Elachista scutulata. Sphacelaria csespitula. S. furcigera. fS. cirrhosa. Cladostephus spongiosus. Desmotrichum undulatum. Punctaria latifolia. fAsperococcus echinatus. fLitosiphon Laminariae. fPhaeostroma parasiticum. fHimanthalia lorea. fHalidrys siliquosa. tCruoriella Dubyi. Possibly many of these species, perhaps almost half of them, are to be found in S. and SW. Iceland. The 26 species marked with a f before the name may possibly be found in Iceland, though with manv of them this is doubtful. \j If it should appear from further investigations that some of the species considered not common to the Faeroes and Iceland are com- mon to them, it should not be assumed from this that the floristic 74 H. JONSSOX resemblance is increased, as it is highly probable that other species which are not common would be simultaneously found. From the coast of Norway I select Finmark for comparison with Iceland, which comparison gives the following figures: Iceland 143 species Finmark 125 species not common common not common 45 (26%) 98 (58°/o) 27 (16%) A comparison between Finmark and SW. Iceland gives the following figures : SW. Iceland 115 species Finmark 125 species not common common not common 29 (19 %) 86 (56 %) 39 (25 %) The following are the 45 species which are found in Iceland and are absent from Finmark: — 1 Porphyropsis coccinea. 2 Chantransia Alarise. 3 Phyllophora membranifolia. 4 Ceratocolax Hartzii. 5 Lomentaria rosea. 6 L. clavellosa. 7 Plocamium coccineum. 8 DeJesseria Baerii. 9 Bonnemaisonia asparagoides. 10 Pterosiphonia parasitica. 11 Callithamnion Arbuscula. 12 Ceramium acanthonotum. 13 C. fruticulosum. 14 C. arborescens. 15 C. atlanticum. 16 Rhodochorton minutum. 17 R. repens. 18 Dilsea edulis. 19 Petrocelis Hennedyi. 20 Cruoria arctica. 21 C. pellita. 22 Rhododermis parasitica. 23 Petroderma maculiforme. 24 Ralfsia ovata. 25 R. clavata. 26 R. verrucosa. 27 Myrionema Lamimariae. 28 M. Corunnae. 29 M. globosum. 30 M. faeroense. 31 Ascocyclus islandicus. 32 Microsjrphar Polysiphoniae. 33 Ectocarpus Stilophorse. 34 E. tomentosoides. 35 E. tomentosus. 36 E. Hinksiae. 37 Sphacelaria radicans. 38 S. olivacea. 39 Ornphalophyllum ulvaceum. 40 Litosiphon filiformis. 41 Phseostroma pustulosum. 42 Phyllitis zosterifolia. 43 Dictyosiphon Mesogloia. 44 Desmarestia ligulata. 45 Laminaria faeroensis. Of these species ten (Nos. 3, 6, 7, 10, 11, 12, 18, 21, 26 and 35) are known from Nordland and therefore might possibly be found also in Finmark. Furthermore, nine of the species (Nos. 2, 5, 9, 25, 33, 34, 35, 40, 44) are known from West Norway, and the majority of these will also be found, without doubt, in Nordland and Fin- MARINE ALGAL VEGETATION 75 mark. Of the remaining 23 species the greater number will cer- tainly be found in Finmark. But 1 think it less probable that Om- phalophyllum aluaceum and Laminaria fceroensis are to be met with there, while one can form no opinion as to whether the endemic Icelandic species (17 and 31) exist in Finmark, as their distribution outside Iceland is unknown. On the other hand I think it somewhat probable that the Icelandic- Fa3roese species Myrionema fceroense may occur in Finmark. The 27 species, which are present in Finmark and absent from Iceland, are the following: Chantransia eftlorescens. Ectocarpus terminalis. C. Daviesii. E. nanus. Harvevella mirabilis. E. ovatus. V Polysiphonia elongata. Myriotrichia filiformis. Spermothamnion Turneri. Sphacelaria racemosa. Furcellaria fastigiata. S. cirrhosa. Polyides rotundas. Phseosaccion Collinsii. Petrocelis Middendorfii. Asperococcus echinatus. Phymatolithon investiens. Delamarea attenuata. Lithothamnion intermedium. Dictyosiphon hispidus. L. fornicatum. Laminaria Agardhii. Lithoderma lignicola. Haplospora globosa. Myrionema foecundum. Halidrys siliquosa. Chilionema reptans. The majority of these species may be expected to occur on the coasts of Iceland, but there is less probability of Laminaria Agardhii being met with there. If we turn to E. Iceland and compare this with the White Sea we obtain the following figures : E. Iceland 80 species White Sea 52 species not common common not common 43 (45°/o) 37 (39%) 15 (16%) The floristic similarity is somewhat considerable, and ten of the species peculiar to the White Sea are known from other parts of the coast of Iceland. These ten species are Dilsea edulis, Cysto- clonium purparascens, Polysiphonia nigrescens, Delesseria a/a/a, Ptilota plumosa, Lithothamnion Lenormandi, Corallina officinalis, Dictyo- siphon hippuroides, Fucus serratus, and Pelvetia canaliculata. Of these species, however, only an extremely small number can be expected to occur in E. Iceland. The five species present in the White Sea and entirely absent from Iceland are: — 76 Ghantransia efflorescens. Furcellaria fastigiata. Polyides rotundus. H. JONSSOX Sphacelaria racemosa. Laminaria Agardhii. With the exception of Laminaria Agardhii all these species certainly may be met with on the coasts of Iceland, hut in E. Ice- land one can scarcely expect to find others than Chantransia efflo- rescens and Sphacelaria racemosa. If we compare Iceland and E. Iceland with West Greenland and East Greenland we get the following figures: — Iceland 143 species not common 62 (36%>) Iceland 143 species not common 79 (48%) E. Iceland 80 species not common 11 (9%) E. Iceland 80 species not common 23 (21 %) common 81 (48%) common 64 (39%) common 69 (58%) West Greenland 108 species not common 27 (16%) East Greenland 85 species not common 21 (13%) West Greenland 108 species not common 39 (33%) East Greenland 85 species not common 28 (26 %) common 57 (53%) The species present in Iceland and absent from West Greenland are the following: Porphyropsis coccinea. Chantransia Alariae. Phyllophora membranifolia. Ahnfeltia plicata. Chondrus crispus. Gigartina mamillosa. Cystoclonium purpurascens. Lomentaria rosea. L. clavellosa. Plocamium coccineum. Delesseria alata. D. sangvinea. Bonnemaisonia asparagoides. Polysiphonia fastigiata. P. nigrescens. Pterosiphonia parasitica. Odonthalia dentata. Callithamnion scopulorum. C. Arbuscula. Plumaria elegans. Ptilota plumosa. Ceramium acanthonotum. C. Deslongchampii. C. fruticulosum. C. circinnatum. C. arborescens. C. atlanticum. Rhodochorton minutum. R. repens. Dumontia filiformis. Dilsea edulis. Petrocelis Hennedyi. Cruoria pellita. Rhododermis parasitica. Phymatolithon polymorphum. Lithothamnion Lenormandi. L. llavescens. L. Ungeri. Lithophyllum Crouani. Dermatolithon macrocarpum. MARIXE ALGAL VEGETATION 77 Corallina officinalis. Dictyosiphon Ekmani. Petroderma maculiforme. D. Mesogloia. Myrionema Laminariae (in E.Greenl.) Desmarestia ligulata. M. vulgare. Leathesia diffbrmis. M. Corunnse. Laminaria saccharina (in E. Greenl.) M. fasroense. L. faeroensis. Ascocyclus islandicus. L. hyperborea. Ectocarpus tomentosus. Alaria esculenta in E. Greenl.) E. fasciculatus. Fucus spiralis. E. Hinksise. F. serratus. Sphacelaria olivacea. Pelvetia canaliculata. Of the 57 species here mentioned three are found in East Green- land; viz. Myrionema Laminarice, Laminaria saccharina and Alaria esculenta. It is most probable that none of these 57 species will be met with in West Greenland, with the exception, perhaps, of some of the M[//'zo/?ema-species. There are 74 species absent from East Greenland which are found in Iceland. Of these, 54 species have already been mentioned (see above), and to these must be added 20 species which are absent from East Greenland but present in West Greenland and Iceland. These species are the following: Bangia fuscopurpurea. Ectocarpus Stilophorae. Porphyra umbilicalis. E. tomentosoides. Chantransia secundata. E. penicillatus. C. virgatula. Sphacelaria radicans. Polysiphonia urceolata. Litosiphon filiformis. Antithamnion floccosum. Phyllitis zosterifolia. Ceramium rubrum. Dictyosiphon Chordaria. Ralfsia ovata. D. con^mbosus. R. verrucosa. D. hippuroides. Microsyphar Polysiphoniae. Castagnea virescens. It seems most probable that these 20 species are to be met with in East Greenland. The 11 species present in E. Iceland and absent from West Greenland are included in the above-mentioned figure. They are: Gigartina mamillosa. Laminaria saccharina (in E. Greenl.) Delesseria sangvinea. L. fseroensis. Odonthalia dentata. L. hyperborea. Dumontia filiformis. Alaria esculenta (in E. Greenl.) Lithothamnion flavescens. Fucus spiralis. L. Ungeri. The 23 species present in E. Iceland and absent from East Green- land have also been alreadv recorded. With the addition of the 9 «/ of the above-mentioned species they are the following: 78 H. JOXSSOX Bangia fuscopurpurea. Porphyra umbilicalis. Chantransia secundata. Polysiphonia nrceolata. Antithamnion floccosum. Ralfsia verrucosa. Ectocarpus Stilophorae. Ectocarpus tomentosoides. E. penicillatus. Sphacelaria radicans. Litosiphon filiformis. Phyllitis zosterifolia. Dictyosiphon Chordaria. Castagnea virescens. These species might possibly also be found in East Greenland, as they are already known from West Greenland (see above), and in that case there could be only 9 E. Iceland-species which were not known from East Greenland. In West Greenland there are 27 species which are not found in Iceland. These are: — 0 Harveyella mirabilis. Callymenia sangvinea. Delesseria Montagnei. Polysiphcnia elongata. Ceramium Areschoughii. C. septentrionale. 0 Rhododermis elegans. Lithothamnion intermedium. Chantransia collopoda. 0 Sorapion Kjellmani. Ectocarpus Pringsheimii. 0 E. ovatus. 0 E. pycnocarpus. 0 E. maritimus. 0 Sphacelaria racemosa. Phseosaccion Collinsii. 0 Symphyocarpus strangulans. Kjellmania subcontinua. Coelocladia arctica. 0 Delamarea attenuata. Dictyosiphon hispidus. Myriocladia callitricha. 0 Laminaria solidungula. 0 L. longicruris. L. cuneifolia. 0 L. groenlandica. Agarum Turner!. Of these species 13 (with 0 prefixed) are found in East Greenland. In East Greenland 21 species are found which are not known from Iceland. In addition to the 13 above-mentioned species, they are the following: Chantransia efflorescens. Dilsea intcgra. Petrocelis polygyna. Cruoriopsis hyperborea. Ectocarpus helophorus. Haplospora globosa. Punctaria glacialis. Myrionema foecundum. The greater number of the West Greenland and East Greenland species here mentioned wTill probably be met with in Iceland, espe- cially on the north and east coasts. It is less likely, however, that the following species will be found in Iceland: - -Callymenia sanguinea, Delesseria Montagnei, Dilsea integra, Petrocelis polygyna, Laminaria solidungula, L. cuneifolia, L. groenlandica and Agarum Turneri. According to the above comparisons Iceland most nearly resembles Finmark, and next to that place the Faeroes: the resemblance to West MARINE ALGAL VEGETATION 79 Greenland is rather less, and to East Greenland is least of all, as the following figures show: common not common Iceland— Finmark 58 % 42 °/o Iceland— the Faroes 55 °/o 45 ° o Iceland-West Greenland 48 °/o 52 % Iceland— East Greenland 39 ° o 61 % S. Iceland is too little known to be compared with other districts, as has already been emphasized above. The resemblance to the Faeroes will certainly prove to be much greater than is shown by the figures now known. common not common S. Iceland— the Faeroes 44 °/o 56 %> S. and S\V. Iceland — the Faeroes . 55 °/o 45 % S. Iceland and SW. Iceland together resemble the Faeroes to the same degree as do the coasts of Iceland taken as a whole. The fact that S. Iceland least resembles the Faeroes is merely due to a de- ficient knowledge of its coasts. As regards SW. Iceland — Finmark the percentage of species common to both places is 56 and that of not common 44. Thus, the resemblance is somewhat less than that between Iceland and Finmark. If wre now turn to E. Iceland we obtain the following figures : common not common E. Iceland— West Greenland 58 °/o 42 °/o E. Iceland— East Greenland 53 % 47 % E. Iceland— White Sea 39 °/o 61 % E. Iceland has thus the greatest floristic resemblance to West Greenland, resembles East Greenland somewhat less and the White Sea least of all, although the floral districts of E. Iceland and the White Sea resemble each other most closely when the species are grouped according to geographical distribution (see above). V. THE VERTICAL DISTRIBUTION OF THE SPECIES. IF we walk along the beach at low-tide we see a belt laid bare, the breadth of which varies according to the degrees of declivity of the coast and according to the tides that is according to whether it is spring-tide or neap-tide. At spring-tide the belt is broad and at neap-tide narrow. The low-water mark of neap-tide divides the belt laid bare into two parts, an upper part which is laid bare during every low-tide and a lower part which is laid bare only at and about spring-tide. The upper part, between the upper limit of growth of the algal vegetation and the low-water mark of neap-tide, which almost coincides with the lower edge of the Fucacece- belt, I call the Upper Littoral Zone. The lower part, from the lower edge of the Fucacece-beli to the usual1 low-water mark of spring-tide, I call the Lower Littoral Zone. Below low-water mark of spring-tide begins the vegetation which is constantly submerged. That the vegetation of the lower littoral zone is closely connected with that in the upper part of the constantly-submerged zone is natural and will be discussed subsequently. In the following table a dash ( — ) denotes a habitat (depth) in which the species has been found many times, and a dot (•) one in which it has been found either once or comparatively rarely. Table 7. The Vertical Distribution of the Species. Litto- • ral zone Depth metre) Q, 3 01 _o m 0 T CO I 0 co 1— t m I •=> co M 7 m 0 1 co M I § 1 co in o 1 o 0 CD CO m m CO ^ CO 0 m co oo 00 Rhodophyceae. 1 Bangia fuscopurpurea si Porohvra miniata . 1 No notice is taken of the extraordinarily low ebb-tides which occasionally occur and by which large Laminaricc are often exposed. H. JONSSON: MARINE ALGAL VEGETATION 81 Table 7. The Vertical Distribution of the Species (continued). Litto- ral zone Depth (metre) ^! «. %. | §• 1 u- 0 1 CO m T ?> CO 1-H tfi C* \ -* G* 0 03 1 ss w i« n -* so = -t- 1 co n i* 'ji •v 0 >o CO -r .-: 0 i o s 1 Jfl .- Ifl 50 o o t- 90 --- » CO 1 Porphyra umbilicalis • — • • — — • • • • • • • s Porphvropsis coccinea s Concliocelis rosea si Ghantransia microscopica • • • • s C. Alariae si C. secundata si C. virgatula si Chondrus crispus si Gigartina mamillosa si Ahnfeltia plicata s Phyllophora Brodisei v. interrupta. si P. membranifolia s Actinococcus subcutaneus s Ceratocolax Hartzii si Cystoclonium purpurascens s Turnerella Pennvi s Euthora cristata s Rhodophvllis dichotoma si Rhodvmenia palmata ^^^~ • s Lomentaria clavellosa s L. rosea s Plocamiurn coccineum si Halosaccion ramentaceum — — • • — • • s Delesseria alata D. Baerii * corvmbosa — • • • • — — — s D. sinuosa s D. sanguinea Bonnemaisonia asparagoides s Polysiphonia urceolata 1 P. fastigiata s P arctica si P. nigrescens — — • • • 6 * s Pterosipbonia parasitica si Rhodomela Ivcopodioides s Odouthalia dentata si Gallithamnion Arbuscula . . si C. scopuloruni si Plumaria elegans s Ptilota plumosa s P. pectinata s Antithamnion Plumula v. boreale. si A. floccosum . si Ceramium Deslongchampii 1 The Botany of Iceland. I. 82 H. JONSSON Table 7. The Vertical Distribution of the Species (continued). Litto ral zone Depth (metre) cc -i co IM I CO CO T T CO — u-3 CS o r- co CO oo oo sl Ceramium fruticulosum si C. circinnatum sl C. arborescens sl C. atlanticum sl C. rubrum sl C. acanthonotum 1 Rhodochorton Rotbii s R. repens sl R. minutum s R. penicilliforme s R. membranaceum sl Dumontia filiformis sl Dilsea edulis s Petrocelis Hennedyi s Cruoria arctica sl C. pellita s Peyssonellia Rosenvingii s Rhododermis parasitica s Lithothamnion glaciale s L. Ungeri s L. tophiforme , s L. flavescens s L. foecundum s L. laeve s L. Lenormandi , sl Phymatolithon polymorphum s Clathromorphum compactum . s Lithophyllum Crouani , s Dermatolithon macrocarpum sl Corallina officinalis 1 Hildenbrandia rosea , Phaeophyceae. s Lithoderma fatiscens 1 Petroderma maculiforme s Ralfsia ovata 1 R. clavata 1 R. verrucosa sl R. deusta sl Myrionema vulgare s M. Corunnae sl M. globosum sl M. fseroense., MARINE ALGAL VEGETATION 83 Table 7. The Vertical Distribution of the Species (continued). Litto- ral zone * Depth (metre) si Myrionema Laminarise si Ascocyclus islandicus si Microsyphar Polysiphoniie s Streblonema a>cidioides si S. Stilophone v. csespitosa 1 Pvlaiella litoralis M s Ectocarpus tomentosoides 1 E. tomentosus si E. confervoides si E. siliculosus si E. penicillatus si E. fasciculatus s E. Hinksiae si Leptonema fasciculatum v. subcy- lindrica 1 Elachista fucicola 1 Sphacelaria britannica si S. radicans s S. olivacea s Chietopteris plumosa s Omphalophyllum ulvaceum si Punctaria plantaginea si Litosiphon filiformis 1 Isthmoplea sphaerophora si Stictyosiphon tortilis si Phaeostroma pustulosum si Scytosiphon Lomentaria si Phyllitis zosterifolia si P. fascia si Coilodesme bulligera . . si Dictyosipbon Ekmani si D. Mesogloia si D. Cbordaria si D. corymbosus si D. hippuroides si D. foeniculaceus s Desmarestia viridis s D. aculeata s D. ligulata si Castagnea virescens si Leathesia difformis si Chordaria flagelliformis s Chorda tomentosa . c r- 00 oo .9 84 H. JONSSON Table 7. The Vertical Distribution of the Species (continued). Litto- ral zone Depth (metre) M to to 00 oo s Chorda Filum s Saccorrhiza dermatodea s Laminaria saccharina s L. fseroensis s L. nigripes s L. digitata s L. hyperhorea s Alaria Pylaii s A. esculenta 1 Fucus spiralis 1 F. inflatus 1 F. serratus 1 F. vesiculosus 1 Pelvetia canaliculata 1 Ascophyllum nodosum Chlorophycese. 1 Chlorochytrium Cohnii s G. inclusum si C. dermatocolax 1 C. Schmitzii , si Codiolum Petrocelidis 1 C. gregarium ... 1 C. pusillum 1 Percursaria percursa 1 Enteromorpha aureola 1 E. Linza 1 E. intestinalis 1 E. clathrata 1 Monostroma groenlandicum si M. Grevillei si M. undulatum si M. fuscum si Ulva lactuca 1 Prasiola pohirhiza 1 P. furfuracea 1 P. stipitata 1 Ulothrix consociata v. islandica . . . 1 U. suhflaccida • i 1 U. pseudotlacca 1 U. flacca 1 Pseudendoclonium submarinum.. . si Entoderma Wittrockii . MARINE ALGAL VEGETATION 85 Table 7. The Vertical Distribution of the Species (continued). Litto- ral zone Depth (metre) » % 11 m 0 7 to 1C 7 0 LC 7 7 •O ,T-, O »t O OO CO ^» J> 1 I C1! CO ~^ I — 0 1 «' 0 I I 0 0 O I— 1 ! oo — 0 •^ to 1 Acrochsete parasitica — v 9 — . • • • si A. repens si Bolbocoleon piliferum 1 Ulvella fucicola si Pringsheimia sctitata 1 Ochloclisete ferox 1 Urospora rnirabilis 1 U Hartzii si U Wormskioldii 1 Cha?tomorpha tortuosuni si C Melagonium 1 Rhizoclonium riparium si Spongomorpha vernalis 1 Acrosiphonia albescens si A incurva si A hvstrix 1 A flabelliformis 1 A penicilliformis 1 Cladophora rupestris 1 C hirta . 1 C sericea 1 C glaucescens . .... 1 C gracilis Sf"lct'r'pnVii 11 m Ouplrptti Cyanophyceae. 1 Pleurocapsa amethvstea . . . 1 Plectonema norvegicum 1 Pliormidium autumnale 1 Spirulina subsalsa 1 Calothrix scopulorum 1 Rivularia atra Fungi. Dothidella Laminariie A. Upper Littoral Zone. The preceding table shows that 18 species of Rhodophijcecc, 18 species of Phceophycece , 36 species of Chlorophycece and 6 species of Cyanophycece grow in the upper littoral zone; that is, 78 species 86 H. JONSSOX in all. Of these species some are common and others rare; some are found exclusively in the upper littoral zone and others extend further downwards. In order to show this more distinctly, each «/ • group will here be dealt with separately and in detail. Rhodophyceae. Of the 18 species mentioned 5 are common, and the remaining 13 rarer. The following are the commonly dis- tributed species: Bangia fuscopurpurea , which is found exclusively in the upper littoral zone; Porphyra ambilicalis which is found also in the lower littoral zone; Polysiphonia fastigiata which occurs ex- clusively on Ascophyllum nodosum in the upper littoral zone; Rho- dochorton Rothii which extends to a depth of 10 metres, and Hilden- brandia rosea which extends to a depth of 5 metres. The three first-named species are common in places exposed to the light in the littoral zone; Rhodochorton and Hildenbrandia , on the other hand, occur most frequently as an undergrowth, or in shady clefts (Rhodochorton), and at the bottom of pools (Hildenbrandia). These two species are met with more rarely in places exposed to the light. The 13 species which occur in the upper littoral zone, but which must be called somewhat rare, are there shade-loving, and are then found either as an undergrowth or in shady clefts and depressions. Some are epiphytes, and are then protected against de- siccation by the host-plant. The greater number of these 13 species are common in the lower littoral zone, and the 3 Ceramium-species which are recorded exclusively from the upper littoral zone may doubtless be expected to occur also in the lower littoral zone. Conchocelis does not occur in the lower littoral zone but is found at a depth of 6 — 35 metres. Pha3Ophyce8e. Of brown algae the Fucacece play the most important role. The species are few in number, but are of social growth, and occur in such quantities that they comprise by far the greater portion of the bulk of the plants in the upper littoral zone. Of the 18 species mentioned above, 12 are commonly distributed, and 6 are more rare, in the upper littoral zone. Of the 12 common species, 9 are found exclusively in the upper littoral zone, viz., 6 species of Facacece, Sphacelaria britannica, Ectocarpus tomentosus and Ralfsia clavata; Pylaiella littoralis, Elachista fucicola and Isth- moplea, on the other hand, range to a depth of about 10 metres. Of the rarer species, Petroderma is found exclusively in the upper littoral zone, Ralfsia uerrucosa is most frequent in the upper littoral zone, but is also met with in the lower littoral zone, as an epi- MARINE ALGAL VEGETATION 87 phyte; the remaining 4 species are common in the lower littoral zone, and occur in pools in the upper littoral zone, consequently they cannot, strictly speaking, be reckoned as belonging to the upper littoral zone. Chlorophyceae. In the table, 36 species of green algae are mentioned from the upper littoral zone. Of these, 15 are charac- terized as common and 21 as rarer. 31 species are recorded ex- clusively from the upper littoral zone; 2 species, Entoderma and Bolbocoleon are more frequent in the lower littoral zone; one species Chlorochytrium dermatocolax is as frequent in the lower littoral zone as in the upper littoral zone, one species Ulothrix flacca, which must be considered a decidedly littoral species, grows to a depth of about 10 metres. Ulvella fucicola also grows to a depth of 5 metres, but must nevertheless be considered littoral. In addition to these 36 species, others may be found in the upper littoral zone, which have a more downward extension (e. g. Monostroma Greuillei var. arctica, M. fusciim, Acrosiphonia incurva, and others), but they generally keep to the pools. Cyanophyceae. All the species hitherto found (6) grow in the upper littoral zone. The species which are recorded from the upper littoral zone may be put into two divisions. The one comprises the species which are common in the upper littoral zone and are adapted to growth in places exposed to the light and the wind during the period of exposure; these species, then, might be termed strictly littoral. The other division comprises species which are common in the lower littoral zone or extend even further downwards. The greater number of these species does not occur in the open littoral zone, but is found as an undergrowth or in shady clefts, or in pools. The real home of these species is lower down than in the upper littoral zone, and for this reason they can scarcely be de- signated littoral species. According to the above, there are in all 58 strictly littoral species: Rhodophycece 5 species, Phceophycece 14 species, Chlorophycece 33 species and Cyanophycece 6 species. The upper littoral zone is thus essentially characterized by a paucity of species of red algae, by many species of green algae and by a preponderance of Fucacece. The number of the species of brown algae is of less importance; it is three times as large as that of the red algae and about half as large as that of the green algae. 88 H. JONSSON * If we reckon, in percentages, the number of species of each group of the total number of species in the upper littoral zone we obtain the following figures. The strictly littoral species (58 in all) are indicated by a, the other species (20 in all) not strictly littoral are indicated by b, but no attention is paid to those species which may be found in the upper littoral zone and appear to grow by preference in pools. Rhodophyceae Phseophyceai Chlorophycese Cyanophycese a 5 9 ° o) 14 (24 ° o) 33 (57 °/o) 6 (10 %) a - b 18 23 ° o; 18 (23 ° o) 36 (46 ° o) 6 (8 ° o) So far as the abundance of species is concerned I lay special stress on the red algae, green algse and blue-green algae. It would be confusing, especially as regards the red algae, to reckon the b- species as strictly littoral, nor can this be done, because they are not adapted for life in the open littoral zone. As regards the brown and green algse, on the other hand, at is of no essential importance whether the b-species are included or not, as they are so fewr in number. If we calculate how great a proportion the strictly littoral species form of the combined number of each group we obtain the following figures: Rhodophycese Phseophyceie Chlorophyceje Cyanophycese Species known at pre- sent from Iceland.. 76 67 51 6 Strictly littoral species 5 (6.6 °/o) 14 (20.9 ° o) 33 (64.7 ° o) 6 (100 ° o) a + b (;see above) 18 (23.7 ° o) 18 (26.8 %) 36 (70.6 °/o) 6 (100 ° o) B. The Lower Littoral Zone and the Belt below down to a depth of about 10 metres. 1. The Lower Littoral Zone. From this are recorded 93 species in all (Table 7), viz. 39 Rhodophycece , 37 Phceophycece and 17 Chlorophycece. Of these 93 species, the upper and lower littoral zones have 15 species in common (8 red, 5 brown, 2 green) which do not extend further downwards; the greater number of these belong, strictly speaking, to the lower littoral zone, as, in the upper littoral zone, they usually occur in pools or very shady spots. 19 species (8 red, 11 brown) have been found only in the lower littoral zone; but the majority of them probably occur also below the limit of low-tide, and, in any case, some have their area of distribution MARINE ALGAL VEGETATION 89 close to this limit; 6 species (1 red, 3 brown, 2 green) have their area of distribution in the upper and lo\ver littoral zones, and to a depth of 10 metres, these species are mentioned under the Upper Littoral Zone, and there, 5 of them are reckoned as littoral. 35 species (10 red, 15 brown, 10 green) grow in the lower littoral zone, and to a depth of about 10 metres. They appear to be about as common in the lower littoral zone as in the belt between the limit of low-tide and the depth mentioned. 18 species (12 red, 3 browrn, 3 green) grow in the lower littoral zone, and to a depth of more than 10 metres; these belong to the species which have a lower downward range, 15 of them have their uppermost limit in the lower littoral zone, and 3 of them in the upper littoral zone. The species characteristic of the lower littoral zone are especially the 19 species which are found there only, and the 35 species wrhich extend to a depth of about 10 metres, for some of these (e. g. Rhodymenia and Halosaccion), by occurring in masses, often cha- racterize large portions of the lowrer littoral zone. 2. The Belt down to a depth of about 10 metres. In the table 103 species are recorded from this belt. Of these, two occur also in the upper littoral zone, and have been previously mentioned (Rhodochorton Rothii and HildenbrandiaJ , 6 occur also in the upper and lower littoral zones and are mentioned above, 35 occur also in the lower littoral zone (see under this heading); while 9 species are found only at this depth, but of these some may be presumed to extend further downwards and some may possibly occur in the lower littoral zone. 19 species extend down- wards, with their uppermost limit in the upper and lower littoral zones, as, for example, Conchocelis rosea which occurs in the upper littoral zone and is absent from the lower littoral zone, besides the 18 species mentioned under the Lower Littoral Zone. 32 species with a downward range have their uppermost limit at a depth of about 10 metres. Besides the 9 species which are found only in this belt, it is especially the 35 species which this belt has in common with the lower littoral zone which characterize the belt, as some of them those mentioned under the Lower Littoral Zone by occurring in masses often characterize large portions of the bottom. By comparison it can easily be seen that the lower littoral zone is much more closely related to this belt than to the upper littoral zone, which, amongst other things, is evident from the great 90 H. JOXSSON number of red algae in the lower littoral zone. This can be dis- tinctly seen from the following figures: — Upper Littoral Zone Lower Littoral Zone 58 species (a) 93 species not common common not common 51 (35.42%) 7 (4.86%) 86 (59.72% Upper Littoral Zone Lower Littoral Zone 78 species (a 4- b] 93 species not common common not common 54 (36.73 %) 24 (16.33 %) 69 (46.94 % Lower Littoral Zone From the limit of low-tide to a 93 species depth of about 10 m. 103 species, not common common not common 34 (24.82%) 59 (43.06%) 44 (32.12 %; The figures showr that the upper littoral zone is very unlike the lower littoral zone, especially if only the species which are characteristic of the latter, or which occur in the open littoral zone (a), are taken into consideration, which is most correct, as the shade-loving species in the upper littoral zone must be regarded as stragglers from associations lower down. As previously mentioned, the lower littoral zone bears the greatest resemblance to the belt which exists Iow7er do\vn (to a depth of about 10 metres). The species which have their lower limit of growth at a depth of about 10 metres and which are commonly distributed in the lower littoral zone I designate semi-littoral because they are found both laid bare during low-tide in the lower littoral zone and con- stantly submerged in the belt belowr; as these species especially characterize the lower littoral zone and the belt below to a depth of about 10 metres I designate this area the semi-littoral zone. It must not be concluded, however, that semi-littoral vegetation covers the bottom everywhere down to a depth of 10 metres; below the limit of lowr-tide the semi-littoral vegetation appears rather to consist of stragglers from the lower littoral zone into the Laminaria-be\i. Thus the semi-littoral zone is situated between the Facus-belt and the La- in inar za-belt. The species which specially occur in the semi-littoral zone I designate on the whole as semi- littoral, also those which are found in the lower littoral zone and are not found below the limit of low-tide; they will probably be found also below this limit. Species which are common in the lower littoral zone and are rare in the MARINE ALGAL VEGETATION 91 upper littoral zone, but not known below the limit of low-tide I have also designated semi-littoral as they might be expected to grow lower down.1 C. The Sublittoral Species. These species play the principal role in the sublittoral vegeta- tion. This may be characterized in a somewhat similar manner as v the vegetation of the upper littoral zone, although conversely as regards red and green algae, viz. by a large number of red algae and an extremely small number of green algae and by the occur- rence of a mass of Laminariacece. In Table 7, these species are marked with an s before the name. Their number is shown by the following figures: Species Littoral Semi-littoral Sublittoral Red2 . . . . 5 30 39 Brown . . . . 14 31 22 Green . . . . 33 15 3 Blue-green . . . . 6 Total... 58 76 64 From a comparison of the number of species of the sublittoral zone with that of the littoral and semi-littoral zones it is seen that it is smallest in the littoral zone, somewhat larger in the sublittoral zone and considerably larger in the semi-littoral zone. Species of green algae occur most abundantly in the littoral zone, and their number is infinitesimal in the sublittoral zone. Red algae are most numerous in the sublittoral zone and verv scarce in the littoral V zone. The brown algae are more evenly distributed in the different zones, yet they are richest in species in the semi-littoral zone. The blue-green algae are found exclusively in the littoral zone. In the semi-littoral zone the large brown algae do not occur in masses like, for example, the belt of Fucus in the upper littoral 1 As regards the majority of the species (see Table 7) it is easy to decide whether they are littoral, semi-littoral or sublittoral. but there are some species, nevertheless, which it is difficult to refer definitely to any one of the three zones mentioned, and therefore it is sometimes a matter of opinion whether they should be reckoned in the one or the other. By perusing the table these species are easily detected. - Two species, Delesseria Baerii and Bonnemaisonia asparayoides are not in- cluded as their habitat is unknown to me (see 31, pp. 140. 141). 92 H. JONSSON zone and the belt of Laminaria in the sublittoral zone. This is possibly the reason why the semi-littoral zone is richest in species. In Table 7, 64 species are recorded as sublittoral. Besides these, semi-littoral species occur in this zone, especially as epiphytes, or, more rarely, as undergrowth. 12 species, almost all semi-littoral, which extend to a greater depth than 10 metres are mentioned in the table. In addition, about 14 species can be regarded as epi- phytic, particularly in the upper part of the sublittoral zone. The sublittoral vegetation is thus composed of 64 sublittoral species and of about 26 semi-littoral species, or about 90 species in all. Lower Limits of Growth. As far as the lower limit of growth is concerned the species which are found below low-tide must be dealt with collectively. In the following table, which shows the lower limits of growth, all the depths at which species have been known to occur have been, as a rule, taken into consideration, and thus the table does not indicate the depth of their general distribution or the depth at which they form associations (see under Vegetation). In the table, 2 red algae are omitted, as I have no further knowledge regarding their habitats (31, pp. 140, 141). These species are Delesseria Baerii and Bonnemaisonia asparagoides. As regards the green algse it must also be stated that I have not taken into consideration the records from greater depths than 10 metres. I myself am responsible for some of these records; the algre often were detached, but sometimes it appeared as if they really had been growing at the depth recorded. I always, however, have entertained some doubt concerning this, and consequently prefer at present not to consider records from such depths. Regarding Chlorochytrium inclusion the record is cor- rect, as it grew in Turnerella, which was attached to the bottom at a depth of 30 metres. For Gomontia and Ostreobium I have given the most common depth, down to about 40 metres, but I have also obtained these species, growing inside an old shell of Mi/a, from a depth of about 60 metres. It should be further stated that I do not know the depth to be given for Desmarestia ligulata. This species has been found by Ove Paulsen between the Vestmannaeyjar and South-Iceland; I presume that it grew at a depth of about 20 metres. MARINE ALGAL VEGETATION 93 Lower Limits of Growth in metres (see Table 7). 1 ^bout 5—10 About 15-20 About 25-30 About 35—45 About r.O and more Red al°'8e 13 10 10 15 7 •1 V A U t Brown algae .... 19 8 4 5 3 Green algae 16 • • 1 2 • • Total. . . 48 18 15 22 10 37 In the above table 113 species are mentioned, of which 48 do not extend to a greater depth than about 10 metres, 18 extend to about 20 metres, 15 to about 30 metres, 22 to about 40 metres and 10 to about 60 metres and more. Thus there appears to be a flori- stic boundary at a depth of about 10 metres; while another boundary can be faintly distinguished at a depth of from about 30 to 40 metres, as there are 37 species which appear not to extend further downwards; and here, also, is situated the lower limit of growth of most of the Laminariacece which play the same role on the sub- littoral bottom as the Fucacece play in the littoral zone. Exactly where the lower limit of growth, as regards the marine algae in Iceland, is situated whether it lies at a depth of about 60 to 80 metres or deeper I cannot at present decide, but it is most pro- bable that the vegetation at greater depths than 60 metres is, in any case, extremely poor in species. According to the above statements the lower limits of the algae are as follows: The littoral limit (Upper Littoral Zone) 53 species 26.77 °/o The limit of low-tide , 32 16.17 ( At about 10 metres 48 24.24 ° o At about 20 metres 18 9.09 °/o At about 30— about 40 metres 37 18.68 °/o At about 60 metres and more ... 10 5.05% 198 species. By future investigations all these figures will undoubtedly be altered and many of them to no inconsiderable extent; but the four V principal boundaries, namely the littoral limit, the 10-metre limit, the 30 — 40 metre limit and the absolute depth -limit will always remain evident. As regards the absolute depth-limit I cannot make any definite statement. The dredgings which I myself have undertaken have all been conducted with a small dredge from a small rowing boat. 94 H. JONSSON : MARINE ALGAL VEGETATION The positive results which are obtained with this small dredge are satisfactory enough when the material is inspected with sufficient criticism , but it should not be concluded with certainty from a negative result (that is, when absolutely nothing comes up from the bottom) that no plants grow there. In great depths, also, dredging is difficult from a small rowing-boat. In some places only, in SW. Iceland and E. Iceland, have I undertaken dredgings in a depth of about 80 metres, and the result has been negative. In those places where the plummet revealed a clayey bottom, it was certain that no plants were growing there. Off S. and SW. Iceland I have met wTith no vegetation at a greater depth than about 40 metres, but in Rey5arfjor5ur in E. Iceland, on the other hand, I have come across plants at as great a depth as about 60 metres. This might indicate that the vegetation extends further downwards in E. Iceland; but as the observations are too few, this point must remain undecided until further investigations are forthcoming. It was a fairly common occurrence, especially in the fjords of E. Iceland, to encounter sunken fragments of algae (often strictly littoral species) and of mosses in depths of more than 22 metres. In Sey5isfjor5ur I came across leaves of Betula and Salix at a depth of 14 — 20 metres. C. H. Ostenfeld (the Ingolf Expedition) found Lithothamnion laeve in great quantities at a depth of 88 metres off the north coast of Iceland, and R. Horring (on board the "Diana," off E. Iceland) found Lithothamnion tophiforme at a depth of 70 metres. In depths of from 60 to about 100 metres Horring found, in addition, frag- ments of various algae, amongst which were strictly littoral species, and of mosses which had evidently fallen to the bottom. In order to draw the attention of future investigators to this matter it must further be mentioned that Horring brought home in spirits a young plant of Laminaria saccharina from a depth of from 81 to 104 metres (Mjoifjordur, 14.5. 1898, St. XIII) and on the label was written "In the trawl were many large Laminarice which had been torn away from the bottom." The Laminaria brought home appeared normal, and, if it had fallen to the bottom could only have lain «/ there for a short time. What is most likely is that the trawl passed over an uneven bottom, and that the Laminarice grew at a lesser depth than that mentioned; or is it possible that the deep-water form of Laminaria saccharina ranges so far downwards? I leave it to future investigation to decide this question. VI. MARINE ALGAL VEGETATION AND SEA- GRASS VEGETATION. THE vegetation of the sea is naturally divided into two principal groups: the Plankton composed of the small plants floating passively in the water, and the Benthos which comprises the species attached to a substratum. In this article only the latter is dealt with. The Benthos1 falls naturally into two divisions, viz. the litho- philous vegetation, the communities of marine algae (suhformation of marine algae, Halo-nereid communities, Warming, 72, p. 170), and the vegetation of loose soil (Enhalid-formation, Warming, 72, p. 177). The Benthos has been divided in different ways. J.G. Aeardh «/ o in 1836 (Novitise Florae Sveciae) divided the marine vegetation into three zones, a green zone above, a brown zone in the middle and a red zone lowest of all. Lyngbye in the same year, also divided the vegetation into three zones (Rariora Codana, printed 1879 — 80), a zone of green algae (Ulvacece) being above, a zone of red algae in the middle and a zone of Laminarice below this. 0rsted (77), like Agardh, also divided the vegetation in the 0resund into three zones, but 0rsted has the merit of being the first to explain that the division of the zones depends upon the depth to which the light penetrates, and upon the colour of the light at the various depths. Kj el 1m an has divided the algal Benthos into regions. Where there is a tide, the littoral region is reckoned as being between the highest high-water mark and the lowest ebb-tide mark; where, on the other hand, there is no tide Kj ell man reckons the littoral region as extending from the uppermost limit of the algal vegetation to a depth of I1/ 2 — 2 fathoms (34, p. 7). The sublittoral region ex- tends from the lower limit of low-tide, or else from a depth of I1/ 2 1 No notice is taken of the bacterial flora of the sea. 96 H. JONSSON -2 fathoms, down to a depth of 20 fathoms; and finally the elittoral region stretches from the 20-fathom contour downwards. This division, unchanged in its main features, is generally employed. The boundary between the littoral and sublittoral regions is, I think, commonly supposed to be rather sharply defined in places with tides, and less sharply defined where tides do not occur. By this division into three regions the algal Benthos is divided into three belts of different depth. Although the boundary lines thus drawn may be described as floristic boundaries, as regards many species, yet they cannot always be regarded as natural limits of vegetation. If natural limits of vegetation are to be drawn, several factors must be taken into consideration. From depth-records, pure and simple, a somewhat clear idea may naturally be formed of the conditions of light, but not of the salinity and warmth. If the conditions of light, salinity and warmth in those layers of water where the algal vegetation lives were known all the year round it would be easy to characterize the limits of the vegetation. The great importance of the salinity to algal vegetation is well-known and is emphasized by Rosenvinge (63), Svedelius (71), Borge- sen (12) and Kyi in (45), amongst others. The ecological factors in the coastal waters round Iceland are not so well known that I am able to draw the natural limits of the vegetation by means of them. My starting point is, therefore, the vegetation itself, and from the appearance of the vegetation it is possible, to a certain extent, to form an opinion as regards the ecological factors, in the same way as an opinion may be formed from these concerning the appearance and composition of the vegetation. The marine algal vegetation divides itself into several zones as the Benthos does in fresh water. The divisions between the zones in the sea are very distinct : some species seem to be spot-bound or very sensitive to changes of level; other species may occur in two or several zones; but it depends especially, I think, on the quality of the water, the intensity of the light and, where there are tides, on the period of exposure (in the littoral zone). By studying the vertical distribution of the species and asso- ciations, I have come to the conclusion that the marine algal vege- tation may be divided into three zones almost parallel one with an- other: the Littoral Zone, the Semi-littoral Zone and the Sublittoral Zone. The Littoral Zone understood in a more restricted sense is identical with the upper littoral zone and extends almost to the MARINE ALGAL VEGETATION 97 low- water mark of neap-tide, and is exposed during each low- tide; the upper littoral zone is, then, the littoral zone proper, which doubtless corresponds exactly with Kolderup Rosenvinge's limitation of the littoral zone in Greenland, but not entirely with Borgesen's limitation of the littoral zone in the Faeroes, as some of the Faeroese littoral associations seem to belong to the next belt. The semi-littoral zone extends from about the low -water mark of neap-tide to a depth of about 10 metres, and thus extends over the lower littoral zone, and even lower than that. In reality this zone comprises the lower littoral zone including stragglers below the low-water mark of spring-tide to a depth of 10 metres. That part of the zone lying in the lower littoral zone is laid bare at and about spring-tide, but is submerged at neap-tide. The sublittoral zone extends from the low-water mark of spring- tide to the absolute depth-limit. There seems to me to be no reason for calling any part of the Benthos elittoral. Even if the 40-metre contour is a lower limit of growth in the case of several species, and is, approximately, the lower boundary of the La/mnarza-community, yet the upper bound- ary of the red-algae communities which extend further downwards than 40 metres lies much higher, and the 40-metre contour thus cuts straight through natural communities. It cannot, therefore, be considered the principal boundary as regards the wrhole of the constantly-submerged vegetation. Stromfelt (I.e.) is of the opinion that, possibly, elittoral vegetation does not exist on the coasts of Iceland. According to Kjellman the elittoral vegetation is extremely poor in species, and probably has a limited distribution everywhere in the northern seas. From what has been said above respecting the 40-metre line, and from a comparison with Greenland (Rosen- vinge, 63) and the Faeroes (Borgesen, 11 and 12), it is obvious that a division of the constantly-submerged vegetation at this depth- line is not quite natural in the northern seas. It is more correct, therefore, to do as Rosenvinge and Borgesen do, and to class the vegetation as sublittoral down to the absolute depth-limit. In the following I do not employ the term "region," which is now generally used to describe somewhat limited subdivisions of The Botany of Iceland. I. 7 98 H. JONSSON the vegetation, as I have thought it better to designate the algal Benthos taken as a whole by this word, and to call it the "algal region." I have also avoided the term "formation." The algal formations established by Kj el 1m an and others are not real formations, but only associations (Warming, 72, p. 171), and the entire marine algal vegetation is given as a subformation of marine algae (Halo- nereid; Warming, 72, p. 169). I do not think that the term "forma- tion" should be used in connection with the algal region in any other sense than that in which it is used in connection with land vegetation. As the term "facies" cannot be used in English to de- note a vegetation-unit (Warming, 72, p. 146, foot-note) I have avoided it. I divide the vegetation in each of the three zones into asso- ciations, and where I think they are closely allied I combine them into communities. Sometimes by the suffix "etum" is meant the subordinate part of an association, and sometimes a vegetation-unit which almost corresponds with an association. ACCOUNT OF THE MARINE ALGAL VEGETATION AND THE SEA-GRASS VEGETATION. A. The Marine Algal Vegetation. a. The Littoral Zone. aa. The Photophilous or strictly Littoral Communities. 1. The Prasiola stipitata-association. 2. The Community of Filiform Algae. 3. The Community of Fucaceae. 4- The Pelvetia-Fucus-spiralis-belt. + + The Fucus-belt. 4. The Enteromorpha-association. 5. The Acrosiphonia-association. bb. The Shade-vegetation. 6. The Hildenbrandia-association. 7. The Rhodochorton-association. cc. The Vegetation of Tide-pools. b. The Semi-littoral Zone. 8. The Monostroma-association. 9. The Chorda-association. MARINE ALGAL VEGETATION 99 10. The Community of Rhodymenia. 11. The Polysiphonia urceolata-association. 12. The Community of Corallina. 13. The Crustaceous Alga-association, c. The Sublittoral Zone. 14. The Community of Laminariaceae. 15. The Desmarestia-association. 16. The Deep-water Community of Florideae. 17. The Lithothamnion-association. 18. The Community of Crustaceous Algae. B. The Sea-grass Vegetation. 1. The Zostera-association. A. The Marine Algal Vegetation, a. The Littoral Zone. The Littoral Communities. The littoral vegetation is com- posed of several communities. These are left dry during low-tide, with the exception of the pool-vegetation; but the period during \vhich the different communities are left dry varies greatly. I am not prepared to state accurately the period of exposure, but I pre- sume that the lowest littoral communities will be laid bare for about 1 — 2 hours under normal conditions, while the uppermost communities will be submerged for about one hour, and will be laid bare for about 11 hours. On exposed coasts the period of ex- posure is shortened by the beat of the wraves. The ecological factors differ not a little in the uppermost and lowermost part of the littoral zone, and because of this difference the vegetation is divided into longitudinal belts along the coasts. The littoral vegetation falls na- turally into three divisions: the light-loving or strictly littoral com- munities, the shade-loving communities, and the pool-vegetation. In the following description of the vegetation, the communities in each of the three zones are arranged as far as possible according to depth, and in such a way, that the uppermost come first and the lowest come last. This rule, however, cannot always be adhered to. aa. The Photo philous or strictly Littoral Communities. These communities are found in places in the littoral zone which are directly exposed to light during low -tide. They are composed of blue-green, green, brown and red algae; the 7* 100 H. JONSSON brown algae are found most abundantly, the green algae come next, red algae occur less frequently, and blue-green algae are found in the smallest quantity. The substratum is either rock solid rock or else talus of debris or pebbles, or in many places consists of gravel, clay or mud. The perennial species prefer almost exclusively the rock-substratum; but they occur, though very seldom, on other substrata, and are then, as a rule, dispersed here and there on small stones, viz. on a gravel-clay soil at the head of the fjords. The annual, short-lived species also occur most frequently on a rock-substratum, but they also occur fairly frequently on the pebbles of the littoral zone. « 1. The Prasiola stipitata-association. This association extends furthest upwards and is found well developed on the rocks in several places round the coasts. It is usually almost on a level with the Verrucaria-maura belt and, up- wards, sometimes approaches the outposts of the land-vegetation. At the highest water-level it is covered by the sea for a short period, or is, at least, washed by the breaking waves ; but such high water occurs only rarely, and in normal conditions this association must, without doubt, be content with the spray from the waves during spring and summer. On less exposed coasts this association is, as a rule, sharply defined from the community of filiform algae which exists below, but on a very exposed coast the boundary is more variable. The dominant species in this association is the small, leaf-like Prasiola stipitata, which grows very socially upon the tops of flat rocks. As it has its distribution almost exclusively in this belt it seems natural to designate the association by its name. It is essen- * tially adapted to live in the air, and is capable of withstanding desiccation well, which may be perceived, inter alia, by the fact that it does not seek clefts and crevices but grows on surfaces which are exposed to light, wind and weather. The density of the vegetation must also afford each individual some protection against desiccation. In spring and summer long periods must occur during which this association is not wetted by the sea, and, during summer drought, I have often seen Prasiola stipitata as dry as a bone upon the rocks. Prasiola furfuracea also occurs side by side with this species, but is rarer. Species and varieties of species, such as Enteromorpha intesti- MARINE ALGAL VEGETATION 101 nalis f. micrococca and Rhizoclonium riparium, which essentially be- long to the lower belts, but can accommodate themselves to life in the air, are also included as members of this association. They are species which, owring to their structure or their manner of growth, are capable of enduring desiccation. Enteromorpha intestinalis f. micrococca has, as is well known, small cells, the walls of which are very thick, the inner walls, in particular, being highly thickened, serving possibly as reservoirs for water. It prefers fissures in the rocks, where, as a rule, it is less exposed than the species charac- teristic of the association. Sometimes, however, I have come across f. micrococca growing, like Prasiola stipitata, on flat rocks manured by birds, but then it had a different aspect and, at first sight, some- what resembled Prasiola. Rhizoclonium also occurs in fissures, where, owing to its manner of growth that is to say its pulvinate form — it is protected from desiccation during the long periods of drought. Calothrix scopulorum also occurs as a member of the Prasiola- association and forms Calothriceta of limited dimensions; the indi- viduals are procumbent and are placed so closely together that the rock is completely covered, and thus they protect each other from desiccation. From a biological point of view, the manner of growth of this species, in the dry condition, is similar to that of the cru- staceous algas. Enteromorpha intestinalis f. minima also occurs in this community. The species in this community grow, as a rule, in small, pure societies which form a narrow, though not a continuous belt along the coast. This belt is situated higher on exposed coasts than it is on those which are less exposed. The Praszo/a-association is undoubtedly commonly distributed in neighbouring countries, but the constituting species may be dif- ferent. In Greenland (Rosenvinge, 63, p. 200) CalothrLv scopulorum, Ectocarpus maritimus, and Rhizoclonium riparium occur only in the uppermost part of the littoral zone. This vegetation, however, scarcely corresponds with the Prasz'o/a- association, but rather with that occurring below. In the Faeroes, however, an exactly corresponding association is found, which Borgesen calls the Chlorophycece- formation (12, p. 712). The Faaroese Chlorophycece-formaiion , how- ever, appears to be more luxuriant and is composed, in part, of other species. A Praszo/a-association (Foslie, 18, p. 127) similar to that of Iceland is evidentlv found in Finmark. 102 H. JONSSON 2. The Community of Filiform Algse. It is not easy to find a suitable name for this community. It is composed of several species which are all equally common and are all dominant to almost the same degree. As almost all the species are filiform and non-branching, it seems to me that the community may fitly be named in accordance with the form of the frond. The community of filiform algae forms a narrow belt, which is often continuous along fairly considerable stretches of the coast, at about the average limit of high water. The vertical height of the belt is inconsiderable, about one foot, but the breadth conforms somewhat to the slope of the coast, and may attain to 3 — 4 feet, or even more. This vegetation is very well developed on the face of vertical rocks, and the various associations of the community can be distinctly seen, one above the other, as parallel bands of varying colour. The species which occur most frequently are the following : Ulothrix flacca. Bangia fuscopurpurea. Urospora mirabilis. Porphyra umbilicalis f. typica. Monostroma groenlandicum. These are all dominant species, and form extensive associations, of which some are pure and others mixed. Other species may also occur, but only in lesser quantities. The Ulothrix-association, as a rule, reaches highest up the cliff. The principal species is Ulothrix flacca, which forms a distinct belt, extending rather far in a horizontal direction. On rock- walls, the filaments are often comparatively long, and are moved to and fro over the entire belt by the beat of the waves or the ripple of the sea at flood-tide; during the period of exposure they hang down, pressed closely against the face of the rocks. The outer filaments protect the underlying ones from desiccation during low-tide, and thus it happens rather frequently that the outer layer is dry while the protected layer that nearest to the rocks is moist. In this way the social growth of the plants protects them against desiccation (cf. Rosenvinge, 63, p. 201), at any rate under normal conditions, and so long as no exceptionally long periods of drought occur. It happens rather frequently, however, that the (7/o//7rzo>vegetation becomes quite dry during low-tide. This is espe- cially the case when the vegetation occurs on boulders in the lit- toral zone where, when the water subsides, the filaments radiate MARINE ALGAL VEGETATION 103 from the highest point of the stone. Such a stone-surface is smooth and dries more quickly and completely than the uneven face of the rock. This vegetation appears to maintain itself well in spite of being completely dried up, day after day, during low-tide, during the nocturnal low-tides, naturally, it is dried up to a much lesser extent. As a rule, such a dried-up (7/o//ir/.r-vegetation is so tightly adpressed to the surface of the stone that it can only be removed by being scraped away with a knife. Thus, owring to de- siccation, it appears, from a biological point of view, to have as- sumed a crustaceous form, which evidently diminishes the evapora- tion from its surface. The other species which occur in great quantities in this as- sociation, such as Urospora mirabilis and Monostroma groenlandi- cum, act biologically in a similar manner to Ulothrix flacca. Both these species are found fairly frequently growing among Ulothrix flacca; and as I think that these species may in several respects be comprehended in one biological unity, I consider them members of the same association, although they both occur in pure growths (Urosporetum, Monostrometum). While Ulothrix flacra and Urospora mirabilis are commonly distributed, the distribution of Monostroma groenlandicum is more restricted, for this species can be reckoned as commonly distributed only in E. Iceland. Of the species which are rare or of local occurrence, and which are reckoned in this association, the following may be mentioned: Ulothrix pseudoflacca, Ulothrix consociata var. islandica and Urospora Hartzii. In addition, Enteromorpha intestinalis f. micrococca and Rhizocloniam occur here as they do in the Praszo/a-community. Of these species Ulothrix consociata var. islandica requires to be described most fully. It is very social, and grows in cushion- like masses higher up, as a rule, than the other species of the as- sociation. It appears to be protected from desiccation both by its manner of growth and by the thickness of its cell-walls. Codiolnm gregarium I have found to be of very social growth, covering comparatively large stones as pure Codioleta. It is most nearlv related to this association. \j The structure of the frond in this association displays various peculiarities, which must be considered as being beneficial to the plant during the period of desiccation; for instance, the interior of the frond of Monostroma groenlandicum is filled with a gelatinous 104 H. JONSSON mass. Rosen vinge (63, p. 201) has pointed out that this mass must be of importance as a reservoir for water during low-tide. Ulothrix flacca also has very thick cell-walls, especially in the fruit- bearing filaments (cf. Rosen vinge 1. c.). The remaining species also are rather thick- walled. Although the £//o//irzo>association is distributed to an extra- ordinary extent, yet it cannot be expected to be found everywhere. It prefers that part of the littoral zone which is rocky and stony, and grows luxuriantly on a somewhat exposed coast, and even, indeed, on one which is very exposed. It also extends right into the fjords, if the nature of the shore is favourable to it. The succession of the associations of filiform algae is seen most distinctly on vertical rocks on somewhat exposed coasts, where the 67o//7rzx-association appears uppermost; but where the substratum is uneven a talus of debris or irregularly heaped-up fragments of rock the zonal division of the community becomes less ap- parent, and it may then well happen that the Ztom/z'a-association extends above the t//o//7rza>association. The Bangia-association prefers the rocky part of the litto- ral zone, and vertical rock- walls in particular, and is only rarely found in the stony part of that zone; it occurs normally below the t//or7zrz;r-assoclation , but where the latter is absent the Bangia- association is not infrequently the algal vegetation which reachest highest. In many places it has an extremely wide horizontal dis- tribution, and is often perceived at a considerable distance as a reddish-brown band in, and at the edge of, the water. This Bangia- belt may attain a considerable breadth, as much as three feet, and it often displays several shades of colour. Thus, in a less exposed spot facing the south, I have seen the uppermost part coloured green, the middle part brownish and the lowest part a fresh reddish- brown. This is certainly connected in some way with the long period of drought w7hich, when the weather is calm or the wind blowing off the land, may wTell continue from neap-tide till about spring-tide. During the period of drought Bangia behaves somewhat similarly to Ulothrix flacca. The filaments are rather long, as long as 10 cm., and are very closely packed together; on vertical rocks they hang straight down during low-tide, and are, on the whole, protected by their structure and manner of growth from too severe desiccation, in the same way as has been mentioned in connection with the t//oz7irzac-association. The Bangia-belt is, as a MARINE ALGAL VEGETATION 105 rule, distinctly separated from the Porp/iz/ra-association which exists below, but sometimes the two belts are concurrent, even over rather large areas. The Porphyra-association is formed by Porphyra umbili- calis f. typica. It might be justifiable to incorporate this association in the fian^za-association, but I prefer to reckon it as a distinct association; partly on account of the difference in the form of its thallus, and partly because Bangia is confined to its narrow belt, while Porphyra occurs also in other associations in and below the Fucus-beli. The Porp/?yra-association occurs on a rocky substratum. On vertical rock-faces it is very luxuriant, and then forms a continuous belt below the £a/i* ac%fe«s»Sts^fe?^\i *s^ • »V * ,\ft 5 ^Ef ' •"* " ' * ' J ' ^ r *>rf» - -^ -^ ' ^j/^t*f*-^ *KCl^t -/ ^V V *s.-iP-i^ Fig. 4. Fucus spiralis. Part of a luxuriant Fucns spfraZis-vegetation below the Pelvetia- association. Reykjavik, Aug. 13, 1909. (From phot, by Hesselbo.) to exposed coasts has already been discussed. It may happen with Fucus spiralis, as with Pelvetia, that it extends to so great a height that at times it is not wetted dailv bv the sea. In such a case the •/ */ individuals are as a rule smaller, and are often rather strongly spirally twisted; they may be found lying quite dried up on the rocks, apparently without being damaged thereby. The twisting of the fronds may possibly result from the drying process. This belt is, as a rule, somewhat sharply defined from the Fucus vesiculosus-heh situated below. Although it does not always happen that there is any distance worth mentioning between F. spiralis and the upper outposts of the Fucus ^eszaz/osus-association, yet the boundary is almost always sufficiently distinct. 110 H. JONSSON This belt behaves similarly in the Faeroes (Borgesen, 12, p. 744). Both Boye (10) and Hansteen (25) mention a Pelvetia-forma- tion from western Norway without mentioning Fucus spiralis. In Finmark Fucus spiralis has the same manner of growth as in Ice- land (F os lie, 18, p. 66). The Fucus-belt. The four associations which belong to this belt are the most extensive in the littoral zone and three of them are found everywhere on rocky coasts. The breadth of the belt naturally is dependent on the degree of the declivity of the coast in addition to the nature of the substratum. This belt occurs everywhere along the coast, but it is not luxuriant to the same degree everywhere. Its vegetation is so dense that the bottom is entirely covered, or almost so, by the Fucus plants which during low-tide partly lie prostrate upon the rocks and partly hang on them; seen from a distance it appears as a brown-coloured belt of varying width along the coast. These asso- ciations are usually found on a rocky substratum, but they may also occur on a fairly firm gravelly bottom; the latter is especially the case at the head of the fjords. In such places their vegetation is poor and the species grow scattered, attached especially to small stones which are somewhat firmly embedded in the gravelly bottom. Fucus plants are also found scattered on wood-work, for instance, on wooden piles. On the whole the species of Facus require a stable substratum. The dominant species are the following: Fucus vesiculosus. Fucus inflatus. Ascophyllum nodosum. Fucus serratus. The first three species are common and grow very luxuriantly along the coast; Fucus serratus, on the other hand, was found only in a few places in S. and SW. Iceland; in Hafnarfjordur it grew very socially, while in the Vestmannaeyjar it had a more scattered growth. The individuals of these species form pure associations which usually occur in regular succession: Fucus vesiculosus growing upper- most, Ascophyllum in the middle and Fucus inflatus (and Fucus serratus) lowest of all. This succession is distinctly observable in places where the bottom is flat and gently sloping. But where the bottom is uneven a talus of debris the divisions between the belts are less regular, but can, as a rule, be discerned. It does not, however, follow that the associations always occur quite regularly; MARINE ALGAL VEGETATION 111 the species being rather frequently found intermingled; the Asco- jD/7z///zz/7?-association, in particular, does not appear to be as spot- n u o o •3 tc = o - c/i i 5" L.O bound as the others. It generally occurs somewhat below the middle of the area occupied by the community, but may occur also higher up, even at the very top. It is, however, a fairly constant fact that 112 H. JONSSON Fucus vesicLilosiis grows highest and F. inflatus lowest. How the species behave when left dry is mentioned in connection with each association. The Fucus vesiculosus-association, as already mentioned, «/ is uppermost and often borders closely on that of Fucus spiralis. It varies extremely in extent, according to the character of the coast. The breadth may vary from about one foot on vertical rocks to several fathoms on a gently sloping coast. Fucus vesiculosus extends right into the innermost part of the fjords, which is generally considered a protected coast; if there is a favourable substratum there, its growth may be fairly luxuriant. It also grows very luxuriantly on slightly exposed coasts, i. e. where a landward wind is sometimes both frequent and tempestuous, but where breakers are extremely rare. If the exposure is increased, it seems that F. vesiculosus not onlv decreases in number of individuals •j but also seeks for shelter between the rocks. Then, gradually, it re- treats lower into the littoral zone, and Ascophyllum, which is evidently better able to withstand the heavy seas, advances. This is seen very distinctly at Reykjavik, where the coast must be considered somewhat exposed, because the south-west, west, and north winds are frequently stormy and occasion heavy seas. If we take a bay which is bounded by a rocky promontory, we see at the head of it a luxuriant Fucus yes/cw/os«s-association; this extends out upon the promontory, growing gradually narrower as it extends outwards; in the vicinity of the point itself, the alga has commenced to seek for shelter, and at the extreme end of the point it has disappeared and Ascophyllum has occupied its place, and then Fucus vesiculosus is either found not at all or onlv a few individuals of it occur \j high up in the Ascophyllum-heli. What is here seen on a small scale is repeated on a larger scale on proceeding from the head of the fjord outwards to the extreme point of the peninsulas. Thus, Fucus vesiculosus was not observed on exposed points at Ondverd- arnes (the extreme point of Snsefellsnes) , while Ascophyllum and, naturally, Fucus inflatus also, were found in abundance. On the most exposed points of the Vestmannaeyjar Fucus vesiculosus was also absent, while both of the others occurred plentifully.1 Thus, Fucus vesiculosus behaves in Iceland precisely as it does in the Faeroes (Borgesen, 12, 11). 1 Strom felt (70) records Ascophyllum as growing above Fucus vesiculosus at Holmanes in E. Iceland. MARINE ALGAL VEGETATION 113 Generally this species forms a pure association, hut yet, some- times, it and Ascophyllum can grow intermixed.1 It is always sub- merged at high-water, but seems well able to withstand desiccation at low-water, when however the upper side of the frond dries up considerably, especially on rocks which face the sun, and may then and this is true of Fucus inflatns also be bent upwards, or even slightly distorted, at the apex, though never to the same extent as is the case with Fucus spiralis. When such a branch is lifted up, it can be seen that the under side is moist, even on a warm sum- mer's day. During low-tide the Fucus plants lie prostrate on the rock, one plant overlying the other, or one branch of the frond covering the other, and in this way some water is retained amongst the plants. Naturally, this applies also to the members of the other asso- ciations in the Fuczzs-belt. Those individuals which hang freely from the rocks are exposed more than others to desiccation. The Ascophyllum -association differs from the other asso- ciations in its light-brown colour, and, in addition, by the occur- rence of Polysiphonia fastigiata, which grows very luxuriantly on it in S. and SW. Iceland; it is often somewhat singular to see the light-yellow Ascophyllum-be\t with the numerous dark reddish-brown patches of Polysiphonia. So far as the behaviour of the species during exposure to the beat of the waves is concerned I must refer to the above-mentioned remarks, and can only add that at times it may also occur in the most exposed places, but no longer in the uppermost belt. Here it gives place to the surf-forms of Fucus inflatus (f. exposita, f. den- droides), which form a narrow belt at the upper boundary of the Fucus-belt. Ascophyllum occurs, then, between this and the real Fucus inflatus-belt, which retains its usual position. In a few places, how- ever, on the most exposed points, Fucus inflatus only is found. The Fucus in flatus-association. When exposed, this spe- cies behaves conversely to Fucus uesiculosus and grows most luxu- riantly on exposed coasts. Thus, it behaves in Iceland as it does in Greenland (Rosenvinge, 63) and in the Fseroes (Borgesen, 12). In Iceland, as in other places, the species varies extremely and must be considered very capable of adapting itself to varying circumstances, especially as regards exposure. The association can — in accordance with the exposure - be divided into three belts: - The Surf-belt, the Wave-belt, and the Calm-belt. 1 The epiphyte-vegetation is mentioned subsequently. The Botany of Iceland. I. 8 114 H. JONSSON The Surf- belt, as previously stated, occurs at the upper boundary of the Fucus-association on rocky coasts which are very much exposed. I have seen such a belt at 0ndver6arnes in SW. Iceland, in the Vestmannaeyjar in S. Iceland and at Vattarnes in E.Iceland. Stromfelt (70) has noticed a similar belt on Seley in E. Iceland. It will probably be found, on further investigation, that the surf-belt is far more widely distributed along the coasts than is recorded above. As a rule, the individuals in this belt are of low growth; the frond is leathery, and very thick in proportion below, and rounded, but tapering evenly upwards and becoming thinner; above, it is often excessively branched. The height of the individuals varies to some extent, f. dendroides is the highest, while f. exposita attains only an insignificant height (5 — 9 cm.). A similar surf-boll occurs in the Faeroes (Borgesen, 12), although the vegetation is possibly more luxuriant there than in Iceland. The W a v e - b e 1 1 (Fig. 6) comprises the ordinary Fucus inflatus-be\[. As a rule, it is exposed to the movement of the waves, a movement which is often very violent indeed; more rarely it is exposed di- rectly to the breakers. Here the species occurs in its typical form which shows a considerable variation in the breadth and consistency of the thallus. In very exposed places the branches of the frond are comparatively long and narrow and leathery, but where the move- ment of the waves is less felt, the frond is usually broader. The vegetation of this belt is generally very luxuriant, and covers the sub- stratum entirely. Often, however, the vegetation occurs in patches, owing to the surface-features of the shore. Such a mode of occur- rence is met with, for instance, where the shore is a wild talus of debris consisting of large scattered blocks of stone, or where the solid rock has a similarly uneven surface owing to erosion by the sea. There a belt is formed around each block of stone, while the hollows between them are occupied by semi-littoral or sublittoral vegetation, or by stragglers from these zones. The Calm Belt is lowest of all, often at the boundary between the constantly submerged and the lower littoral vegetation, and some- times forms small offshoots of vegetation1 below that boundary. Here the movement of the waves is least felt, and the frond of the 1 The species varies from the principal form to two kinds of dwarf-forms - a small "surf-form" of tough texture, and a small "pool-form" of delicate texture. The pool-form-association (F. in flatus] f. linearis) may be regarded as the fourth belt, which should then be termed the Delicate Belt (see under The Vegetation of Tide-Pools, p. 125). .MARINE ALGAL VEGETATION 115 plants is also usually broader. In this belt the broad forms are to be found; they may occur both with and without air-bladders. This belt is as a rule poorly developed, and is often non-continuous. Within Hvammsfjor5ur, in bays protected from the surf, but where current-movements are felt to a certain extent, I have observed, below low- water mark, semi -littoral stragglers from the Fncns- comm unity (i. e. from the Fucus zV?/7afos-association). In a similar locality I saw Ascophyllum occurring semi-littorally. mmWmwi*- Fig. 6. Fucns inflatus (sterile plants) lying prostrate upon the rocks. Above, hanging Ascophyllum. Reykjavik, Aug. 13, 1909. (From phot, by Hesselbo.) The Fucus serratus-association was especially luxuriant in Hafnarfj6r5ur on both sides of the fjord, some distance from the head of it, and occurred lowest of all in the Ffzracece-community. In the Vestmannaeyjar the species grew scattered here and there, but was not found in the most exposed places. Sometimes on ex- posed coasts it was found growing in pools in the middle of the littoral zone. The Epiphytes and Intermixed Species of the Fucus- belt. As is well known, an abundance of epiphytic vegetation exists in the Fucus-beli, at any rate at times. Many of the species grow socially and often occur in such quantities, that they set their stamp upon the vegetation. These species may be divided into those 8* 116 H. JONSSOX which grow exclusively, or by preference, on the 'Fucacece, and those which occur as frequently on other substrata and must consequently be considered chance visitors. Only one single species, Polysiphonia fastigiata, is exclusively confined to the Fucus-be\l (Fig. 7). It is well known that this species grows only on Ascophyllum nodosum, into the frond of which it puts its rhizoids,1 although its distribution does not coincide entirely with that of the latter. While Ascophyllum is common everywhere along the coast of Iceland, Polysiphonia fastigiata is common only in S. and SW. Iceland; it has been found, also, in a single place in NW. Iceland. Thus it keeps to the warmer parts of the sea off the coasts of Iceland, but even within this area it may be absent from coasts where the sea-water mixes abundantly with the fresh water. It did not occur for instance at Borgarnes, nor on the nearest islands, although Ascophyllum occurred in great quantities together with Fucus vesiculosus and Fucus inflatns. There the sea-water is freely mixed with water from the large glacier-torrent of Hvita in Borgar- fjor5ur. Polysiphonia fastigiata does not seem able to endure such water, but further out, along the fjord, where the water becomes more saline, it grows in the ordinary manner. Of those species which grow by preference on Fucacetv, Ela- chista fucicola and Uluella fucicola may be mentioned. The first of these is the most important and often occurs in wonderful abun- dance, Ulvella forms a much less important part of the vegetation, more especially on account of its minute size. Besides these two, other species frequently occur which just as often, or even more often, grow on other substrata. Of these Pij- laiella littoralis plays a very important part in the vegetation, espe- cially in the spring; then it sometimes occurs in such quantities that it almost covers large stretches of the coast. Ulothrix flacca also frequently occurs in great abundance, as may also be the case with Isthmoplea sphcerophora. In spring and early summer Ecto- carpus tomentosns and Ectocarpns fascicnlatns are found growing to- gether socially on Fucus in/latus (both of them in S. and SW. Ice- land). Ectocarpns confervoides is also a rather frequent epiphyte. In addition, Monostroma and Enteromorpha intestinalis may be mentioned, on which, again, epiphytes can grow (as for instance Chantransia- species); also Ulothrix pseudoflacca, Acrosiphonia, Ceramium rubrum, Ralfsia verrucosa, Conchocelis rosea, Porphyra umbilicalis and others. 1 Gertrud Tohler- Wolff, in Beihefte z. Botan. Centralbl.. Bd. 24, 2. Abt.. 1909. MARINE ALGAL VEGETATION 117 It is usually the case that it is the older parts of the fronds which are most covered by epiphytes; this agrees well with the fact that the older fronds have a rougher surface than have the younger ones, and therefore retain the spores better. The epiphytes Fig. 7. Polysiphonia fastigiata on Ascophijllnni which is hanging on the face of the rock. Reykjavik, Aug. 13, 1909. (From phot, by Hesselbo.) on the Fucacece do not appear to arrange themselves in strata as do the Laminaria-epiphytes. The intermingled species do not play any special role in the Funis-belt, as regards the vegetation; they consist partly of species which grow on the Fucacece and partly of species which ordinarily grow outside the Fucus-heli. Of these species the fol- lowing may be mentioned : 118 H. JONSSON Monostroma Grevillei. Ulothrix flacca. Monostroma groenlandicum. Pylaiella littoralis. Ghordaria flagellifbrmis. Cladophora rupestris. Acrosiphonia. Enteromorpha intestinalis. Porphyra umbilicalis. Chaetomorpha tortuosa. The Under-vegetation. This is found commonly distributed, and varies considerably, both as regards luxuriance and the species composing it. The luxuriance seems to increase with the degree of exposure (except perhaps in the most exposed places), and then the composition of the species is also changed, as species which pri- marily belong to a lower belt extend higher up, probably on ac- count of the frequent movements of the sea which cause the de- siccation-period to last but a short time. The undergrowth must be regarded as a kind of shade -vegetation; during low-tide it is completely covered by the Fucacece, and at high-water the intensity of the light is also subdued by them, as they float on, or rise with, the wrater, and are moved backwards and forwards by its ripples. The greater part of the species of the undergrowth do not occur at the height of the Fiicus-beli in places open to the light, but occur frequently and abundantly in shaded localities, although these may be found at the upper boundary of the F^ais-belt as, for in- stance, in depressions in the talus of debris, upon the under side of overhanging blocks of stone. This vegetation may therefore be justly termed the shade-vegetation of the littoral zone. The species are normally developed and cannot be compared with the shade- forms of light-plants belonging to the land-vegetation. On the other hand, the shade-vegetation of the littoral /one actually corresponds with the shade-vegetation of the land, such as the fern-vegetation and the liverwort-vegetation in the lava-fissures. The under-vegetation belongs in part to the shade-vegetation (see p. 123) and is divided into several associations; here it is dealt with under the Fuciis-beli as its strictly littoral distribution almost coincides with that of the latter, and it is just as dependent on the upper-vegetation, or even more so, as is the undergrowth in a coppice. As regards the relation between the upper-vegetation and the undergrowth in the F«ctzs-belt the main advantage is, I think with the undergrowth ; although during low-tide, the upper-vegeta- tion may gain some advantage from the fact that the layer of living plants under it retains more water than does the bare sur- face of the rock. MARINE ALGAL VEGETATION 119 The following species of the under-vegetation occur in such quantities that they form associations of considerable extent which are usually pure, but may sometimes be mixed: Hildenbrandia rosea.1 Gigartina mamillosa. Rhodochorton Rothii. Ceramium rubrum. Sphacelaria britannica. Callithamnion Arbuscula. Polysiphonia urceolata. The Hildenbrandia-associaiion is most extensively distributed, and is found everywhere along the coasts. It usually occurs as a pure association, though, in many places, Verrncaria mucosa is inter- mingled with it. Hildenbrandia may also occur at the same height outside the Fucus-covering, but is then if the light is not sub- dued by other plants or projecting rocks usually of a darker colour. The Rhodochorton -association often covers extensive areas, like a dense, red carpet of felt. Most often it is pure, and appears, as far as its littoral distribution is concerned, to be confined mainly to the shady parts of the littoral zone. A few of the other shade- species may be found intermingled with it in small numbers, which fact is of minor importance, whereas the occurrence of the epiphyte Pleurocapsa amethystea is of great importance, as this species is found only on Rhodochorton Rothii. R. Rothii does not occur below the Fzicus-belt of the littoral zone until it occurs again at a considerable depth on Laminaria stems; thus it is found in two distant belts: the shade-belt of the littoral zone and the Lami/iarza-belt. A Sphacelarietum (S. britannica) occurs almost quite pure in many places over rather large surfaces; it forms a dense covering on the rock as does R. Rothii. Sphacelaria radicans, Polysiphonia urceolata and Rhodochorton Rothii occur intermingled. Polysiphonietum. P. urceolata occurs in a similar manner, covering rather large surfaces of rock. In places where the species forms extensive mats, it is usually very low in growth, yet alwrays considerably higher than Rhodochorton Rothii. A thin layer of fine clay may often be seen to have accumulated between its basal parts. A Gigartinetum occurs as under-vegetation, especially in more exposed places, and is then generally found in the lowest part of the Fncns-be\i or in the Fucns zn/7a/us-association ; it is a direct continuation upwards of the G/^ar//na-association below. 1 Arranged in accordance with the frequency, the most frequent coming first. 120 H. JONSSON In addition Ceramieta formed by Ceramium rubrum, and Cal- lithamnioneta formed by Callithamnion Arbnscnla, occur distri- buted in patches, especially in the Vestmannaeyjar. The following species, growing in a more scattered manner, occur also in the under- vegetation of the F«c«s-belt: Delesseria alata. Plumaria elegans. Lithothamnia. The species of the under-vegetation thus belong principally to the red algas; only two species being brown, one blue-green and one being a lichen. The majority of the species belong to the shade- vegetation of the littoral zone. 4. The Enter omorpha-association. Of the Enteromorpha-species E. intestinalis is the most common and the most variable. The variability, no doubt, principally de- pends upon local conditions, possibly also, on the age of the species. The typical form is extremely common in the upper littoral zone, but occurs most luxuriantly in the water-filled depressions. As a rule, the plants have a social growth, although they rarely cover large areas. E. prolifera and E. clathrata, on the other hand, occur in such abundance that it might be justifiable to speak of an asso- ciation formed of branching Enteromorphas. These species are found in greatest abundance in the middle littoral zone, often in empty depressions in the lower part of the Fzious-belt proper, and may sometimes extend beyond the boundary of the lower littoral zone, where they then encounter semi-littoral associations, for example, Dictyosiphonetum, Chordarietum and others. While the typical form of E. intestinalis seems to flourish during the spring, the branching forms E. prolifera and E. clathrata are noticeable in the summer. I have known these two species to occur predominantly in June, July and August on the same substratum upon which Urospora Wormskioldii was growing in great quantities in the month of May. Enteromorpha Linza is also a summer species, and, although it forms in several places pure Enteromorpheta of lesser extent, plays only an inferior role in the E/?/e/*o/norjo/7a-association. In E. Iceland, on the point between Seydisfjordur and Lo5mundar- fjordur, at the height of the lower part of the Furns-belt and a little lower, I found a C/i/orop/iycece-vegetation which must be con- sidered to belong most nearly to this association. A Monostro- MARINE ALGAL VEGETATION 121 metum of Monostronm groenlandicum was here found in abundance, often entirely covering the boulders and occurring on them, like U. flacca, closely pressed to the stone-surface with a radiating ar- rangement. In addition, an abundance of E. intestinalis f. compressa and a few Ulothrix flacca were found. A little lower down, a Uro- sporetum (U. Wormskioldii) occurred, which can scarcely be separated from this association; it seemed to form a connecting link with the semi-littoral communities. Cladophora gradlis occurred intermingled in the E. clathrata- association, yet without playing any important role in the vegetation. A similar association, especially an Enteromorphetum of E. in- teslinalis, is, beyond doubt, very common in other countries. It seems to be more poorly developed in Greenland (Rosenvinge, 63, p. 205) than in Iceland; in the Fasroes (Borgesen, 12, pp.714, 715), on the other hand, the Enteromorpha intestin a Us- vegetation seems to be more luxuriant than in Iceland. 5. The Acrosipho nia-association. This association consists of decidedly filiform, much branched alga3, which grow very socially. During the period of desiccation they retain the water as a sponge retains it. This circumstance is mentioned with regard to some of these species by Rosenvinge (63, p. 202), who says of Cladophora arcta (= Acrosiphonia incurva, cf. Jonsson, 32, p. 43) that in Greenland it behaves during low- tide like a sponge saturated with water.1 The principal species in this association are Acrosiphonia albe- scens and A. incurva. They form individually, pure Acrosiphonieta in the upper and lower littoral zones, where they cover flat stones with a densely matted vegetation. On flat rocks the association may be somewhat widely distributed, in other places it has more the character of scattered Acrosiphonieta. The density of the vegetation is due to the structure of these species. As the basal part of the principal axis is too weak to carry the plant when its branch- system has developed, lateral rhizoids, which slope downwards, are developed at an early stage from the principal axis itself and from the lower branches. These rhizoids often form creeping filaments from which arise erect shoots, which can be detached from the parent-plant and thus become independent individuals. The principal 1 Possibly this is the case in an even higher degree with Callithamnion Ar- buscula (cf. Borgesen, 12, p. 726). 122 H. JONSSON axis dies awav from below and in this way branches, or svstems »- v */• of branches, with well developed rhizoids are detached from the parent-plant and become independent individuals (cf. Jonsson, 32). The rhizoids of one individual become entangled with those of others, and in this way the basal parts of the entire vegetation be- come matted together. In addition to this, the lower branches are hook-shaped, or bent downwards and outwards; thus it easily hap- pens that they are entangled between the branches of other indi- viduals, which further increases the matted condition. The upper branches are directed obliquely upwards, without being entangled with the branches of other individuals. An A. albescens-associaiion occurs everywhere along the coasts, but on very exposed coasts the species generally grows dispersed. Here it occurs in separate tufts which are, no doubt, composed of several individuals, densely matted together below, but branching upwards in various clusters, which are themselves densely matted together by their hook-shaped branches ; the branch-systems above are, as usual, free. These characteristically matted branch-clusters a) are probably an effect of the heavy beat of the waves; the matted Acrosiphonia "tufts" will not be able to resist the force of the waves and, even at an early stage, will become divided into very densely matted branch-clusters. Acrosiphonia albescens occurs both in the upper and the lower littoral zone, while A. incarva belongs to the lower littoral zone, but yet also extends upwards into the pools of the upper littoral zone, and below low-water mark. S p o n g o m o r p h e t u m . Spongomorpha vernalis grows very so- cially, but never forms such a dense vegetation as does Acrosiphonia, although it is sufficiently dense to characterize the spot. The species has been found in only a few places; it occurs in abundance at Grotta, where it grows both on a stony substratum and also as epiphytic vegetation in the Corallina-Gigartina-beli and in the Poly- siphonia urceo/ata-asspciation of the lower littoral zone. A Cladophoretum formed of Cladophora sericea f. (see under Tide-Pools) also belongs to this association. It forms a densely matted belt in pools high up in the upper littoral zone. As a rule it follows the edge of the water round the entire pool. Cladophora rupestris ought also to be considered to belong to the Am>szp/7o/na-association. An association similar to the /Urosz/j/zo/i/a-association is, no MARINE ALGAL VEGETATION 123 doubt, very common in adjacent lands. During low-tide the upper- most free branches appear lo be more exposed to desiccation and possible death in Greenland than in Iceland; in the latter place it occurs rather frequently. Both in Greenland (Rosen vinge, 63) and in the Faeroes (Borgesen, 11, 12) the Am)s//)/?o/na-association is well represented. bb. The Shade-vegetation. To this vegetation are referred certain littoral associations which occur exclusively in shady places. The shade is produced by the light being subdued both by a covering of living plants and by projecting blocks of stones. Thus the greater part of the under- vegetation-associations of the Fucas-belt mentioned above belong to the shade-vegetation. In addition, the shade-vegetation grows, as already mentioned, on the under side of overhanging rocks on their in-sloping sides. The vegetation is divided into several associations which have already been mentioned in connection with the under-vegetation of the Fucus-belt (cf. p. 118). The frond differs in form and structure in the various species of this vegetation. It is a feature common to the whole of the shade-vegetation to be low and dense in growth. The frond of Hildenbrandia is a crust which covers the rocks; the others are finely branching; some, as Sphacelaria and Polysiphonia. are bush-like and are richly and finely branched; Rhodochorton has a similar mode of branching, but to a considerably less degree. These three species often propagate vegetatively by means of run- ners, which increases the density of the vegetation. Plumaria is closely and distichously branched ; it is tightly adpressed to the surface of the stones during the time of lowr-tide and retains much water betwreen its branches, considered from a biological point of view it might almost be regarded as leaf-like during the period of desiccation. 6. The Hildenbrandia- association belongs almost exclu- sively to the under-vegetation of the Fucus-beM, and of the shady localities in the littoral zone (see above, p. 118). 7. In shady places the Rhodochorton -association, the Sphacelarietum britannici, and the Polysiphonietum ur- ceolatse behave in a similar manner outside the Fucus-beli as they do within it (see above). They may occur either as pure as- 124 H. JONSSON sociations, and are then dominant here and there in patches, or the species may be found intermingled with each other. Besides these associations a Plumarietum, consisting exclusively of Plu- maria elegans, is found in many places. This association is darker in colour than the other red-alga3 associations of shady places; it occurs frequently in S. and SW. Iceland. Intermixed in it occur Callithamnion scopnlornm and Chantransia virgatnla. Rhodochorton Rothii is sometimes found high up in the littoral zone on flat rock-surfaces exposed to the light and to the heat oi' the waves, but then it grows in small globular cushions (f. globosa). Consequently, this globetum of Rhodochorton Rothii does not be- long exclusively to the shade-vegetation. On flat surfaces in the shade the usual arrangement is that the decided shade-associations occur at the bottom, where the light is feeblest, e. g. the Sphacelarietum, Rhochortonetum, Polysiphonietum, Plumarietum; at the top, where the illumination is stronger, light- loving species occur, e.g. Pylaiella, Ulothrix, or others. Near Rey- kjavik a vertical section of such a surface showed uppermost, at the edge, Pylaiella littoralis, next Rhodochorton Rothii f. globosa, and lowest of all Plumaria elegans. In a grotto in the Vestmannaeyjar, where the illumination was very feeble, Enteromorpha intestinalis f. micrococca occurred on the roof, Ceramium acanthonotum grew rather high up on the walls, and Plumaria elegans, together with Delesseria alata, formed a belt lower dowrn the walls. A vegetation corresponding to the shade-vegetation appears to occur in Greenland where Hildenbrandia rosea, Ralfsia clauata and Verrucaria mucosa form the undergrowth in the Fucus-belt and in other places (Rosen vinge, 63, pp. 198 and 203). Rhodochorton Rothii and Sphacelaria britannica also appear to grow in a similar manner in Greenland (Rosenvinge, 63, p. 205). A comparison with the Fasroes showrs some difference. The vegetation in the grottoes in the Faroes (Borgesen, 12, p. 739), however, resembles in its main features the Iceland shade-vegetation, and, in addition, a similar vegetation is beyond doubt to be found in fissures and clefts of the rocks in the Faeroes, but the mode of occurrence of the species is not the same. Thus, in the Faroes (Bor- gesen, 12, p. 711), the Hildenbrandia-association appears to have a much wider distribution upwards and to occur in fully illuminated localities. Rhodochorton Rothii seems to occur in a similar manner MARIN7E ALGAL VEGETATION 125 in the Faeroes (Borgesen 1. c., p. 718) as it does in Iceland, with this difference, however, that the species extends higher upwards in the Faeroes and, also, frequently grows in fresh-water in the latter place.1 Sphacelaria britannica grows similarly in the Faeroes (Borgesen, 13, p. 432). According to Foslie (18) a similar shade-vegetation appears to occur in northern Norway. This vegetation is a natural upward continuation of the semi- littoral Polysiphonia urceo/afa-association mentioned below. cc. The Vegetation of Tide-Pools. Where there is a rocky coast, depressions of varying sizes occur everywhere in the littoral zone. These depressions may be divided into two groups: The tide-pools of the upper littoral zone, and the tide-pools of the lower littoral zone. The plant-growth in these tide- pools does not form a vegetation-unit, and is chiefly composed of species either scattered or of social growth which belong to the littoral or semi-littoral communities ; they are, however, here treated separately in order to further characterize the illustration of the littoral zone given in the above description. Tide-Pools of the Upper Littoral Zone. Of these the uppermost, which are almost on the level of Fucus spiralis, are of the greatest interest, as they often contain species, or forms of species, which do not occur elsewhere in the littoral zone. These species are Cladophora sericea f. and Fucus inflatus f. linearis. The water in the uppermost pools is not replenished with the recurrence of each high-tide, and this is especially the case in calm weather during neap-tide; if, simultaneously, dry weather occurs, the water evaporates, and this the vegetation cannot endure for any length of time. In rainy weather, on the other hand, the pools are filled with fresh water, and should this happen during neap-tide the salinity would be insufficient for any of the species to exist. At spring-tide the water in the pools becomes mixed with a fresh supply of sea-water, and the vegetation then lives, for a time, under good conditions. Thus, there are times during which the conditions in the pools are not favourable to algal life. Cladophora sericea grows very socially in many places and, in 1 When I visited the Faeroes in October, 1897, I was at first surprised to find R. Rothii growing in streams at and above the upper limit of flood-tide; such a situation for it in Iceland was unknown to me. 126 H. JONSSON small pools, often forms a continuous fringe, the upper branches either reaching to or lying on the surface of the water. In such pools Chcetomorpha tortnosa sometimes occurs in fair abundance loose upon the surface of the water; its filaments are usually densely matted together. Fucus inflatns f. linearis grows even more socially. At times it is almost the dominant species in the smaller pools, although, rather frequently, several other species are found intermingled with it. This form is a biological variety of Fucus in flatus which, in the pools, exists evidently under less favourable circumstances; this is one cause of its small size and feeble structure, although another is that it is not exposed, to any extent worthy of mention, to the beat of the waves. Rosen vinge has especially shown this to be the case as regards Greenland. Rosen vinge explains the frequent occurrence of this form in the pools by the fact that the eggs of F. inflatns are carried into the water-filled depressions by the move- ment of the water and accumulate there. This explanation is un- doubtedly correct, and, as Rosen vinge points out, all intermediate stages between the feeble pool-form and the typical form can be demonstrated. To any one who has seen this endless variation in a/ nature, it seems so certain that it is due to the influence of outside factors, that experimental proof is almost superfluous. The uppermost pools are generally extremely poor in species ; and, besides those already mentioned, only Enteromorpha inlestinalis occurs in any great abundance. Where the coast is exposed , the uppermost pools may, however, have a resemblance to the lower ones of the upper littoral zone a natural consequence of the exposure. On more exposed parts of the coast a fringe of small Monostroma Grevillei is often found, almost on the surface of the water itself. At one place in E. Iceland I took the temperature of such a submerged plant-covering, and the thermometer showed 20° G. (June 13). The vegetation was also somewhat injured, and evidently did not prosper under these conditions. High up on a rocky coast to the south of Vattarnes in E. Ice- land, I came across a pool-vegetation. I did not measure the alti- tude of the spot, but I do not think that I was greatly mistaken in estimating it at 70 — 100 feet above sea-level. As regards the place, I noted in my dairy "High rocky coast, land-plants grew round the alga-pools. The water in the pools must be replenished with rain and heavy surf, which sometimes fail for long periods during MARINE ALGAL VEGETATION 127 the summer. Many pools filled with decaying and dying algae." The occurrence of algae so high up can be imagined only on a very exposed coast facing the open sea, and even there it would be exceptional. The dominant species of the vegetation in these high-lying pools consisted of Fncns in flatus f. linearis. Intermixed occurred Chceto- morpha Melagonium which I never before had seen so high above sea-level Acrosiphonia sp., Dictyosiphon foeniculaceus , Pylaiella littoralis and Enteromorpha intestinalis f. typica. The undergrowth consisted of Hildenbrandia rosea. In another somewhat elevated and exposed place in the vicinity of Vattarnes, a number of small pools were found just below the Verrucaria maura-belt. Hildenbrandia and Verrucaria mncosa grew on stones at the bottom of these pools; in addition, Fucus in flatus f. linearis occurred fairly abundantly, although dying in the pools, which were deficient in water. The lower pools in the upper littoral zone have a much more luxuriant vegetation. Here various littoral and semi-littoral species can grow in fair abundance. The water in these pools is replenished at each high-water, so here, the plants exist under fairly favourable conditions. The following species may occur dominantly or abun- dantly: Fucus in flatus f. linearis, Halosaccion ramentaceum, Mono- sir oma fuscum, Enteromorpha intestinalis, Monostroma Gremllei, Dic- tyosiphon foeniculaceus and Castagnea virescens. Many species are found growing scattered, of which the fol- lowing are most frequently met with: Polysiphonia urceolata. Phyllitis fascia. Porphyra miniata. Pylaiella litoralis. Rhodomela lycopodioides. Scytosiphon Lomentaria. Rhodymenia palmata. Chsetomorpha Melagonium. Chorda Filum. C. tortuosa. Chordaria flagelliformis. Cladophora gracilis. Coilodesme bulligera.1 Monostroma undulatum. Ectocarpus confervoides. Ulothrix flacca. Elachista fucicola. Urospora Wormskioldii. The under- vegetation often consists of Ralfsia densta , Litho- thamnion circumscriptum and Hildenbrandia rosea. Taken on the whole, the majority of the littoral species un- doubtedly may be met with in the pools, and it should be especially 1 It may be observed, as regards Coilodesme, tbat it is tbe more inflated the lower the water is in the pools. 128 H. JONSSON noted that, when the semi-littoral species occur in the upper littoral zone, they are usually confined to the pools, a fact which proves that they do not essentially belong to the upper littoral zone. b. The Semi-littoral Zone. The upper boundary of this vegetation is just at the lower boundary of the Fucws-belt. The vegetation covers the lower littoral zone, and extends, usually in the form of stragglers, below the limit of low-tide, down between the uppermost extensions of the vegeta- tion of the Lamz/?arza-belt, which in many places reach right up to the limit of low-tide. The semi-littoral communities consequently grow side by side with the Lamz'nana-associations below the limit of low-tide, to a depth of about 10 metres; they also occur as epi- phytic vegetation on the Laminarice at the depth mentioned. Thus, the semi-littoral zone is that between the lower edge of the Fucus-be\i and the Laminaria-belt. The breadth of the zone depends on the slope of the coast. If the coast is steep the zone is narrow, but if the coast slopes gently, it is broad. In this zone no single dominant community of extremely social and large species occurs, like, for example, the Fucacece-community in the littoral zone, and the Laminariacece-commumly in the sublittoral zone. The IS zone appears to lie too low for Fncacece and too high for La- minariacece. Green and brown algae occur in abundance in the semi-littoral zone, but the Rhodophycece are most richly represented. The semi- littoral associations occur both laid bare during low-tide and also continuously submerged to the depth mentioned. It is easy to follow the upper boundary of the zone, even in places where the Fncacece are absent. The lower boundary can also be ascertained with comparative ease, if the large Laminarice only are followed, and if the observer is not led astray by the small La- minarice, which may occur in the depressions and pools of the lower littoral zone. The semi-littoral zone comprises several associations which play a rather considerable part in the vegetation. By authors who describe the algal vegetation on coasts where there is a change of tide these associations are generally considered to belong partly to the littoral and partly to the sublittoral vegeta- tion. On coasts where there is no tide, the littoral vegetation has, MARINE ALGAL VEGETATION7 129 however, been subdivided. Reinke (58, p. 10) divides the littoral region near Kiel into an upper zone which is always laid bare at low-water, and a lower zone at a depth of 2 --4 metres. So far as I can judge from the species enumerated, Reinke's upper zone corresponds approximately with the previously mentioned upper littoral zone, while the lower zone corresponds only in part with the semi-littoral zone here described. Gran (24, p. 11) records under the term "sublittoral vegetation" transitional formations which evi- dently, for the most part belong to the semi-littoral communities. On the west coast of Swreden (Kylin, 45) the border-line be- tween the littoral and sublittoral algal vegetations is, on the coast of Bohuslan, at a depth of 3 — 4 metres, and at Halland at a depth of about 5 metres. The upper part of the littoral zone seems to correspond to some extent with the above-mentioned upper littoral zone, while the lower part has something in common with the semi-littoral zone. Where there is no tide, the boundary between an upper and a lower littoral zone is not, I think, so sharply defined. 8. The M onos troma-associati on. This vegetation is composed of relatively large, membranaceous, green algae. Monostroma fuscum and M. Grevillei form associations in a rather considerable number of places in protected localities in the fjords, below the limit of low-tide, and to a depth of at least 4 — 6 metres. There, the association is most sharply defined, although it is fairly frequently found intermingled with other semi-littoral associations. In the lower littoral zone also, an abundant Mono- stroma- vegetation frequently occurs, both on a rocky substratum, and also very often as epiphytic vegetation in other associations, as, for example, the ffa/osacczo/?-association, the Co7'rt//z/?a-association, the Polysiphonia urceola ta -association and others. In water-filled depressions in both the upper and lower littoral zone the Mono- stroma species often occur abundantly. Monostroma Grevillei var. arctica seems to belong more imme- diatelv to the littoral zone. In many places it is found entirely laid »/ */ j. bare, sometimes on a clayey substratum, when it often attaches itself to small individuals of Mytilns ednlis, and sometimes where the substratum of the littoral zone is pebbly. During April and May this vegetation is rather characteristic, because then, the species occurs for the greater part in the inflated stage. The bladders vary The Botany of Iceland I. 9 130 H. JONSSON greatly in size and form and may be at times considerably elon- «/• •/ «J gated, bearing then a striking resemblance to an Enteromorpha. Under calm conditions the inflated stage continues until the spores at the apex of the frond are ripe. In seyeral places in the fjords Monostroma Grevillei var. typica forms a luxuriant, characterizing vegetation at a depth of about 3 — 5 metres. Monostroma fuscum very frequently forms associations of con- siderable extent in the fjords at a depth of about 4 metres. The specimens of this species which occur there are generally very large (f. grandis). These large specimens are rather frequently found at low-water mark, detached or floating at the water's edge, and then it may generally be taken for granted that a Monostrometum exists further out at a depth of about 4 metres.1 Both in the fjords of E. Iceland and in Eyjafjor5ur in N. Iceland this Monostrometum oc- curred in the same characteristic manner, viz. alternating with a Chordarietum, a Dictyosiphonetnm (D. foeniciilaceus) , a Halosaccione- tum, and a Rhodymenietum. Generally the order was that Rhodymenia grew deepest (as deep as 12 metres), and M. fuscum most frequently uppermost (at about 4 metres). Alaria and Laminaria saccharina, in addition, may be found growing scattered in such places, which makes the character of the vegetation still more heterogeneous. Monostroma undulatum does not occur so abundantly as do the other Monostroma species. It is found growing rather luxuriantly, however, in pools in the lower littoral zone, and on the stems of Laminaria in comparatively low water. Ulva Lactuca, also, is most nearly related to this association. This association is very common in the Faroes (cf. Borgesen, 12, pp. 731 and 764), and occurs also in Greenland (Rosenvinge, 63). 9. The Chorda-association. This association is composed of rather large, brown algae. The fronds are either non-branching, thick filaments (Chorda, Scytosiphon), or else branched, as in almost all the others; one, Coilodesme, how- ever, is almost leaf-like. This association has an insignificant distribution, and is found usually in patches, where the substratum is clayey or somewhat muddy. It occurs both in the lower littoral zone above low-water mark, and to a depth of at least 4 — 6 metres. The members of the 1 Stromfelt (70, p. 11) mentions this Monostrometum at Eskifjor5ur. MARINE ALGAL VEGETATION 131 association are as follows: Chorda Filum, C. tomentosa, Chordaria flagelliformis and Dictyosiphon foeniculaceus. Scytosiphon Lomentaria, Castagnea virescens and Coilodesme bnUigera are also most nearly related to this association. Some of the species occur in such abun- dance that they may be said to form an association; this applies more particularly to Chorda Filnm. A pronounced Chorda Filuiu- association occurs in many places at a depth of almost 3--4 metres (measured during low-water of spring-tide). The individuals are ex- tremely long, about 6 metres and, to a great extent, float on the surface at low-tide. The substratum is, as a rule, clayish with pebbles here and there. This association is often almost pure, but scattered individuals of Laminaria saccharina are not rare, although they play no essential role in the vegetation. Chorda Filum also grows above low-water mark in the lower littoral zone, but there it does not really form associations, although it may grow luxuri- antly in many places. Chorda tomentosa is also social and grows more luxuriantly below low-water mark than above it. It often occurs in abundance as a component of the sublittoral vegetation. Chordaria and Dictyosiphon foeniculaceus are species which grow socially, and rather frequently a Chordarietum or a Dictyosipkonetum is found, usually with a limited distribution both below low-water mark and above it, especially in water-filled depressions in the littoral zone. The remaining species which have been mentioned are also of rather social growth in many places, both below and above low-water mark, especially in pools. Coilodesme, for example, was rather frequent in tide-pools in E. Iceland, and sometimes at a very high level, but then the individuals were frequently inflated ; in SW. Iceland the species grew at and below low-water mark in company \vith Chorda Filum and Saccorrhiza. With the exception of the above-mentioned Chorda Fz7u/?i-asso- ciation, this vegetation occurs usually in patches, distributed be- tween other associations. Thus it is closely connected with the Enteromorpha clathrata-association at the boundary between the upper and lower littoral zones, and with the /?/?o(/yn?e/na-association and the Monostroma /izscz/m-association at a depth of about 4 metres, as has been previously mentioned. In the Faeroes a similar vegetation occurs as a part of the Stictyosiphon-Sissociation (Borgesen, 12, pp.762 — 763). 132 H. JONSSON 10. The Community of Rhodvmenia. «/ \j The species are Rhodymenia palmata and Halosaccion ramen- taceum, both of which occur along the coasts in great abundance as pure associations. These two associations most frequently ac- company each other, and seem to demand somewhat similar con- ditions of life. Halosaccion, however, extends the higher up in the littoral zone, but is then almost always submerged in water-filled depressions in that zone, while Rhodymenia extends the further down below the limit of low-tide. In Rhodymenia the form of the thallus is leaf-like and branching; in Halosaccion it is round and excessively or slightly branching. The Rhodymenia-association. On regarding a steep rocky coast, where the succession of the associations is always most easily observed, it is seen that a luxuriant Rhodymenia- vegetation com- mences even at the lower boundary of the Fuciis-belt. In many places the species occurs so socially that it predominates as a characterizing plant over relatively large stretches, both in the lower littoral zone and far below the limit of low-tide, to a depth of about 12 metres. On a gently sloping rocky coast, the upper boun- dary of the association is not so sharply defined, but if proper allowance is made for the depressions and elevations in such a littoral zone the boundary is fairly easily ascertained although it may then lie in very many curves. On a gently sloping pebbly coast, the same regular boundary is found as on a steep rocky coast and at about the same height, although the Fncacew-comm unity is not present there. Various algae may occur on stones under the fronds of the Rhodymenia, as, for instance, Hildenbrandia rosea, Ralfsia clauata, Sphacelaria radicans and others; also a number of epiphytes, espe- cially Myrionemacece and Ectocarpacece, often grow on old individuals of Rhodymenia palmata. Rhodymenia occurs also epiphytically on the stems of Laminaria in the upper part of the Laminar za-belt. The Rhodymenia- vegetation does not seem to be so luxuriant in Greenland (Rosen vinge, 63) as in Iceland. Possibly this is true also of the north of Norway (Foslie, 18). In the Faeroes the Rhodymenia-\egeialion is luxuriant, but seems to differ somewhat from that of Iceland, as it consists of a form with lower and more dense growth and narrower branches; this form also extends further up (Borgesen, 12, p. 727). MARINE ALGAL VEGETATION 133 The Halosaccion -association also commences below the lower boundary of the Fuciis-belt, and extends from there to a depth of about 5 metres below low-water mark. The species grows very socially, entirely covering the rocky substratum upon which it grows: it is very variable, and, while f. densa seems to extend rather high up, in pools in the littoral zone, it is f. robusta which, as a rule, reaches to the greatest depth. F. subsimplex is very com- mon just below the Fuazs-belt, where it grows so densely that it covers the substratum completely. It is very often interwoven with byssal-threads of Mijtilas edulis. In this form, as in the laid-bare /'. robusta, inflated shoots occur fairly frequently. As is the case with Rhodymenia palmata, this species is very frequently decoloured in the lower littoral zone; yet in both species the lower part of the frond is, as a rule, reddish in colour. Of the epiphytes, Elachista fncicola is especially frequent, and various other species may also occur closely applied to Halosaccion, such as Porphyra miniata, Monostroma fuscum, M. Grevillei and others. At times the Monostroma species may be so abundant that they may be said to play a distinct role as epiphytic vegetation. Among the Halosaccion individuals, Damontia filiformis, Chcetomorpha Mela- gonium and others often occur growing scattered at the very limit of low-tide. A similar Halosaccion- vegetation occurs in northern Nor- way (Foslie, 18), and probably on the arctic coasts as well (Kj ell- man, 36; Rosen vinge, 63). In the Faeroes the association is poorly represented (Borgesen, 12, 13). 11. The Polysiphonia urceolata-association. At and about low-water mark in the lower littoral zone, an association is found consisting principally of excessively branching, red algae which I will name after the species Polysiphonia nrceolata, which is dominant in the association. This vegetation frequently occurs on a flat or convex substratum of rock (lava-substratum) and often forms, especially in SW. Iceland, for example at Grotta near Reykjavik, a broad belt around Corallina- pools, near low- water mark. The species which occur in greatest abundance in this asso- ciation are Polysiphonia nrceolata, Cystocloninm pnrpurascens and Rhodomela lycopodioides. The Polysiphonia urceolata-association is very luxuriant 134 H. JONSSON in many places at about low-water mark, and often covers rather large stretches of rock with a dense but, at times, low vegetation. As previously mentioned, it grows luxuriantly on flat rocks, without any protection whatever during the period of exposure, which is cer- tainly short, and indeed exceedingly short during a heavy sea. Here, as in the upper littoral zone, the association occurs also on the face of the rocks, and the species seems as a rule to be larger than when on flat rocks. The association appears to have an extended vertical distribution since, as previously mentioned, it occurs as shade- vegetation or undergrowth in the upper littoral zone; further- more it must be assumed that it has a larger sublittoral distribu- tion than is shown by the dredgings, as it has been found outside the 10-metre contour (see below). From what I have seen, the stretch of shore just above and somewhat below low- water mark is essentially the home of this association. It usually grows on a rocky substratum and on Laminaria stems at no great depth. Cystoclonium purpurascens grows socially in many places both on a rocky substratum and on Gigartina. Sometimes it occurs in such abundance as an epiphytic vegetation in the Gigartina- Corallina-beli, that it is the alga which characterizes the vegetation, and not until the plants are moved aside is it seen that they are attached to a living substratum. On Cystoclonium various epiphytes occur, for example, Chantransia, Monostroma Grevillei and others. Rhodomela lycopodioides very frequently grows scattered, and then plays only an insignificant role in the vegetation ; but Rhodomeleta of limited extent also occur, often in contact with the Polysiphonia wrceo/a/a-associations. Ceramium rubrum, C. acanthonotum and Callitham- nion Arbuscula really belong to this association. In many places these species grow fairly socially and Ceramieta of C. rubrum and C. acanthonotum and Callithamnioneta of limited extent occur both on rocky substrata and on one of Gigartina; however, I think that these species occur too sparsely to be termed association-formers. Sphacelaria radicans also belongs to this association. Often, even at low-water mark, it covers flat stones with a dense vegeta- tion, but though these small Sphacelarieta are considerably distri- buted, yet they cannot be called associations. In the Faeroes a similar Polysiphonia urceolata-\egeiaiion occurs (Borgesen, 12, p. 731). MARINE ALGAL VEGETATION 135 12. The Community of Gorallina. The members of this community are Corallina of/icinalis, Gigar- tina mamillosa, Chondrus crispus and Ahnfeltia plicata. These species occur as a rule in luxuriant and more or less sharply defined as- sociations, which very frequently occur together, and may therefore be regarded as nearly related to each other. The fronds are branched, and their consistency is on the whole V firm because, as is known, Corallina is encrusted with calcium car- bonate, Ahnfeltia is horny and Gigartina and Chondrns are car- tilaginous. Gigartina and Corallina generally occur in a belt at the limit of extreme low7- water; in my diary I have always called this belt the Gigartina-Corallina-beli. Gigartina, however, extends higher up; on exposed coasts right up to the Fzzczzs-belt, and even into that as under-vegetation ; but, where Gigartina and Corallina meet, there is most frequently a mixed belt. These associations belong to S. and SW. Iceland. The Cora///7?a-association occurs most luxuriantly about low- water mark. The species grows extremely socially, and entirely covers depressions in the littoral zone which are more or less filled with water. The association occurs on somewhat exposed and also on very exposed coasts, but almost always in depressions surrounded by rocks upon which the waves break; yet I have seen it, where the exposure is but slight, covering the face of rocks below low- water mark. Onlv rarelv have I obtained Corallina bv dredging in V */ \s greater depths, as, for example, on the north coast, at a depth of about 14 metres, where it seems to be present in abundance. It may be assumed with certainty, however, that this association has a much wider distribution below the limit of low-tide than is shown by the dredgings, and writh exceptional low -tides it can be seen that in many places the Co/'a//z/?a-vegetation covers the rocks as far as the eve can reach. \j On a very exposed coast I have met with Corallina growing socially in water-filled depressions at a considerable higher level, sometimes on a level with the upper part of the Fzzczzs-belt. Here, however, with the exception of the lowest part of the frond, it is quite white in colour, and has evidently strayed outside its real domain. The spores have been carried to this height above the true Cora//z'/? a -be It bv the beat of the waves, and have been retained in */ the depressions. It can also thrive fairly well at this height during 136 H. JONSSON the autumn and winter, both because the light is feeble, and be- cause the increased high seas following stormy weather to some extent compensate for the difference in height. But it is in the spring and summer that it is most evident that Corallina has really extended too high up. According to Borge sen's description (12) it appears to extend still higher up in the Faeroes. In the Cora//z/?a-association an abundant epiphytic vegetation may occur, both of small algae like Chantransia and of larger algae such as Monostroma Grevillei, Spongomorpha vernalis, Acrosiphonia albescens and also the Lea//?esza-associations. Furthermore, Ceraminm rubrnm, Cystocloniam, Delesseria sinaosa and D. sanguined occur, and other red algae, often in great quantities. When to this is added the fact that Gigartina is often abundantly intermingled with those already mentioned, it is easily seen that this mixed vegetation as- sumes quite a different character, according as to whether the brown-red Gigartina and Ceraminm or the light-green Monostroma and Spongomorpha predominate. This characteristic appearance dif- fers entirely from the usually monotonous appearance of the Coral- lina-belt. This epiphytic vegetation must be considered to form as- sociations which do not really belong to the CoraZ/zna-vegetation in any other respect than that of having it for a substratum. The Gigartin a-associatioii is widely distributed on very exposed coasts; it often is of very great breadth and, as already frequently mentioned, extends in under the Fncacece as undergrowth. Where the exposure is less, it does not reach so high; but yet, in the lower littoral zone there is, as a rule, a distinct Gigartina-be\t< which most frequently occurs above the limit of low- tide. The GzV/ar/z/ia- vegetation extends also below the limit of low-tide, but it seems to belong most closely to the lower littoral zone. The belt is quite characteristic, and can often be seen from a distance, as the colour is in strong contrast to that of the Fncaceoe-co mm unity. The species is of very social growth and is usually dominant where it occurs; not rarely, however, a few species occur intermingled with it, especially in the Vestmannaeyjar and at OndverSarnes, two very exposed localities. In the Gigartina-be\\. Corallina may occur lo\vest, Callithamnion Arbnscnla is sometimes plentifully intermixed with the Corallina and occurs also in abundance as an epiphyte, and Ceraminm acanthonolnm occurs rather frequently and often abundantly. In the Vestmannaeyjar Callithamnion scopnlornm occurs rather frequently as undergrowth. In addition, Ceraminm rnbrnm, MARINE ALGAL VEGETATION 137 Delesseria alata, Acrosiphonia, Polysiphonia urceolata, Plnmaria elegans and Rhodymenia palmata occur. Porphyra umbilicalis occurs as an epiphyte in the Vestmannaeyjar, and also Cystoclonium, Polysiphonia urceolata and others. The Gigartina-be\i is very luxuriant in the Vestmannaeyjar and at Ondver5arnes; it may be said to be, on the whole, luxuriant in S. and SW. Iceland. As regards the relations between Gigartina and Corallina, the following observation from the Vestmannaeyjar may be recorded. At Brimsur6, on the south-east side of the inhabited island, where the littoral zone consisted of large, although not par- ticularly high boulders, these were entirely covered above by a V V V dense GzV/ar/z/?a-vegetation, while Corallina formed just as dense a belt around them below. Chondrus crisp us also grows socially in S. and SW. Iceland. It is true that Chondreta of lesser extent occur usually at or near low-water mark , but nowhere does the species occur nearly as luxuriantly as does Gigartina. The C/zonrfriis-vegetation is most luxuriant in the Vestmannaeyjar and at Eyrarbakki. Between the skerries off the latter place, the broad-fronded form grew in abun- dance, both above and a little below the limit of low-tide. In both the places mentioned, which belong to S. Iceland, it grows in such abundance that the vegetation might almost be termed an associa- tion; in SW. Iceland on the other hand, it seems to occur more sparsely, and the small Chondreta may then be considered to be- long to the Gz the depth and the exposure. The Lamina ria digitata-association. This association is very common everywhere along the coasts where there is a rocky substratum, from a depth of about 4 to about 25 metres. On rocky coasts, however, small specimens occur just at the limit of low-tide, and there represent a kind of boundary. Small individuals may occur also in water-filled depressions in the lowest part of the littoral zone. The tendency of Laminaria digitata is to vary in the same manner as L. saccharina, and thus the character of the association differs according to the depth and the degree of exposure, as the forms, individually, grow socially. The typical appearance of the association is determined by the deep-water form, or the typical form, which seems to grow most luxuriantly at a depth of about 10 — 20 metres. Here the species attains its greatest length and, as a rule, the stipe is so strong that it is able to raise the much- divided lamina from the bottom. Just as the forms are connected with one another by inter- MARINE ALGAL VEGETATION 145 mediate forms, so are the various associations connected. If the deep-water character of the association is taken as a starting point, we notice that it changes gradually with decreasing depths on ex- posed coasts, and that somewhere near the limit of low-tide it assumes an entirely different character, which is displayed in the leathery, long-stemmed form with a narrow and slightly-divided frond (f. stenophylla). On an exposed coast this variety might be termed the shallow-water type or perhaps, rather, the surf-type, in conformity with the surf-form of Fucus inflatas. The surf-character \vas very beautifully and typically developed in the Vestmannaeyjar; here f. stenophylla grewT very socially, forming a continuous belt, the upper boundary of which occurred almost at the limit of low- tide. When during low^-tide the waves receded it was verv inter- V esting to see how7 easily the leathery, narrow, slightly-divided laminae moved with the waves, and everywhere, as far as the waves receded, the rocky substratum between the Sfeno/)/ij///a-individuals was quite reddish in colour from the encrusting Phymatolilhon polymorphum. The Laminaria plants were very firmly attached to the rock a fact evidently well known to the fishermen, as they fastened the boat to a Laminaria while we went ashore. The stipe is leathery and pliable, and the plants cling closely to the substratum when the waves recede. I have found /'. stenophylla in other places, although not in such abundance, and not quite so typical. Here the same rule applies as with regard to Fucus inflatus and Laminaria saccharina, that the tendency to vary seems to depend on the greater or lesser movement of the water; on coasts which are somewhat exposed, a Laminaria t%z7ata-association is rather frequently found at about the limit of low-tide, with a character midway between the surf- character and the deep-water character. If we again take the deep-water character of the association as a starting point, and move inwards towards the protected coasts, we see that the character changes again, but in another direction. The stipe becomes shorter and the frond much broader and slit into fewer and broader lobes. Within the fjords, in W. and E. Ice- land especially, the character is entirely different from the deep- water character. Here occur forms with very broad fronds which are either undivided or divided into two, or a few, very broad lobes (f. cuciiHataJ. Generally the depth is about 4 — 20 metres, even deeper occasionally. I have found associations with this character The Botany of Iceland. I. JQ 146 H. JONSSON well developed in E. Iceland. At an insignificant depth, where I was able to see the sea-bottom, the individuals were not really closely placed, but lay on the bottom, some quite flat, and others slightly obliquely with the hollow surface turned upwards to the light. In this way the sea-bottom was almost entirely covered by the broad fronds. This character might be termed the calm-w7ater character, and an association-character exactly corresponding with it is found in Laminaria saccharina and Fucus inflates. Laminaria digitata f. cucullata occurs also scattered among La- minaria nigripes v. atrofalva and Alaria Pylaii. The Laminaria hyperborea-association. It is developed luxuriantly in S. and SW. Iceland, and occurs also in E. Iceland and N. Iceland, in those places which I have visited. I can pro- nounce no opinion upon its general distribution in N. Iceland, as dredgings have been undertaken there in a few places only; yet I think that it occurs everywhere there. In E. Iceland, on the other hand, where I have done a great deal of dredging, I have only found it at the mouth of BerufjorSur. It is very luxuriant in the Vestmannaeyjar, forming a continuous belt round the inhabited island. The association grows on a rocky substratum, from a depth of about 4 metres to about 30 (or 40) metres, and occurs both on exposed and on slightly exposed coasts; close to the limit of low- tide and in water-filled depressions in the lowermost part of the littoral zone small specimens may occur. In shallow water, with exceptional low-tides, the upper part of the stipe is frequently seen rising above the surface, raising the lowrer part of the frond obliquely above the water. Borgesen (12, p. 755, Fig. 160) has reported and illustrated this from the Faeroes. In Iceland L. hyperborea does not vary in the same manner as do L. saccharina and L. digitata, the fact being that it does not occur in protected places, and on exposed coasts does not extend so far up as the two species mentioned. Those specimens which occur close to the limit of low-tide, or in pools in the littoral zone, are quite as typical as the large, deep-water individuals. The asso- ciation seems to thrive best at a depth of about 20 to 30 metres, but on somewhat exposed coasts it also thrives fairly well at lesser depths, and is then very frequently mixed with the other members of the community; while at greater depths it is generally pure. As MARINE ALGAL VEGETATION 147 is generally known, the species has very strong haptera, often ar- ranged in rows, which issue from the lower part of the stipe, so that the haptera appear one above the other in vertical succession. They then look like a vertical row of obliquely placed props, one above the other; such rows issue in all directions from the stipe. By this it must be understood that, as the plant grows, new hap- tera appear, usually in the regular succession mentioned, until such time as the plant attains its normal size. The development of the haptera must necessarily keep pace with the rest of the growth, because the larger the plant becomes so much the more is it moved by the waves, and so much the stronger must the props become if the plant is not to be torn up. The youngest props are the longest, and are situated at the extreme (upper) end of the row. That the growth of the organs of attachment is contemporaneous with the increase in size of the individual applies also, of course, to the other Laminar iacece , but scarcely anywhere is it seen so distinctly as in this species. In its main features L. hyperborea behaves in Iceland at any rate in S. and SW. Iceland in the same manner as it is recorded by Borgesen (12, p. 755) to behave in the Faeroes. The Under- vegetation. No doubt an abundant under- vege- tation occurs everywhere, chiefly formed of crustaceous algae as a lower layer, and of branching or membranaceous Rhodophycece and a few Phceophycece in addition as an upper layer. Here, as in the Fucus- belt, it must be assumed that the under-vegetation is of no real value to the upper; \vhile, on the other hand, the latter affords protection to the under-vegetation in the Fucus-be\i against de- siccation and too strong light, and in the Laminariacece-communily against strong movements of the water. The subdued light caused by the Laminaria fronds is, no doubt, also of importance in the upper portion of the Laminaria-he\i , in so far, at any rate, that the species with a more downward range may endeavour to attain greater heights. In the lower portion of the Lamina ria-beli the sub- dued light does not seem to affect the under-vegetation very much which seems to thrive there just as well as in open places where Laminariacece are absent; to have the waves moderated may be beneficial to the under-vegetation, especially in shallow water. During excessive ebb-tides opportunities may occur in many places of seeing how the Laminaria fronds moderate the motion of the waves, that is when this is not too violent, as, for example, when it approxi- 10* 148 H. JONSSON mates to surf. In deep water the Laminaria fronds will always moderate the motion. In shallow water, especially where the bottom can he seen, it is easy to observe the under-vegetation, but in deep water one has to rely upon the specimens brought up by the dredge. The under-vegetation is not divided according to the various associations of the upper vegetation, but seems to possess the same character wherever it is observed. In S. Iceland (the Vestmannaeyjar and Eyrarbakki) Phymatolithon poly morph urn occurs in great abun- dance, covering the bottom entirely for large distances near the coast as an undergrowth. It is mainly composed of crustaceous, membranaceous, and more or less branching Floridece. A. Crustaceous Algae. In several places in N. and E. Iceland I found a luxuriant under-vegetation formed of Lithothamnion Iceve, L. circnmscriptam and L. glaciale. Among these species the first men- tioned especially seemed to be widely distributed both on the bottom of an Alaria and a Laminaria association. In E. Iceland Lithothamnion flavescens and L. foeciindum in addition, like Clathromorphum compac- tnm, were very frequent on a La/7?z'/?a/'z'a-association-bottom. Together with these species occurred Peyssonellia Rosenvingii, Craoria arctica, Lithoderma and others, as in the crustaceous algal vegetation (see p. 155). B. Membranaceous and Branching filamentous spe- cies. The upper layer, with which are also associated inter- mingled species, varies considerably according to the depth. In addition to a number of the species which extend further down- wards, various semi-littoral species may thus occur in the upper part of the Laminaria-be\i. A number of the species which grow on the Laminaria stems may also occur on the bottom between the Laminaria3. Here it is really a question of several associations; the lowest layer, as mentioned, is the crustaceous alga-association; the second layer is composed in its upper part of semi -littoral associations, which meet the associations which extend deeper and are mainly associations of Rhodophycew, and the uppermost layer is the Lami- naria-association. The species are named where these associations are mentioned, and are therefore omitted here. The Epiphyte- vegetation. Epiphytes very frequently occu r on the stipes and laminae of the Laminariacece, and it is a parti- cularly common occurrence for old stems of Laminaria hyperborea to be entirely overgrown, for the epiphytic vegetation is much more . MARINE ALGAL VEGETATION 149 luxuriant on this species than on the remaining Laminariacece. It happens rather frequently, however, that a rich epiphytic vegetation occurs on L. diyitata, L. saccharina, Alaria esculenta and Saccorrhiza dermatodea. The epiphytes generally occur most abundantly on the older individuals. The vegetation on the stems is usually composed of species which may be found growing among the Laminarice, and then belong to the second layer of the under-vegetation, which can thus be raised upon the Laminaria stems; consequently, it is very natural that the composition of the species of the epiphytic vegetation should vary according to the depth. The following distinction is evident: that the semi-littoral species which grow epiphytically in shallow water vanish as the depth increases; while, on the other hand, some of the deep-water species may occur at a relatively lesser depth. The following species have been found to occur on stems of Laminariacece, and may almost all be found on Laminaria hyper- borea: — Ahnfcltia plicata. Antithamnion floccosum. A. Plumula v. boreale. Ceramium rubrum. Delesseria alata. D. sanguinea. D. sinuosa. Dermatolithon macrocarpum. Euthora cristata. Gigartina mamillosa. Lithophyllum Crouani. Lithothamnion circumscriptum. Lomentaria clavellosa. L. rosea. Odonthalia dentata. Petrocelis Hennedyi. Peyssonellia Rosenvingii. Plocamium coccineum. Polysiphonia arctica. P. parasitica. P. urceolata. Porphyra miniata. Ptilota pectinata. P. plumosa. Rhodochorton repens. R. Rothii. Rhododermis parasitica. Rhodophyllis dichotoma. Rhodymenia palmata. Desmarestia aculeata. D viridis. Ectocarpus confervoides. E. fasciculatus. E. Hinksiae. E. penicilliformis. E. siliculosus. E. tomentosoides. Isthmoplea sphserophora. Laminaria digitata. L. hyperborea. Leptonema fasciculatum. Litosiphon filiformis. Sphacelaria olivacea. S. radicans. Monostroma fuscum. M. Grevillei. M. undulatum. Ulothrix flacca. Ulva Lactuca. Ulvella fucicola. Urospora Wormskioldii. 51 species in all. In addition, the fungus Dothidella Laminarice must be men- tioned; it is a very common endophyte in the stipe of various 150 H. JONSSON LaminariaceoB. It occurs most commonly at depths of from 6 — 20 metres, but has also been found as deep as 30 metres. Many of these species grow very socially and often cover the stipes completely, or nearly so. At Reykjavik, it is very usual in the spring to find Laminaria hyperborea in shallow water (4 — 10 metres) with the stipes entirely overgrown by Rhodochorton Rothii, Antithamnion floccosum and Polysiphonia urceolata, each on its own particular stipe, or else intermingled. Petrocelis and Rhododermis occur also in great abundance, almost covering entire stipes. In the Vestmannaeyjar and at Eyrarbakki it was also a fairly common occurrence to find stipes of Laminaria hyperborea completely over- grown by Dermatolithon macrocarpiim. In the northern part of the country Lithophyllum Crouani also sometimes covers the stipe of L. hyperborea over its entire length. Many of the remaining species may also occur abundantly, but most frequently the vegetation upon the stipes is mixed, and crustaceous species grow side by side with branching and membranaceous species. The epiphytic vegetation seems to be most luxuriant and richest in species at lesser depths, where the semi-littoral associations meet the associations which extend deeper down, and where, in addition to the Rhodophycece which always predominate, both brown and green species occur. At a greater depth the species are few, and are almost exclusively Rhodophycece. Generally, the rule seems to hold good that while the species with a more downward range occur on the haptera or on the lower part of the stipe, the more light-loving species occur on the upper part of the stipe; the green algae, however, are by no means always uppermost. Thus in the upper portion of the Laminaria-be\i Euthora, Rhodophyllis, Odon- thalia and others frequently occur among the haptera, but in the lower portion of the belt they may be found on the stipe almost everywhere, especially Eathora. This agrees with Berth old's ob- servation of the succession of epiphytes on Cystosira in the Medi- terranean, and with Borgesen's observation of the order of the epiphytes on Laminaria hyperborea in the Faeroes. The vegetation on the stipe of L. hyperborea is extremely luxuriant in S. and SW. Iceland and is considerable both in NW. and N. Iceland, but is poor in E. Iceland. Epiphytes frequently occur also on the fronds of the Lami- nariacece , especially in the upper portion of the Laminaria-belt. Here the brown algae predominate, while only a few red and green MARINE ALGAL VEGETATION 151 algse occur. The following species are common or, at least, occur very socially: Chantransia Alarise. Myrionema Corunnse. Rhodochorton memhranaceum. Phseostroma pustulosum. R. penicilliformis. Pylaiella litoralis. Ascocyclus islandicus. Streblonema Stilophorae. Ectocarpus tomentosoides. Acrosiphonia incurva. Litosiphon filiformis. Ulothrix flacca. All the remaining Ectocarpiis-species which are found on the stipes occur in addition. Of the species mentioned there are three in particular which grow very socially: Chantransia Alarice covers the entire frond of Alaria esculenta from tip to base; Ectocarpus tomentosoides also grows very socially on the fronds of Laminaria */ \f hyperborea and L. digitata which it frequently, entirely or nearly, covers during spring-time, at which time, also, Litosiphon filiformis often covers large portions of the lamina of L. saccharina. Myrionema Laminaric? and Streblonema wcidioides, in addition, grow as endophytes in the Laminaria fronds. Thus, at least 62 species of marine alga3, or about 59 % of the algal species (113) which grow below the limit of low-tide, occur on or in Laminariacece. On other coasts, those of the Faeroes, for example (Borgesen, 11 and 12), and those of Norway (Boye, 10) a similar epiphytic vegetation occurs in the La/nmar/a-association. On the coasts of Greenland (Rosenvinge, 63) the epiphytic vegetation is much scarcer in the La/nz/?arza-belt, which is possibly due, in part, to the absence of Laminaria hyperborea from that country. 15. The Desmarestia-association. Desmarestia aculeata is very common and grows both scattered and socially; vertically it is widely distributed, as it has been found at depths of from 4—60 metres (in E. Iceland). It seems to grow most luxuriantly at a depth of about 6 — 30 metres, and then is frequently found in associations of lesser extent. Only rarely is this association found dominant on the bottom, and even then only in small patches. Most frequently it occurs intermingled with other associations; thus, when dredging on a Laminaria bottom, it very frequently happens that Desmarestia aculeata is brought up, and as frequently as not it is intermingled with the associations which extend deeper down. It often grows among the Laminarice, 152 H. JOXSSON where there are openings in the Lf//m/?a/'za-vegetation, and beyond the Lrt/nzno/'za-belt it is very frequently found on a sandy or pebbly substratum, at any rate at inconsiderable depths. In the Laminaria- belt proper it plays the role of a kind of "underwood," but beyond the belt, at greater depths, it protrudes far above the associations of red algse. Desmarestia viridis occurs in a similar manner, very often with the other species, both inside and outside the Lamzharia-belt. It is of less importance, however, as it is much less common. At depths of between 20—30 metres it may also occur dominantly in patches. In E. Iceland, at a depth of 20 — 30 metres, Chorda tomentosa occurs growing very socially with the Desmarestia species. Of the epiphytes on Desmarestia acnleata the small Porphyropsis coccinea is of most importance (in S. and SW. Iceland). In several respects this association recalls the semi -littoral G/?ort/a-association. Similar Desmarestia- vegetation occurs in the Faeroes (Borge- sen, 14), in Greenland (Rosenvinge, 63) and in northern Norway (Foslie, 18, p. 100). 16. The Deep-water Community of Fl or ideas. A mixed society, which consists mainly of a fewr species of red algae, generally occurs at a depth of about 15 — 40 metres (over 50 metres in E. Iceland). To judge from the dredgings, the plants some- times seem to grow7 socially at any rate over small areas and sometimes to grow7 scattered, then, as a rule, much intermingled with one another. The species which seem to have an abundant local distribution in fairly many places are the following: Delesseria sinuosa at depths of from 14 to nearly 40 metres and some\vhat deeper in E. Iceland; Ptilota plnmosa from 16 metres to about 40 metres; Odonthalia dentata from 15 — 30 metres, and Polysiphonia arctica from 16 — 40 metres (10 to about 60 metres in E. Iceland). Thus each of these species forms associations, but these are often of inferior extent, with the exception, however, of that formed by the last mentioned species. The Pol ysipho nia arctica-association. Polysiphonia arc- tica is of extremely social growth in E. Iceland, in Sey5isfjor5ur and in Rey5arfjor5ur. It grows most luxuriantly at depths of from 16 — 40 metres and forms an enormous, continuous belt along large MARINE ALGAL VEGETATION 153 stretches of the coast. It is the only species of this community which forms a large, continuous and almost pure association, at any rate over considerable areas. The upper part of the association is, however, generally mixed with Delesseria sinnosa, Ptilota, Odon- thalia, Rhodophyllis and others, and then these species often occur in such abundance that the community acquires its usual mixed character; that is, several species occur dominantly side by side, though no single species can be said to predominate. Among the remaining species of the community Rhodophyllis dichotoma is often rather social. It is very common for Euthora cristata to occur intermingled, but to judge from the dredgings it seems to have a scattered growth. Ptilota pectinata may be of fairly social growth in E. Iceland. Delesseria sanguinea and Polysiphonia urceolata occur also in this community, the first mentioned ap- pearing to be most frequent while, in a few places, the latter has been found in abundance. Of the species with a scattered growth which belong to this community may be mentioned: Lomentarm dauellosa (20 — 40 melres), L. rosea (20 — 40) and Plocaminm coccineum (20 metres), all in the Vestmannaeyjar; and also Tarnerella Pennyi and Omphalophyllnm iiluaceum in E. Iceland. In many places the under-vegetation of the community is formed of crustaceous species of Lithothamnion. Of the intermixed species, Desmarestia aculeata and D. viridis are very frequent. In E. Iceland Chorda tomentosa has been found intermingled in this community. In addition, Laminariacecc of scat- tered growth such as Laminaria hyperborea, L. digitata, L. saccharina, Alaria esculenta, f. pinna ta and Alaria Pylaii occur very frequently; they are the outposts of the Laminariacew-commumiy . It has been previously mentioned that many of the members of this community grow on Laminaria stems, and that in several places the community forms, together with species of Desmarestia, the second layer of the under-vegetation of the community of Laminar iacece. The community grows both on somewhat exposed and also on exposed coasts, and occurs both on a rocky and on a pebbly sub- stratum, and even on sand. The semi-littoral Polysiphonia u/Teo/a/a-association has much in common with this communitv. V Rosenvinge (63) and, following him, Borgesen (12) call this 154 H. JONSSON community or quite similar communities: the sublittoral Floridece- formation. 17. The Lithothamnion-association. In this association I include only the highly branched species, Lithothamnion Ungeri and L. tophiforme, as these differ from the other calcareous algal vegetations by their characteristic and very social growth. Within the fjords, at a depth of about 12 — 25 metres, these algae occur in such abundance that there might be good reasons for calling it a submarine reef of calcareous algse. Within ArnarfjorSur in the vicinity of Bildudalur there was, for instance, such a luxuriant vegetation of L. Ungeri that the dredging-bag was filled time after time, nothing being found in it save this species. Mr. B. Ssemundsson has also found a similar Lithothamnion vegetation in several of the small fjords at Isafjardardjup; and as, moreover, there are specimens to hand from several other fjords in NW. Iceland, this association seems to be luxuriantly and com- monly distributed in this part of the country. A similar vegetation, mainly composed of L. tophiforme, occurred also in abundance in EyjafjorSur in N. Iceland. I have also noticed a similar vegetation though not so luxuriant in several of the fjords of E. Iceland. Horring collected L. tophiforme in HvalfjorQur in SW. Iceland, and Sa?mundsson also found it there, apparently growing very socially. In this association very few epiphytes occur, though Turnerella Penniji ought to be mentioned in N. and E. Iceland; on the other hand animals generally occur in abundance, especially Ophiurida and snails and other smaller molluscs which project everywhere from between the Lithothamnion-branches. When dredging on such a bottom rather large pieces are hauled up, which cohere, usually, by reason of the numerous branches being matted together. Rather large globular masses, which are sometimes hollow, but which are often filled with comparatively thick interwoven branches, are also frequently obtained. The hollow masses must be supposed to have grown on the outer side of some substratum which has disappeared. This ^Egagropila-form is gene- rally known. Rosen vinge mentions it from Greenland, and assumes that it lies loose upon the bottom; the masses must then be illu- minated all round by being rolled about as, for instance, by the action of the undercurrent. From what I have seen, it seems to me that a point of attachment can be perceived on entirely fresh MARINE ALGAL VEGETATION 155 forms of Mgagropila, and in my opinion they are outgrowths upon old "blocks" of calcareous algae. The action of the current probably loosens them, and they then roll about on the bottom. Undoubtedly they can live fairly long in that condition, but if they roll about much, they will surely by degrees go to pieces. This association recalls the semi-littoral Cora/Zzna-community, particularly the Cora//zY?a-association. The Lithothamnion-associalion occurs in Greenland (Rosen- vinge, 63) but not at the Faeroes (Borgesen, 12). 18. The Community of Crustaceous Algae. The characteristic life-form of the crustaceous algae which is so essentially different from that of the rest of the marine algae seems to justify the idea that they all belong to one community. With all of them the thallus is flat and, like the crustaceous lichens on the rocks, adheres by the whole of its lower surface to the sub- stratum. As the form and the manner of development of the thallus in the different species are identical in their main features, I think that the community may appropriately be named after the crusta- ceous growth. The substratum of the community consists of rocks, pebbles, mussel-shells and the like; also of other algae, especially species of Laminaria. The community has a very large distribution both in a hori- zontal and in a vertical direction, and possibly it is more particu- larly members of this community which we may expect to find in the vicinity of the absolute depth-limit of growth of algal vegeta- tion. The community has already been mentioned as the under- vegetation in the Laminariacece -community; it occurs also as an undergrowth in communities which extend to a greater depth and thus, partly as a dominant growth on the bottom and partly as undergrowth, it reaches from the great depths right up to the limit of low-tide. The semi-littoral and the littoral crustaceous alga- asso- ciations should also be regarded as part of this community although, for practical reasons, they have been dealt with earlier in this paper. The community is pure, that is to say it is composed of only cru- staceous algae ; there occur, it is true, various intermingled species, of which the majority are Floridece, though some are Phceophycece, but these I consider unessential and almost irrelevant to the crustaceous alga- community proper. They have their homes in other communities, 156 H. JONSSON and sometimes, perhaps, are in the act of forming an upper vege- tation ; sometimes they may be individuals which have ''strayed" bevond the real limits of their community. «/ +s The community is divided into various associations, according as to whether the one or the other of the species is dominant over considerable areas. A widely distributed and typical Phymato- lithon-poly morphum-association occurs, thus, in S. Iceland, and Lithothamnion Lenormandi has a fairly social growth at Reykjavik. L. Iceve and Clathromorphum compaction also form associations in several places. Judging from the dredgings, Lithothamnion flavescens and L. foecundum have a more scattered growth while, on the other hand, L. glaciate often occurs abundantly. Otherwise, it is very common for the Lithothamnion species to grow intermingled with, and at times upon, one another. The crustaceous, calcareous algae form the greater part of this community in Iceland. The remaining crustaceous species, such as Peyssonellia, Cruoria arctica and Litho- derma fatiscens, are found more scattered, although the last-named species forms associations in shallow water right up to the limit of low-tide. These species must, however, be much more common on the sea-bottom than is shown by the dredgings. It may be taken for granted, also, that Petrocelis Hennedyi and Rhododermis parasitica occur on a stony substratum in deeper water, seeing that they are so common on the stems of Laminaria hyperborea at considerable depths. Both Hildenbrandia and Petrocelis occur on a stony sub- stratum at shallower depths, and also Ratfsia ovata. The species which has, with absolute certainty, been found growing deepest is Lithothamnion lawe (see Part V), and it occurs in masses at a depth of 88 metres, that is, it forms a Lithotham- nionetum at this depth. Thus, of all the marine algal communities in Iceland, this community extends deepest. A similar vegetation occurs in Greenland (Rosenvinge, 63, p. 223), and in other places in the Arctic Sea (Kjellman, 36), but in both these places it differs in the wider distribution and greater luxuriancy of the Lithoderma /o/z'scefls-association. In the Faeroes (Borgesen, 12) the conditions seem to be somewhat similar, simi- lar in any case to the conditions in S. and SW. Iceland, although Lithoderma fatiscens seems to occur there more sparsely than in Iceland. MARINE ALGAL VEGETATION 157 B. The Sea-grass Vegetation. The Zostera-association. This association differs so much from the other marine com- munities in Iceland the marine algal communities that it must be regarded as not being in any way connected with them. The present community occurs especially on a substratum of muddy clay, which the algae avoid. The "roots" of the algae - - the haptera - are organs of attachment only, their sole function being, in most cases, to attach the plant to the substratum the stony substra- tum -, while Zostera has true roots which obtain nourishment from the substratum. For this reason Zostera requires a good nutrient substratum which is generally soft. It is rare, indeed it must be reckoned exceptional , for Zostera to be found growing on a hard clay-substratum. I understand such cases to indicate that formerly the substratum had been softer and then became more compact owing to the deposition of clay and sand, and that the Zostera is consequently about to disappear from such a spot. In places where there is no danger of either sand or clay being deposited from brooks or rivers the substratum will, nevertheless, scarcely remain •> unchanged, as the mud and ooze which are thrown down every- where in the ocean, when once they have been brought to a place, readily accumulate and remain, in the shelter of the dense vegetation. The Zosfera-vegetation occurs widely distributed, especially in SW. Iceland where, in many places, both inside the smaller arms of the fjords and in the large fjords such as Brei5ifjor5ur, a sub- stratum of muddy clay occurs between the coast and the skerries */ t/ which lie nearest. At Brei5ifjordur, during low-tide, the pale-green Zostera-be\i may be observed stretching for miles along the coast. At Faxafloi also the Zosfera-vegetation has an extensive distribution. Zostera is found, most certainly, on other parts of the coast of Ice- land, but nowhere have I seen such extensive "meadows" of it as in SW. Iceland. The vegetation is generally pure and there is hardly an asso- ciation of species in the sea off these coasts which is less mixed than the Zostera-association. In other places Zostera is a fa- vourite substratum for epiphytes, but however much I searched I did not find anything worth mentioning on the Zostera plants here; in this respect the Zosfera-meadow of Iceland agrees with that of the Faeroes. At times, species belonging to the semi-littoral commu- 1 58 H. JONSSON nities occur intermingled in the Zosfera-belt; these then grow in small depressions, where the subsoil, usually a solid, clayey and pebbly substratum, appears. Such species are: Chorda Filum, Chor- daria flagelliformis , Castagnea uirescens, Dictyosiphon foeniculaceus, Pylaiella littoralis, Ceramiam rnbrnm, Cystoclonium purpiirascens and others. They should not be classed in the Zosfe/'a-association, and «/ are mentioned only to explain the appearance of the Zosfera-meadow. The Zos/mj-association is sublittoral, but hardly extends as far down as do the semi-littoral communities. It cannot endure pro- tracted exposure, and thus a substratum which would adapt itself well to Zostera may be found completely devoid of plants in places which are exposed for a long time during low-tide. From time to time, during extreme ebb-tides, the upper part of the Zosfera-meadow may, however, be seen quite dry. But here two points have to be taken into consideration : the first being that the period of exposure is extremely short, and the second being that the extreme ebb-tides occur so seldom that thev ought not to be taken into calculation. V As a rule, the Zostera substratum is always submerged during low- tide. The water is so low, however, that the leaves float on the surface of the water, giving it a greenish tinge. When wading in a Zos/era-meadow during low-tide the water reaches to about the knees. In SW. Iceland the time of fruiting is during August — October. On Zos/era-soil there is in most places a very rich animal-life, but whether this has any influence upon the vegetation or on the nutrient substratum needs further investigation. In connection with the Zosfera-association I will just mention the Brackish- water- vegetation. It is so little known that there is nothing to be said about it, except that I have found Ruppia maritima in one solitary spot, where it grew so luxuriantly that, although scattered, it characterized the bottom. VII. DIFFERENCES IN THE VEGETATION IN EAST AND SOUTH ICELAND. East Iceland and South Iceland - - at the south-eastern corner of the island, at about the stretch of coast from Vestrahorn to Eystra- horn are divided by a rather sharp boundary both as regards the hydrography and the composition of the vegetation , as has been mentioned above. On the other hand, E. Iceland is connected with S. Iceland by a large transitional area (see p. 67) which stretches further along the north and north-west coasts and a part of SW. Iceland. The difference as regards the vegetation is therefore greatest between E. Iceland and S. Iceland; so it is these coastal districts which will exclusisely or almost exclusively be treated of in this part of the present paper. Where a great floristic difference exists between the different parts of the coast as, for instance, between E. Iceland and S. Ice- land1 (see Part III) it is to be expected that there will be differences in the vegetation, more especially as some of the species which are not common to all the coastal districts grow socially and form associations. The majority of the communities and the associations are however common to all the districts and are somewhat similar in appearance, as is also seen from the above description (Part VI), where the differences are alwavs mentioned. 9f As regards the communities and associations common to both districts it is enough to refer to the above description. Here, only those communities and associations will be mentioned which are found in the one district but are absent from the other. 1 The greater part of the coast of S. Iceland is a sandy coast or a barren "desert;"1 in this part of the present paper, by S. Iceland is meant only that part of the coast where vegetation occurs • the Vestmannaeyjar and the stretch of coast from Reykjanes in the direction of Thjorsa or somewhat more to the east than Stokkseyri. 160 H. JONSSON Occurring in E. Iceland Occurring in S. Iceland and absent from S. Iceland. and absent from E. Iceland. Arctic Associations. Boreal Communities and Asso- The Monostroma groenlandicum- ciations. association. The Peluetia-Fucus-spiralis-be\i. The Polysiphonia arc/zca-associa- The Community of Corallina. tion. Subarctic Association. The Laminaria fceroensis-associa- The Facas se/ra/izs-association. The Phymatolithon-polymorphiim- association. tion. Further it should be pointed out that the epiphytic vegetation on Laminaria hyperborea is quite infinitesimal in E. Iceland, but very luxuriant in S. Iceland. Of E. Iceland species which are important to the vegetation, Laminaria nigripes, Turnerella Penny i and others are absent from S. Iceland. Of S. Iceland species which play a prominent part in the vegetation a great many are wanting in E. Iceland (see Part III). The Zostera-association requires to be described separately. It belongs properly to SW. Iceland. Zostera is also known to occur in E. Iceland and it is possible that it forms associations there, but they are probably far more limited in extent than those in SW. Iceland. Zostera has not been found in S. Iceland itself, which is probably due to the fact that a favourable substratum for it is wanting there. If we now leave the Zos/era-association out of consideration, as the latter does not occur in S. Iceland, and confine our attention to the above-mentioned communities and associations which are found in the one district but are absent from the other, then it is seen that at any rate some of them characterize the vegetation to a considerable extent. The Monostroma g r o e n 1 a n d i c u m - a s s o c i a t i o n, as already mentioned, is peculiar to E. Iceland and has a considerable exten- sion in several places there. It is not found in S. Iceland. Mono- stroma groenlandicum occurs sparingly both in N. Iceland and NW. Iceland, therefore it is possible that this association is not exclu- sivelv confined to E. Iceland. As the community of filiform algae is «/ ^/ o commonly distributed both in E. and S. Iceland and green filiform algae occur very luxuriantly in S. Iceland, the absence of M. groen- landicum from the latter place is of no essential importance to the vegetation as regards appearance. MARINE ALGAL VEGETATION 161 The Polysiphonia arctica-association is of importance as regards the appearance of the sublittoral vegetation in E. Iceland. This association does not occur in S. Iceland, but a corresponding one occurs, which is however far less luxuriant, composed of Poly- siphonia urceolata. The Laminaria fasro en sis- association is known to occur only in E. Iceland. This association is probably more widely dis- tributed along Iceland than is at present known; the species occurs at any rate in N. Iceland. But I think that this species will hardly be found along the coast of S. Iceland owing lo the fact that shel- tered localities are wanting there. The Pelvetia-Fucus-spiralis-belt is most commonly dis- tributed in S. and SW. Iceland, and composes there the upper part of the Fiicacece-community. As this belt is absent from E. Iceland there is a considerable difference in the appearance of the upper- most part of the Fucacecp-community in the coastal districts in question. Fncns spiralis is, however, found in E. Iceland. The Community of Corallina is also peculiar to S. and SW. Iceland and absent from E. Iceland. This community, or the */ f Corallina-Gigartina belt, is very luxuriant and often of considerable extent in S. Iceland (and SW. Iceland), owing to which the semi- littoral vegetation in E. Iceland and in S. Iceland differs highly in character. The Fucus serratus-association is poorly represented in S. Iceland, but it is luxuriant in a single locality in SW. Iceland. As Fucus serratus is rare, and somewhat resembles in appearance the large, broad-leaved forms of Fucus inflatus which are common everywhere, it plays only an inconsiderable part as regards the appearance of the F«ca cere-community. The Phymatolithon polymorphum-association is pe- culiar to S. Iceland, but as other crustaceous, calcareous algae occur in E. Iceland in a similar manner though less luxuriantly, the ab- sence of Phymatolithon polymorphum is of no essential importance as regards the appearance of the crustaceous-alga- vegetation. From what has been stated above it is evident that the occur- rence of the Peluetia-Fucus-spiralis-beli and the Corallina- Gigartina- belt in S. Iceland (and SW. Iceland) gives to the littoral and semi- littoral vegetation of the southern district a character different from that of E. Iceland. The Botany of Iceland. I. H 162 H. JONSSON If we now turn to the individual species which (besides those already mentioned) are found in the one coastal district but are absent from the other, and which are important as regards the appearance of the vegetation, we see, as already mentioned, that such species are few in E. Iceland and numerous in S. Iceland. Some of these species have a fairly social growth without, however, forming independent associations. In the following, only those species are given which occur most abundantly. East Iceland. Lithothamnion flavescens. L. foecundum. Laminaria nigripes. Turnerella Pennyi. Ptilota pectinata. Peyssonellia Rosenvingii. Goilodesme bulligera. Ulothrix consociata var. islandica. South Iceland. Lomentaria clavellosa. Plocamium coccineum. Ghantransia Alariae. Callithamnion Arbuscula. Plumaria elegans. Ceramium acanthonotum. C. rubrum. Polysiphonia fastigiata. Rhododermis parasitica. Cystoclonium purpurascens. Ptilota plumosa. Petrocelis Hennedyi. Ectocarpus fasciculatus. E. tomentosus. Cladophora rupestris. Enteromorpha Linza. It should moreover be noted that all the Ce/Y/m/iz/??-species are absent from E. Iceland. In S. Iceland, besides those mentioned above, Ceramium atlanticum occurs, and it will no doubt be possible to find several more Ceramzum-species in S. Iceland. The Epiphytic Vegetation on Laminaria hyperborea, as mentioned several times, is very luxuriant in S. and SW. Iceland; in NW. and N. Iceland it must also be said to be fairly luxurianl, but in E. Iceland it is quite infinitesimal in amount, which is pro- bably connected w7ith the fact that Laminaria hyperborea is rare in E. Iceland. The epiphytic vegetation on the stipes of Laminaria hyperborea is very luxuriant and finely developed in S. Iceland. The following from the Vestmannaeyjar are given as an example : dominantly on the stipe of L. hyperborea occurred Rhodymenia palmata, Delesseria alata and Plocamium coccineum; less abundantly than the three above-mentioned species occurred Delesseria sanguinea, Lomentaria clavellosa, Lomentaria rosea, Euthora cristata, Giqartina mamillosa, Ahnfeltia plicata, Petrocelis Hennedyi, Dermatolithon macrocarpum and MARINE ALGAL VEGETATION 163 Pterosiphonia parasitica. To show how rich in species the epiphytic vegetation occurring on a single individual of L. hyperborea may be, the following species, also from the Vestmannaeyjar, may serve: Desmarestia viridis, Ptilota plumosa, Delesseria sinuosa, Delesseria alata, Lomentaria clauellosa, Polysiphonia urceolata, Plocaminm coccineum, Delesseria sanguined and Euthora cristata. So luxuriant and finely developed an epiphytic vegetation gives to the La/mnar/ame-community of S. Iceland a character different from that which it has in E. Iceland. On the stipes of Alaria and Laminaria diyitata, on the other hand, a similar epiphytic vegetation occurs in both the coastal districts. The zonal division of the marine algal vegetation is in its main features similar in E. Iceland and S. Iceland, as is shown by the following examples which have been taken straight from the diaries. East Iceland. Vattarnes, steep clifts, highly exposed. l4/7. I. Ulothricetum U. flaccae. II. Bangietum B. fuscopurpurae. III. Porphyretum P. umbilicalis. IV. Fucetum F. inflati, in the most exposed localities, consisting only of f. exposita. V. Rhodymenietum. Halosaccionetum. Acrosiphonietum. VI. Sublit. Alarietum. Borgarnes, sloping rocky coast, considerably exposed. 13/e. I. Prasioletum P. stipitatae. II. Ulothricetum U. flaccae, in great abundance. Rhizoclonium in crevices. III. Bangietum B. fuscopurpureae, in wonderful abundance. IV. Porphyretum P. umbilicalis, of great extent. V. Monostroma groenlandicum. VI. Fucus vesiculosus, sparse and miserable. VII. Fucetum F. inflati, abundantly. The specimens very variable. VIII. HalosaccionetumH.ramentacei. Urospora Wormskioldii. Polysiphonia urceolata. Bhodomela lycopodioides. Chorda tomentosa. IX. Alarietum. Holmanes, somewhat exposed, sloping rocky coast. 18/7. I. Ulothricetum U. flaccae. II. Enteromorphetum E. intesti- nalis. III. Fucetum, uppermost, narrow margin of F. vesiculosus, be- low that, broad belt of F. in- flatus, intermixed here and there sparsely with Ascophyl- lum nodosum. Under-veg. Hildenbrandietum. IV. Halosaccionetum, intermingled with Monostroma fuscum, Por- phyra miniata, Rhodymenia palmata and a few Fucus in- flatus. V. Alarietum, composed of Alaria esculenta and A. Pylaii. VI. Laminarietum L. saccharinae. VII. Laminarietum L. digitatae. 11* 164 H. JONSSON South Iceland. Vestmannaeyjar, much exposed, steep, rocky coast. 20/5. I. Ulothricetum U. flaccae. II. Porphyretum P. umbilicalis. III. Ascophylletum A. nodosi. IV. Gigartinetum G. mamillosae, broad belt, in it Corallina offici- nalis, Ceraraium acanthonotum, Gallithamnion, Delesseria alata. V. Corallinetum G. officinalis, dis- persed in it Laminariae. Vestmannaeyjar, Vikin, considerably exposed, sloping rocky coast. 21/5. I. Ulothricetum U. flaccae. Enteromorphetum E.*micrococ- cae. Acrosiphonietum . II. The Fucus-belt. Ascophyllum nodosum. F. vesiculosus. F. inflatus. III. Gigartinetum G. mamillosae floating IV. Corallinetum together. C. officinalis V. Laminarietum L. * stenophyllae. VI. Alarietum A. esculentae. VII. Laminarietum L. hyperboreae. on flat rocks between crevices. Vestmannaeyjar, the skerry, considerably exposed, sloping rocky coast almost destitute of phanero- gams. 18/5. Uppermost, Gochlearia officinalis and rosettes of Plantago maritima. At the same level, in crevices: I. Enteromorphetum E.* micrococ- cae, 4- Cladophora sericea. II. Ulothricetum U. flaccae Prasioletum P. stipitatae III. Pelvetia canaliculata. IV. Fucus spiralis. V. Ascophylletum A. nodosi. F. vesiculosus. F. inflatus. Under-veg. Callithamnionetum. Antithamnion. Gigartina. VI. Gigartinetum G. mamillosae. VII. Laminarietum L. * stenophyllae. Under-veg. Phymatolithon po- lymorphum. South Iceland, sou according to C. H. Stadur, wide foreshore, considerably exposed. 12/6. 1. Pelvetia canaliculata. -r stunted F. vesiculosus. 2. Fucus spiralis, with stunted Ascophyllum and Cladophora rupestris. 3.a Ascophyllum + Polysiphonia fa- stigiata, broad belt. 3.b Fucus vesiculosus formation, in it F. inflatus, widely distributed, 3.c in it Gigartina. 4. Here and there in depressions the formations: Monostroma Grevillei and M. fuscum. th side of Reykjanes, Ost en f eld's diary. 4- Halosaccion. Cystoclonium. Ahnfeltia. Dictyosiphon foeniculaceum. Rhod3Tmenia. 5. Laminaria + Alaria. The foreshore between StacHir and Reykjanes. 13/e. 1. Uppermost, Porphyra umbili- calis. F. spiralis. F. vesiculosus f. sphaerocarpa. Enteromorpha compressa. 2. F. inflatus-formation, widely dis- tributed and in it Gigartina. MARINE ALGAL VEGETATION 165 3. Gigartina, widely distributed. 4. Corallina widely distributed, in Rhodymenia. a single pool Halosaccion. Plum aria elegans. Delesseria alata. Acrosiphonia. Monostroma Grevillei. Chsetomorpha Melagonium. 5. Laminaria 4- Alaria. In another part: Ascophyllum in quantities between 1 and 2. Delesseria sanguinea and others. It would carry us too far to give several more examples from the diaries, but on regarding the material taken as a whole it is distinctly seen that there is no other difference of importance be- tween the two districts with regard to the zonal division of the algal vegetation beyond the fact that some of the communities and associations occur in one place and are absent from the other. The division of the belts varies somewhat in both places which is chiefly due to the greater or smaller degree of exposure of the locality (see Part VI). The most conspicuous difference in the division of the belts in E. Iceland and S. Iceland is due to the enormous size of the Cora//zna-belt in the latter place. This community (Gigartina, Coral- lina and others; see Part VI) occurs just below the Fucus-helt where, for instance, in the Vestmannaeyjar it is the dominant one; this also applies to Eyrarbakki and the south side of Reykjanes, but perhaps to a somewhat less degree. This leaves less room for the /?/ioGh/me/?za-community (Rhodymenia, Halosaccion ; see Part VI), which also forms a belt below the Fucus-belt; consequently it is not so large there as in E. Iceland where it is extremely common and in several places widely extended. In many places in SW. Iceland the /?/7OG?z//T7e/?za-community is as large as in E. Iceland. The sea off the coast of S. Iceland is in movement everywhere, and calm water is almost unknown. The calm-water-vegetation proper, which is so common writhin the fjords of E. Iceland, is therefore absent from S. Iceland, but is again found richly repre- sented in SW., NW. and N. Iceland. The luxuriancy of the vegetation is somewhat similar in both places; it appears, however, to be greater in the Vestmanna- eyjar and the western part of the south coast. Depth-limit. It appears that there is reason to believe (see Part V) that the algal vegetation extends to greater depths in the fjords of E. Iceland than on the south coast, but as this cannot be regarded as sufficiently proved I shall not enter into it more fully. VIII. SOME NOTES ON THE BIOLOGY OF THE ALG.E ALONG THE COAST OF ICELAND. Ar present very little is known with regard to the biology of the marine algae along the coast of Iceland. What is known, on the whole, regarding this point is for the most part mentioned in the "Marine Alga? of Iceland" (31) under each species. Some ob- servations which have been made subsequently will be mentioned in the following pages. Of these, I regard those which have been made during winter as the most important, although they are very incomplete owing to the fact that during winter I have only rarely been able to make investigations, and then have had access to the littoral zone only; thus, with the exception of a few species, the winter-habit of the sublittoral species is not known. Consequently I can, by no means, treat of the biological conditions exhaustively, but must content myself with giving a few incomplete contributions. On the whole, the behaviour of the species is best known in spring and summer, less well in autumn, and least well during winter. From most of the coastal districts there are observations to hand either only those of a single season of the year, or at most of two or three seasons. From Reykjavik we have observations of all four seasons, but those of the winter are sparse, and only a few species have been observed throughout the year. The observations to hand are so few and insufficient that a comparison of the biological conditions in the five coastal districts in question cannot be made; here, therefore, Iceland is treated for the most part as an entirety. 1. Duration of Life. The life-period of the algal species is of very varying length ; in this connection the alga? may be divided into two groups: annual algae and perennial algae. H. JONSSON: MARINE ALGAL VEGETATION 167 A. Annual algae. The annual species are especially the Green Algae which grow in the upper littoral zone and are exposed during each low-tide. The upper littoral zone is that part of the algal region where the change of seasons is most felt and where winter prevents many species from continuing life. Such species then produce spores which live through the winter as such, or in the early stages of germination. Of annual species the following may be mentioned : Codiolum gregarium. Ulothrix-species. C. pusillum. Ulvella. Percursaria. Pringsheimia. Enteromorpha-species. Urospora-species. Monostroma-species. Chsetomorpha tortuosa. Prasiola-species. Spongomorpha vernalis. Cladophora-species. In addition, some of the endophytic species must be regarded as belonging to the annuals, although some of them can be met with at all seasons of the year. As examples of such species may be mentioned : Chlorochytrium-species and Codiolum Petrocelidis. The life-periods of the species mentioned above are probably of different lengths and it is possible that some of them can pro- duce several generations during one summer (cf. Borgesen, 11 and 12). The majority of these species grow luxuriantly during spring (March — May) and summer (June — August), produce spores at the end of summer and then die. Some of them, however, con- tinue life into the autumn (September — November), or at any rate until September. A few may also be met with during winter (De- cember— February), e. g. Enteromorpha intestinalis f. prolifera, Mono- stroma fuscum (sterile and fruiting), Cladophora rupestris (abundantly) and Cladophora sericea (sparingly). Of the above-mentioned endophytic species I shall refer to Chlorochytrium inclusion and Codiolum Petrocelidis only. Both these species occur at all seasons of the year. They are found most fre- quently in the host-plants in the sublittoral zone, where the conditions of life must be considered to be more stable than in the littoral zone. I regard such species as short-lived. They are found all the year round, as probably several generations are produced during the year. Among the annual Brown Algae the following must be in- cluded: M(/rzo/?e/77a-species, Ascocyclus and the majority of the Ectocarpaceoc , Leptonema, Lilosiphon, Isthmoplea, Phwostroma, Ca- 168 H. JONSSON stagnea and Leathesia (living from June to September). In addition there are species which may be supposed to be annual, as for example, Punctaria, Stictyosiphon , Scytosiphon (?), Phyllitis, the ma- jority of the Dictyosiphonacece, Chorda-species, etc. Of Red Algae the following must be presumed to be annual: Bangia, Porphyra-species, Porphyropsis, Conchocelis, Chantransia- species, Cera/nzum-species and possibly several more. As regards Porphyra ambilicalis it should, however, be stated that it has been found at all seasons of the year and at Reykjavik it occurs as luxuriantly in December — January as during the spring. B. Perennial Algae. With regard to some of the species it is difficult to decide whether they are perennial or annual, as our knowledge of them is incomplete; consequently it is sometimes a matter of opinion whether they are to be included in the one or in the other group. Only a few of the Green Algse are perennial, as for instance, the majority of the Acrosiphonia-species. The latter, besides being reproduced by spores, have also abundant vegetative reproduction by means of offshoots; and some of them, as for in- stance, A. albescens and others, live all the year round in the semi- littoral zone. Of the Brown Algae the Fucacece and the Laminariacew are perennial. It is, however, doubtful whether we are justified in in- cluding Saccorrhiza dermatodea among the perennials. At Reykjavik I have seen only old fruiting specimens in the winter, and judging from their appearance it is very probable that they die during the winter; nothing can, however, be stated with certainty regarding this point. In the fjords of E. Iceland large individuals of this species were growing in the sublittoral zone; I believe they were more than one year old, but I could not prove this. In Greenland (Rosen vinge, 61, p. 852) perennial or upwards of a year old individuals of this species occur. With regard to several other species of Brown Algae it is not easy to say at present whether they are annual or perennial. I think, however, that the following may be classed as perennial :- Lithoderma, /?a//sza-species, Sphacelaria-species, Chcetopteris, Desma- restia acu/ea/a, D. ligulata, Chordaria flagelliformis (?). I think that the majority of the Red Algae are perennial or can, at any rate, live through more than one growth-period. I shall, MARINE ALGAL VEGETATION 169 in the following, name some species which I am fairly certain may be included among the perennials: Gigartina mamillosa. Halosaccion ramentaceuni. Ahnfeltia plicata. Polysiphonia urceolata. Euthora cristala. Rhodomela lycopodioides. Rhodophyllis dichotoma. Odonthalia dentata. Rhodymenia palmata (?). Ptilota plumosa. Rhodochorton Rothii. The crustaceous algae such as Hildenbrandia, Petrocelis, Cruoria, Peyssonettia and Rhododermis, and by far the greater number of the calcareous algae must also be classed as perennial. 2. Periodical Changes. The Period of Activity. The Period of Rest. The period of activity of the annual species is identical with their period of life and it extends, probably as regards the majority of the species, over the spring and summer months. The perennial species and the species which can live more than one year, have a very long period of activity which extends over the greater part of the year with the exception of the darkest part; consequent^ these species have a very short period of rest. Although from the observations to hand it is not possible to fix the length of the period of rest, yet they indicate that it must be short. The Fucacece may be men- tioned as an example. Their vegetative growth appears to be very slight during December — January, and in the case of the older individuals there is probably none at all at that time; but although the majority of the individuals of the Fucacece are sterile during winter yet, even in December, reproductive organs are developing here and there. Young plants of the summer or autumn appear to have vegetative growth also during the winter. In the Fucacece-belt, taken as a \vhole, the period of rest is consequently extremely short. In the sublittoral zone I think that entire rest must be of extremely short duration. Renewal of the lamina. The young shoots. As is well known, a renewal of the lamina takes place yearly in the Laminaria- species. At what time this takes place in Iceland cannot be stated with absolute certainty, but the observations seem to indicate that the new7 lamina begins to grow even in the winter time, as the light increases. I believe that in SW. Iceland a general renewal of the 170 H. JONSSON lamina takes place in February — March. In April individuals with large, new laminae, with the old laminae or portions of them still attached, are frequently found at Reykjavik, but the majority of the Laminaria individuals have renewed their laminae by that time. A few individuals are however met with until June in SW. Iceland in the act of lamina-renewal. With regard to some of the most common species the following may be noted: Laminaria saccharina in SW. Iceland (1897) was frequently found in the lamina-renewal stage in April, while some individuals were renewing their laminae in May. Laminaria digitata: some individuals were renewing their laminae in April and until June in SW. Iceland (1897). Laminaria hyperborea in SW. Iceland (1897), some were re- newing their laminae in May — June. L. hyperborea was found renewing its laminae in July on the north coast of Iceland. This was observed only once, therefore it cannot be concluded from this that the lamina-renewal stage com- mences later or lasts longer on the north coast than in SW. Iceland. In the Faeroes (Borgesen, 11 and 12) the lamina-renewal stage occurs at the same time as in SW. Iceland. On the west coast of Sweden the renewal of the lamina takes place during winter (Kyi in, 45) and in Laminaria saccharina and L. digitata the young leaf, in December, is a quarter the size of the old one; in April it is only exceptionally that individuals are found with a portion of the old lamina attached. The renewal of the lamina in L. hyperborea takes place later, and in April the new lamina is a quarter the size of the old one, and in the beginning of July a portion of the old lamina is still present. In connection with the lamina-casting species the following may be mentioned: Desmarestia aculeata has been collected bearing the brown, as- similatory hairs in March — May in SW. Iceland, in May in NW. Iceland, in June in E. Iceland, and in July in N. Iceland. This appears to indicate that the hair-bearing stage occurs later on the north and east coasts, or that it lasts longer. The species behaves in the same manner in Greenland (Rosenvinge, 61, p. 857) where the hairs are cast off, at the latest, in June in South Greenland, while hair-bearing individuals are found in July and August in North Greenland. In the Faeroes it has been observed with hairs upon it in May — June (Borgesen, 13, p. 445), but hair-bearing individuals were, however, rare in June. On the west coast of MARINE ALGAL VEGETATION 171 Sweden (Kyi in, 45) the hairs are cast in June, but (young) hair- hearing individuals may, however, be met with in July. Polysiphonia urceolata has hair-leaves in the spring, summer and beginning of autumn, but individuals without hairs are found side by side with hair-bearing ones from May to August. In De- cember only hairless individuals have been observed at Reykjavik, and in January — February only hairless plants have been collected in N. Iceland. Rhodomela lycopodioides has been collected with hair-leaves in March — July, and hairless in June — August in SW. Iceland; with hair-leaves in April — December, and hairless in July — August in N. Iceland; with hair leaves in June — July and hairless in May — July in E. Iceland. In the autumn this species had cast off its leaves and branches in SW. Iceland. In addition to this, the following instances of the occurrence of young shoots may be mentioned: Odonthalia dentata. The young shoots in this species are readily recognized by their paler red colour. Material from January- February bears young shoots. The latter are easily recognizable in January, and have probably begun to grow out in December. The young shoots then increase in size, and the colour becomes gradu- ally darker month by month. In material collected in June — July the length of the year's shoot may even then be determined in some of the individuals, but I believe, however, that it is in July that the shoot is almost full-grown. Polysiphonia fastigiata has young shoots in December at Reykjavik. Odonthalia shows distinctly the period of development of the vegetative shoots in the sublittoral zone, and I presume it may be taken for granted that the other sublittoral species do not differ very much from it as regards this point. In Greenland the formation of the new shoots begins in Feb- ruary—March (Rosenvinge, 63, p. 239), and the growth is con- tinued until August or during the whole summer. In the Faeroes the new shoots begin to appear in the latter part of autumn (Borgesen, 12, p. 828). From Spilzbergen (Kj ell man, 36) some sublittoral species are known which form new shoots during the winter, as for instance, Delesseria sinuosa in January, and Rhodymenia palmata from No- vember to May. 172 H. JONSSON The Fruiting Period. In the table given below is indicated the time at which the species have been found in fruit. A + sig- nifies that the greater part of the individuals in the samples gathered were fruiting; a -*- signifies that fruiting and sterile individuals occurred in almost equal abundance, or sometimes that only a few fruiting individuals occurred ; a signifies that only sterile indi- viduals of the species were found. In the majority of the annual species the fruiting period coincides with the vegetative stage, and thus growing vegetative shoots and sporangia are frequently found on the same individual. These species fruit comparatively quickly and the young, or purely vegetative, stage is of short duration. The fruiting period extends over spring and summer probably in the case of the majority of the species. They do not, however, behave similarly in this respect in the different coastal districts. Urospora Wormskioldii, Monostroma Grevillei, M. imdiilatum, Ectocarpus tomentosoides and Litosiphon fili- formis are all decidedly spring plants at Reykjavik, but in E. Ice- land they have been found bearing fruit far into the summer. Leathesia difformis is a decidedly summer species at Reykjavik, it has been observed fruiting in June, July, August and even into September, but it was dying away in the middle of September. At Reykjavik its life-period coincides with its fruiting-period, but in N. Iceland it has- been gathered in a sterile condition in September. This species appears to behave in the same manner on the west coast of Sweden (Kylin, 45) as at Reykjavik. Moreover the fact may be emphasized that at the latter place Enteromorpha Linza is usually a summer and autumn species. With regard to the perennial species, it happens both that the vegetative growth and the fruit-formation is simultaneous, and also that the two stages occur at different times. A purely vegetative, young stage, more prolonged than in the annuals, is found in several of the perennials; thus, I think that I have seen indications of Alaria and Laminaria species being in a purely vegetative stage throughout the first year and perhaps longer. Kylin (45, p. 274) divides the perennial species into three groups according to their life-activity: Group 1 includes species which carry on vegetative and repro- ductive work all the year round. Group 2. Species which carry on vegetative work the whole year, but reproductive work only for a part of the year. MARINE ALGAL VEGETATION 173 Group 3. Species which carry on vegetative and reproductive work only during a part of the year. In referring the Icelandic species to these groups the difficulty at once arises, that we lack knowledge regarding the behaviour of a number of the species during winter. Such species cannot there- fore be grouped with any certainty at the present time. True, we may judge with some probability how they behave here during winter by a comparison of their winter-activities in other places, provided these are known ; but as it has been shown that one and the same species often behaves differently in two distantly situated places, a satisfactory result could not be arrived at through such a comparison which has, for that reason, been omitted. I therefore mention a few species only, which I think I can group with some certainty. Group 1. The following species belong to this group: Hilden- brandia rosea which fruits all the year round. Rhodymenia palmata; it might appear doubtful whether this species should be classed as a perennial. K jell man (36, p. 150) regards it as an annual plant which forms tetraspores twice, once as a young plant and the second time just before it dies. It appears to me that the new shoots which arise early in spring from evidently old fronds, show that it lives through at any rate more than one period of growth. Peluetia canaliculata bears fruit and carries on vegetative work the whole year at Reykjavik. In December — January the fruiting indi- viduals were comparatively few, but there was a quantity of young plants almost all of which were in the "rosette" stage. It is per- haps doubtful if this species belongs to this group at all. Group 2. The Facacece belong to this group (with the exception of Pelvetia (?)). Fncus spiralis produces fruit in spring, summer and autumn. In December (1911) it was sterile, and young plants were found in quantities. Ascophyllum nodosnm, Fncus vesiciilosus and Fncus inflatus, all these behave almost similarly. In December and January I saw, here and there, individuals with very young receptacles, but by far the greater part were sterile. In spring and early summer these species are found everywhere with full-grown receptacles. In the latter half of August they are sterile everywhere and at that time only a few individuals, which must be regarded as stragglers, are found bearing receptacles. Of F. inflatns I saw no stragglers in Au- gust. In September also they are sterile. In the first half of October 174 H. JONSSON both Ascophyllum and F. inflatus are sterile, but at that time a few individuals of F. vesiculosus bear young receptacles. From the middle of October until December there are no observations to hand from Reykjavik. The usual course, with regard to these species, appears to be as follows: In the latter part of the autumn the development of the sexual organs begins, and is continued throughout the winter; in March the species are found everywhere with ripe sexual organs and the spores continue their development until the latter part of the summer. Kj ell man (36, p. 195) records with regard to Ascophyllum no- dosum in Finmark, that it had numerous receptacles in July and the first half of August, was sterile in the latter part of August, and that new receptacles had begun to appear in October. Foslie (18, p. 64) records with regard to the same species in East Finmark that it has abundant receptacles in July and is sterile in August. In the Faroes (Borgesen, 12, p. 830) Ascophyllum ngdosum has young receptacles in December and fruits during the whole summer. On the west coast of Sweden the sexual organs begin to de- velop late in August or early in winter (Kylin, 45, p. 106), and by the beginning of June the receptacles have fallen off. Fucus inflatus probably behaves in East Finmark in a similar manner as at Reykjavik. Foslie (18, p. 67) mentions receptacle- bearing individuals in June — July and the first part of August. At Spitzbergen it bears fruit during winter and Kj ell man (36, p. 204) found germinating spores in December, January, February and March. To Group 2 belong, in addition, probably all the Laminariacece, Rhodochorton Rothii, Polysiphonia urceolata, Halosaccion , Gigartiua mamillosa, Odonthalia and probably many more. Petrocelis Henuedyi belongs also most nearly to this group; it has been found producing fruit the whole year, but in spring, summer and autumn only a few fruiting crusts are found, while the winter appears to be the ordinary fruiting period. Chcetopteris plumosa I include, although with doubt, in this group. Group 3. To this belong Desmarestia aculeala, which has been found fruiting in October, Rhodomela lycopodioides, Delesseria san- guinea and others. MARINE ALGAL VEGETATION 175 The Fruiting-period of the Species. Bangia fuscopurpurea -j- Porphyra umhilicalis -|- P. miniata PorphjTopsis coccinea . . . Conchocelis rosea Chantransia microscopica C. Alariae C. secundata -j- C. virgatula -|- Ghondrus crispus -}- Gigartina mamillosa Ahnfeltia plicata Phyllophora Brodisei * interrupta P. membranifplia Actinococcus subcutaneus Geratocolax Hartzii -j- Cystoclonium purpurascens Turnerella Pennyi Euthora cristata _L Rhodophyllis dichotoma . . Rhodymenia palmata Lomentaria clavellosa L. rosea , Plocamium coccineum ... Halosaccion ramentaceum Delesseria alata D. sinuosa J- D. sanguinea Pterosiphonia parasitica -(- Polysiphonia urceolata . P. fastigiata P. arctica P. nigrescens Rhodomela lycopodioides -{- Odonthalia dentata Gallithamnion Arbuscula C. scopulorum Plumaria elegans Ptilota plumosa J- P. pectinata Antithamnion Plumula * boreale. A. floccosum Geramium acanthonotum G. Deslongchampii C. fruticulosum . Spring Summer Autumn Winter 4- _L _L -f I 4- 176 H. JONSSON The Fruiting-period of the Species (continued). Spring Summer Autumn Winter Ceramium circinnatum G. arborescens G. atlanticum C. rubrum Rhodochorton Rothii -)- R. repens -j- R. minutum R. penicilliforme -j- R. membranaceum -\- Dumontia filiformis Dilsea edulis Petrocelis Hennedyi _L Gruoria arctica C. pellita Peyssonellia Rosenvingii , Rhododermis parasitica -j- Lithothamnion glaciale L. Ungeri L. tophiforme L. flavescens L. foecundum L. laeve L. Lenormandi Phymatolithon polymorphum Clathromorphum compactum Lithophyllum Grouani Dermatolithon macrocarpum Corallina officinalis Hildenbrandia rosea -|- Phseophyceae. Lithoderma fatiscens Petroderma maculiforme Ralfsia ovata R. clavata R. verrucosa R. deusta Myrionema vulgare M. Corunnae -(- M. globosum -|- M. faeroense ] -j- M. Laminariie Ascocyclus islandicus Microsyphar Polysiphonise Streblonema aecidioides . + MARINE ALGAL VEGETATION 177 The Fruiting-period of the Species (continued). Spring Summer Autumn Streblonema Stilophora? v. crespitosa -J- Pylaiella littoralis -\- Ectocarpus tomentosoides -j- E. tomentosus -j- E. confervoides -|- E. siliculosus -(- E. penicillatus -(- E. fasciculatus -j- E. Hinksise -)- Leptonema fasciculatum v. subcylindrica. . -|- Elachista fucicola J- Sphacelaria britannica S. radicans S. olivacea Chsetopteris plumosa Omphalophyllum ulvaceum Punctaria plantaginea Litosipbon filiformis -(- Isthmoplea sphrerophora Stictyosipbon tortilis Phaeostroma pustulosum Scytosiphon Lomentaria -[- Phyllitis zosterifolia -[- P. fascia _l_ Goilodesme bulligera . -|- Dictyosiphon Ekmani D. Mesogloia D. Chordaria I), corymbosus D. hippuroides D. foeniculaceus Desmarestia viridis D. aeuleata D. ligulata Castagnea virescens Leathesia difformis Chordaria flagelliformis Chorda tomentosa C. Filum Saccorrhiza dermatodea Laminaria saccharina L. faeroensis L. nigripes L. digitata L. hyperborea The Botany of Iceland. I. Winter _L + + + + + _j_ + + -f + + JL 12 178 H. JONSSON The Fruiting-period of the Species (continued). Alaria Pvlaii . V A. esculenta Fucus spiralis -|- F. inflatus -J- F. serratus -}- F. vesiculosus -}- Pelvetia canaliculata -J- Ascophyllum nodosum -|- Chlorophyceae. Chlorochytrium Cohnii C. inclusum C. dermatocolax C. Schmitzii Codiolum Petrocelidis C. gregarium G. pusillum Percursaria percursa Enteromorpha aureola E. Linza E. intestinalis -j- E. clathrata Monostroma groenlandicum M. Grevillei _j_ M. undulatum ... _j- M. fuscum _L Ulva lactuca _j_ Prasiola poh'rrhiza P. furfuracea P. stipitata Ulothrix consociata var. islandica U. subflaccida U. pseudoflacca -|- U. tlacca _[_ Pseudendoclonium submarinum Entoderma Wittrockii Acrochsete parasitica A. repens Bolbocoleon piliferum Ulvella fucicola Pringsheimia scutata Ochlochsete ferox Urospora mirabilis U. Hartzii -f- U. Wormskioldii -[- I Spring Summer Autumn Winter 4- f MARINE ALGAL VEGETATION 179 The Fruiting-period of the Species (continued). Spring Summer Chsetomorpha tortuosa C. Melagonium -j- Rhizoclonium riparium Spongomorpha vernalis Acrosiphonia albescens -|- A. incurva A . hvstrix .... 4- »/ A. flabelliformis A. penicilliformis Cladophora rupestris C. hirta C. sericea C. glaucescens C. gracilis Gomontia polyrrhi/a Ostreobium Queketti Cyanophyceae. Pleurocapsa amethystea Plectonema norvegicum Phormidium autumnale Spirulina subsalsa Galothrix scopulorum Rivularia atra. Autumn Winter By looking through the ahove table it is seen that in the case of a number of species the fructification-period is of long extent. Some nine species have been found fruiting all the year round, but I presume that by future investigations it will be proved that many more species bear fruit during the whole year. Kj ell man (36) mentions 11 species which bear fruit all the year round in the Arctic Sea, and Rosen vinge (63) mentions 6 species which fruit all the year round along the coast of Greenland, and he adds that probably there are many more. If we consult the table as to which season of the year is richest in fruit-bearing species it is easily seen that summer stands highest, with 64 °/o of the total number of species; then conies spring with 42 °/o, and after that autumn with 33 %>. As the conditions during winter are so very little known no percentages can be given for that season. It is well-known (Kjellman, Rosen vinge, Borgesen) that 12* 180 H. JONSSON in the case of many species both the vegetative period and the period of fructification is longer in the Arctic Sea and the northern part of the Atlantic Ocean than in the remaining more southern part of the boreal area of the same Ocean. It holds good also for Iceland, that both these periods are prolonged. Iceland, the Faeroes and Greenland agree also in the fact that the summer is richest in fruit-bearing species. 3. Littoral Winter-vegetation at Reykjavik. The littoral vegetation changes its appearance according to the season, and this is especially owing to the annual species. The vegetation is most luxuriant, and richest in species during spring and early summer; in the latter part of the summer the annual species decrease in number and a quantity of them disappear, and in the autumn only a small number of them is left. During winter only a few short-lived species are found, and some of them play either no part, or only an unimportant one in the vegetation which by that time is usually composed of perennial species. Thus, the number of the species of the winter- vegetation is much less than of the summer-vegetation because the Chlorophycece , which is the group in the littoral zone that is richest in species during summer, are few in number during winter; also the number of species of the Phceophycece is greatly reduced in the littoral zone during winter. It is the Fucacece which form by far the greater part of the mass of plants in the littoral zone, during the winter as in the other seasons of the year. In December and January, in the winter 1911--1912, the vege- tation of the upper and lower littoral zones was composed as follows : Highest of all a Prasioletum stipitatae occurred in patches. Prasiola stipitata grew luxuriantly and had a normal appearance. It was not injured by the winter climate. Below that came a well -developed Porphy return umbili- calis, occurring also in patches on account of the surface -form of the coast. The Porphijra was both sterile and in fruit and had an entirely normal appearance. Below that again came the Fucacece-comm unity which, as is usually the case there, consisted at the top of a Pelvetia-Fucus spiralis-be\i and at the bottom of the usual Fucus-bell (Fucns vesi- MARINE ALGAL VEGETATION 181 culosns, Ascophyllum nodosiun and Fucus in flatus). The vegetation was as luxuriant during winter as in the other seasons of the year when, however, the epiphyte-vegetation of the Fncns-beM is excepted. For further information concerning the Fucacece see p. 173. Epi- phytes, for instance such as Pijlaiella littoralis and Elachista fucicola, which at the other seasons of the year are common everywhere in •/ «/ the Fzzcus-belt, were not observed. On the other hand, Polysiphonia fastigiata occurred in abundance on Ascophyllum. Of intermixed species in the Fzzczzs-belt Cladophora rnpestris may specially be mentioned ; it occurred abundantly as well-developed specimens with the uppermost apex of the shoots destroyed. A Hildenbrandietu m occurred everywhere in the littoral «/ zone, of similar appearance and extent as at the other seasons of the year. In rock-crevices at the level of the Fzzcz/s-belt a well-developed Rhodochortonetum Rothii \vas found. The Rhodochorton was sterile and Pleurocapsa amethystea was growing upon it. In such crevices were noted in addition: Pijlaiella littoralis, sterile and in very small quantity; Acrosiphonia sp. , a few filaments; Clado- phora rnpestris, abundant; Polysiphonia urceolata, sterile and with- out hairs and sparse, and Callithamnion sp. The vegetation of the Pools was on the wrhole very poorly developed. The upper pools, at the level of Pelvetia and Fucus spiralis, contained a few small individuals of Cladophora sericea f., a fe\v filaments of. Pylaiella littoralis, a few small individuals of Monostroma fuscnm, and Diatoms in abundance. In such pools Cladophora sericea forms a dense and luxuriant vegetation in spring and summer. Fucus inflatus f. linearis grows also in these pools in spring and summer, but was not found there in winter. It appears as if this form is annual, at any rate in the upper pools. The lowrer pools, almost at the level of the lower part of the Fucus- belt, contained only a poor vegetation. Monostroma fnscum, how- ever, occurred fairly abundantly, but Halosaccion ramentaceum was found more sparsely; on Halosaccion was growing Elachista fucicola v. lubrica with unilocular sporangia and somewhat sparse assimila- tory filaments, and also Ceramium sp. Below the Fucus-belt, in the lowrer littoral zone (upper part of the semilittoral belt) were found at Efferseyjargrandi , Rhody- menieta distributed in patches here and there, many of them of rather considerable size. The majority of the individuals were old, 182 H. JONSSOX with abundance of new shoots. The Rhodgmenia^egetalio'n is never really luxuriant in this place; it attains, however, to a greater luxuriancy in spring and summer than during winter. Halosac- cioneta occurred also here and there, but sparsely and not nearly as abundantly as in spring and summer, but the same applies to it, as to Rhodymenia, that it never grows really luxuriantly in this place. There were noted in addition: Monostroma fuscum, Py- laiella littoralis (extremely sparsely) and Enteromorpha intestinalis f. prolifera, also extremely sparsely. Lithodermeta occurred abun- dantly and were well -developed, but the species was sterile. A Sphacelarietum composed of Sphacelaria radicans was found here and there. The plants were low in growth, sterile and without hairs. A Gigartinetum occurred here and there of about the usual extent. Just below the limit of low- tide I gathered the uppermost stragglers from the sublittoral zone. There Laminaria saccharina, L. digitata, L. hyperborea and Alaria esculenta f., were growing. All the Laminarias were sterile and the newr laminae had not begun to grow. The frond of the Alaria was usually torn in pieces; in the middle of the stipe there were old sporophylls. The uppermost part of the stipe, just below the leaf-base, was young and evidently growing. Lowermost in the part that was growing, small projections could be seen on the two sides, which were evidently the begin- nings of new sporophylls. The following species were found thrown ashore: Lami- naria saccharina, L. digitata, L. hyperborea, Alaria esculenta f., Sac- corrhiza derrnatodea, Dcsmarestia aciileata, Odonthalia dentata, Petro- celis Hennedyi richly fruiting (on L. hyperborea), Rhodochorton Rothii (on L. hyperborea), Delesseria sanguinea with the small tetraspore- bearing fronds, and Ptilota plumosa with tetraspores. For the further understanding of the winter-life of the algae it may be mentioned that the winter of 1911- -1912 was unusually mild at Reykjavik. BIBLIOGRAPHY. 1. Agardh, C. A., Icones algarum ineditse. Fasciculi qui exstant duo. Editio nova, Lundae, 1846. 2. Species algarum rite cognitae. I-II, Gryphisvaldiae, 1821 — 1828. 3. Agardh, J. G., Bidrag till kannedomen af Spetsbergens alger, med Tillag. Kgl. Sv. Vetensk. Akad. Handlingar, Bd. 7, N° 8, Stockholm, 1868. 4. De Laminarieis symbolas offert. Lunds Universitets Arsskrift. Tom. IV, 1867. 5. Species genera et ordines algarum. Vols. I-III, Lundae et Lipsite, 1848 — 1876. 6. Ahlner, K. , Bidrag till kannedomen om de svenska formerna af algslagtet Enteromorpha. Stockholm, 1877. 7. Areschoug, J. E., Observationes Phycologicae. Particula prima, de conferva- ceis nonnullis. Upsaliae, 1866. 8. Phyceae scandinavicae marinae sive Fucearum nee non Ulvacearum, quae in maribus paeninsulam scandinavicam alluentibus crescunt, descriptiones. Up- saliae, MDCCCL. 9. Berthold, G., Ueber die Vertheilung der Algen im Golf von Neapel. Mitth. aus der zool. St. zu Neapel, III Bd., 4. Heft. 10. Boye, P.. Bidrag til Kundskaben om Algevegetationen ved Norges Vestkyst. Bergens Mus. Aarbog, 1894 — 95, NO XVI. 11. Borgesen, F., Om Algevegetationen ved Faeroernes Kyster. Kjobenhavn og Kristiania, 1904. 12. — The Algae- Vegetation of the Faeroese Coasts. Botany of the Fseroes. Part III, Copenhagen, 1905. 13. The Marine Algae of the Faeroes, Botany of the Faeroes, Part II, Copen- hagen, 1902. 14. and Helgi Jonsson, The Distribution of the Marine Algae of the Arctic Sea and the northernmost part of the Atlantic. Published Aug. 1905, Botany of the Faeroes, Part III, Copenhagen, 1908. 15. De la Pylaie, Flore de 1 ile de Terre-Neuve et des iles St. Pierre et Miclon. Paris, 1829. 16. Farlow, W. G.. Marine Alga? of New England and adjacent coast. Washington 1881. Reprinted from Report of U. S. Fish. Commission, 1879. 17. Falkenberg, P., Die Meeresalgen des Golfes von Neapel. Mitth. aus der zool. St. zu Neapel, I Bd., 2. Heft. 18. Foslie. M., Contribution to the Knowledge of the Marine Alga? of Norway. I, East-Finmarken. Tromso Mus. Aarshefter 13, Reprint, Tromso, 1890. 19. The Norwegian forms of Lithothamnion. Det Kgl. norske Vidensk. Selsk. Skrifter, 1894. Trondhjem, 1895. 20. Ueber die Laminarien Norwegens. Kristiania Vidensk. Selsk. Forh.. 1884, Nr. 14. Kristiania, 1884. 184 BIBLIOGRAPHY 21. Foslie, M.. Om nogle nye arctiske havalger. Kristiania Vidensk. Selsk. Forh 1881. Nr. 14. 22. — Nye havsalger. Tromso Mus. Aarshefter. Tromso, 1887. 23. Gran, H. H., Algevegetationen i Tonsbergfjorden. Kristiania Vidensk. Selsk Forh. for 1893. Kristiania, 1893. 24. Kristianiafjordens Algeflora. Vidensk. Selsk. Skrifter, I Mathem.-nat. Klasse, 1896, NO 2, Kristiania, 1897. 25. Hans teen, B., Algeregioner og Algeformationer ved den norske Vestkyst. N}rt Magazin for Naturvidenskaberne. 1892, Kristiania. 26. Harvey, W. H., Phycologia Britannica, Vols. I-III. London, 1846—1851. 27. Helland-Hansen, Bjorn and Naiisen, Fridtjof, The Norwegian Sea Re- port on Norwegian Fishery and Marine-Investigations. Vol. II, 1909, N° 2. Kri- stiania, 1909. 28. Huber, J., Contributions a la connaissance des Cha'tophorees epiphytes et endophytes et leur affinites. Ann. scienc. nat. VII Serie, 1892, vol. 16. 29. Imhauser, L., Entwickelungsgeschichte und Formenkreis von Prasiola, Flora, 1889. 30. Icones Florae Danicae, Vols. 1-16, Haunia?, 1766—1877. 31. Jonsson, Helgi, The Marine Algae of Iceland, I-IV, Bot. Tidsskrift, 24-25. Bd. Kjobenhavn, 1902—1903. 32. The Marine Algse of East-Greenland, Meddel. om Gronland, XXX. Copen- hagen, 1904. 33. Om Algevegetationen ved Islands Kyster. Kobenhavn, 1910. 34. Kj el 1m an, F. K., Uber Algenregionen und Algenformationen im ostlichen Skager Rack. Bih. till K. Svensk Vet. Ak. Hdl. Bd. 5, N° 6. Stockholm, 1878. 35. Handbok i Skandinaviens hafsalg-tlora, I, Fucoidese. Stockholm, 1890. 36. The Algte of the Arctic Sea. Kungl. Svenska Vetensk. Akad. Hdl. Bd. 20, N° 5. Stockholm, 1883. 37. Zur Kenntniss der marinen Algentlora von Jan Mayen, Arkiv for Botanik Bd. 5, NO 14. Upsala & Stockholm, 1906. 38. Ueber die Algenvegetation des Murmanschen Meeres an der Westkiiste von Novaja Semlja und Wajgatsch. Upsala, 1877. 39. Bidrag till kannedomen om Islands hafsalg-tlora, Bot. Tidsskrift 3. Raekke 3. Bd. Kjobenhavn, 1879. 40. — Om Spetsbergens marina klorofyllforande thallophyter. Bih. till Kung. Sv. Vet. Akad. Hdl. Bd. 3, Nr. 7, 1875, and Bd. 4. Nr. 6, 1877. 41. Studier ofver Chlorophyceslaegtet Acrosiphonia J. Ag. och dess scandinaviske arter. Bih. till Kung. Sv. Vet. Akad. Hdl. Band 18, Afd. Ill, NO 5. 42. Blastophysa polymorpha och Urospora incrassata. Bih. till Kung. Sv. Vet. Ak. Hdl. Bd. 23, Afd. Ill, NO 9. 43. Kleen, E., Om Nordlandenes hogra hafsalgar. Ofversigt af Kung. Vet. Akad. Forh., 1874, NO 9, Stockholm. 44. Knudsen, Martin, Contribution to the Hydrography of the North Atlantic Ocean. Meddelelser fra Kommissionen for Havundersegelser. Serie: Hydrograii Bind I, Nr. 6. Kebenhavn, 1905. 45. Kylin, Harald, Studien fiber die Algenflora der schwedischen Westkiiste Upsala, 1907. 46. Kuckuck, P., Beitriige zur Kenntniss der Meeresalgen. Sonderabdruck aus : \\'issenschaftliehe Meeresuntersuchungen, herausgegeben von der Kommission BIBLIOGRAPHY 185 zur Untersuchung der deutschen Meere in Kiel und der Biologischen Anstalt von Helgoland. Neue Folge. II Bd., Heft. I. Kiel und Leipzig, 1897. 47. Kuckuck, P., Bemerkungen zur marinen Algenvegetation von Helgoland. Sonderabdruck aus: Wissenschaftliche Meeresuntersuchungen, herausgegeben, etc. Neue Folge. I Bd., 1894, II Bd.. 1897. Kiel und Leipzig. 48. Beitrage zur Kenntniss einiger Ectocarpus-Arten der Kielerfohrde. Separat- abdruck des Bot. Gentralblattes, 1891, Bd. 48. 49. Le Jolis, Auguste, Examen des especes confondues sous le nom de Lami- naria digitata Auct.N. Act. Ac. Leop.-Car. Vol.25, 1856. 50. — Liste des Algues marines de Cherbourg. Paris, 1863. 51. Lyngbye, H. C., Tentamen Hydrophytologiae Danicae. Hafniae, 1819. 52. Niel sen, J. N., Contribution to the Hydrography of the Waters North of Ice- land. Meddel. fra Kommissionen for Havundersegelser. Serie: Hydrografi, Bind I, Nr. 7. Kebenhavn, 1905. 53. — Contribution to the Hydrography of the North-eastern Part of the Atlantic Ocean. Meddel. fra Kommissionen for Havundersogelser. Serie: Hydrografi, Bind I, Nr. 9. Kebenhavn, 1907. 54. Oltmanns, F., Morphologic und Biologic der Algen. Zweiter Band. Jena, 1905. 55. Paulsen. Ove, Plankton-Investigations in the Waters round Iceland in 1903. Meddel. fra Kommissionen for Havundersegelser. Serie: Plankton, Bd. I, Nr. 1, 1904. Kjebenhavn, 1904. 56. Plankton-Investigations in the Waters round Iceland and in the North At- lantic. Meddel. fra Kommissionen for Havunders0gelser. Serie: Plankton, Bd. I, Nr. 8. Kjebenhavn, 1909. 57. Petersen, Henning E., Ceramium-Studies land II. Saertryk af Bot. Tidsskr. 31. Bind. Kebenhavn, 1911. 58. Reinke, J., Algenflora der westlichen Ostsee deutschen Antheils. Kiel, 1889. VI. Bericht der Kommission zur Untersuch. d. deutschen Meere in Kiel. 59. — Atlas deutscher Meeresalgen. Im Auftrage des K. preuss. Minist. fur Land- wirtschaft, Domanen und Forsteu, herausgegeben im Interesse der Fischerei von der Kommission zur wiss. Untersuch. d. deutschen Meere. Kiel, 1889 — 92. 60. Beitrage zur vergleichenden Anatomie und Morphologic der Sphacelariaceen. . Bibliotheca Botanica, Heft 23. Cassel, 1891. 61. Rosenvinge, L. Kolderup, Grenlands Havalger. Meddel. om Grenland, III. Kjebenhavn, 1893. 62. Deuxieme memoire sur les Algues marines du Groenland. Meddel. om Gren- land, XX. Kjebenhavn, 1898. 63. • Om Algevegetationen ved Grenlands Kyster. Meddel. om Grenland, XX. Kjebenhavn, 1898. 64. On the Marine Algae from North-east Greenland. Danmark-Ekspeditionen til Grenlands Nordestkyst, 1906—1908. Bind III, Nr. 4. Meddel. om Grenland, XLIII. 65. The Marine Algae of Denmark, Part I. K. Danske Vidensk. Sclsk. Skr. 7. Rsekke. VII. Nr. 1, Kjebenhavn, 1909. 66. Sauvageau, C., Sur quelques Myrionemacees. Ann. d. sc. nat. 8. Serie, Botanique. Tome V. Paris, 1907. 67. Remarques sur les Sphacelariacees. Journal de Botanique. Tome XIV, 1900, and Tome XV, 1901. 68. Schmidt. Jobs., Danmarks blaagreune Alger (Cyanophyceae Daniae), I, Bot. Tidsskrift. Kjebenhavn, 1899. 186 BIBLIOGRAPHY 69. Schmitz, Fr., Die Chromatophoren tier Algen. Verb. d. nat. Ver. d preuss. Rheinl. Jahrg. XXX, 4. Folge, X Bd. 70. Stromfelt, H. F. G., Om Algevegetationen vid Islands kuster. Goteborgs Kgl. Vetenskaps och Vitterhets Samhalles Handlingar. Goteborg, 1887. 71. Svedelius, Nils, Studier ofver Ostersjons hafsalg-flora. Upsala, 1901. 72. Warming, Eug., Oecology of Plants. Oxford, 1909. 73. Wille, N., Studien iiber Chlorophyceen, I-VII. Vidensk. Selsk. Skr. I. Mathem. naturvidensk. Klasse, 1900, Nr. 6. Kristiania, 1901. 1A. Bidrag til Algernes physiologiske Anatomi. Kgl. Sv. Vet. Akad. Hdl. Bd. 21. Stockholm, 1885. 75. Om Fucaceernes Blaerer. Bib. till Kgl. Sv. Vet. Akad. Hdl. Bd. 14, Afd. Ill, NO 4. Stockholm, 1889. 76. Wittrock, V. B.. Forsok till en monografi af algslagtet Monostroma. Stock- holm, 1866. 77. 0rsted, A. S., De regionibus marinis, Elementa topographic historico-naturalis freti Oresund. Haunia?, 1844. ERRATA. P. 3, line 6, for of read off. P. 29, line 15, for Kylin (43) read Kylin (45). THE BOTANY OF ICELAND EDITED BY L. KOLDERUP ROSENVINGE PH. D. AND EUG. WARMING PH. D., SC. D. PART I 2. AN ACCOUNT OF THE PHYSICAL GEOGRAPHY OF ICELAND WITH SPECIAL REFERENCE TO THE PLANT LIFE BY TH. THORODDSEN PROFESSOR, PH. D. • (PUBLISHED BY THE AID OF THE CARLSBERG FUND) COPENHAGEN J. FRIMO LONDON JOHN WHELDON.& CO. 1 914 PRINTED BY H. H. THIELE 2. AN ACCOUNT OF THE PHYSICAL GEOGRAPHY OF ICELAND WITH SPECIAL REFERENCE TO THE PLANT LIFE BY TH. THORODDSEN PROFESSOR, PH. D. WITH 36 FIGURES IN THE TEXT The Botany of Iceland. I. 13 CONTENTS. Page I. General Topography. Geology 191 Glaciers 200 Snow-line 202 Rivers 205 Lakes 209 Geology 211 Volcanoes and Lava-streams 219 II. Conditions pertaining to Surface and Soil 232 III. Climate 265 IV. Account of the General Distribution of Plant-life 292 V. A Sketch of the Chief Plant-formations of Iceland. 317 13* I. GENERAL TOPOGRAPHY. GEOLOGY. I CELAND is a large island in the North Atlantic. It stretches from 63V2° to 66V2° N. lat. and from 13° 27' to 24° 30' W. long. fr. Gr. ; consequently, the Arctic Circle touches its northernmost points, and as a result, along the northern coast, the midnight sun turns night into day for a short time during summer. In the most northerly districts the sun is above the horizon for a week; at Reykjavik the longest day is 20 hours 56 minutes, the shortest 3 hours 58 minutes. The distances from Iceland to the neighbouring countries are as follows: - to Norway 950 km., to Scotland 900 km., to the Faroes 450 km., to the east coast of Greenland 330 km., and to Denmark 1500 km. The length of the island from east to west is 490 km., and the breadth from north to south is 312 km. Its area is about 104,000 sq. km. Iceland is a very mountainous country although it has not any true mountain-chains; it is most properly described as a continuous table-land with an average height of 700 — 1000 metres above sea- level: besides, there are only narrow borders of coastal land, valleys which cut into the table-land from all sides, and a few small areas of level land towards the south and west. Scarcely one-fifteenth of the country can be reckoned as lowland. Owing to its northern situation, its height above sea-level, and the resulting severe climate, only a relatively small part of the country is inhabitable. More than two-thirds of the entire area of the country is situated at so great a height above sea-level that almost no vegetation can thrive there. The sandy and stony deserts of the interior plateau, the lava tracts, and the glaciers are not fit dwelling places for human beings: therefore it is almost exclusively the coasts and the valleys which are inhabited. 192 THORODDSEN The coast of Iceland is upon three of its sides indented by fjords and bays, while the south coast is almost wanting in inden- tations — fjords, harbours, etc.; the coast-line is best developed to- wards the north-west, where a peninsula with many fjords projects from the main land. The coast-line of Iceland is about 6000 km. in length, of which one-third belongs to the north-western peninsula. The indentations of the coast may be divided, according to their origin and size, into two groups, the large bays which are probably formed by the sinking of the earth's crust, and the fjords proper which appear to owe their origin chiefly to erosion. Along the south coast the great Jokulls (snowfields) descend toward the sea and the masses of gravel carried down from the glaciers have, in the course of time, filled up all the fjords, causing the sea outside the fjords to be so shallow that it is dangerous for vessels to approach the coast. The breakers throw up bars of coarse shingle which dam the waters of the numerous glacier-rivers, and so shoals and shifting lagoons are common along this coast; the largest of them are to be found towards the south-east (Alftafjorflur, Papos and Hornafjor5uij. In the western part of the south coast the fjords and bays of olden times have been filled up not only with the sediments of glacier- rivers, but also with great lava-streams. From the west two large bays, FaxafjorSur and Brei5ifjor5ur, extend inwards, they are separ- ated by the mountainous and volcanic Snaefellsnes. The former of these bays is 68 km. in length and 90 km. in breadth, the latter 124 km. in length and 74 km. in breadth; from them several small fjords extend inwards, especially from BreiSifjorSur, which at its head divides into two large arms, GilsfjorSur and Hvammsfjordur. The north-western peninsula (VestfirSir), as mentioned above, is in- dented by numerous fjords, and there are several excellent harbours, trading-stations and fishing villages in those fjords which turn their mouths towards the north-west; of these fjords the largest are Arnar- fjor5ur and Isafjardardjup, both of which have many branch-fjords. On the north coast there are also many large indentations which are separated by mountainous peninsulas; of these bays and fjords the following are the largest, proceeding from west to east: Huna- floi, Skagafjor5ur, Eyjafjor5ur, Skjalfandi, Axarfjordur and Thistil- fjor5ur; and from these larger bays several smaller fjords extend further into the country. Of the peninsulas the following may be mentioned: Skagi, Tjornes, Melrakkasljetta and Langanes; the two large peninsulas on either side of Eyjafj6r5ur have no distinctive PHYSICAL GEOGRAPHY 193 names. In East Iceland there are also many larger and smaller bays and fjords; the largest and best- known are Vopnafjor5ur, Hjeradsfloi, Sey5isfjor5ur, ReydarfjorSur and Berufjordur. Although the coast of Iceland is so cut up by fjords, it is not especially rich in islands; only in Brei5ifjor5ur is there a number of islands and islets, in two large groups or collections; otherwise, the islands along the coast of Iceland are few and scattered, and are usually high and rocky. To the SSW. of Iceland are situated the volcanic islands, the Vestmannaeyjar and Fuglasker, and north of Iceland, in the Arctic Ocean, the Isle of Grimsey, 45 km. from the coast. The north, east and north-west coasts of Iceland, which abound in fjords, are everywhere rocky, and rise steeply from the sea like black walls 300 — 500 metres in height, composed of from 30 to 100 layers of basalt, distinguishable as narrow ledges or steps. Numerous small streams have excavated channels in the rock-sides, and leap in small cascades from ledge to ledge. The foot of the mountain and the narrow coastal-land are usually green and grassy in places \vhere the rocks are not too steep, but the mountain itself is chiefly of dark rocks, covered with gravel, and writh white patches of snow in the higher regions. At the head of the fjords, whence the various valleys branch off into the interior of the country, there usu- ally occur several or a fe\v groups of mountains with crests, ridges and peaks often of the most fantastic form, while the edge of the mountains along the fjords resembles w'alls with bastions and battle- ments. At many of the fjords there are trading-stations with the wooden houses painted white or red, while scattered under the sides of the mountains the white gables of the farm-houses peep forth amidst the sap-green home-fields. On the south coast the mountains retreat and the strand is bounded by sandy and pebbly flats; along these tracts the mountains are usuallv more rounded and softer in %J outline, as they are composed of tuff and breccia. The highest parts of the plateau are covered by snow-fields (Vatnajokull, Myrdalsjokull) from \vhich large and small glaciers come down through every cleft, and extend to the level country. Iceland consists of two distinct table-lands, one large (about 88000 square km.) and one much smaller (about 9000 square km.), the north-wrestern peninsula, which is attached to the mainland by a narrow7 neck of land only, forming a table-land by itself. The large table-land which almost entirely occupies the remaining part of the island is highest towards the SE. where the sno\v-masses of 194 THORODDSEX Vatnajokull cover an area of about 8000 square km. The axis of elevation lies from NW. to SE., from somewhere near the head of Hvammsfj6r5ur to Hornafjor5ur; it does not, however, consist of one continuous ridge, but of a series of snow-covered, dome-shaped mountains separated by broad stretches of more level ground. These snow- and ice-covered domes are strictly speaking a series of small plateaus which rise from the main plateau to an absolute height of 1400-2000 metres, as compared with 600 — 1000 metres above the plateau. The most easterly of the great glacier-bearing moun- tains is Vatnajokull which is separated from the much smaller Tungnafellsjokull (100 square km.) by Vonarskard (1000 metres); be- tween Tungnafellsjokull and Hofsjokull (1350 square km.) lies the broad stretch of level ground, Sprengisander (650 metres); west of Hofsjokull and between the latter and Langjokull (1300 square km.) lies Kjolur or Kjalvegur (600 metres); and between Langjokull and Eiriksjokull (100 square km.), Flosaskar5 (800 metres). The plateau north of the last-mentioned ice-mountains abound in lakes and bogs. The interior plateau consists chiefly of deserts almost destitute of vegetation, but the surface varies somewhat in character in accord- ance with the geological nature of the underlying rock. Where basalt or dolerite forms the substratum, the surface is strewn with innumerable angular blocks of rock, split asunder by the action of frost; where tuff and breccia form the foundation, the surface is usually covered with gravel and fragments of slaggy lava which have been loosened from the breccia bv the action of weathering. •/ More than one- half of the plateau is overlain by more recent formations lava, blown sand, volcanic ashes, glacial forma- tions, clay and river-gravel. The lava-fields, taken together, cover a very large area in the interior and present a most desolate scene; as far as the eye can reach only a black, hardened mass is seen, and the dark colours are only here and there interrupted by mounds of reddish slag, smoking craters, scattered snow-drifts, and in the distance by glistening, white Jokull-domes; there is no sign of life and an oppressive silence reigns over the land. The interior plateau is trenched by many valleys, chiefly towards the north and east; between these valleys long mountain spurs the skeletal ribs of the eroded plateau branch outwards toward the sea. Of these, the mountain- mass which extends towards the south, and is crowned at the top by Myrdalsjokull , is the most considerable. Towards the west the two volcanic mountain-chains PHYSICAL GEOGRAPHY 195 extend outward upon Reykjanes and Snaefellsnes. The mountain- chain on Reykjanes is broad and flat at the top; it is divided into several smaller plateaus (800 — 600 metres) with rows of craters and volcanic mountain-tops, and it sinks down more and more towards the west so that the extreme end of the peninsula consists chiefly of low-lying lava-fields with a few low, isolated mountain- tops. The mountain-chain which extends toward Snaefellsnes is higher (600 — 900 metres), but much narrower; it is also very vol- Fig. 1. Snsefellsjokull (West Iceland). canic and terminates in the ancient, ice-capped volcano, Snaefells- jokull (1446 metres). In North Iceland several great mountain-masses proceed from the high land outward upon the peninsulas, and from the head of the fjords long valleys extend into the country; by the extensive branching of the valleys, the mountains are divided into a number of ridges and peaks, which however when examined more closely, prove to have been cut out of an originally continuous plateau. Several valleys lead up from Hunafloi, of which the western are narrow, but the eastern (Vi5idalur, Vatnsdalur, Blondudalur) are broader and more fertile; these latter open out toward the low land (Thing) at the head of Hunafloi. The large peninsula between Skaga- fjorSur and Eyjafjorftur is occupied by mighty, steep mountain- masses which are intersected on both sides by numerous valleys, the largest (Oxnadalur, Horgardalur, Svarfaftardalur) being on the 196 THORODDSEN east. On this peninsula occur the greatest heights in northern Ice- land (Vindheimajokull, 1445 metres; Kerling 1350 metres; Rimar, 1261 metres). Broad, fertile and thickly-inhabited main valleys stretch up from the head of Skagafjordur and Eyjafjor5ur; east of Eyja- fjor5ur, between the latter and Skjalfandi, the stern, wild mountains rise to a height of 1200 metres, and there Fnjoskadalur, which cuts through these mountains towards EyjafjorSur from the east, opens out. These mountains fall abruptly towards the east down towards Bar5ar- dalur (70 km. long), but east of that valley the country becomes lower, and has a different geological structure and character; hitherto, basalt has been the dominant rock, but hereafter, for a long distance, the substratum is formed of tufY, breccia and recent volcanic forma- tions. East of Bar5ardalur the plateau gradually sinks down to- wanjls the sea, and inhabited spots occur, not only along the coast and in the valleys, but also here and there upon the high land itself (Myvatnssveit, 300 metres above the level of the sea ; Fjallasveit, 450 — 500 metres). Upon these areas of the plateau there are many isolated mountains and mountain-ridges, volcanoes and lava-streams. The eastern part of Iceland is called collectively Austfirdir. There the country is very mountainous towards the coast and in- *, «, dented by deep fjords which are bounded by high, precipitous rock- walls, basalt being now again the dominant rock. From HjeraSsfloi two long valleys, 80 — 90 km. in length, extend towards the south. These valleys, which unite lower down, are called Jokuldalur and Fljotsdalshjera5; the latter lies behind the fjord district of eastern Iceland and cuts it off from the interior plateau which is here somewhat lower than the coastal mountains, which reach a height of 10 — 1100 metres and even more. Towards the south there are two smaller Jokulls, Thrandarjokull and Hofsjokull f Loni , with an area of 100 and of 80 square km. and a height of 1100—1200 metres respectively. Near Hornafjordur the character of the coastal land changes completely, because there the plateau, to its extreme edge, is buried beneath the snow and ice fields of Vatnajoknll. The low, narrow and sandy coastal border is irrigated by glacier- water streaming down from the numerous glaciers which advance through every cleft and valley. Here and there, along the edge of the snow-field, mountains and rocky promontories appear, and of the former the imposing volcano of Orrefajokull (2119 metres), Iceland's highest moun- tain, is the most noticeable; other promontories and mountains near the southern edge of Vatnajokull have a height of 1100 — 1500 metres. PHYSICAL GEOGRAPHY 197 North of the mountain of Lomagnupur (770 metres), which is a breccia-promontory rising steeply from a sandy flat, the glaciers recede from the coast and the plateau is continued as a low hilly edge as « far as Myrdalsjokull, the snow-masses of which cover the upper part of the above-mentioned southern spur of high land and which has an area of about 1000 square km., reaching a height of 1666 metres in the volcano of Eyjafjallajokull. Another very active and dangerous volcano, Katla, also hides itself beneath the glaciers of Myrdals- jokull. West of this snow-field the plateau retires again from the coast; the deep valley of Markarfljot separates Myrdalsjokull from Tindfjallajokull (1462 metres), and north of the latter lies the oblong, volcanic Torfajokull (1400 metres) and the famous volcano Hekla (1447 metres). North-west of Hekla and near Geysir, the plateau retires to its greatest distance from the coast (about 80 km.), but bends thence again towards the south west, outwards towards the peninsula of Reykjanes. The lowlands of Iceland cover onlv a small area; in the north V and east there is no low land with the exception of the larger valley- mouths, e. g. at the head of Hunafloi, Skjalfandi and Fljotsdals- hjerad and the extreme end of the peninsulas of Skagi and Mel- rakkasljetta. Larger lowland areas occur only in South and West Iceland; they are however small in extent compared with the mountainous country and the table-lands. Although the lowlands only comprise one-fifteenth of the entire area of the country, yet together with the valleys they are of very great importance because tolerably favourable climatic conditions and a closer vegetation, per- mitting fixed habitations and the rearing of cattle, are found almost exclusively there; the inhabitants of the few dwellings which occur scattered in the lower parts of the plateau must struggle with severe conditions and are almost entirely reduced to sheep-rearing alone. But by no means all the lowlands are grass-covered; large tracts of lava, glacier-gravel and blown sand are extremely poor in plant-life. On the south coast the country nearest to the sea is quite flat, from Hornafjor5ur to Reykjanes, but in many places this level land is so narrow that it consists only of an insignificant coastal border; in other places it widens out into larger plains and extends further into the country. Between Hornafjor^ur and Myrdalsjokull, below Vatnajokull, the coastal country is formed by deposits from the numerous glacier-rivers and consists exclusively of sand and gravel ; the lowland here is often flooded by branching torrential glacier- 198 THORODDSEN rivers, so that no vegetation can thrive here. Below the southern edge of Vatnajokull the farmsteads are therefore confined to oases separated from one another by gravel-deserts and swollen glacier- rivers. In many places the people have been obliged to move their houses up on the mountain sides in order to avoid inundations caused by the glacier-waters, and here the mountain-sides facing south are covered with a luxuriant plant-growth, while the level country below is devoid of vegetation. The sandy tracts have different names, such as Breidamerkursandur , Skeidararsandur, Brunasandur, Myrdalssandur, Solheimasandur, etc. Of these the largest is Skei5ararsandur (about 900 square km.), and taken to- gether they cover an area of about 2700 square km. Between L6- magnupur and Myrdalsjokull the lowland becomes broader and is covered by extensive lava-fields, blown sand, gravel and volcanic ashes, as large volcanoes occur in the neighbourhood. The largest area of low land in Iceland lies between Eyjafjallajokull and the peninsula of Reykjanes and is about 4000 square km. in extent; it is hemmed in by tuff-mountains which in many places fall abruptly towards the plain. The lowest part of the lowland is only slightly raised above sea-level, but it rises gradually towards the interior where it ultimatelv branches off into different valleys; near Geysir */ •. «/' it reaches its greatest height above sea-level, about 150 metres. The low land is not flat everywhere, some parts of it are hilly, and a few isolated mountains also rise from the plain. West of Hekla the lowland, by means of a gentle rise, is in direct connection with the interior plateau, to the great danger of the inhabited districts, as the blown sand, volcanic ashes and pumice dust which cover large areas of the interior have thereby free access to the lowlands; there- fore, in these regions, during north-westerly storms, large tracts of grass-covered and inhabited land have been overwhelmed in the course of time. The lowlands consist chiefly of grassland, and no- where in Iceland do farmsteads occur so closely together as here, these districts being well-suited to cattle -rearing. Three very well- supplied rivers run through the lowland, viz. Markarfljot with a large delta-land (Landeyjar), Thjorsa and Olfusa, and besides these there are many smaller ones. The eastern part east of Thjorsa is called RangarvallasVsla, the western Arnessysla. At the head of Faxa- floi there is another low area (about 1000 square km.) which on an average has a height of only 20 — 30 metres. It is bounded by steep, basalt mountains which are arranged in a semicircle around PHYSICAL GEOGRAPHY 199 the lowland which is very swampy and above which project numerous low, isolated, basalt hills. The eastern part is called BorgarfjorSur and the western Mjrar. From the low land several wide valleys extend inwards between the mountains. The north-western peninsula, as mentioned above, constitutes a small plateau by itself and is separated from the mainland by Breidifjordur and Hunafloi; the isthmus which connects the peninsula with the mainland has a breadth of only 7 km., and rises to a height of 228 metres. The coast of north-west Iceland, which is much in- dented by tjords, is bounded everywhere by steep, dark mountain- sides which often rise abruptly, even vertically, from the sea to a height of 400 — 500 metres. The mountains are everywhere composed of horizontal, or slightly inclined, basalt-layers which are highly denuded by erosion, so that numerous small valleys, cirques, clefts, ridges and bastions occur as in other basalt regions of Iceland. By climbing the mountain-edge up in the highlands, and ascending to a sufficient height it will be seen that fjords and valleys have trenched the plateau with great regularity. The eye wanders freely over wide wastes, the valleys and fjords either disappear or are seen as insignificant clefts, and a monotonous table-land lies stretched out to view, the surface of which is broken only by low ranges of hills, and ice-striated basalt-masses with large stretches covered by angular blocks of rock and scattered snow-wreaths. It is often ex- tremely difficult to traverse this kind of ground, across an ocean of rocky blocks, wrhere the clay and the sravel between the blocks ft/ V often occur as a slushy mass owing to the thawing of the numerous patches of snow. Only now and then can a solitary, stunted Alpine plant be seen maintaining a miserable existence in the shelter of the large blocks of rock. The plateau of the north-western peninsula has an average height of only about 600 metres, and where it is highest (800—900 metres) the snow drifts and consolidates into neve- domes - - Glama (about 60 square km.) towards the SW. and Dranga- jokull (about 350 square km.) towards the NE. The numerous valleys which lead up from the fjords and cut into this plateau are similar in character to the other Icelandic vallevs in the basalt districts, *. so it will be convenient to describe their physiographical charac- ters collectively. The bottom of the valley rises gradually in broad rock-terraces, upon the surface of which bogs and small lakes often occur, and it terminates amphitheatrically in a large cirque, down the steep rocky sides of which several streams fall in cascades. 200 THORODDSEN The mouths of the contributory valleys are often situated at a higher level than the bottom of the main valley and along the mountain-sides a series of cirques are often found. Hundreds of streams carry gravel and rock-fragments down to the foot of the mountain and to the valley, and below each notch in the mountain there is, therefore, a flat cone of gravel which extends down the side of the mountain to the bottom of the valley; each little cirque and each notch has, like an hour-glass, according to the law of gravitation, emptied its contents upon the level land below. The foot of the mountain is covered with plants, but any aggregate of vegetation has rarely been able to extend higher than half-way up the mountain -side; only on ridges between clefts and hollows, where neither floods, rock-slips nor avalanches can do harm can the plant- covering extend upwards in longer tongues, while the upper half of the mountain-side consists but rarely of anything ex- cept bare rock-ledges or rock-faces or heaps of stones. On a closer investigation, a few individual plants will however be found seeking cover, shelter and foothold among the blocks of rock and in the crevices. In places where springs are trickling out in a row from between rock- strata there is often a luxuriant vegetation of yellowish- green mosses which form soil and pave the way for the higher plants. Even on the most precipitous valley-sides, sheep are seen scattered about seeking the mountain-plants which peep forth be- tween the stones. Upon the north-western peninsula there are no lowlands, but only a narrow7 border of coastal land which is due to the action of the breakers at a time when the sea-level was higher than at present. Only the narrow coastal land along the sides of the fjords is inhabited, and the inhabitants are chiefly de~ pendent upon the sea for subsistence. Where the land which fringes the coast becomes somewhat broader and the valleys more grassy, as along the north coast of Brei5ifjor5ur, the inhabitants' chief means of sustenance is sheep-rearing; where the fjords are small, the mountains steep, and the coastal land has disappeared, as along the coast south of Cape Nord, the inhabitants maintain themselves almost entirely by the catching of birds upon the steep sea-cliffs. Glaciers. The snow- and ice-covered mountains (Jokulls)1 of Iceland, taken together, cover an area of 12700 square km. and 1 In Iceland, by 'MokulT' a glacier-bearing mountain is usually meant, but sometimes the term is used for the masses of snow and neve upon the mountain. PHYSICAL GEOGRAPHY 201 through their glaciers and glacier-rivers they exert a great influence upon the surface and climate of the country and upon the condi- tions of life of the inhabitants. The Icelandic climate is specially adapted to the development of large glaciers, for the air is raw and cold and moist, the amount of rainfall considerable and the summer- heat slight. The amount of precipitation is greatest towards the south-east and there the interior tahle-land is covered bv the sreat •/ o Vatnajokull. The altitude of the snow-line varies from 400 to 1400 metres in the different parts of the country, and the level above the sea at which the glaciers end differs greatly: in the north on the north-western peninsula, and in the south near Vatnajokull, the glaciers descend almost to the sea, to 25 metres and 9 metres above sea-level respectively at the lowest points to which they descend. The great ice-mountains of Iceland are without exception closely associated with the plateau. Large areas of the highest part of the plateau are covered with neve which occurs as slightly-arching domes or undulating snow-fields of great thickness. Prominent mountain-peaks are rare; the latter do not appear until near the edges of the snow-fields and usually as outstanding summits of the underlying rock. The surface of these snow -fields is devoid of gravel; this does not appear until it does so at the extremities of the glaciers which are often quite black with it and with sand and blocks of rock. The large glaciers which descend from these /Re- covered flats have., on an average, a very slight declivity ; only in places where precipitous mountain-peaks project from the edge of the snow-field, do steep glaciers of small dimensions occur. The large glaciers of Iceland closely resemble the glaciers typical of Arctic countries; but there are a great man}7 small glaciers which resemble those of the Alps. Several of the broad glaciers which descend from Vatnajokull cover a very considerable area (e. g. Dyngju- jokull 400 square km., Bruarjokull 500 square km., etc.). Peculiar to Iceland are the so-called "glacier-torrents1' (Jokul-hlaup). When the glaciers, by the eruption of volcanoes hidden under the ice, are broken to pieces and melt, the large stretches of land beneath them are inundated by a roaring sea of dirty water with swirling ice- bergs. Such catastrophes may cause great changes in the surface- features of the surrounding country, as the waterfloods often carry along with them an incredible quantity of gravel and rocky blocks. In this way the volcano of Katla especially has caused considerable changes the course of rivers are constantly changed, the smaller 202 THORODDSEN fjords have been filled up even within historic times, and several large parishes and districts have been destroyed. Oraefajokull has, in the same manner, caused considerable destruction; while the waterfloods of Skei5ararjokull, which were especially frequent during the 19th century, have done less material damage, as only unin- habited sandy wastes were inundated. Minor glacier-bursts are also occasionally due to lakes and rivers, which have been dammed by glaciers, suddenly breaking through their barriers and inundating the district. These glacier-bursts have a great effect upon the plant- life, because no permanent vegetation can exist upon gravelly and sandy flats which are constantly inundated by waterfloods carrying large pieces of ice; therefore very large stretches of lowland in the neighbourhood of such volcanoes and glaciers are destitute of plant- growth, for what little vegetation appears in the period between the glacier-bursts is quickly destroyed. The general physiographical con- ditions pertaining to the glaciers of Iceland may be best seen by a study of the table opposite.1 Snow-line. In Iceland it is not easy to determine the snow- line, owing to the great variability of the climatic factors. Because of the great annual and periodical variability of temperature and of the circumstances connected with precipitation, the snow-line also varies. In Iceland three kinds of height-levels connected with the vertical distribution of the snow, and dependent on climatological and orographical conditions, may be supposed to exist. The snow7- line proper, which signifies the lowest limit of the permanent, con- tinuous snow-covering of the mountains, is not subject to very great changes from year to year. Below7 this comes a zone of detached, •/ */ more or less closely placed, patches of neve and wreaths of snow7; these never melt entirely, but either are added to or else diminish according to the character of the year. Below this zone comes the most variable of the snow-coverings: scattered snow-drifts which to a greater extent than the others are dependent on orographical conditions. These snow-drifts may persist through a series of damp and cold years, but dwindle almost to nothing or disappear entirely in warm and dry years. The snow at Drangajokull upon the north-western peninsula may serve as an example. Here, on 1 Here it should be noticed that many of the figures for the area of the Jokulls are approximate and given as a rough estimate, because the maps of a great part of Iceland, and especially of the plateau are still very imperfect. PHYSICAL GEOGRAPHY 203 Names of Jokulls *J C3 r- '"" r- • " r* X r~ JZ o £ . O *t \ * — - v f c. - 1- £ ^ *^ ^ •+ |^I % Altitude of snow-line in metres * o u •- = 9) V S — 'o o E j3 = Z 5^N ^^ O •- — <« 3 o i> -Si . 'o -b Glama ... cir. 60 901 cir 650 Drangajokull 350 890 7 east side 400 30 west side . . ... . 650 25 Snrefellsjokull 23 1446 2 560 north-east side . south-wrest side. Langjokull 1300 1400 650 1000 31 south-west side . . . east side 900 1000 600 435 north-wTest side .' . . . north side — — 1000 1100 — 628 Eiriksjokull . 100 1798 o 606 Ok 15 1198 900-1120 Hlooufell 1 1188 960 Arnarfellsjokull (Hofsjokull) east side 1350 1700 19 552 north side .... Myrdalsjokull 1000 1666 1200 20 910 north side 1100 700 south side 8-900 97 east side 6-700 205 west side 1000 220 Torfajokull 100 1200 950 Tindfjallajokull 22 1462 950-1000 5 570 Vatnajokull . 8000 2119 41 west side 1000 635 south side — north side — — 900 1300 — 9 600 east side Hofsjokull i Loni 80 1200 950 850-950 1 700 Thrandarjokull 100 1200 900 Siicpfell 15 1822 2 Tungnafellsjokull 70 1600 1300 2 Dyrfjoll 5 1131 600 Fonn . . ... 10 900 _ Kaldbakur 5 1161 Vindheimajokull 10 1466 1 _ Mvrkarjokull 10 1200 1 Tiinahrvggsjokull 40 1200 4 Unadalsjokull 20 1200 12686 — — 142 — The Botany of Iceland. I. 14 204 THORODDSEN the eastern side the snow-line stands at a height of 400 metres ahove sea-level, on the western side at an altitude of 650 metres. Upon the plateau itself, around the base of Drangajokull, there are numerous large, scattered wreaths of snow which taken collectively would cover a large 'area; they occur at an altitude of 250—500 metres. In the summer of 1886 — 87 there occurred, in addition, numerous wreaths at a lower level — in sheltered places even close to the sea-shore, especially at Snaefjallastrond, where very long snow- drifts were lying on the terraces of the basalt beds. In these districts old banks of snow are usually very frequent at a height of 50- 100 metres above sea-level; the climate also is very raw and the precipitation considerable; there are occasional snow-storms in the middle of summer, and drift-ice with cold, damp fogs and drizzling rain is a frequent visitor. The sea cuts into the land from all sides. Above the snow-line proper there is naturally no vegetation, with the exception of a few individuals which maintain their ex- istence on projecting rocks and "Nunataks" in the ice. In the region of permanent snow-drifts there occurs, only here and there, a poor and very scattered rocky-flat vegetation, but in the zone of the variable snow-drifts there may often be a vegetation of different species w^hich is rather luxuriant, considering all things. According to my measurements in the years 1882 — 1898 the snow-limits were about as follows (the snow-limits given in the table on p. 203 may serve for comparison): Snow-limit in metres Lowest limit of the permanent snow-drifts in metres Lowest limit of the variable snow-drifts in metres Northern part of the north-western pen- insula ... . 400-650 250-500 50-100 Snaefellsnes 800-1000 cir 700 500-600 Oda5ahraun 1400 1000 Mvvatn 700-800 • Peninsula NE. of Evjafjordur 1000 450-500 200 Austfirftir .• 900-1000 500-550 300-400 Plateau between Vatnajokull and Myrdals- iokull . 1100 600-700 ArnarvatnsheiSi NE. of Langjokull 1000-1100 600-700 500-600 The neighbourhood of the southern lowland 1000-1100 800-900 PHYSICAL GEOGRAPHY 205 Rivers. Iceland possesses numerous rivers of considerable size; their volume is due to the damp climate of the land and to the great number of glaciers. The Jokulls serve as huge water-reservoirs, from which the majority of the larger rivers are supplied, and the glacier-plateaus also attract rain and other atmospheric precipitation, so that the supply is always uniform. The volume of water is greatest in the lower-lying districts because the plateau above 500 -600 metres, to a great extent, is covered with lava, gravel and loose sand which absorb all the moisture; but it reappears further down, and below the 500 metre line there are extensive bogs both in the lowest part of the plateau, in the lowlands and in the val- leys. In the lava-deserts both on the plateau and in the lowlands, the rain and melted snow disappear immediately, and large tracts are quite destitute of water; but at the edge of the lava numerous springs bubble forth from the earth. The water which at first is usually turbid and intermingled with glacier-mud, becomes filtered by passing through the lava, and is therefore very pure and clear in the springs. When there is no outlet, water is sometimes found in the lava-field itself, at the bottom of deep clefts; as for instance, the renowned, cold, crystal-clear water in the lava-clefts near Thing- vellir. In early summer, during sudden thaw, wiiile the frost-layer still persists in the ground, extensive gravelly and clayey flats on the plateau and in the lowlands are turned into morasses, but when the ice of the subsoil melts later in the summer outlets are again opened for the surface-water so that the gravel-fields are drained. In the val- leys numerous springs make their way through the basalt along the mountain-sides, often in long rows between the layers, and can be detected by the green mosses which grow luxuriantly around them. Icelanders draw a distinction between "bergvatn" (mountain- water) which is clear, and "jokulvatn" (milky- white glacier- water) which may have a muddy, yellowish colour or a chocolate-brown »/ V S W colour according to the amount and kind of glacier-clay carried in it. The amount of clay in the glacier-water is larger in summer than in winter and, again, the glacier-rivers are clearer and of less volume in the morning than later in the day. Differences in the weather cold or warm or wet or dry years - have the greatest influence as regards the volume of water in the glacier-rivers. In dry and warm summers the clear rivers are but small, while the V water carried by the glacier-rivers increases to double or three times the usual volume. As the glacier-rivers are so dependent upon the u* 206 THORODDSEN melting of the snow their size varies from day to day and from year to year; many glacier-streams which disappear entirely during winter, carry in warm summers an immense volume of water. Al- most all Icelandic glaciers rest on soft rocks (tulT and breccia) upon which erosive action is very active; therefore, the Icelandic glacier- rivers carry dowrn an immense amount of rock in the form of mud, sand, gravel and blocks of stone; for this reason they are hardly ever found entering fjords or deep bays, these having quickly be- come filled up in cases where they formerly so entered, while the Jokulls (glacier-bearing mountains) are surrounded by large sandy and gravelly tracts which for the most part owe their origin to the rivers. Taking the whole of the island into account, rivers containing glacier-water are decidedly in the majority. South of Vatnajokull clear water is almost unknown, as all rivers and brooks originate in the glaciers. There all the rivers flow down to the coast by short courses in torrential current, and during summer some of them are so broad that it lakes hours to cross them but then it must be remembered that it is necessary, in the middle of the river, to go a long w^ay round, on account of the current and depth. On the flat, sandy tracts the rivers are constantly changing their course, and greater and smaller changes take place daily. All glacier- rivers branch abundantly. In accordance with the slope of the land, the longest and largest rivers flow in South Iceland towards the south-west and south, and in North Iceland towards the north; the majority of them rise on the plateau at a height of 600 — 900 metres above sea-level, a fact w7hich should be correlated with the limits of the glaciers in the interior. Although the Icelandic rivers carry a com- paratively great volume of \vater yet they are not navigable, because of their usually steep fall, their torrential current, and their tendency to spread out and subdivide into numerous branches in the low land. The largest rivers of Iceland are as follows: From the south edge of Vatnajokull rise Jokulsa f Loni, Hornafjar5arfljot, Jokulsa a Brei5amerkursandi, SkeiOara and Niipsvotn; the last three are considered to be the most dangerous glacier-rivers of Iceland. From the west edge of Vatnajokull rise Hverfisfljot, Skafta and the well-supplied Ku5aflj6t; this last also receives a large supply of water from M\rdalsjokull. From the latter another river flowrs down the short, but torrential Jokulsa a Solheimasandi, also called PHYSICAL GEOGRAPHY 207 Fulilaekur, which often has minor "glacier-torrents" that carry down ice-pieces and stones. On the east side of the southern lowlands, Markarfljot flows into the sea; it rises in Torfajokull, but also receives well-supplied feeders from Myrdalsjokull ; in the low land it divides into four arms which enclose the largest delta-land of Iceland, the so-called Landeyjar. Thjorsa, Iceland's longest river »* *'-3v * Fig. 2. The river J6kuls:'i a Solheimasandi. A ford. (200 km.) rises in Arnarfellsjokull, hut receives about one-half of its water supply from Vatnajokull through its tributary Tungna; Thjorsa carries an immense volume of water and, in the cultivated district, is in several places one km. or more broad; above its mouth it forms an expansion or a saccate lagoon and is joined by a well- supplied arm of Markarfljot, named Thvera. The third and most westerly large river in the southern lowlands is Olfusa which, for the longest part of its course (until the mouth of the tributary river Sog), flows under the name of Hvita and issues from Hvitarvatn near Langjokull receiving, both on the plateau and in the cultivated district, many large affluents from both sides. In the neighbourhood 208 THORODDSEN of Geysir, Hvita forms the large waterfall Gullfoss. Towards the west, another river named Hvita. flowrs down through the district of Borgarfjord; it carries a great volume of water and ils lower course is navigahle. In the north-western peninsula there are no large rivers, but in North Iceland many such occur, among others Blanda, which issues from Arnarfellsjokull and flows into Hiinafloi. From the same Jokull issues also Hjeraflsvotn which flows through the district of Skagafjord and empties itself into the fjord by two Fig. 3. The river Hvita flowing through the district of Borgarfjord. mouths. Three rivers of rather large volume Horga, Eyjafjar5ara and Fnjoska - -enter Eyjafjordur. Then there is Skjalfandafljot which issues from Tungnafellsjokull and flows through the Bardardalur; it has many waterfalls, among which Godafoss is the best-known. Jokulsa a Fjollum, one of the best-supplied rivers of Iceland, empties into Axarfjord and in its lower course falls through a deep cleft and forms Iceland's grandest waterfall, Dettifoss. In East Iceland two large rivers are noticeable, Jokulsa a Bru nnd Lagarfljot, both of which issue by many arms from the north edge of Vatnajokull and fall into Hjeradsfloi, after the latter river has expanded into a deep, oblong lake. Besides the numerous cataracts in the large rivers, there are also beautiful cascades (Fossar) in the smaller streams; of these the better-known are Hengifoss in Fljotsdalshjerad, Glymur in Botnsdalur, Dynjandi in ArnarfjorSur in the north-western pen- PHYSICAL GEOGRAPHY 209 insula, and Skogaibss and Seljalandsfoss below Eyjafjallajokull. These cascades have a height of about 100 metres and more. The largest rivers, only, have been mentioned above, but in addition to these, hundreds of streams of greater or lesser volume occur, often with beautiful waterfalls and cascades in connection with picturesque clefts and rocks. As mentioned above, the torrential and changeable glacier-rivers have a destructive influence upon the cultivation and vegetation of the plains. The greensward is torn off and large areas are covered by gravel, therefore the level country south of Vatnajokull is in several places turned into a desert almost destitute of vegetation except where special natural conditions afford a shelter from the destructive effect of the rivers. Where the action of the glacier-rivers is suddenly arrested by any natural phenomenon the level country again becomes quickly covered with plants. As an example may be mentioned the fact that Hverfisfljot, in the year 1783, was forced out of ils bed by a great lava-stream, and a considerable stretch of land — Brunasandur - which had previously been irrigated by cold and torrential river-branches was freed from these, only a few clear streams of filtered glacier-water with a slight current issuing from the edge of the lava-streams and flowing down the level country; so that where in 1783 there wras a gravelly and sandy flat without plant-life and without means of sustenance for human beings there is now a parish with seven farmsteads and abundance of meadows and pasture-lands for the sheep and cattle of the in- habitants. In itself the glacier-water is not inimical to vegetation; it is only the torrential current, the changeableness of the water- courses, and the lo\v temperature of the water which have a de- structive effect upon plant-growth; where these factors are not active, the glacier-water, with its contents of fine clay, is on the contrary a fertilizer; therefore in the neighbourhood of the mouths of the largest glacier-rivers, where there is only a slight current and the water has become warm on the way, fertile tracts of meadows are often found where the glacier-water is profitably utilized for irriga- tion. Water from rivers such as Thjorsa and Hvita has, by analysis, been proved to contain an unusually large quantity of alkali and phosphoric acid. Lakes. There are many lakes in Iceland, but the majority of them are of small size. The largest lakes Thingvallavatn and 210 THORODDSEX Thorisvatn cover an area of only about 100 square km. each: the lake-surfaces occupy therefore only a very small part of the entire area of the country. The lakes are of very diverse origin, the basins having been formed by tectonic movements, ice-erosion, volcanic action and other natural agents. On the plateau where the outlet is slight, especially in the neighbourhood of the large Jokulls (ice-mountains) many lakes occur, in some places gathered in large groups, as Fiskivotn on Arnarvatnshei5i NW. of Lang- jokull, and VeiSivotn W. of Vatnajokull; the melting snow and ice from the glacier-edges disappear in the nearest lava-streams and sandy tracts and then reappear and gather into basins many kilo- metres away from the glaciers from which they originated. In other places the lakes occur in the immediate neighbourhood of the glacier- edge, as Hvitarvatn and Hagavatn near Langjokull and Langisjor near Vatnajokull; the glaciers project into these lakes and calve their ice-bergs there; the water in these lakes is milky-white as in the glacier rivers. In some places lakes occur in between the glaciers (Graenalon near Skeidararjokull) or are dammed up in valleys by glacier-tongues. The best-known lake in Iceland is Thingvallavatn (105 square km.); it is situated in a new volcanic district bounded on the S. and W. by steep tuff-mountains and on the N. and E. by lava -streams which originate from the volcano of Skjaldbrei5; these streams have afterwards flowed down between the two well- known fissures Almannagja and Hrafnagja. It was here that the Icelandic Althing met in the time of the Republic. It appears as if the basin of the lake of Thingvalla was originally formed by sub- sidence along lines of fracture from SW. to NE.; this lake has a depth of 110 metres. In North Iceland Myvatn is the best-known lake; it is formed in a depression in the lava-stream and has a depth of only 2 — 7 metres; its bottom is lava and several craters project above its surface like islands, while the surroundings are very volcanic. Myvatn has received its name from the mosquitoes (my) which are often quite a plague there. As in Thingvallavatn, trout are plentiful in this lake, but it is especially known as the abode of numerous birds, especially many different species of ducks. The group of lakes called Vei5ivotn consists for the most part of crater-lakes, of which the largest is called Storisjor. In other places valley-lakes occur deeply hollowed basins in the basalt as Skorradalsvatn in Borgarfjorflur and Lagarfljot in East Iceland, the surface of which latter lies 2(> metres above sea-level while its bottom PHYSICAL GEOGRAPHY 211 lies 84 metres below the level of the sea. Lagoon-lakes, situated close to the coast, occur also, especially in North Iceland; the largest of them are Hop, H6f5avatn and Miklavatn. According to the condition of the bars and the outlets the quality of the water of the lakes may often change quickly; sometimes they contain fresh water, sometimes brackish water, sometimes salt water. The fauna in these lakes is also subject to periodical changes. Some- times marine animals immigrate through the outlets; at other times they disappear, and a freshwater fauna becomes dominant. Fishing in the Iceland lakes is of great economic importance to the inhabitants. In the larger rivers and in many smaller streams, salmon is caught, and now the fishing is often rented out to Eng- lish sportsmen, especially in south-west Iceland. Many rivers, and most of the lakes, abound in trout and salmon-trout (Salmo alpinus and 5. tratta) which play a great role as a means of support of the inhabitants. The quantity of living organisms varies greatly according to circumstances; there appears to be most life in the shallow lakes, especially in those with lava-bottoms, in which warm or luke-warm springs are also sometimes found at the lake-bottom, around which plant- and animal-life collect. Some Phytoplankton (Diatoms) occurs in the lakes in South Iceland, but on the plateau and in North Ice- land only Zooplankton (Daphnias) is found, and this is abundant. In Thingvallavatn, Chara and Nitella grow at a depth of 15 — 30 metres; in Myvatn there is a quantity of Nostoc which is found thrown up in great masses along the shores. In the lakes, there occur in addition, several species of Limmea and Pisidium, and Lepidiirns is frequently found in great quantities; there is, also, an abundance of gnat-larvae and other larvae which serve as food for the trout. Geology. Iceland is almost entirely built up of volcanic rocks, none of which appears however to be older than Tertiary times. The foundation of the island is a depressed and broken basalt- plateau similar to the other Tertiary basalt-plateaus on both sides of the Atlantic Ocean, in East Greenland, the Faeroes, Antrim in Ireland and on the islands of Mull and Skye. It is therefore as- sumed that in Tertiary times a continuous basalt-land extended across the Atlantic Ocean, that it subsided in Mid-Tertiary times and that Iceland and the Faeroes are the remains of this land. In Greenland and on Skve the basalt rests on Jurassic strata, in Mull 212 THORODDSEN and Antrim upon chalk; in Iceland, where the basalt formation has a thickness of at least 3000 metres, the underlying rock has not vet been found. \s The principal rocks of which Iceland is composed are two, basalt and palagonite breccia; more than one-half of the surface and the rock-foundation of the country consists of basalt, but the palagonite formation, which is composed of breccias, tuffs and conglomerates of different age and which, taken as a whole, is younger than the basalt formation, forms an irregular band across the country, occupying an area somewhat smaller than that occu- pied by the basalt. Compared with these two formations, all other rocks and formations have quite an unimportant distribution. The basalt mountains, the precipitous walls of which often rise from the sea to a height of 600 - 1000 metres, are composed of layers of varying thickness, wedged in between each other: the thickness of the individual layers often decreases rapidly along a short distance until the layer disappears and gives place to another. In the basalt formation, beds of tuff and breccia sometimes occur between the basalt layers, but their amount is inconsiderable compared with that of the basalt. Dykes are frequent; the majority of them pierce down through the entire series of layers. Seen from a distance, the basalt- mountains with their steep, terraced walls, have a monotonous and gloomy appearance, but on closer inspection exhibit rather great variation. Some of the basalt layers are compact, others are coarsely crystalline, doleritic, porphyritic, amygdaloidal (with more or less completely filled vesicular cavities), slaggy, banded, etc. In some districts the basalt is cleft into beautiful columns; in others into more or less irregular, angular blocks; in others it has an almost slaty appearance. In the vesicular cavities of the basalt zeolites, quartz, chalcedony and calc-spar are often found. As a general rule the basalt layers have a slight inclination (3° — 5°) from the coast inwards towards the tuff and breccia formations, which appear to fill a flat, saucer-like depression in the underlying basalt plateau; but many local deviations occur owing to dislocations and sub- sidences of larger or smaller areas of the underlying rock. In the middle of the basalt-formation in Iceland (as also in the Fasroes and in Ireland) rather considerable clay deposits are found with the impressions and remains of plants of Tertiary times; also lignite and compressed tree-trunks, all called in Icelandic usur- tarbrandur." This plant-bearing formation attains its greatest thick- PHYSICAL GEOGRAPHY 213 ness (20 — 50 metres) in the north-western peninsula and was origi- nally deposited at one level, but was afterwards broken up by dis- locations, so that it now occurs at different levels above the sea. The surtarbrand formation consists of diversely coloured layers of clay and tuff with intercalated layers of lignite and coal-slate; in many places leaves and fruits are excellently preserved in it, espe- cially near Brjanslsekur on the northern side of Brei5ifj6ri)ur, and at Trollatunga and other localities near Steingrimsi]6r5ur; in this Fig. 4. Mountains near KolviSarholl (tuff and breccia). last place it can be seen that large Tertiary woods have been des- troyed by pumice-eruptions and lava-streams. According to O. Heer the mean temperature of the year in north-wrest Iceland, at the time of these Tertiary woods, was at least 9° C probably somewhat more (11° — 12°); now the average temperature for the year in these places is only 2°. The most common tree in western Iceland at that time was Acer otopterix; its leaves are found in abundance and excellently preserved in the clay-layers; there occurred in ad- dition Sequoia Sternbergi, Pinus Steenstrnpiana , P. microsperma , P. aemula, P. brachyptera, Betula prisca, Alnus Kefersteinii, Ulmus diptera, Quercus Olafssonii, Liriodendron Procaccini, Vitis islandica, Rhus Brun- neri, Dombeyopsis islandica, etc. The Palagonite formation is composed of different kinds of tuffs and breccias, and in its upper divisions there is much moraine material and scattered glacial gravel, as also some ice-striated lava- 214 THORODDSEN streams. The tuffs have usually a brownish-yellow colour owing to the intermixture of palagonite, a brown dully-lustrous alteration- product of tachylyte or basalt-glass, which constitutes the greater portion of the ground-mass of the rock ; therefore the whole forma- tion is often called the palagonite formation. The tuffs consist of lava-dust and lava-fragments with an abundance of glass-pieces (pala- gonite, tachylyte), slag and bombs; loose anorthite-crystals are often abundant. The breccias are distinguished from the tuffs bv beins U %s Fig. 5. Cliffs 011 KlificS (tuff and breccia). The Veslmannaeyjar. more coarse-grained and containing larger, angular lava-pieces: the fragments consist of compact basalt, dolerile, pumice, slaggy lava and volcanic bombs: the separate fragments are often covered with a glassy crust. The palagonite formation is sometimes arranged in layers; sometimes not a trace of regular arrangement can be dis- cerned. The tuffs and breccias have been formed by volcanic erup- tions which chiefly ejected ashes and lava-fragments, and produced only a few lava-streams; the ruins of the numerous large volcanoes from which these eruptions proceeded may still be demonstrated. The palagonite formation is traversed by thousands of basalt dykes which branch and send out apophyses and intrusive sheets; very often the breccia and the tuff is filled with numerous hollow nodules and balls of basalt with a radiallv-columnar structure inside and PHYSICAL GEOGRAPHY 2 If) covered on the exterior with a crust of tachvlvte. The tuff-formation v */ appears to consist of several divisions the mutual relation of which has not, however, yet been elucidated. All round Iceland, in both the basalt and the tuft' formations, small patches of liparite occur. This rock occurs in small intrusive beds and dykes which, on account of their light colour, are dis- tinguishable from the dark basalt and can therefore often be seen from a distance. The liparites vary very greatly both in colour and in structure, and in places where larger sections are exposed, the colouring is often richly variegated. Liparites are almost always accompanied by many closely allied, glassy rocks, especially pitch- stones, which occur as dykes, perlites, sphaerulites, obsidian and pumice. In south-east Iceland and on Snsefellsnes, veins of grano- phyre are found in some places. The rocky promontories on each side of the Bay of Lon in south-east Iceland consist of gabbro, probably of Tertiary age; this rock is also found under the neves of Vatnajokull, because many of the glacier-rivers carry down an abundance of pebbles of gabbro. The liparites and allied rocks, which on the whole cover an area of about 800 square km., are distributed all over the island, but most frequently in larger quan- tities in East Iceland. Liparite eruptions have taken place at all periods from the earlier Tertiary times to the present day; some volcanoes which have been active within historic times have ejected liparitic pumice. In the neighbourhood of Husavik in North Iceland are found, near Hallbjarnarstadir, marine deposits with abundant shells of mussels and snails dating from the end of the Tertiary epoch, from the period called in England the Red Crag; these for- mations are found nowhere else in Iceland. In central Iceland, where tuff and breccia form the foundation, large areas are covered by old ice-striated lava-streams. These lava- streams are distinguished from the lavas of the present day by their colour and structure. The modern lavas are usually dark in colour with a compact basaltic structure, while these ice-striated lavas have lighter colours and a doleritic structure. These dolerite- lavas, during the Glacial period and immediately after, flowed from dome-shaped lava-cones with large crater-openings, here and there still extant; or sometimes from large "Bedded Volcanoes" (Strato- Vulkane) the ruins of which half destroyed by erosion - - are still to be found here and there within the area of the palagonite forma- tion. These ice striated lavas are of different ages; some of them 216 THORODDSEN have been produced before the surface assumed its present form, others have been formed after the country had in all essentials «, acquired the sculpture it has to-day; several of them have flowed down through valleys and hollows. Reykjavik is built upon such a doleritic lava-stream, and there the dolerite is much used for building purposes. In several places the glacial lavas are of con- siderable thickness (100 — 200 metres and more). Some larger vol- canoes, which have been in eruption as late as within historic times (Eyjafjallajokull, Snsefellsjokull, Oraefajokull), began their activity even during the Glacial period and at that time discharged doleritic lava-streams. In several places in West and South Iceland large deposits of conglomerates occur, with rolled gravel and sand, alter- nating now and then with moraine material and ice-striated lavas; these resemble the "Nagelfluh" of the Alps and were perhaps formed in a similar manner. During the Glacial period the whole island \vas wrapped in a sheet of inland-ice through \yhich only a few small peaks projected here and there near the edges. The Jokulls (sno\v-fields) of the mainland probably extended on all sides down to the sea, for the bottom-moraines of that time are found everywhere, both on the plateau and in the lowlands; the lateral and terminal moraines which occur in the valleys and lowlands, date from a later period when the ice was retiring; it appears also that large masses of moraine-material occur here and there on the sea-bottom off the mouths of the fjords. The north-western peninsula was probably covered with a separate ice -sheet, from which numerous small glaciers, with intervening ridges free from ice, descended to the sea. The ice-sheet of the Glacial period had, on the interior plateau, a thickness of 700—800 metres. Ice-striated rocks occur all over the island, both in the high land, in the valleys, and in the lowlands, as also on islands and skerries. Large and small erratic blocks are found in thousands scattered over the whole of Iceland. As mentioned above, it is assumed that Iceland, in the begin- ning of Tertiary times, was connected by a broad land-bridge with Greenland, the Faeroes and Scotland; this land-bridge was a volcanic highland or plateau-land formed by innumerable lava- streams which originated principally from rows of craters and from fissures. The plateau, which had a height of 3000 — 4000 metres above sea-level, wras towards the end of the Miocene period broken up and depressed; by this subsidence, perhaps in conjunction with PHYSICAL GEOGRAPHY 217 abrasion, the countries were separated and have never since been connected with each other. But Iceland was then, after the separation, considerably larger than it is now, the land extending 50 — 100 km. O) — r y; ^ •— c a o o o — o i S I § O P 2 5 o « o C3 •3 O C O 0) I/! else have been effaced or destroyed by erosion. All the numerous volcanoes have, in the course of time, since the Glacial period, dis- charged enormous quantities of lava, and the post-Glacial lava-fields of Iceland cover an area of about 11200 square km.; they occur over the country in vast expanses around the volcanoes from which they have originated. The majority of these lava-fields have been produced not by one but by numerous eruptions at various times, and the greater part of the lava dates from pre-historic out- bursts. The largest lava-stream which has been poured forth during a single eruption within historic times, is that which issued from the craters of Laki a row of craters formed in 1783; this stream covers an area of 565 square km.; and its mass occupies about 12Vs cubic kilometres; it is probably the largest lava-stream upon the surface of the earth which, within historic times, has been known to flow out during a single eruption. The largest continuous lava-field in Iceland is OdaSahraun on the plateau north of Vatna- jokull (600 — 1200 metres above sea-level); it was produced by nu- merous eruptions from more than 20 volcanoes, and covers an area of about 4000 square km. The lava-field, next in size, which origi- nated from the many large craters near Vei5ivotn, west of Vatna- 15 220 THORODDSEN jokull, and extends down to the south coast near Eyrarbakki, covers an area of 1550 square km. In the low land this lava-field is covered by a thick layer of soil upon which dense settlements have arisen. In addition, extensive lava-fields (1530 square km.), which originated from 28 volcanoes, occur on Reykjanes; and upon the plateau on either side of Langjokull, there are also other large lava- fields (1030 square km.) of which Hallmundarhraun is the best known; in it is the lava-cavern, Surtshellir, which is P/z km. in Fig. 7. Lava-stream (apalhraun) ; upon the outer south-west slope of Dyngjufjoll. (Phot. Heinrich Erkes.) length. Large tracts of lava occur also around Hekla, near Myvatn, Kelduhverfi and in several other places. The surface of the lava-streams varies greatly ; often it is very rugged and jagged and is then, in Iceland, called apalhraun, and in the Sandwich Islands, aa. Such streams consist exclusively of porous and brittle lava and slaggy fragments heaped together pell- mell. Such lava-streams are comparatively narrow, with high edges which, seen from a distance, look like fences or ridges upon the level land. A lava-stream of this description is very difficult to cross, owing to the fragments being put together so loosely that they are disturbed by the slightest touch. Other lava-fields, especially the PHYSICAL GEOGRAPHY 221 large ones, have another variety of surface lava-sheets which may sometimes be level, but are more frequently broken and cracked in numerous directions; in Iceland they are known as hellnhraun, in the Sandwich Islands as Pahoehoe. Upon the smooth surface, numerous tangled and twisted lava-ropes may be seen, bent in long curves following the undulating surface of the viscous lava. Some- times this lava is compact, and without great irregularities of sur- face, but more frequently, by cooling, the surface has subsided and Fig. 8. Sheet-lava near Frambruni. Slope of Trolladyiigja on Odadahraun. (Phot. Heinrich Erkes.) broken into large pieces, forming a number of hills, ridges, embank- ments and cauldron-shaped depressions, giving to it the aspect of a rough sea with high waves. Sometimes the surface of this sheet- lava is arranged in knots, curls and folds, all as smooth as hard- ened pitch. Beneath the lava-sheets there are often empty spaces, like drain-pipes and tunnels, and sometimes large caverns. Both forms of lava are sometimes found combined in the same stream. Long clefts often occur in the lava-streams, and sometimes enormous cracks, wrhich are due to the subsidence of the substratum. Of lava- clefts of this description Almannagja near Thingvellir is the most famous. In Odaftahraun there are also lava-clefts, 10 — 15 km. in 15* 222 THORODDSEN length. Groups of small slag-cones (craters) and lava-kettles (Hornitos) are very common in the lava-fields; they are often gathered together in hundreds upon a relatively small area and without any regularity Fig. 9. Lava streams in vertical section (Almannagja). of arrangement. These slag-cones are secondary formations, usually associated with lavas which have overflowed marshy ground and lakes, and have therehy absorbed quantities of water-vapour; they are of frequent occurrence near Myvatn, on Reykjanes, and espc- ciallv at Landbrot. PHYSICAL GEOGRAPHY 223 Large areas of Iceland are, moreover, covered with volcanic ashes, slags and bombs ejected by volcanoes. During the eruptions, the fine ashes are often spread out over a large part of the country and are sometimes carried by the wind across the Atlantic Ocean. During the Katla eruption in 1625, the ashes were carried to Bergen in Norway, and in 1845, ashes from Hekla were carried as far as to Germany, and during the Askja eruption in 1875 the volcanic dust was carried to the west coast of Norway in eleven hours forty «/ V minutes, and in another ten hours they travelled as far as Stock- holm. Ashes and slag are thrown up into the air to a great height; on April 21, 1766, the ash-column of Hekla had a height of 5000 metres above the summit of the mountain, and it has often been higher still. Lava fragments and bombs are shot into the air to a great height and often fall at a distance from the place of eruption; during the Hekla eruption in 1510, a man was killed by a volcanic bomb at Skalholt, 45 km. from the volcano; during the eruption of the same mountain in April 5, 1766, a volcanic slag, as big as a man's fist, was hurled to Vidivellir in Skagafjor5ur, 165 km. from the mountain. The fine dust which fills the air during great erup- tions, causes peculiar refraction-effects in the air; thus, during the Laki eruption of 1783, dust-clouds and unusually brilliant sunsets were common over the whole of Europe, North Africa and a part of western Asia. The ashes fell in such quantities in Caithness in Scotland as to destroy the crops; that year is still spoken of by the inhabitants as "the year of the ashie." The shower of ashes, together with the red-hot scoriae ejected in an eruption, often causes considerable damage to the inhabited land. Pastures are buried beneath them or are scorched, and the coppice-woods often suffer severely. Whether the damage done by the ashes to pastures is permanent or not greatly depends upon their nature; the heavy basaltic ashes are especially injurious, as they can only with diffi- culty be carried away by the rain or by irrigation. When, however, the layer of ashes is thin it gradually is absorbed into the soil, by the grass growing up above it. In the neighbourhood of larger volcanoes, several layers of ashes, one above the other, are found in the soil. The light, liparitic pumice-ash, which is rarer, is less injurious, as it is quickly carried away by water. Sometimes the ashes discharged by a volcano contain a great quantity of acids; during the Laki eruption of 1783, the ashes were so acid that they burnt holes in the burdock leaves, and left black patches on the 224 THORODDSEN sheep's skins, and the hoofs of the sheep turned yellow when they walked amongst them; while the rain which fell from the dust-clouds, is said to have been so sharp and biting that it was painful where it fell on the hands and face. The volcanic eruptions, on the whole, have had a very injurious influence upon the plant -distribution in the volcanic regions of Iceland. During the eruptions a great number of sheep and cattle die from want of food or from its un- wholesome nature, and of various diseases caused by the ashes eaten with the fodder. No eruption has however been so disastrous as the eruption of the Laki crater-rows in 1783. In the winter which followed, and in the spring of 1784, the sheep and cattle suffered from all kinds of diseases owing to the unwholesome food, and died by scores. On many farmsteads the entire live-stock died out, and the following year there died in the whole of Iceland 11,500 cattle, 28,000 ponies and 190,500 sheep about 53 per cent of all the cattle, 77 per cent of the ponies and 82 per cent of the sheep. Then came a famine, which carried off 9500 of the inhabitants, about one-fifth of the total population of Iceland at that time. In Iceland there are several types of volcanoes. Usually, by a volcano is understood a large hill or mountain, more or less conical in form, which is ignivomous, discharging lava-streams and ejecting ashes and fragments of lava. Volcanoes of this description occur in Iceland, but are not common; seven or eight such volcanic moun- tains are known, which resemble externally the well-known Italian volcanoes of Vesuvius and Etna, without however being so regular in form; they are built up of alternating layers of ashes, lava and slag, and usually resemble truncated cones with a crater at the top, and a considerable angle of inclination (at the base 8° — 15°; at the top 20° — 35°); the majority of them have their summits covered with snow and glaciers. Of these volcanoes the largest and best known are Oraefajokull, Snsefellsjokull and Eyjafjallajokull. Hekla is also built up of alternating layers of lava and tuff, but is not conical. Its shape conforms to an elliptical ridge, rent down its major axis, and studded with a row of craters along the line of fissure. Another type of Icelandic volcano are the dome-shaped ulava-cones" (dyngja, pi. dyngjur) larger or smaller volcanoes, built up entirely of lava-streams, without any intermediate layers of tuff or slag. Volcanoes of this description, which are also found in the Sandwich Islands, are distinguishable from the country sur- rounding them as shield-shaped cones, and their altitude is low PHYSICAL GEOGRAPHY 225 compared with their extent. The largest have an altitude of 1400- 1600 metres, and a diameter of 10 — 15 km. On such dome-shaped lava-cones the angle of inclination at the top of the mountain is only slightly greater than near the base, seldom exceeding 7° -8°, and more frequently still less, often only 1° — 2°. The summit of such volcanoes consists of a circular or elliptical mouth or depres- sion, often of large dimensions, some having a diameter of 1000 metres or more. The walls of the depression (crater) are usually Fig. 10. Row of craters. cleft by concentric fissures, so that the descent from the lip to the bottom of the crater is, as it were, a series of steps. The sides of these volcanoes are entirely covered with knotty sheets of lava, and long tunnels and caves are very common in the sides of the vol- cano. In some cases the depression is filled with lava to such an ex- tent, that the only indication of the circumference is a ring of small lava-pinnacles and lava-ridges. The immense lava-waste of Oda5a- hraun was mainly formed by outpourings from this type of volcano, of which Trolladyngja (1491 metres) is the largest. Another well- known volcano of this description is Skjaldbreid near Thingvellir. Of post-Glacial lava-cones, 16 are known from Iceland. These lava- cones were also very common during the Ice Age. The greatest amount of lava which has been poured forth in 226 THORODDSEN Iceland, issued from volcanic fissures and crater-rows; these are not volcanic mountains, but rows of low craters established along the direct line of a fissure, more frequently upon level land. Of this kind of eruptive vent, 87 are knowrn from Iceland at present. Each of the craters in such a chain occurs independently and is built up of scoriae and lava. They are usually low7, rarely exceeding a height of 100 — 150 metres, while the majority of them are even considerably lo\ver; they are often very irregularly formed and com- posed of several rings. Some crater-ro\vs are very long; they often attain a length of 5 — 10 km., and some are longer still, as for in- stance, the Laki crate r-row of 1783, which has a length of 30 km., and contains about a hundred separate craters of various sizes. Some crater-ro\vs are so small that they resemble tov-volcanoes. \J *, In some places the lava has wrelled up out of fissures in large streams without any visible craters. The largest of these fissures is Eldgja, north of Myrdalsjokull ; it has a length of 30 km., and has poured out lava-streams sufficient to cover an area of about 700 square km. In some places "explosive craters" occur, cauldron- shaped depressions produced by a single volcanic explosion. The best known of these craters is Viti, on the side of Mount Krafla, north of Mvvatn. It wras formed by a sudden outburst on May 17, 1724. For a long time afterwards there w^as a large, boiling slough at the bottom of the crater, but this is now converted into a greenish- coloured quiescent lake. The majority of the volcanoes in Iceland are basalt volcanoes, and have poured out streams of basaltic lava, and ejected basaltic slag and ashes. Only in the neighbourhood of Torfajokull liparitic volcanoes occur of post-Glacial origin, and of peculiar aspect wiiich have poured out lavas rich in silica. The interior of the lava-mass is grey or reddish brown, but the surface of it is jet-black, as it consists of obsidian, covered here and there by light, almost w'hite, pumice. The largest liparitic lava-streams are those called Hrafntinnuhraun and Domadalshraun. In some places streams of liparitic blocks in a half-melted condition have flowed out from craters in the mountain-sides, and several volcanoes have ejected liparitic pumice, as for instance, the volcano of Askja, 1875. Many volcanoes in Iceland are buried beneath the sno\v and */ the glaciers, and as mentioned above, when they break out into eruption , large masses of ice are melted and the glaciers burst, which causes the neighbouring level lands to be inundated by enormous floods of water, with huge fragments of ice tossing on PHYSICAL GEOGRAPHY 227 its surface. The Katla eruption, in this way, converted thickly in- habited and fertile tracts into deserts, and in 1362 Oraefajokull de- stroyed, in the same way, two parishes, sweeping away forty farm- steads with their inhabitants and live-stock and all else, out into the sea. The best-known Icelandic volcano is Hekla. It has been the scene of 21 eruptions during historic times; next in importance comes the glacier-volcano Katla of which 13 eruptions are recorded. Fig. 11. Kverkfjoll, a volcano on the northern edge of Vatnajokull ; seen from Hvannalindir. (Phot. J. P. Koch.) Submarine eruptions have taken place some ten lo twelve times near Eldeyjar off Reykjanes, whereby new islands have had their origin; but these have disappeared again. Several volcanoes .are present beneath the ice-cap of Vatnajokull, but the foci of eruption are not known for certain. During the last two centuries, from 30 to 40 eruptions have been recorded from the snow-fields of Vatnajokull. During several of these eruptions the snow and ice on SkeiSarar- jokull, on the southern side, partially melted, and enormous torrents of water were discharged. Occasionally Bruarjokull, on the northern side, has been very active. On Reykjanes there are numerous pre- historic volcanoes and rows of craters, and in three or four places eruptions have taken place since there have been settlements on 228 THORODDSEN the island; one of the most active volcanoes on Reykjanes is called Trolladyngja. The best-known volcano near Myvatn is Leirhnukur, but besides this, many smaller crater-rows and separate craters occur; the volcanoes near Myvatn were particularly active during the years 1724 — 30. In the centre of OdaSahraun rises the volcanic mountain-group Dyngjufjoll with the crater-valley of Askja; it is one of the largest volcanoes of Iceland. The crater-valley, which is surrounded by circular mountain- walls, covers an area of about 55 square km. In the south-eastern corner of it there is a deep volcanic depression writh a lake; at the edge of the latter a new crater opened on March 29, 1875, and discharged an enormous quantity of pumice over the eastern part of Iceland and, as men- tioned above, the dust was carried as far as to Scandinavia. The greatest eruption which has taken place during historic times in Iceland was the eruption of the above-mentioned crater- rows of Laki in 1783. The lava which poured forth filled valleys, altered the course of rivers and destroyed several farmsteads, fertile meadows •/ and extensive pastures. North of Iceland submarine volcanic erup- tions have occasionally taken place. Almost all the volcanoes of Iceland are associated with fissures in the tuff and breccia areas of the palagonite formation. In the southern part of Iceland all the mountain-ridges, valleys and rivers exhibit a decided dependence on tectonic lines of deeply situated fracture from SW. to NE. Open fissures in the surface, all the numerous crater-rows, and the lines joining the volcanoes, have a similar direction. Moreover, hot springs both alkaline springs and sulphur springs - are arranged along the same lines. In North Iceland, on the other hand, the tectonic lines and the fissures and volcanoes, have generally a direction from S. to N. Both these directions probably are combined in a curving band of fracture- lines which lies across the island. In the basalt plateaus of the west coast there are several cauldron-like fissures and concentric fractures, and along the southernmost of these depressions, which extend over both the tuff and breccia areas around Faxafloi, the volcanoes and hot springs are arranged in a semicircle. Earthquakes are very frequent in Iceland not only in con- nection with volcanic eruptions, but also apart from them ; in the latter case they are chiefly confined to three districts with well-marked natural boundaries. All the greater earthquake shocks are tectonic PHYSICAL GEOGRAPHY 229 in origin; that is to say, they are due to movements and subsi- dences of large tracts of land bounded by dislocations and fractures of the ground. In North Iceland, between Skjiilfandi and Axar- fjorflur, where the new volcanic tuff district extends to the coast, violent earthquakes are frequent, especially in the neighbourhood of the trading-station of Husavik. The earthquakes of 1755, 1872 and 1885 were especially serious and did great damage in these districts. At Faxafloi there is another earthquake-area where minor shocks are very common ; they are usually most violent on Reykja- nes, especially in the neighbourhood of Krisuvik, and at the extreme point of the peninsula, near the lighthouse. The third earthquake- area comprises the southern lowland area between Reykjanes and Eyjafjallajokull. This district has frequently suffered from violent and destructive earthquakes which have caused great loss of human life and of property. In modern times the earthquakes of 1784 and 1896 have been especially destructive. The former (Aug. 14--16) com- pletely ruined 92 farmsteads, and damaged 372 houses and 11 churches. In August and September, 1896, the earthquake shocks wrere even more violent. Great chasms were rent in the earth, some as long as 15 km.; watercourses were altered and the position of hot springs changed; quantities of hugh fragments of rock were loosened from the mountain-sides; 161 farmsteads were hurled down, and 155 more were greatly damaged; in fact, every house in this area sustained some damage. By each of these violent earthquake shocks a limited tract of land was put in movement. Occasionally, North Iceland has been shaken by volcanic eruptions which origi- nated under the sea off the north coast of the island; this was the case in the years 1838, 1899 and 1910. « Hot alkaline springs occur in hundreds in Iceland, scattered all over the country, sometimes singly, sometimes in groups. At the present time 677 hot and boiling springs are known in 162 locali- ties, and the majority of them are closely dependent upon the fracture-systems of the island. Earthquakes exert great influence upon these springs; many disappear or are altered, and new ones are formed. The surfaces of the springs have any temperature up to boiling point. Tepid springs which can be used for bathing are called "laugar," boiling springs "hverar." Some of the latter throw up jets of water as Geysir does; but otherwise the boiling springs may be divided into five classes: (1) springs which are constantly 230 THORODDSEN spouting, (2) intermittently spouting springs, (3) alternately spouting springs, (4) constantly boiling springs which do not spout, (5) springs with a high temperature and a quiet surface or which boil quietly in the middle. All the boiling springs deposit siliceous sinter. The most famous hot spring is Geysir, in the vicinity of Haukadalur in South Iceland, in the centre of a group of other boiling springs. This group of springs was mentioned for the first time in 1294 and has often undergone alteration by earthquakes, especially in 1630 and 1789. Geysir's eruptions now take place very irregularly and many days may intervene between them. At the end of the 18th and the beginning of the 19th century, Geysir's activity was at its maximum, and it could throw up fountains of water to a height of 50 — 60 metres, wrhereas now the water rarely rises above 30 metres. The neighbouring hot spring Strokkur, began its activity in 1789 and ceased during the earthquakes of 1896. At first Strokkur threw up higher fountains of water than did Geysir, not only boiling water and steam, but also cold water. Other large groups of hot springs are found near Revkir in Olfus, in Revkholtsdalur, on Hveravellir », «, on the plateau NE. of Langjokull, and in many other places. Sulphur springs occur in abundance in the volcanic districts, but not outside the palagonite-formation; also alkaline springs are common in the basalt districts. The alkaline springs are found upon the level land, in valleys and upon mountain-sides where there is much underground wrater, but the sulphur springs commonly occur upon mountain ridges and other dry localities where the water has an outlet through the underlying lava, etc. Several of the solfataras deposit a considerable amount of sulphur in small heaps wrhere the sulphurous vapours rise from the soil. Sulphur from Iceland had a commercial importance even in the 13th century, and the trade in sulphur wras especially lucrative in the 16th century. Since that time the export of it has gradually decreased and now has entirely ceased. The sulphurous vapours which rise through the clefts and cracks in the earth have a great effect upon the neigh- bouring rocks, which are transformed and decomposed in various ways coloured clays, gypsum, iron-alum (Halotrichit), etc. being formed. The mountains which have been penetrated by the hot sulphurous vapours are easily recognizable at a long distance, owing to their naked and discoloured appearance; they arc always light- red, yellow and white in colour and are entirely destitute of plant- PHYSICAL GEOGRAPHY 231 growth. While plant-growth is abundant in the neighbourhood of the alkaline hot springs, from the vicinity of the sulphur springs it is almost absent. In places where surface-water or underground water is found, mud-holes are formed, or larger or smaller sloughs, in which clayey mud of various colours boils and bubbles; it is sometimes ejected a few feet upwards, whereby crater-like mounds are formed around the larger pits, bearing a weird resemblance to Fig. 12. Solfataras of Krisuvik. large cauldrons of boiling porridge. The majority of the sulphur springs occur on Reykjanes, in Hengill, near Krisuvik and at Cape Reykjanes where the large slough of Gunna is well-known. More- over, extensive sulphur-spring districts (Namufjall, Krafla, Fremri- namur) occur near Myvatn, and Kerlingarfjoll near Arnarfellsjokull is another, upon the interior plateau. Carbonic acid springs (olkeldur) and Mofettes occur here and there, especially on Snse- fellsnes in western Iceland; the best-known carbonic acid spring is found near Raudimelur in Hnappadal. Ores, metals and stones of any great commercial value are not found in Iceland. Here and there some lignite occurs which is utilized by the neighbouring inhabitants. The gathering of sulphur is no longer lucrative, but, THORODDSEN as already mentioned, in East Iceland calcareous spar is quarried at Helgusta5ir near Reydarfjorftur and is used by the makers of optical instruments. 1 1 For fuller notes on the Physical Geography and Geology of Iceland the reader is referred to the following papers by Th. Thoroddsen : Island. Grundriss der Geographie und Geologic: Justus Perthes, Gotha. 1906 (358 pages). Lysing Is- lands, Kobenhavn, 1908—1911, in 2 vols (365 and 673 pages). Explorations in Ice- land during the years 1881 — 98 (The Geographical Journal. London, 1899. Vol. XIII, pp. 251—274, 480—513). Vulcane im nordostlichen Island (Mitteilungen d. k. k. Geogr. Gesellschaft, Wien, 1891, pp. 117—145, 245—289). Geological Map of Iceland. Sur- veyed in the years 1881—98. Edited by the Carlsberg Fund. Copenhagen, 1901, Scale 1 : 600,000. II. CONDITIONS PERTAINING TO SURFACE AND SOIL. AFTER having thus given a brief, general survey of the orographical and geological conditions and having described the substratum and general structure of the island, we will now pass on to a de- scription of the surface itself, with which plant-growth is more parti- cularly associated. As mentioned above, Iceland is built up of basalt, tuffs and breccias, but basalt is the fundamental rock; the tuffs and breccias are, for the most part, nothing else but basalt split and pulverized. The mineralogical and chemical composition of the soil is therefore essentially the same over the entire island, provided the siliceous liparites are excepted which have no effect of any importance to the whole. Seen from a distance, the basalt mountains usually appear to be steeper than they are in reality, and the small terraces or steps of the layers of basalt are not distinguishable in the higher part of the mountain from a distance except when they are snow- covered or when, as rarely happens, a scanty vegetation (especially mosses) has been able to gain foothold upon the narrow ledges. The rule is that the steps of the basalt mountains become broader as the base is approached. At the top the separate layers project as a narrow ledge which is only half, one, or two metres broad, but lowest down in the valleys, and nearest to the sea the separate layers form enormous terraces which may attain a breadth of *•/*- *} *• l/2 km. or more. The upper surface of these broad terraces is covered with gravel and clay, and sometimes with a scattered plant-growth, or sometimes with a continuous vegetation, with bogs or swamps; there, enormous, elongated snow-wreaths may persist far into the summer. On basalt mountains erosion is more active on the sunny side, therefore the other side is steeper and more sparsely covered with plants. On the sunny side the average inclination is usually 234 THORODDSEN only 20°— 25°, but upon the other side 30°— 35°. There are, how- ever, a few basalt mountains which are much steeper than this: for instance Skessuhorn near Borgarfjor5ur, which has an inclination of 48°. In the numerous erosion-channels on the mountain sides, where gravel and stones are constantly rattling down and avalanches are frequent, it is difficult for the plants to gain foothold. The ridses between the mountain streams are therefore more closelv •/ covered with plants but, as already mentioned, a continuous plant- Fig. 13. Basalt mountains with snow-wreaths (IsafjorSur). covering rarely extends higher than half-way up on the basalt mountains. In olden times the mountain sides were in very manv */ */ places clad with coppice woods, but these disappeared at an early date, partly owing to the havoc wrought by sheep and partly to man's lack of foresight. Now only some stunted shrubby birches are to be seen upon inaccessible cliffs, where they are beyond reach of man and beast, even with the utmost exertion. Upon mountain- sides deprived of their birch-copses, avalanches of snow and stones have suffered no hindrance, so that all soil and plant-growth have disappeared, thus turning the mountain-sides into naked, gravelly and rocky slopes. In some districts such changes have taken place PHYSICAL GEOGRAPHY 235 even as late as in the 19th century. Where the hasalt mountains are not too steep nor the mountain-streams too torrential, the flat gravel-cones upon the valley sides, below the notches in the moun- tain, are often overgrown with plants. These gravel -cones often underlie the home-fields of the farmsteads. In the fjord districts of Iceland the vegetation upon the basalt mountains differs considerably in passing from the sea inwards. Owing to the effect of the sea-water, the violent storms and the rawness of the climate, the outermost points are comparatively poor in plants, while the vegetation increases inwards towards the valley, and in the bottom of the valleys, especially on the north-western peninsula, remains of coppice woods are often found; but woods could not thrive out along the fjords, still less at the extreme points. Where the basalt does not occur as steep cliffs and is not covered by loose layers of clay, glacial gravel and soil it is usually strewn with loose sharp-edged fragments, split and torn off by frost. The severance of these fragments usually follows the cleavage of the basalt, and they are sometimes slaty and in thin plates, a condition which is especially common in the uppermost part of the basalt formation. Upon the split and torn basalt in the uppermost part of the moun- tains, plants have difficulty in gaining foothold, especially when the climate is as raw and stormy as is the case in Iceland. There- fore, large areas of the higher-lying basalt districts are extremely poor in plant-life even in places, where according to the situation, the conditions might be expected to be somewhat more favourable. The landscape in the tuff and breccia districts has a different appearance. Basalt mountains usually have sharp, and breccia mountains soft outlines. Those areas of the cultivated districts and on the lower spurs of the highland wrhich are built up of tuff and breccia have often a more or less undulating appearance; the moun- tains are broken down into numerous rounded ridges and pro- tuberances with intervening stretches of level ground and valleys of irregular shape; but here and there are seen tabular mountains or promontories with steep sides and a flat surface, where the basalt or dolerite has covered and protected the tuff and the breccia. On the plateau, where through centuries storms have been continually altering the sculpture of the surface, the soft tuff-mountains have suffered in particular; here the tuff-ridges are connected into irregular chains which have been eroded in every possible way, and often resemble fantastic ruins with numerous sharp peaks, protuberances The Botany of Iceland. I. 16 236 THOHODDSEN and knolls separated by sandy areas and labyrinths of branched1 clefts and small valleys. Tuff mountains, owing to the loose nature of the rock, are relatively poor in water as the latter often dis- appears into the ground, and does not appear until at a distance from the mountain. This is especially the case with the tuff-moun- tains on the plateau where, during summer, not a single stream or spring is met with for long distances, but only large, deep, dry river-beds and water -courses filled with coarse gravel and large Fig. 14. Remaining portion of a "mohella" upon a wind-eroded gravelly ilat (orfoka). boulders. These river-beds are due to the melting of the snow in spring or during periods of thaw in winter when, for a short time,, they are all filled with torrential floods of water. The surface of the breccia mountains is usually concealed by loose, angular and porous fragments of lava which have been dis- integrated from the breccia; the nature of the rock is often seen only in clefts and in a few7 prominent protuberances and projecting rocks. The power of resistance of the rock against the action of water and wind differs however greatly, because tuff and breccia are of all possible degrees of hardness, although a loose texture is the most common. Tuff is easily disintegrated, and water and wind PHYSICAL GEOGRAPHY 237 carry the finer palagonite-dust down into the valleys or to distant quarters of the island where it is deposited and retained by the vegetation, frequently forming thick layers (mohella). The heavier lava-fragments which have thus been deprived of matrix are left behind. On stretches of level ground the lava-gravel, thus loosened, sometimes attains a thickness of several metres. As a general rule the surface of the tuff-mountains is much affected by the action of water and air and along the sides of the fissures the effects may be traced far down. Sometimes these fissures are filled with zeolites, calcite or gypsum; and sometimes the mass which fills the fissures is harder than the surrounding rock so that the surface presents the appearance of a network of raised lines, while the tuff in the intervening spaces has been disintegrated and carried away. On the ridges and peaks of several tuff mountains the surface is, as it were, pock-marked with numerous small irregular hollows, chan- nels and pot-holes which are probably due to the combined action of water and drifting sand. There are often a great number of clefts and fissures in the tuff-mountains which can sometimes be seen from a great distance because of the plant-growth which retreats to them to find shelter from the storm. On the whole the varied forms of surface in the tuff districts greatly influence the details of the distribution of the plants of the place. As tuff is far more easily decomposed than basalt, soil is formed more quickly upon tuff-mountains in cases where external factors such as sand-drifts and storms do not interfere. Therefore on the tuff and breccia mountains of southern Iceland there is a thick coating of soil and a luxuriant plant-growth right to the verge of the mountain; this is rarely the case on basalt mountains. Even on steep mountain-sides of tuff and breccia there exists a luxuriant vegetation of various species. This is especially conspicuous in Myr- dalur (south of Myrdalsjokull) where, for instance, the extreme point of Reynisfjall is densely covered with plants. A luxuriant vegetation is also found in Vikurklettar and in several other places. On the whole, as mentioned above, it is characteristic of the lower moun- tains, south of the great Jokulls, to be covered by a comparatively luxuriant plant-growth, while the level country is barren, owing to the destructive action of the glacier-rivers. In the case of the sea- fowl cliffs, where manure is supplied by the sea-fowl, such tuff- mountains are far more densely covered with Cochlearia, Archangelica, etc., than are the basalt mountains. 16* 238 THORODDSEN In Iceland all which lies above 500 metres is a complete de- sert; this is also true of great parts of all that lies between 300 and 500 metres; at this altitude there are, however, rather ex- tensive bogs covered with Carices and Eriophorum, especially to- wards the west (N. W. of Langjokull). Larger and smaller desert- areas are also found lower down; in some places they extend even to the sea, but in such cases they owe their origin to special cir- cumstances — the destructive influence of glacier- rivers, volcanic eruptions or blown sand. In many inhabited districts the greater part of the surface consists of a rocky flat with scanty vegetation; a dense plant-growth such as that found in meadows, bogs and heather-moors covers only a very small part of the entire surface of the island, perhaps not more than 1500 — 2000 square km. ; but the amount so covered cannot be stated with any certainty. The interior plateau owing to its height above sea-level and its climatic conditions will probably never be of any greater importance, as regards the livelihood of the inhabitants, than it is at the present time. Considerable tracts of the lower-lying parts of the plateau (afrjettir), in spite of the very poor herbage, are used as summer- pastures for the hardy Icelandic sheep which are driven up into the mountains about the end of June and fetched home again in the middle of September. No small percentage, however, of these sheep perishes yearly by venturing too far into grassless wastes, falling into rivers and down clefts, being overcome by snow-storms, becoming a prey to foxes, etc. The snow-line and the glaciers form the upper limit of plant- growth ; but from thence down to about 500 metres above sea-level, individual plants usually occur widely separated; there are, there- fore, virtually no habitations on the plateau. As the height of the snow-line in the different parts of the island varies greatly, so similar laws 'govern the occurrence of the habitations, which are closely associated with the plant-growth. The highest snow-line occurs in north-east Iceland, and there the habitations also extend furthest upwards. Three parishes are found on the plateau itself, viz. My- vatnssveit at 300 metres above sea-level, Fjallasveit at 400 — 500 metres, and J6kuldalshei5i at 500—530 metres. In the last two the number of the farmsteads and of the people has been subject to very great fluctuations. The inhabitants of these parishes main- tain themselves almost entirely by sheep-rearing. In Myvatnssveit the conditions are however more favourable, as it is situated at a PHYSICAL GEOGRAPHY 239 lower level, is sheltered by several high mountains and has more- over a rich fishery in Myvatn. The district of Fjallasveit is chiefly covered with blown sand, and the plants growing on the sand-dunes and sandy flats, viz. Elymus arenarius, Salix glauca and S. lanata and Carex incur ua, serve as fodder for the sheep, which thrive well. These plants are cut and gathered during autumn for winter-use. It is, however, difficult to keep cowrs; and on account of the weather, neither potatoes nor root-crops will thrive. The climate is also far more severe than at the coast; the annual mean temperature in Modrudalur is 0.8° C. In these districts the snow-line has a height of 1300 — 1400 metres above sea-level, and the highest limit at which habitations occur is 530 metres. On the north-wTestern peninsula, to the extreme north the habitation-limit occurs at only 80 metres above sea-level, and the snow-line at 400 metres; more to the south, on the same peninsula, at about 130 metres, and the sno\v-line at 650 metres. For comparison with the height of the snow-lines given above I give in the following pages a list of the inhabited farmsteads situated at the highest levels in the different parts of the country, because they give an indication of the limit of the more densely plant-covered, inhabited country upwrards to- wards the wride expanses, poor in plant-life, in the interior. The settlements in the uppermost valleys and in some parts of the plateau are by no means fixed ; in years when severe weather occurs, with cold and damp summers, some of these farmsteads are aban- doned, but are again inhabited when more favourable wreather sets in. The areas above the populated districts, between these and the desert proper, are too poor in plant-life to be inhabited, but they are of great importance as summer-pastures for sheep and ponies. Heights of the uppermost farmsteads above sea-level in metres. Upon the northwestern peninsula. Smidjuvik SE. of Cape Nord 80 metres Skogar in Mosdal at Arnarfjordur. . . . 82 Hlidarsel at Steingrimsfjordur 128 Near Faxafloi. Fornihvammur in Nordurardal 170 metres Gilsbakki in Hvitarsida 175 Fljotstunga 232 Kalmannstunga 218 Haell in Flokadal. 174 240 THORODDSEN Upon the peninsula of Reykjanes and in its immediate neighbourhood. Vigdisarvellir NE. of Krisuvik 122 metres Kolvidarholl1 near Hengill 262 Nesjavellir S. of Thingvallavatn 172 Svartagil N. of the above lake 182 Near the Southern Lowland. Austurhlid on the way to Geysir . . . . 163 metres Haukadalur N. of Geysir 104 Tungufell S. of Gullfoss 148 Storinupur in Hreppar 135 Galtalaekur W. of Hekla 142 Selsund SW. of Hekla 155 Raudnefsstadir NW. of Tindfjallajokull. 239 Barkarstadir in Fljotshlid 134 South of Myrdalsjokull and Vatnajokull. Hofdabrekka in Myrdalur 128 metres Svartinupur in Skaptartunga 273 Holt in Sida 172 Eintunahals in Sida 227 Skaptafell (Bolti) in Oraefi 161 Vioidalur in Lon2 E. of Vatnajokull.. 439 In the valleys of East Iceland. Myrar in Skriddal 183 metres KJeif in Fljotsdal 126 Hakonarsta6ir on Jokuldalur 310 HauksstaMr in Vopnafjor5ur , 195 Upon the plateau between Jokuldalur and Bardardalur. ViSirholl on Jokuldalsheidi 533 metres Lon 531 Moth'udalur on M65rudalsfj611 469 Vididalur 462 GrimsstaSir 432 Vi6irholl 415 Reykjahlid at Myvatn 292 Gautlond ". 279 Krakarbakki 363 Svartarkot. 409 • In the valleys of North Iceland. Isholl in Bardardalur 368 metres Reykir in Fnjoskadal 223 Tjarnir in E\rjafj6rdur 223 Bakkasel in Oxnadalur 350 Gilhagi in Skagafjor6ur 227 Mardargnupssel at Svinadalur 395 In reality only an inn for travellers. Not built until 1840; abandoned in 1842—1882: rebuilt 1883. PHYSICAL GEOGRAPHY 241 The surface of the interior plateau, with the exception of the glacier-covered areas, consists of deserts of stones, lava, gravel and blown sand. Where the underlying rock is basalt, the surface of the plateau is usually strewn with angular blocks of basalt, often in irregular heaps. Occasionally ridges are found, covered with gravel and blocks of glacial origin, but blocks split by frost are decidedly the more frequent upon these ridges. On the plateaus above the fjord districts of East Iceland, extensive areas are covered with aneular blocks of basalt, but usuallv this laver of loose blocks v v is relatively thin. Similar conditions are met with in the north- western peninsula and on the mountains of North Iceland; but in the interior of the country glacial materials and the more recent formations preponderate. In that part, therefore, the greater part of the area consists of ice-striated ridges of dolerite, post-Glacial lavas, old bottom moraines and blown sand. Ice-striated streams of old dolerite-lava occupy vast areas in the interior. North of the great Jokulls the dolerite lavas may be traced continuously from Arnarvatn in the west to Snsefell in the east. Here and there the dolerite is covered with recent lava, and the closer the Jokulls are approached the thicker becomes the surface-layer of glacial gravel. As already mentioned, these monotonous, bluish-grey, stony deserts present an extremely desolate appearance. The only points upon which the eye can linger are scattered snow-drifts and large erratic blocks lying scattered upon the ridges. The landscape has undoubtedly remained for centuries unaltered in appearance; it must have looked as it does now, immediately after the snow-fields of the Ice Age had retired. There is no sign of life, and deep silence reigns over the land. The dolerite ridges, as has been already said, are covered with blocks rent by frost between which ice-striated domes protrude here and there. All the upper faces of the stones are dully polished and seamed by wind-abrasion. For days the traveller may see nothing else but gravel and ridges of rocks in endless succession like waves upon the sea; while as regards plants, only at intervals of 10 to 20 metres may a few stunted specimens of Armeria mari- tima, Salix herbacea and Cerastium alpinum be met with; the first- mentioned plant occurs most frequently. Here and there partially dried-up water courses and river-beds are found which are filled during the thaws of spring; and pools and small lakes are also occasionally met with hidden away in the low-lying ground be- tween the ridges. In the immediate neighbourhood of the Jokulls, 242 THORODDSEN where numerous glacier-rivers branch out upon the plateau, there are flats of rolled gravel and clay which are sometimes so steeped in water from the melting glaciers that they cannot be traversed. On the interior plateau large areas (about 6500 square km.) are occupied by lava-streams - usually sheet-lava, with intervening tracts of slaggy lava. These lava-deserts are very poor in plant-life, and in the most highly-situated districts, they are almost entirely destitute of vegetation. Water is also very scarce, as the rain-water and the melting snow from the snow-drifts penetrate into the lava and do not reappear until far away as springs. The surface of these deserts consists of a hard, stiffened stony mass without a vestige of soil- covering, the hollows often filled with volcanic ashes and blown sand. The vegetation upon the lava-streams differs greatly on the plateau from what is found in the low-lying district, even if the lava- streams are of the same age; in the latter locality the oldest post- glacial lava-streams are often covered by a thick layer of soil, sup- porting heather-moors and coppice- woods ; while on the plateau^ lavas of the same age are quite bare. From Trolladyngja, a volcana in Odadahraun, an enormous lava-stream (Frambruni), 110 km. in length, has flowed dowrn through Bardardalur to Ullarfoss. Up at the volcano itself the lava, to a height of 1000 metres above sea- level, is entirely destitute of vegetation ; lower down from Dyngju- fjoll to Bardardalur (800—200 metres), the vegetation becomes denser each step. The blown sand which has accumulated in the low-lying tracts, is here and there covered with lyme grass, which is soon followed by dwarf willows and heather. Down in Bardardalur (160 — 100 metres) the lava is entirely covered with greensward and river-gravel so that only a few lava-peaks protrude. Near Lunda- brekka a rather thick layer of peat is formed on the surface of the lava; otherwise the large lava- waste of Odadahraun (600 — 1200 metres) is almost destitute of plant-growth. The fewr plants which occur are especially met with where hollows in the low-lying tracts are filled with blown sand; most frequently some tufts of Elymus arenarius can be seen, and here and there a few specimens of Silent maritima, Cerastium alpimim and Armeria maritima. In places where the moun- tain streams from Dyngjufjoll have carried down a little clay, soil- formation has taken place to a small extent and extremely small specimens of Salix herbacea and Pohjyonnm viviparum exist there. Lichens, which occur so frequently on low-lying lava-streams, are very rarely met with on Odadahraun, so the lava-surface is usually PHYSICAL GEOGRAPHY 243 quite destitute of covering; on the other hand, on lava-streams near Myvatn (above 300 metres above sea-level) lichens are very frequent. There, many species of phanerogams have also made their appear- ance, and both there and in Kelduhverfi the lava is often found to be covered by a thick carpet of moss. On Reykjanes scores of square km. of lava are entirely hidden beneath a soft greyish carpet of Grimmias. Between Jokulsa a Fjollum and Jokulsa a Bru, and between Kaldakvist and Skafta, more recent tuffs appear upon the surface of very considerable tracts of the plateau, not covered by ice-striated dolerite-lavas or bv modern, basaltic lavas. Where this tuff does «/ not appear as bare peaks or steep mountain-chains, the surface is covered by lava-gravel disintegrated from the breccia, or else it is covered by blowrn sand, wrhich is widely distributed, not only on the interior plateau but also in the lowlands. Blown sand is of varied quality and origin. It may be coarse or finer; it is some- times so fine that it penetrates everywhere. During violent storms in sandy districts the fine dust is carried to the most remote quar- ters of the island and is deposited as a fine layer all over the sur- face: it even falls on vessels in mid-Atlantic. But naturally most dust falls in districts bordering on the tuff-belt or situated in it, as the dust chiefly originates from the tuff. The atmosphere in distant regions is often yellowish-brown because of the fine dust suspended in the air, and this dust-cloud is known in Iceland as "mistur." This tuff-dust has played a very important part in the formation of the Icelandic soil and subsoil, and it can be demonstrated almost everywhere. In the blo\vn-sand districts proper, and in the neigh- bouring regions, where also larger particles of stone are put into motion, wind-blown sand has a great mechanical influence and is a mighty geological factor; its denudating effect upon the tuff moun- tains has been very great. Harder rocks are also affected by the sand ; dolerites acquire a dull polish with irregular depressions, striations and furrows, while basalt is likewise sand-polished though to a less extent. Tuff and breccia mountains are always more highly disintegrated on the windward side, and the isolated fragments of basalt embedded in the breccia project further from the ground- mass on that side than to leeward. When the wind is stormy, great masses of blown sand are constantly driven through the narrow valleys, which occur between the numerous sharp tuff-ridges east of Tungna, so that no vegetation can thrive there; only here and 244 THORODDSEN there upon the highest ridges and peaks, which cannot be reached bv the coarser grains of the drifting sand, are seen small patches of soil supporting mosses and a few phanerogams. In Iceland blown sand consists almost invariably of decomposed volcanic rocks; quartz-sand does not occur in Iceland. The most common blown sand is palagonite-dust usually of a yellowish-brown colour, which when examined microscopically, is seen to consist of glass-particles, tachylite, palagonite, plagioclase, augite and various Fig. 15. Mohella in Kroksdalur, not far from Sandmiiladalsa, (Phot. Heinrich Erkes.) finely decomposed zeolitic alteration-products. Volcanic ashes of recent date often occur as blown sand especially in the interior of the lava-wastes; they are heavier and less mobile, consequently, they are not dispersed in quantities beyond the volcanic districts. In the neighbourhood of the great glacier-bearing mountains, considerable tracts of level land are often covered with glacial clay, which when dried, crumbles into dust and drifts beyond the nearer surroundings. Around Dyngjufjoll, especially south-east of Askja, large areas are covered with blown sand, consisting of liparitic pumice-dust which all dates from the eruption of 1875. Moreover, stretches of blown sand consisting of decomposed mussel shells, i. e. calcareous dust, occur here and there along the coast of the north-western peninsula. PHYSICAL GEOGRAPHY 245 Owing to the variability of the wind-conditions, the dunes in the blown-sand districts are usually small and irregular in form; they are rarely higher than 3- -4 metres, usually much less, and they are bound together by lyme grass and a few creeping dwarf willows. Sandy levels with low waves of sand are of general occurrence, and when moisture comes into play, the surface is cracked into numerous polygonal cakes by the action of desiccation or frost. The cracks are filled with drifted sand, so that the surface resembles a kind of mosaic. Fig. 16. Soil torn up by the wind. Large tracts in Landsvcit are occupied by these loess-like formations. Here several square miles of land, which were formerly wood-covered, are torn up by the north-east wind. The district of Landsveit, west of Hekla. (Phot. Th. Thoroddsen.) Ill tuff districts proper older and younger seolian formations are the thickest and most widely distributed, and often alternate with volcanic and glacial formations; but the tuff-dust is also car- ried to the basalt districts, where they initiate the formation of the loess-like layers known in Iceland as "mohella." Smaller layers of "mohella" occur everywhere in valleys and lowlands alternating with older and more recent glacial formations, with peat and lava- streams, but they decrease in thickness the further they are away from the large stretches of blown sand in the tuff districts. "Mo- hella" usually resembles a fine, easily crumbled, yellowish-brown or grey tuff, which is often traversed by stems of plants and red tuff-tubes which have been formed around the decayed stems; they 246 THORODDSEN often alternate with layers of wind-polished stones, gravel, scoriae or pumice, sometimes with clay. Where the blown sand is conti- nuously moving, no vegetation can thrive, but when the fine dust and sand has blown away as far down as to the coarse gravel, Icelanders say that the sand is "orfoka," i. e. it cannot drift any longer (see Fig. 14). Then plants are again able to take root and new soil is gradually formed - until that also is blown away. The phenomenon of alternating periods of sand-drift and of vegetation, which has lasted through centuries, is nowhere so distinctly trace- able as in Rangarvellir. Here the substratum is exclusively formed by "mohella," the thickness of which is unknown, but it must be considerable, probably 100 metres or more. Here the lowland plain abuts on the lava-fields of Hekla, whence quantities of volcanic ashes are blown down into the cultivated land. The lowland plain is intersected by deep, branching valleys, which are usually dry, but during the thaws of winter and spring large quantities of water have an outlet through these channels. From the plain a series of small terraces leads down to the bottom of these valleys, which «/ often consists of a grass-covered, level stretch of land. The valley- sides offer favourable opportunities for studying the composition of the mohella; fine bluish-grey layers of sand alternate with reddish sand-layers penetrated by compounds of iron, and the embedded stones of varying sizes bear testimony to the strong erosive action of blown sand. In some layers soil and remains of plants occur, also clay-tubes formed around haulms of grasses. Here and there layers of pumice and scoriae are also seen. No inhabited district at the present time is so exposed to being attacked and overwhelmed by blown sand as Landsveit in the southern lowlands. Here, during the nineteenth century, large stretches of grassland and many farmsteads were overwhelmed by drifting sand, especially in the years 1836 and 1880 — 1881. The substratum consists of old lava which formerly had a covering of mohella and greensward, now to a great extent torn up and destroyed by the masses of blown sand from the north-east. Sand storms cause deep channels and furrows in the soil, which constantly enlarge and by combining with others, gradually destroy the entire layer of soil, so that only a few massive fragments of mohella with hollowed sides and covered with greensward traversed by the fibres of plants, are left behind until they also succumb to the universal destruction. In large stretches of this district all greensward and soil have been torn off down to the naked lava-rock. PHYSICAL GEOGRAPHY 247 From here proceed broad, ramified channels containing blown sand, which are continued and widened, and which constantly encroach on the remaining piece of grassland. The north-eastern part of this district was formerly covered with heather and coppice wood, which the inhabitants, with incomprehensible lack of foresight, destroyed and used for fuel. Skarftsfjall, which stands in the centre of Land- sveit, has protected the areas situated in its shelter towards the south-west. Some streams have also checked the advance of the drifting sand and have thus acted as a protection. Most of the blown sand which in various ways devastates the cultivated districts, originates from the wastes of the plateau. There is indeed enough and to spare of it, at least 3 — 4 thousand square km. of the interior being covered by blown sand of various thickness. On the plateau the blown-sand tracts appear rarely or never to become "orfoka," therefore they are almost ahvays quite bare of vegetation with the exception of the scattered tufts of lyme grass and a few small wil- lows in more favourable localities. In the elevated districts, the surface of which we have been describing, many plants cannot be expected to thrive. As we shall have occasion to mention later on, the highest part of the interior of Iceland, at a height of 650 — 1000 metres, is a waste extremely poor in plant-life. The lowlands, as mentioned above, cover only a very small area (Vis) of the entire surface of Iceland; and a considerable part of this small area consists moreover of barren soil, of lava-streams, of stony rocks and ridges poor in plant-life, and of glacial and blown sand. Therefore the area which may properly be regarded as densely covered with plants, is very small compared with the entire area of the country, and with the present method of cultiva- tion it could scarcely maintain the rural population if the moun- tains and parts of the plateau could not be utilized as pastures for sheep during the summer. The extent of plant-distribution differs, however, greatly in the different parts of the lowlands. While some of the districts are almost entirely or for the greater part covered by a dense vegetation of grass, sedges, heather or coppice-w^ood, as Olfus, Floi, Skei5, Landeyjar, Myrdalur, etc., great parts of other inhabited districts have not even half of their area grass-covered. Several inhabited areas in Skaftafellssysla, Mulasyslur and Thing- eyjarsyslur contain very large tracts of rocky flats, poor in plant- life, lava-streams, sandy wastes, etc.; and in some districts only a very small fraction of the surface is of any use for the sheep- and 24(S THORODDSEN cattle-rearing of the rural population. The peninsula of Reykjanes,, which is of no great elevation, more than half of its surface being below 100 metres, and which has a comparatively mild climate with a considerable rainfall, is, however, so poor in plant-life, that only 4 per cent of the area is grass-covered. The area bounded on the west by Lagaskar5 and on the south by Hafnarfjor5ur is about 1635 square km., but of this only at most 69 km. is occupied by grassland. The greater part of the area of the peninsula of Reykjanes is covered by more recent lava-streams, and has a scanty vegetation; the inhabitants along the coast maintain themselves chiefly by fishing. As it has already been remarked, the surface of the lowlands varies in character. It usually consists of loose masses, but some- times also of solid rock which projects here and there through the more recent formations and the older and more recent lava-streams, as crests, ridges and hills. As mentioned above, in South Iceland large areas (2000 — 3000 square km.) are covered with glacial and volcanic sand, through which branching glacier-rivers flow. Although these sandy tracts originate mainly from river-gravel and sand, other constituents are also found intermixed in them, for instance tuff-dust, and volcanic scoriae and ashes, where active volcanoes occur in the neighbourhood. On Myrdalssandur volcanic slags and ashes pre- dominate. River-gravel and glacial clay occur only upon the surface of changing river-beds. The vast SkeiQararsandur, on the other hand, is formed almost exclusively of rolled glacial-gravel mingled with fine sand and clav, which increases in amount the nearer the coast •/ ? is approached. The various sandy tracts differ naturally somewhat as regards the quality of the material and the size of the grainsr etc. Old lava-streams, also, extend over large areas of the lowlands; they are usually covered with soil which supports a luxuriant vege- tation with heather-moors, coppice-woods and grasslands. In thickly inhabited districts such as Floi and Skei5, the substratum is of lava, and in the former district it is marshy, as it lies so low (at the level of the sea), that the water cannot drain off. Floi is jammed in between twro river-deltas so that the rain-water cannot be drained away owing to the pressure of the bottom water, and in rainy years this district suffers greatly from water which has no outlet, so that the ground is quite boggy. The underlying lava protrudes from every hill and the soil is mixed with lava-fragments. The low- land tract of Floi gradually merges into the district of Skei5 which PHYSICAL GEOGRAPHY 249 is situated upon the same lava-stream, hut as the level above the sea is somewhat higher, the water most frequent!}' penetrates into the lava, and the soil is sandy, hard and dry and covered with a good and vigorous growth of grass. The sedges disappear or retreat to small patches where the local conditions allow the accumulation of a greater amount of moisture. The thicker soil-layers upon the lava-streams, usually originate from tuff-dust (mohella), which has been carried thither and has gradually filled up all the depressions, and from glacial clay deposited by rivers. Several lava-streams with a thinner layer of soil support coppice woods - - e. g. on Thingvalla- hraun, and Hvitarhraun - - and heather, as on Reykjaheidi near Keldu- hverfi, and others. In the lowlands the quantity of the plant-growth upon the lava-streams is closely connected with their age, and by the end of a century, a number of species has already settled down on a lava-stream, as may easily be seen upon the Skafta-lavas of 1783 and the Leirhnuk-lavas of 1724 — 30; on the other hand, the lava-streams of Sveinagja, which date from 1875, are still very poor in phanerogams. In the most thickly inhabited districts the substratum of the soil generally consists of older and more recent glacial and alluvial formations, very often in connection with "mohella," volcanic ashes and lava-gravel. In the lowlands, which were covered by the sea at the close of the Glacial period, marine sand and clay layers are most frequently found immediately upon the basal rock. The clay, which was [deposited by the glacial rivers of the Ice Age, often occurs in layers of considerable thickness ; it is most frequently bluish-grey in colour and turns blackish-blue on being wetted; it is very tough and dense, and can often become rather hard. The clay contains a very insignificant amount of carbonate of lime, usually only 0.1 — 0.2 per cent, while the Danish Yoldia-clay, according to Johnstrup, contains 5 — 15 per cent of carbonate of lime. Along the rivers the banks of clay may often be traced for several kilo- metres without any disturbance being observed in the position of the layers, which is extremely regular and nearly always horizontal. The thickness of the clay-formations varies greatly; in the most highly situated parts of the lowland area and in the valleys, it is sometimes as much as 20 — 30 metres, further down from 5 to 15 metres. The thickness diminishes the nearer the coast is approached ; but it may vary greatly. The clay occurs not only where rivers cut through, but also as a substratum belowr morasses. Marine sand 250 THORODDSEN often occurs above the clay, and then more recent river-deposits, alternating with mohella, volcanic gravel, etc. Here and there, these formations are exposed at the surface and form an almost barren gravelly flat (melar), which, far into spring, is quite slushy, owing to the melting snow, which cannot drain away on account of the sub-surface ice and the clayey subsoil. In the eastern part of the southern lowland tract, under the boggy ground, occur enormous river-deposits (Landeyjar) — delta formations from Markarfljot and other rivers. But higher up, where the soil is drier (Rangarvellhj, the subsoil is composed of thick mohella-formations; nearer to the sea occur fine sand and downs. In several districts in the lowlands there are a great many rocky ridges (holt), which protrude through the morasses and grass- land. They vary in nature according to the character of the underlying rock, and generally consist of basalt or dolerite, rarely of palagonite- breccia. These ridges usually bear clear evidence of the action of the glaciers during the Ice Age. They are highly ice-striated and often have two distinct sides, one bearing traces of having been exposed to the direct force of the ice, while the leeward side is comparatively destitute of such marks. Their external form is some- times dependent upon the direction of the inclination of the basalt- layers. In some places (Myrar and Breidifjor5ur) they originate from fragments of a deeply sunk basal rock. The surface is usually stony, with solid rocks, larger loose stones, and smaller gravel; sometimes a great many erratic blocks are scattered upon the ridges. As re- gards plant-growth, these ridges should be characterized as rocky flats more or less covered with vegetation, and protruding like small islands above the grass-covered, usually boggy level lands. In several places the ridges, in olden times, have been clothed with coppice woods, but it is centuries since the coppice has been destroyed by sheep and goats. Such ridges are found scattered over a great part of the island in the lowlands and valleys, and often impart a cha- racteristic feature to the landscape. The farm-buildings are often built upon them. In the low-lying marshy land of Myrar, at the edge of Faxafloi, they occur scattered in hundreds in the bogs, and in the southern lowland district they are frequent in the eastern part of Floi and the upper part of Holt, while elsewhere large stretches of this lowland area are completely flat. In other districts they are so large that they are most properly described as small moun- tains (fell, halsar). In the valleys the old moraines are often trans- PHYSICAL GEOGRAPHY 251 formed by the rivers, and converted into terraces, which form the substratum of bogs and grassland. The glacial moraine-gravel often extends far up the mountain-sides and forms here a substratum for soil and plant-growth. In other places in the valleys are steep rock-faces, stony slopes, heaps of large fragments of rock (urd), and the conical heaps of finer and coarser gravel brought down by the mountain-streams, which all help to give variety to the plant-formations. While basalt -mountains are slightly and slowly disintegrated, tuff-mountains are extremely liable to disintegration, hence the products of the latter, combined with the action of wind, glaciers and rivers, play a more important part. The contribution of the basalt towards soil-formation dates mainly from the Glacial period. As we have seen from the above, the substrata which support plant-growth are (1) firm ground, having a rocky base (basalt, liparite, breccia and lava); (2) loose soil, consisting of mo- raines, river-gravel, sand, clay, blown sand, volcanic ashes and tuft- dust (mohella); and (3) the products of the plants themselves: boggy soil composed of peat and humus. The character of the subsoil below the humus-layer and the plant-covering is consequently in close relation to the chemical and mineralogical composition of the underlying rock. Over the greater part of Iceland the inorganic soil consists of decomposed basaltic rocks, the main mineralogical constituents of which are plagioclase (especially lime-felspars) and augite, but magnetite and olivine also occur, often in great quantities, and apatite and a small quantity of titanic iron. The chemical composition of the Icelandic basalts is rather uniform. On an average they contain 43 — 53 % of silica, 11—18 °/o of alumina, 11—22 % of iron (Fe 0 and Fe2 ()3), 8—13 % of lime, 2—9% of magnesia, 0.2—2% of potash and 1—4% of soda. Because anorthite, of the plagioclases, is very largely distri- buted in the Icelandic rocks, not only in the basalt, but also in the recent lavas and tuffs, these Icelandic rocks often contain a comparatively small amount of silica and a very considerable amount of lime and also alumina. The reason why the Icelandic soil is nevertheless poor in carbonate of lime may be found in the fact that the lime can only with difficulty be separated from its siliceous compounds, and because in the whole of the island, no sedimentary calcareous rocks are found, though such are of common occurrence in other countries. In districts where sulphurous acids sent out from fumaroles have affected the rocks, as is common in tuff-districts, The Botany of Iceland. I. 17 252 THORODDSEN calcium sulphate (gypsum) is very common in the soil. The Icelandic basalts have not been investigated with regard to the amount of apatite contained in them, but judging from the abundance of phos- phoric acid which often occurs in the waters of the large rivers, it must, in some places, be very considerable. In the districts where liparite is the main rock, the soil has not yet been investigated, but it must vary somewhat, on account of the varying composition of the rock, which contains much larger amounts of silica (65- 78 %>), potash (2—4 %), and soda (3—6 °/o) than the basalts. For the rest, the Icelandic liparites show evidence of their connection with the basalt-area in which they occur, by the fact that almost •/ •/ all of them are soda-liparites. From Iceland there are as yet only a few soil-analyses to hand, and from a few districts only.1 In calcined samples of fine soil from dry grasslands the main mass consisted of silica (37 — 48 °/o), alumina and peroxide of iron (38 — 50°/o), while lime, magnesia, alkalies, present as silicates and other compounds, were found in quantities of from 7 to 14 °/o. The amount of carbonate of lime Avas but small, and varied from 0.5 to 1.7 °/o; in home-fields most frequently 1.5 — 1.7 %. On the other hand, the amount of phosphoric acid was larger (0.3 — 0.4 °/o) than in ordinary Danish soil. The soil samples were rich in humus and contained an unusually large amount of nitrogen considering the amount of humus - from 7 °/o to as much as 24 °/o. The amount of humus and also of iron com- pounds is larger than in ordinary Danish soil. Under unfavourable conditions of humidity the abundance of the organic substance found in the soil constitutes a danger, on account of the formation of pro- toxide of iron; and climatic conditions make the chemical changes in the materials of the Icelandic soil difficult and slow in wet tracts. The amount of the inorganic substances in the sand-samples gives a correct idea of the chemical composition of the solid basaltic rock. Some analyses have been made of Icelandic plants. Firstly, of Icelandic hay, both hay from home-fields (tun; see Fig. 17) and hay from dry and from wet meadows. The analyses show that the Icelandic hay resembles mountain hay from the Alps. It con- tains a larger amount of fat than does the Danish hay; less cellulose; L Analyses of Icelandic soil are found in P. Feilherg: Bemaerkninger oni Jordbund og Klima paa Island (Tidsskrift for Landekonomi. 1881) and by A. Tor- fason in Bunadarrit, Reykjavik, XX (1906), pp. 173—184; XXIII (1909), pp. ,">2— 54. Also, in tbe recently published work by M. Gruner: Die Bodenkultur Islands, Berlin, 1912. PHYSICAL GEOGRAPHY 253 and a larger amount of ash.1 Afterwards, many different species of plants were analyzed and compared with Swedish plants and the main result arrived at was very similar.2 P. Feilberg writes: "Iceland is the land of the Cyperacese and of the coarser species of grass, hut the sheep and the cattle which through generations have accustomed themselves to this coarse food, thrive well on it. The chemical contents of the fodder show that this is also possihle, nor is there any reason why it should be otherwise." In a damp and cold climate such as the Icelandic, the chemical changes in the material of the soil take place more slowly than the formation of vegetable matter, which accumulates and absorbs water where this is copiously present; thus the entrance of air is prevented, and heat is not generated. These circumstances give rise to the production of acid, boggy humus as in other northern coun- tries with a cold and damp climate. Considerable areas in Iceland are covered with boggy soil, and there are also the very best con- ditions for the formation of peat and bogs. In some parts of the lowlands there are vast extents of bogs and swamps, as e. g. in Myrar and Andakill at the head of Faxafloi, and in Floi, Olfus, Holt, etc. in the southern lowland tract; while larger and smaller swampy areas are found almost everywhere. In the lower-lying parts of the plateau there are also wide stretches of boggy land, e. g. Tvidsegra, north-east of Langjokull; Miklumyrar, north of Hreppar, and many other places. In the majority of the districts the area of wet grassland, covered with Cyperacea? and mosses, exceeds by far that of the dry grassland, but unfortunately as yet no measurements are to hand as regards the extent of the bogs.3 There are considerable peat-formations in the Icelandic bogs, but their thickness, distribution, plant-remains, etc. have not yet been investigated. In Skaftafellssyssel where glacier-rivers and volcanic 1 Josep J. Bjornsson: Um heygsedi (Thj<>561fur, 1886, No. 40). P. Feilberg: Graesbrug paa Island, Kbhavn., 1897, pp. 14 — 17. 2 St. Stefansson and H. G. Soderbaum: Islandska foder- och betesvaxter (Meddelanden fran kgl. Landbruks-Akademiens experimentalfalt. Nos. 74 og 83, Stock- holm, 1902 and 1904). Also in Bunadarrit, XVI. 1902, pp. 179—196; XVII, 1903, pp. 25-66; XXIV, 1910, pp. 1—48. 8 M. Gruner's "Die Bodenkultur Islands," Berlin, 1912, came to my notice after I wrote the above; in it he estimates the entire bog area of Iceland at 10,000 square km. or about 10 % of the entire area of the island. According to this, as regards the extent of its bog area, Iceland is reckoned to be third among the Scandinavian countries (Finland 27.2 %, Sweden 12.6 %, Iceland 10 °/o, Den- mark 5 °/o and Norway 3.7 %). 254 THORODDSEX eruptions have constantly been influencing the soil-formation, peat is rare. In the peat there often occur quantities of stems, roots and branches of Betula odorata, and Francis J. Lewis1 has found remains of Betnla verrucosa in peat-bogs in South Iceland. Even in districts where birch-copses no longer thrive, as e. g. on Horn- strandir north of Jokulfirdir, birch-stems are found in the peat bogs. The Icelandic peat is largely utilized as fuel, and as regards its fuel-value it compares favourably with peat from other countries, * only the amount of ash is rather considerable, especially in peat from South Iceland, perhaps on account of the presence of volcanic ashes and blown sand. In the Icelandic peat-bogs there usually occur several, or a few, bands of volcanic ash of various thickness, usually basaltic ashes, but yellowish liparitic pumice -ash occurs also. In districts situated in the neighbourhood of active volcanoes the ash-bands in the bogs are very numerous. Bog iron-ore occurs also in the Icelandic bogs; in some places in rather considerable quantities. In olden times it was much used by the inhabitants, and tor the smelting of the iron many coppice woods were destroyed. Parts of various boggy stretches wet meadows have great economic importance, and the hay (uthey) is largely used as fodder for sheep and cattle, especially as winter fodder for horses and sheep; the best part of it, only, is used for the cows, which live mainly upon hay from the home -fields (tun-hay). In 1910 about 14,300 tons (each ton 1000 kg.) of this uthey were harvested. Of the largest and most productive meadows of this kind may be mentioned Hvanneyri in the district of Borgarfjord, Forin in Olfus and Safamyri in Holt. At the mouths of various large glacier-rivers are large and very fertile stretches of meadow-land, which either constantly or else at times are flooded by glacier water, e. g. in Lon in East Iceland, at Hvanneyri in Borgarfjord, and in several other places where a kind of marsh is formed by the deposited glacier-clay which contains fertilizing substances; in such places the river-water is no longer very cold, and has a very slight current or none at all. In water from Icelandic rivers the following sub- stances of use for plant-food were found: the figures given are those for five million pounds of water (one pound = half kg.).3 F. J. Lewis: Stratification of Peat Deposits in Iceland (Transactions of the Koyal Society of Edinburgh, Vol. 47, Part IV (No. 26), Edinburgh, 1911. pp. 827 — 831). Analyses of Icelandic peat by A. Torfason are found in Eimreidin, XI (190!i), pp. 40—41, and in Bunaoarrit, XX (1906), pp. 116—119. The analyses are found in P. Feilberg: Gnesbrug paa Island. 1897, p. 22;. PHYSICAL GEOGRAPHY 255 1 2 3 4 5 6 7 8 3i "Z « CS t, ^ — ^^2 I « *o .5. .rt •_ +J cs vjj ^ 1 r* ^2 vt; £' > S S * ^ H --' -^ * V5 . 7" ' VC S O ul 3 rt W ffl | ^ -3 JS Nitrogen .... Potash 15-20 10-15 1.37 13 50 4.65 11.50 1.35 11.00 110.00 15.50 4.9-14.12 12.5-17.5 4.5- 6.5 11.0-15.5 3 140 Phosph. acid. Lime 0.6-1.0 100-700 1.05 33.02 5.00 35.00 4.35 41.50 35.00 55.00 1.5- 3.5 21-37 5.6-19.5 49-82 157 46 In several places river-water is utilized for irrigation, and ir- rigated meadow -lands were calculated to cover an area of 28.4 square kilometres in 1909. The only cultivated soil in Iceland is that of home-fields (tun) around the farm-buildings. These home-fields are manured and levelled, but generally are not ploughed. The extent of the cultivated areas (tunes) of the whole of Iceland was in 1909 calculated to be 187.8 square kilometres; to this should be added 2.8 sq. km. for the cab- bage and potato plots. There are often numerous knolls in the tunes, which render haymaking very difficult (Fig. 17). Therefore the im- provement of the soil consists in the levelling of these knolls, which, however, reappear in several places after a time. The nature of the tunes and the quality of the soil naturally differ greatly according to cultivation and situation. The grass (tada) from the tunes con- sists mainly of Gramineas, and, as already mentioned, is used as winter fodder for the cows. In 1910, 5145 tons of tun-hay (tada) were cut. Outside the home-fields there is also a great deal of dry grassland (harflvelli) covered with Graminese, which is chiefly used as pasture-land for sheep and cattle. In the soil of the tunes and the dry grassland a larger quantity of lime and phosphoric acid is usually found than in that of the wet meadow-tracts. l As in other arctic and subarctic regions, "soil-flows" (Soliiluk- tion) are a common phenomenon in Iceland, and they exercise, especially in mountainous regions, no slight influence upon the soil and plant-growth. The upper layer of the soil upon slopes and and in Bunadarrit. XXII, 1908, pp.265 and 266. No. 1 is by Feilberg and YV e- stermann, Nos. 2. 3, 4, 5 and 8 by Detlefsen and Meyer; Nos. 6 and 7 by A. Torfason. Nos. 5 and 8 appear somewhat doubtful. 1 P. Feilberg in Tidsskrift for Landokonomi, 1881, pp. 8 — 12. 25(> THORODDSEN mountains-sides is saturated with water from melting snow during spring, and slides slowly downwards; very often gravel and clay is by this arranged in bands down the slopes. Below large snow-drifts which persist till far into summer, or during the whole of summer^ there is always water which soaks into the soil, and upon many mountain-sides, slow-flowing mud-streams are formed, which in shape and movement recall small glaciers. In other places the flowing Fig. 17. An old Tun. K.jos near Reykjavik. (Phot. A. HesselboJ soil forms small terraces, which are partially transformed into rows of knolls overgrown with plants. Sometimes the soil is loosened from the solid rock, or slides upon the ice of the subsoil; some- times clayey streams flow down into depressions and valleys, and occasionally fragments of greensward, which had been resting upon a saturated substratum of sand and clay, are loosened. Frequently it can also be seen how water, flowing upon the ice of the subsoil down the slope, undermines the soil so that large pieces of green- sward are put into motion, give way, and are torn asunder. It may also happen, where there is a thick layer of loose soil, that the PHYSICAL GEOGRAPHY 257 downward-flowing water, during lhaw, tunnels below the ice-layer and forms subterranean channels, causing the upper layer to col- lapse and fall down into these hollows. In this way, after the vol- canic eruptions of Askja in 1875, great damage was done to the soil on Jokuldalur; its enormous moraine -terraces were covered with pumice-gravel which froze into a thick layer that melted but slowly, because the white, glistening gravel reflected the rays of the sun ; belowr this layer, the water dug out channels, 20 — 30 metres deep, through earth, sand and gravel, and caused catastrophes such as that mentioned above. In this connection may be mentioned the influence often exerted upon the soil by the numerous avalanches and rock-slips of different kinds, and the very slow, creeping movement which may be ob- served in connection with gravel, stones and rocky blocks upon the mountain sides which, in the course of years, may become of very great morphological importance. During earthquakes it may happen that mountain-sides clad with grass and coppice are suddenly de- nuded of their surface soil, which slides down into the lowland plain. During the earthquake of 1896 a piece of swampy soil, 10,000 sq. metres in area and 2 — 3 metres thick, at Thjorsa in the neigh- bourhood of Krokur, thus slid down, being thrown into wavy folds and hummocks, although the slope of the ground was only 1 — 2°. The mountain of SkarSsfjall, which rises 227 metres above the plain, had, before the earthquake, a thick coating of soil, and was grass- covered to its verge; but after the earthquake it resembled a fruit which has been peeled. Thirteen landslips descended on the western side, leaving behind them large surface depressions, and strewing below mighty mounds of soil, clay, gravel and stones together with larger and smaller fragments of torn greensward. These landslips must, in the course of time, have had an enormous influence upon the soil and the plant-growth of many districts where earthquakes are very frequent. All over the island are often seen, along the mountain-sides, marks of ancient huge earth- and rock-slips that could only have occurred during earthquakes. Level tracts with their surfaces cracked into polygonal cakes (rudemarks) are extremely common in Iceland, and fine specimens of such may be seen as, for instance, in the neighbouring districts of Reykjavik. They have a peculiar effect upon plant-distribution on the rockv flat. -kRudemarks" are usually formed only on flat », •/ •/ land where the soil consists of gravelly clay especially clay inter- 258 THORODDSEN mixed with a large quantity of tuff-dust (mohella). The surface is divided into squares or more or less regularly formed polygons, by bands of small stones or gravel, while the clay of the interior of the squares or polygons is destitute of stones. The surface resembles a net the meshes of which are formed by irregular bands of gravel. Usually it is only the gravel which supports plant-growth; there the plants can find shelter between the stones, while the middle of the clay-cakes is too wet for plant-growth. But when the "rude- mark" becomes drier, vegetation may gradually extend over the wr, ?3»' Fig. 18. Portion of a '"rudemark" in the neighbourhood of Reykjavik, showing the position of the stone and gravel bands. (Drawn by Th. Thoroddsen.) cakes of clay, first forming a scattered growth upon them and ulti- mately soil and a plant-carpet, especially when tuff-dust and drifted soil have settled on the surface. The polygonal cakes vary greatly in form and size, but generally the}7 have a diameter of only 1/2- 1 metre. The knolls (Jmfa, pi. fmfur; see Fig. 17), which play such an important part with reference to Icelandic vegetation and agri- culture, stand in close genetic relation to the "rudemarks" and we will therefore try to give an account of the way in which they both have probably originated, but questions bearing on this point require to be elucidated by the experimental investigations of persons living on the spot. My investigations in Iceland have confirmed me in my opinion that the Icelandic pufur as already mentioned stand in a close genetic relation to the "rudemarks" and that sub- surface ice is an essential condition for the formation of both; PHYSICAL GEOGRAPHY 259 Avhere for some reason or other no ice-layer is formed in the ground, neither arudemarks" nor knolls seem to occur. The depth at which this ice is present in spring differs greatly in the different parts of the island, and - according to the weather - in the different years. In the northernmost districts this ice may remain throughout the summer for years during cold and damp periods. As a rule, in the first half of June, frozen ground is met with at a depth of 1 — I1/? metres over the greater part of the island; on the plateau a thicker or thinner layer of this sub-surface ice is no doubt always present throughout the summer, and there, in several places, it gives rise to the formation of swamps, hogs and lakes, as the melting snow and ice cannot drain off. In some districts with very warm ground, where hot springs or other secondary effects of volcanoes occur, the ground is never frozen. As is well-known clay and clay-soil develop, by contraction, numerous intersecting clefts; such cracks are also formed during Avinter by the action of frost, and in severe winters loud cracks are constantly being heard, announcing the rending of the ground. The surface layer of soil is therefore traversed by a network of numerous rents and cracks which divide the clay -soil into irregular frag- ments or a number of prisms. On closer investigation these cracks are not only found to occur on gravelly and clayey flats, poor in plants, but also in the clay-humus of the home-fields ; some of them are as fine as hairs, others have a breadth of 2 — 3 cm. Both these factors, the sub-surface ice and the cracks in the ground, are neces- sary to the formation of urudemarks" and knolls, and to these should be added two other very important factors, viz. frost and unequal surface-evaporation. When, in spring, the snow melts on the cracked and netted surface of a flat consisting either of clayey gravel or of plant-covered clay-soil, and the flat itself thaws at the surface, the water percolates through the ground and the cracks, but cannot escape on account of the sub-surface ice so that the entire laver of soil becomes *. saturated with water. Where the flat consists exclusively of sand and gravel, without any mixture of clay, the entire soil-layer is evenly saturated and the surface of the water can sometimes rise tj to a level with that of the ground. Evaporation then takes place evenly over the entire surface, and when the sub-surface ice melts, the surface-water drains away or evaporates, and nothing further happens. But on a clayey gravel-flat intersected by a network of 2(50 THORODDSEN cracks, the circumstances are somewhat different. During spring the ground is partially thawed ; it freezes in the night and thaws during the day. The sub-surface ice forms a downward limit which does not permit the water to drain away, and uniform circulation and evaporation at the surface are prevented in a "rudemark" by ice- formation in the many cracks, originally full of water, which to- gether with the sub-surface ice as a base form a vascular- or cell- system over the entire flat, and this system lasts at least for some time. The water from the melting snow and ice on the surface col- lects mainly in the cracks and depressions, where it freezes during the night; this is best observed on a knolly flat, which during the thaw of spring produces an entire network of small water-canals. The water cannot penetrate downwards on account of the ice in the cracks or, if they should be free from ice, it wrill vet remain / *^ »/ for a long time in them, for as the water-layer is thicker there than upon the polygonal cakes, the evaporation is slower. The heating of the sun and consequent evaporation of water is therefore greater on the cakes, so that the wet from below, from the slowly thawing parts of the ground and the ice of the subsoil is drawn up into the centre of the cakes. A clayey soil with particles of a certain size has great capillary power and can absorb water and draw it up in great quantities.1 The power of absorption of the clay-soil is greatly increased when it is covered by soil, humus and plant remains. During spring, frost and thaw alternate constantly and daily. The urudemark" freezes in the night, at least partially, and thaws in the morning; in the course of the day the water rises in the individual clay-prisms owing to the rapid evaporation from the surface, but in the night it freezes, expands and raises the central part of the cake. This occasions a constant wandering of the particles of the clay soil upwards into each clay-prism, and by the constant pres- sure exerted by the frozen soil throughout a long period, the heavier particles are sorted out, and as they are less mobile, they are left behind or pushed to the sides. This sorting-out of the coarser ma- terial is the most characteristic feature of the "rudemark." The enormous pressure due to the freezing process is well known. As will be mentioned again later on, in several places in the Iceland mountain-bogs there are opportunities of observing how the frozen The air contained in sandy clays in a dry condition may amount to 40 °/o of their volume, and by infiltration, as large a volume of water may replace the air. See A. G. Hogbom in Geol. Forhandl., Stockholm, 1905, XXVII. p. 22. PHYSICAL GEOGRAPHY 261 water inside the knolls can rupture a greensward, 10 — 20 cm. thick, and traversed with plant-roots. This pressure from below, repeated for years in a "rudemark," must gradually push and force the gravel aside so that it lodges at last in the cracks which, while they are filled with ice, form a kind of wall around each clay-prism. Thus the stones are placed in the neutral territory between the small centres of power, and form a boundary to each cake, the upper edge of which boundary appears upon the surface while the lower reaches down to the ice in the subsoil. Below the level of this ice the gravel is irregularly dispersed in the clay; it is regularly arranged only in the surface-layer above the ice. In the summer, when the soil has thawed and the sub-surface ice melted, the water drains off, and the "rudemark" dries. Everybody who has travelled in Iceland during spring knows what an enormous difference there is between the clayey gravel-flats in which the horses sink deep down while the ice of the subsoil still hinders the draining off of the water, and the same flats in summer when they are dry, so that horses can gallop across them. During summer the clay-polygons be- come somewhat depressed. Many of them are however slightly arched during the summer also and retain for a long time a considerable amount of wet in their interior. Clay which easily absorbs water and expands is well known to Swedish geologists1 who call it "jaslera," and recently it has been connected with "rudemarks." In the neighbourhood of Reykjavik (Melar) some well-defined "rude- marks" have developed in clay soil where a water-containing layer at a depth of about I1/!' metres rests on a thick "mohella" through which water can penetrate only with difficulty, and which therefore freezes in winter into a plate of sub-surface ice. Where the ground consists of clayless sand no "rudemarks" are developed, nor where the subsoil is so porous that water cannot accumulate and form sub-surface ice proper. In my opinion the knolls which are of such common occur- rence in the home-fields of the farmsteads (see Fig. 17) are developed in a similar manner. These knolls are usually larger or smaller elevations of earth which occur together in numbers: the surface- layer consists of humus and plant-remains, but the interior is formed 1 A. G. Hogbom: Om s. k. jaslera och om villkoren for dets bildning (Geol. Foren. Forhandl., Stockholm, XXVII, 1905. pp. 19—36). - E. Bergstrom: En marklig form af rutmark fran barrskogsregionen i Lapp- land. Geol. Foren. Forh., XXXIV, 1912, pp. 339—340. 262 THORODDSEN of the clay soil, which projects into them like a kernel and forms the main part of their volume. The knolls have usually a diameter of i/2 — 2 metres and a height of 1/4 — 1/2 metre; sometimes they are somewhat smaller; or they may he larger. The form varies, but is usually oblong. When the knolls are large the channels (kargatthyfi) between are but narrow. The thicker soil-layer together with the greensward-covering and the vegetation, are intersected by a net- work of cracks similar to that in the "rudemarks:" These cracks divide the ground into numerous pieces of various shapes which behave, as regards the sub-surface ice and the moisture, as the "rudemark." The knolls are dependent on the crack-systems of the substratum. In the knolls the ascending stream of clay and humus particles must be stronger than in the cakes of the rudemark as the greensward, rich in humus, has a very great capillary attraction and an enormous water-capacity it can absorb water to the extent of 50 — 60 °/o of its volume; therefore during the evaporation from the surface it absorbs water vigorously, not only from the thawing ground below, but also laterally from the channels between the knolls filled •/ with water during the spring thaw: in spring the greensward upon the smaller knolls is as saturated with water as a sponge! Bands of volcanic ashes, which were present in the ground or in the greensward before knoll-formation began, become bent upwards in curves according to the form of the knolls — a fact which, among others, is a proof of the local pressure from below in each knoll. In spring, during the melting of the snow, the channels between the knolls are often half filled with water which cannot escape, while the tops of the knolls are dry, sometimes even very dry owing to evaporation, so that the vege- tation upon them is totally different from that upon the sides of the knoll. Sometimes large knolls have a kernel of ice far into the summer. The formation of knolls does great damage in the peasants' home-fields and great trouble is taken in levelling them, but they may reappear comparatively quickly if the ground is not thoroughly drained, so that the surface-water and the water from the melting ice of the subsoil are immediately carried away. In this connection it should be mentioned that where a snow-covering during spring protects the ground for a long time against the action of frost and thaw, no knolls are formed. Knolls of the kind described above occur in thousands also in uncultivated grasslands with clayey mohella-soil, on heather moors and on grass "moar", and here, also exclusively upon somewhat flat PHYSICAL GEOGRAPHY 263 land1 and in places where the water, for some reason or other, cannot drain off. Here the soil is usually much thinner than in the home-fields hut nevertheless it lias a very great power of ahsorbtion during the process of surface-evaporation ; the greater part of these knolls consists of niohel la-clay. On an entirely hare clay-Hat no knolls are found, and the separate cakes in the ,,rude- mark" rise but slightly in the middle during spring, hut not until they are plant-covered and clad with greensward do they bulge and retain their convexity. On a closer investigation it is probable that several transitional forms will be found between "rudemarks" and knolly flats, but investigators have not as yet made this matter a subject for study. Where similar plant- formations (heather and "moar") occur upon slopes or mountain sides upon which the water easily finds outlets, knolly flats are absent, and k'rudemarks" are never found in such localities. Upon mountain-sides small eminences of various kinds may sometimes be observed which are due to under- lying stones or ridged mud-flows, etc., but never "thui'ur" proper. Nowhere have I seen any indications of "rudemarks" or uthufur- marks" being at all connected with mud-flow phenomena. On the plateau peculiar knolls of usually large size are found which the inhabitants of the different districts call "riistir," "haugar" or udys.'' These knolls are, as a rule, of irregular, oblong form, and are bare of vegetation on the top, where they consist only of humus and clay. It can be seen that the ground has bulged and the knolls are cracked at the top. In some places there are evident proofs of this being the case: bands of the original soil are seen to lie upon the top of the knoll while the clay and gravel within have poured out of the cracks between the bands. In the neighbourhood of Ullsvatn on Tvidaegra, at a height of 460 metres above sea level, I investigated such knolls in 1898; they were 1--11/* metres in height, 15 — 20 metres in length and 8 — 10 metres in breadth. The sides were covered with heather, but Cyperacese grew in the intervening spaces. Similar knolls occur in several other places in the interior of Iceland, but only in one more place did I observe 1 Dr. H.Jo us son describes a heather-covered "rudemark ' where the cracks between the cakes were covered with Grimmici hypnoides, but the cakes themselves with Calluna and Empctrum. This peculiar vegetation the brownish heather- vegetation divided into numerous polygons by the greyish moss-bands - - occurred only upon level ground (Botanisk Tidsskrift. XXVII, 1905, pp. 43 and 44). 264 THORODDSEN them to be as large as those at Ulfsvatn, viz. in East Iceland on the plateau south-east of Sna^fell, at a height of 690 metres above sea-level. This was in 1894, but the inclemency of the weather pre- vented me from making a halt at the place. Here, also, several of the knolls had burst and discharged a large quantity of clay-soil. This form of knoll occurs especially at the edge of the bogs where the foot of the knoll stands in water during the melting of the snow: where the supply of water is too abundant these knolls do not occur. It is a fact peculiar to all knolls, both in the lowlands and on the plateau, that they consist of mohella-soil and humus intermixed with clay; this is true also of the "rudemarks" with the exception that more gravel and stones are found in them than in the plant-covered ground. As mentioned above, there is no direct connection between the mud-flow phenomena and the "rudemarks" and knolls, but it may happen that a flat with either of these sur- face-forms is afterwards affected by mud-flows; this is then a se- condary phenomenon. Where subsoil-ice is absent, as for instance in the neighbourhood of the hot springs which are scattered in hundreds all over the country, I do not remember to have seen "rudemarks" or "thufur" proper. Upon the plateau itself "rude- marks" are rather rare, but, on the other hand, various mud-flow phenomena are common; thus, stones are often seen arranged in bands on the slopes, and in various other ways which, however, I regret to say, have not as yet been more closely investigated. On basaltic plateaus covered with coarse basalt-gravel and blocks of rock, patches of clay are now and then seen which appear, in the middle, to have thrust up from below and spread out slightly on all sides. Besides these knolls there are many other similar eminences in the gravel-covered ground which are called knolls, but are due to quite different causes and have a different origin. Sometimes nodules on the greensward are due to the form of the substratum, as where a thin layer of soil occurs upon lava, or upon a stony and rugged bottom. In blown sand, knolls are formed around roots of Ely mus arenarins, Halianthus peploides and other plants. Where the ground- water stands very high, as in many extensive boggy tracts, so that the entire surface freezes into an ice-cake, only small knolls of organic material are seen, formed of moss and Carices. On hard grassland, on gravelly flats and similar places, small PHYSICAL GEOGRAPHY 265 knolls are formed of the rootlets, leaf-rosettes, etc. of various plants. Sometimes knolls are formed of organic material in places where a small spot is constantly manured, e. g. the so-called "bird-knolls" on higher levels, where bird of prey are in the habit of alighting, and the characteristic Icelandic "dog-knolls" (hundapufur) along the roads where the dogs are wont to stop, which almost all travellers in Iceland have for companions. III. CLIMATE.1 AONG the coast of Iceland, generally speaking, a decidedly insular climate prevails, the conditions of which are determined by the wind-distribution over the North Atlantic and by oceanic cur- rents. The climate has an oceanic character, the summer being cool and the winter usually mild but it may become very cold when northern winds prevail and the Polar ice blockades the coasts. The air is usually damp, and storms are very frequent and violent. The climate varies, howrever, rather considerably in the different parts of the coast, and there is also a considerable difference between that of the coast and the interior. 1 Meteorological observations from Iceland are published annually in "Mete- orologisk Aarbog," 2. Del. Bilandene. Kobenhavn (Annuaire Meteorologique. Public par I'lnstitut meteorologique Danois. Deuxieme partie. Les colonies), in Danish and French. See also. Elements meteorologiques des iles Feroe, de 1'Islande et du Greenland. Copenhague, 1899. The climatological observations from Iceland have, as yet, been worked out to a small extent only, and there exist no larger and more exhaustive accounts of the climate of Iceland; only some small, but valuable, papers are to hand by V. Willaume- Jantzen Climat du littoral islandais. Con- gres maritime international de Copenhague. 1902; and some articles in "Salomon- sens Leksikon,'' 1899, and in "Atlanten, " 1904); use has been made of these in this paper. The climatological means which are given in this paper have been calculated and classified at the Meteorological Institute in Copenhagen in 1910, and were previously published, 1911. only in my book on Iceland "Lysing Islands,'1 in which there is a section (vol. II, pp. 327— 394) giving an account of our present knowledge of the climate of Iceland and its weather-conditions during historic times. Among older papers on the climatology of Iceland may be mentioned .I. Thorstensen. Observationes Meteorological, 1823-1837, in Islandia factae. Hafniae, 1839; J. F. Schouw, Nogle Bermerkninger om Vejrliget paa Island i Vinteren 1824 — 25 (,Tids- skrift for Naturvidenskaberne. Kebenhavn, 1826, IV, pp. 259—262); Mag. Pedersen, Undersogelse om Barometrets daglige Middeloccillation paa Island (Overs. Vid. Selsk.. Forh., 1845, pp. 65— 69); and lastly observations by H. .). Scheel from Akureyri, 1811 — 13 (Annals of Philosophy. Edited by Th. Thomson, Series 1, Vol. XI, London, 1818, pp. 96—103 and 169—175), and observations by A. O. Thorlacius from Stykkisholm, published in "Journal of the Scottish Meteorological Society," 1869. 1873, etc. Articles on the winds, drift-ice and other more special subjects will be mentioned subsequently in footnotes to the text. PHYSICAL GEOGRAPHY 267 The climate of Iceland, owing to oceanic currents, is much milder than could be expected from the position of the island. As is well-known the temperature of the air varies greatly at the same latitude on the eastern and western sides of the North Atlantic. Slykkisholm in Iceland lies about 65° N. lat. as also Brono in Nor- way and Angmagsalik on the east coast of Greenland, but the tem- perature varies greatly in these three places as shown in the fol- lowing table: February July Average for year Brono — 1.4° C. 12.8° C. 5.2° C. Stykkisholm — 2.7° C. 9.7° C. 2.8° C. Angmagsalik — 10.8° C. 5.4° C. — 2.6° C. The above table shows, among other things, the great influence exerted by the cold, ice-carrying current in Denmark Strait. On the whole, oceanic currents have a great effect upon the climate of Ice- land. The west coast has its temperature raised by the Gulf Stream. One branch of this passes Cape Nord and continues its course along the north coast where it becomes cooled, but has still a compara- tively high temperature which is distinctly felt at Grimsey. Another branch of the Gulf Stream comes from the south up towards the south and east coasts where it meets the cold Polar current which comes dowrn along the east coast of Greenland and at Iceland di- vides into two branches. Of these branches the one broad branch turns dowrn through Denmark Strait, while the other flows down towards Langanes, and thence southwards along the east coast of Iceland to South Iceland where it passes between the coast and the warm current outside. The conditions connected with these currents exert a great influence upon the temperature of the ocean off dif- ferent parts of the coast and thus upon the flora and fauna of the ocean, which differ greatly according to whether they are under the influence of cold or warm water. Thus, the algal vegetation along the north-eastern coast of Iceland has a different character is more Polar than towards the south-west ; this applies also to the fish fauna, and the deep-water, spot-bound fauna, as well as to the plankton. The temperature of the surface-water of the ocean varies there- fore in no slight degree off the different parts of the coast. During wanter the temperature of the ocean-water off the east coast, where the influence of the Polar current is greatest, is on an average 0.8° 1 Julius Hann: Handbuch der Klimatologie. Stuttgart. 1908, I, p. 181. 18 268 THORODDSEN (Papey); off the north coast 2° (Grimsey); off the west coast 1° (Stykkisholm); and off the south coast 4.1° (the Vestmannaeyjar); but it varies rather considerably with the increased or decreased cm — • ° n o> a o OQ ss o =• o_ o' o D3 fr n> n ** a a o ^B' ^^ 5" O o proximity of the Polar ice. The mean summer temperature of the ocean round Iceland is 5.8° at Papey, 6.1° at Grimsey, 9.7( at Stykkisholm, and 10.4° at the Vestmannaeyjar. For further parti- culars consult the following table: PHYSICAL GEOGRAPHY 2(59 Table I. Mean Temperature of the Surface Water of the Ocean round Iceland for the Period 1872 — 1906. (Centigrade scale used.) 0 Name £ >> ^^j i- — * "3 r "-a ^1 t/3 •^ O O 0 CJ C ^ oc 3 d j- T 0 0 I \ 0 It o u o 0 (i II 0 0 o 0 M U Stvkkisholm . . 0.9 0.3 0.4 1.7 4.7 8.2 10.4 10.6 9.1 6.3 3.7 1.8 1.0 2.3 9.7 6.4 4. The Vest- ma nnaeyjar. 4.1 4.1 4.5 6.1 7.7 9.5 10.9 10.8 9.1 6.8 5.1 4.1 4.1 6.1 10.4 7.0 6. Papev 0 7 04 0.5 1 5 3.1 4.8 6 0 6,6 6 2 43 2 6 1 3 08 1 7 58 4,4 3 Grimsev 1 8 1 4 0 7 1 3 ?! 6 4? 6 6 7 4 69 5,3 4 1 ? 9 ? 0 1 5 61 5.4 3. The temperature-conditions of the coastal districts correspond very closely \vith those of the ocean. The north-east coast has, on V *• an average, a mean winter-temperature of from - 2° to 4° C.? a summer temperature of 6° to ll/2Q C., and a mean for the ^yhole year of 1° to 2°C. ; while the south-west coast has a winter %J temperature of 0° to - - 2° C., a summer temperature of 9° to 10° C.,> and one for the whole year of 3° to 4°. When the Greenland drift- ice blockades the coast, the difference between these temperatures is considerably greater, in the severe ice-year, 1881, the difference between the temperature on the isle of Grimsev and in the Vest- mannaeyjar, from January to March, was 10V20 to 14° C. , while normally in these months it is only 3l/2° to 51/2°; in the same months (in 1881) the difference of temperature between Stykkisholm and the Vestmannaeyjar was 7° to 10° C., while normally it is only 3° to 4°. Therefore the temperature varies very greatly from year to year, and so also does the mean of the different months. For instance, take the month of March: at Stykkisholm the highest mean temperature in a period of 33 years was 2.9°C. ; and the lowrest mean in the same period w^as 13.3° C. From this it will be seen that the Polar current, and especially the drift-ice, exercises a great influence upon the climate in Iceland and thereby upon the vegetation and the means of sustenance of the people. When the Polar ice arrives at the north coast the temperature immediately falls: when in May and June the people who live in South Iceland see that it is snowing on the mountains, they at once take it as a sign that the dreaded drift-ice is about to blockade the coast of 18* 270 THORODDSEN North Iceland. So long as the ice drifts backwards and forwards along the coast, the weather continues to be very changeable and stormy; but once the ice has been grounded on the land, the weather becomes more settled, although colder. The parts of the coast which are most subject to be blockaded by the drift-ice are the north and east sides of the north-western peninsula, especially Strandasysla, and Langanes and Melrakkasljetta. In these districts the effect of the presence of the drift-ice is shown both in the wider extension of the snow-wreaths and in the character of the vegetation. On the east coast of the north-western peninsula, south of Cape Nord (also called Cape Horn) there is no vestige of coppice-wood, while this occurs luxuriantly on the western and south-western side of the peninsula, and at the heads of the fjords. Cabbage and potato plots are also absent along this coast, north of 65 ° 40 ' N. lat. ; while small plots of cabbages occur frequently in ASalvik on the north- western coast of the peninsula at 66° 25' N. lat. Even at the head of Hunafloi the influence of the drift-ice is evident: Chamcenerium latifoliam which flowers every year on the plateau, 600 — 700 metres above sea-level, flowered only twice during eleven years (1878 — -88) at Midfjordur. In the districts which are most exposed to the effects of the Polar ice, the herbage is extremely poor owing to the con- stant coldness of the springs and the ra\vness of the summers; the frequent snow-falls even at midsummer, make hay-making precarious, so that the sheep and cattle and thereby the inhabitants are liable to suffer want. During the nineteenth century the coast of Iceland has been free from ice on an average about one year in every four or five; but no rule can be formulated for the arrival of the Polar ice on the coast of Iceland; sometimes it remains absent for many years: «. •/ sometimes it visits the coast several years in succession. The ice arrives at different times as a rule from January to April; if it arrives early, it usually drifts away soon without doing any great damage, but if it arrives near the time of spring, it often remains till far into summer, and causes much inconvenience, hindering navigation and fishing, spoiling the hay-harvest, etc. But it almost invariably drifts away at the end of August, and for the last four months of the year the coast is almost always free from ice. As a rule, the Polar current first carries the drift-ice to the north-western peninsula of Iceland, off Cape Nord, and the main mass is carried out through Denmark Strait; but that part of the ice which besets PHYSICAL GEOGRAPHY 271 Hornstrandir (the coast down either side of Cape Nord), is carried by the coastal current (the Irminger Current) along Strandasysla into Hunafloi, from thence outwards along the east coast of this fjord, then further eastwards along the entire north coast, past Langanes and on, down to the east and south coasts, wrhere it either melts or else drifts away into the open sea. The ice wrhich comes to South Iceland always conies from the east, never from the north-west. In severe ice-years the drift-ice may blockade the whole of the south coast as far as to Cape Reykjanes; on the other hand, it very rarely happens that any drift-ice enters Faxafloi and more rarely still that any icebergs drift past the north-western fjords into Brei5ifjor5ur; the ice is not known to have been grounded on the land, down past Patriksfjordur, for the last 200 years. The most severe ice-year known was the year 1695: in that year drift-ice surrounded the whole island with the exception of Snaefellsnes - - a fact unparalled in the history of the country. In most of the places the ice occurred in such quantities in that year that open water \vas not visible from the highest moun- tains. As usual the ice drifted from North Iceland to South Iceland and then southwards and reached Thorlakshofn as early as April: from thence it drifted into Faxafloi as far as Hitaros, and from the north-wrest the ice drifted past Latrabjarg into Brei5ifjor5ur. In the beginning of May it was possible to walk and ride everywhere out- side all the fjords of North Iceland. It sometimes happens that the drift-ice drifts to Langanes and then dowrn to the east coast without touching the north coast.1 The Polar current bears great quantities of drift-wood to the northern coast of Iceland, most frequently to the neighbourhood of Cape Nord and Langanes. The greater part of this drift-wood starts probably from Siberia; that of most common occurrence is Larix sibirica, Picea obovata, Abies sibirica, Finns cembra, Pinus silvestris, Popnlns tremiila, Salix vitellina, and others.2 1 Th. Thoroddsen: Den gronlandska drifisen vid Island (Ymer. Stockholm, 1884, pp. 145—160). C. Ryder: Isforholdene i Nordhavet, 1877—1892 ;Tidsskrift for S0va?sen, 1896). V. Garde: Isforholdene i de arktiske Have in Nautisk Mete- orologisk Aarbog of 1898 and subsequent years (in Danish and English). W. Meinardus: Periodische Schwankungen der Eisdrift bei Island (Annalen der Hydrographie und Maritimen Meteorologie. 1906). • J. G. Agardh: Om den Spetsbergska Drifvedens ursprung (Overs, af Kgl. Vet. Ak. Forh., 1869, No. 2, pp. 97—119). A. G. Nathorst: Tva somrar i norra Is- hafvet. Stockholm, 1900, I, p. 345. Th. Thoroddsen: Ferdasaga af VestfjorSum (Andvari, XIII, 1887, pp. 164—168) and Ferdabok, II, 1914, pp. 63—70. Compare Geografisk Tidsskrift, IX, p. 45. E. Olafsson: Rejse gennem Island (1772), I, pp. 508 — 513. O. Olavius: Oekonomisk Rejse gennem Island (1780), pp. 126—146. 272 THORODDSEN The drift-ice also carries to the coast of Iceland fragments of rock which are not found in Iceland (gneiss, granite, talc-slate, mica- slate, etc.), and also gravel, clay and earth from other Arctic coun- tries. Moreover, the warm water of the Gulf Stream sometimes carries drift-material with it from southern parts, e. g. mahogany, sugar- cane, "nuts" of Entada Gigalobinm, Miicuna urens, Guilandina Bon- diicella, etc. The weather of the North Atlantic Ocean depends upon the winds, which also exercise a great influence upon the currents around Iceland. The mean atmospheric pressure in the North At- Fig. 20. Seeds of Entdda fni/alohiuin (1), Mucitna urens (2), and Gnilandhut Bonducdla (3). lantic is generally lowest to the south-west of Iceland but there is, in addition, another centre of low barometric pressure north-east of Iceland; on the whole, the direction of the winds over Iceland is determined by this. In West and North Iceland the majority of the winds blow from SE., E. and NE., while on the east coast winds from the NE. and NW. are nearly equally numerous (20 % and 17 % relatively) because here both the above-mentioned centres of low atmospheric pressure try to prevail. This is true of the average condition only, because in reality the wind-conditions from day to day are very variable, as the ocean here is constantly visited by moving centres of depression. The barometer in Iceland is subject to very great changes. In Reykjavik the barometer, during a period of 28 years, has oscillated between 789 mm. and 697 mm.; at Stvkkis- * holm, dining one of 33 years, between 789 mm. and 700.3 mm.; in the Veshnannaeyjar during a period of 29 years, from 789 mm. to PHYSICAL GEOGRAPHY 273 o -r LO ^H CO •«* X LO t^ co r- r- 0 LO CO CM CM LO CM ,-H CM Si SO CvS .- CO LO 1-H jaqcxpo UO f"1^ L^ *n^ __ CM SO X CM LO CO ^ X ^ LO CO uO CO $££ LO LO ^* I"""* *^* CO T X i— 1 uO Si T* CO X >* r- CO X I-H LO CO CO l~- CO LO 1-* CO C^l -f J9qui9}d9s LO CM r— £^ LO CO Si ^ X *H< r>- i— i LO ^f UO CO X CO »— 'H. t^ CM >o CO X !>• co T r^* X T* Si LO co c t- 1-H I-H CO CO •**• X LO X LO Si • .» — . ^^ CO isngny 5 CO O co LO CO "* [>• CM • ~ — LO CO O LO CM LO r- co LO CO ^+* - - __ LO t^ co l^ CO LO l^ X CM "* LO ** f^ r*1* co co ^ si r- t^ CM CO Si Si X LO CO CM oo A*plf LO CM I-H CO LO ^t* CO l^ CO uO CO ?•! ^H CO LO t^ co l^ CM uO CO ^ C^l >0 LO t^ CO uO C^l i— CO CO Si CM -* X CM •^* Si LO Si co UO Si Si C— CO -t 1-H C^l CO o 91111 f OC CO LO X LO t— CM w. ^5" LO CO •* r^ co uo r* CM Si LO LO co co r- I-H LO t- CO 5S CO LO 1- CO CM l> r- co uO CO *H-I CO Si SO CO 1-H ^f ^f CM I-H uO X 1-H Si 1-H ten o CO LO CO O x co O '* CO CO 't CO CO LO X CO 1-H l-O CM CO X uO X CM TZT CO CM CO CO LO X 0 CO (M CO CO -* 0 CO Si LO CM O ^1 ^H LO LO CO uO CO CM X CO tudy uO 2S Si X r-- I-H CO — LO CO r^ CO X "* X 1-H X -H. LO CO I— CO X -t X i— X LO co rh 06 — rf ** LO LO co ~ X Si ^~* r^*" Si X x r- co co l^ CO CM CM ¥ T ^ T"P TiT 1 1 «_/«.l O l'\ S CO O CO *H- X CM X ^H CO CO UO CO r~ Si O CO X I-H •^t" LO ^? X ^H LO CO CO X CO -H" X LO f-H CO LO Tl- ^^ **^ Ci X CO LO C^l CO CM •* t^ Si CO ^^ p*^ ^^ t^ AJBU.KPJ 1—1 CO CO Si CO X sg 1^- I>- Si CO X Si CO CO f- LO CO Si X ^ CO X CO LO t^ X Si CO CO X LO O X X CM F- CO Si 1- LO ^« rt Si 1-H X X l> — — CO 00 AJBUUBf 5 CO CO OC .— fc - — •*4l l~ CO X CO CO X CO LO LO CO CO X co LO X X LO CO CO X X m . - . f o 4V C3 4-> t/: 1 C5 so (j • ^ 3 * < 1 — " £1 •o ° S £ § -* S "E T" 50 C of maximum pressure of minimum pressure. te maximum pressure te minimum pressure |f. -H o'" « O p • p « "^"^ r- 0 " Atmospheric Pressure of maximum pressure of minimum pressure, te maximum pressure, te minimum pressure . CO ^ ? ? 3 " «r 2. -S^^P '"s^ -^ *"" ~~ of minimum pressure, te maximum pressure te minimum pressure ~ " 2 o c/: < £ "d 5 £ 4) "^ .- 3 S § <• ^ ? ^^ |>. ^, > E •M p < of maximum pressure of minimum pressure, ite maximum pressure ite minimum pressure £j CS C5 _ — "5 'c y: T: a CJ C & — ~ x S s — — J -2 -C c ^ Pu p s II! _, "o 05 274 THORODDSEN Table III. Prevailing Winds expressed as percentages. Name of station >> — rt S f-< OS "-s P'ebruary »-* 1 rt s •IH (H a < >> rt & w e 2 >-5 >> "s "-a 4-> CO S tc s < September October November December t-, « s §?>> S 2 > zs < -- £ Vestmannaeyjar (V? 1877 to 31/i2 1906). N 13 13 18 13 10 4 7 10 11 18 17 17 12 NE 3 4 3 3 2 2 1 2 1 4 4 3 3 E 19 22 24 27 27 24 20 23 26 21 20 18 22 SE 9 8 g 10 10 14 13 9 10 9 8 8 10 S 10 10 9 9 8 9 5 5 9 8 9 10 8. SW 16 13 9 7 8 10 9 9 11 8 12 14 U W 8 g 7 7 8 9 13 8 6 B g 7 8 NW 3 3 4 5 4 4 7 4 4 4 4 3 4 Calm 19 18 18 19 23 24 25 30 22 22 20 20 22 Papey (Vs 1873 to 31/i2 1906). N 11 15 13 12 10 5 5 8 7 12 14 11 10 NE 11 13 18 26 27 24 26 24 19 16 15 12 20 E 4 7 6 8 9 13 10 10 8 8 6 6 8 SE 5 5 5 6 5 6 6 5 6 5 4 5 5 S 7 7 6 8 7 10 10 8 8 5 6 7 8 SW 19 17 12 12 14 18 13 12 '>0 15 17 15 W 9 6 5 3 4 5 5 3 5 5 6 7 5 NW 24 20 94 15 13 5 6 10 24 25 27 17 Calm 10 10 11 10 11 14 19 20 13 U 9 8 12 Stykkisholmur (V6l873to31/12 1906). N 5 5 5 (5 6 5 5 9 g 8 (3 7 < & NE 14 17 18 17 16 10 10 12 15 19 20 14 15 E 21 93 24 25 '>! 18 21 21 19 99 21 24 22 SE 16 15 15 15 14 14 10 10 15 17 16 14 14 S 13 11 9 10 11 13 9 9 13 11 9 10 11 SW 16 15 12 6 fi 8 « 5 10 9 12 16 10 W 5 4 5 6 8 12 15 11 7 4 5 4 7 NW 1 1 1 1 4 7 8 6 2 1 1 1 3 Calm 9 9 11 14 14 13 16 17 13 9 10 10 12 G r i m s e y C/7 1873 to 31/i2 1906). N 8 Q 9 7 8 5 8 7 6 6 8 9 7 NE 16 18 18 19 0 ?6 27 27 28 97 23 91 90 16 22 SE *>() 21 20 16 10 11 9 19 17 1 S 1 ^ 20 16 S 7 7 r; 4 9 Q 9 O ^ i; i; 8 5 SW 9 g 5 3 2 3 9 9 5 6 7 5 W 14 12 12 10 u 10 10 10 13 U 13 12 11 NW 5 5 5 5 9 10 11 12 « g 6 6 7 Calm 6 4 « 10 13 19 17 16 10 7 4 5 10 PHYSICAL GEOGRAPHY 275 699.2 mm.; at Beruijord during a period of 34 years, from 788.7 mm. to 704.9 mm.; and at Akureyri, during one of 33 years, from 789 mm. to 705.7 mm. The paths of the moving centres of depression lie especi- ally frequently south of Iceland so that one centre of depression after the other crosses the Atlantic from the west with a main direction of east or north-east; sometimes nearer, sometimes further from Iceland. This movement of low pressure centres causes the wind to veer to- wards the sun, especially from S. or SE. through E. and NE. Table III shows that at Stykkisholm, very frequently the winds blow from S. and SW. (such form 11 °/o and 10 °/o respectively) which is due to the fact that Denmark Strait also provides a path lor the moving centres of depression ; they come from the south and cause on Iceland a veering of the wind with the sun from SE. or S. through SW. It is probably these centres of depression which cause the rather frequent winds from the W. and SW. at Grimsey and Papey respectively. There may, however, occur longer periods in which the winds of Iceland are rather constant, the lowest atmospheric pressure over the North Atlantic remaining at the same place. The wind-conditions are gen- erally favourable for Iceland, the warm air from the Atlantic Ocean is carried in over the land, while it is rare for cold winds to be thus brought during winter. According to observations taken at Stykkisholm on the west coast, the warmest winds there are the SE., S. and SW. while the greatest cold is caused by the N., NE. and E. winds. The difference between the temperature induced by the warmest and bv the coldest winds is on an average 9° — 10° C. */ in the months of December to April, and 4° — 6° C. in the other months. The temperature may show great differences according to whether northerly or southerly and easterly winds are the prevalent ones for any length of time; thus, the mean temperature at Stykkis- holm in March 1856 was 4- 4.3° when southerly winds prevailed, and in March, in the ice-year 1881, the mean temperature was — 13.3° when easterly winds were prevalent.1 Conditions pertaining to the atmospheric pressure over the Atlantic, which is specially 1 V. Willaume-Jantzen : Climat du littoral islandais, 1902. N. Hoffmever: +f Die Vertheilung des Luftdruckes uber den nord-atlantischen Ocean wahrend des Winters und deren Kinfluss auf das Klima von Europa (Zeitschrift fur Meteorologie. Wien, 1878. Bd. XIII. Nr. 22). N. Hoffmever: Etude sur les tempetes de TAtlantique septentrional. Copenhague, 1880. G. Rung: Repartition de la pression atmos- pherique sur 1'ocean atlantique septentrional. Copenhague, 1894. V. Garde: Vind- kort over den nordligste Del af Atlanterhavet ' Nautisk-meteorologisk Aarbog for 1899. Kebenhavn, 1900. S. 25-46). 276 THORODDSEN Fig. 21. The mean annual temperatures of 28 years (1874 — 1901). (V. Willaume-Jantzen.) Fig. 22. The meanspring temperatures for the period 1874 — 1901. (V. Willaume-Jantzen.) PHYSICAL GEOGRAPHY 277 Fig. 23. The mean summer temperatures for the period 1874—1901. (V. Willaume-Jantzen.) Fig. 24. The mean winter temperatures for the period 1874 — 1901. (V. Willaume-Jantzen. 278 THORODDSEN expressed in the height of the barometer at Iceland and in the Azores, have, as is well-known, a very great influence upon the climate of western Europe.1 The wind blows with great force over Iceland and storms are frequent, especially in the winter half of the year. At Stykkisholm there are, on an average, 50 days of storm annually; in the Vest- mannaeyjar 25; and on Grimsey 11. On the west coast the majority (60 %>) of the storms are from the NE., while in East Iceland the NW. wind is the most stormy, causing 50 %>. As the fisheries along the coast of Iceland are carried on especially in the winter half of the year (the fishing season along the south-west coast begins in February), the storms constantly cause a great many disasters at sea. During the years 1850 — 1877 (with the exception of 1853 and 1875, for which no data are to hand) 2008 people were drowrned off Iceland, forming 3 °/o of the total number of deaths during those years; of these, 60 % were drowned in March. During the years 1881- -1910, 2096 persons were drowned out of a population of 70,000 — 80,000. Both in northern and eastern Iceland the Fohn wind, which blows down from the Jokulls of the plateau, is fairly com- mon during winter and causes a sudden rise of temperature, re- sulting in the melting of the snow in the lowlands and of the ice upon the surface of the lakes and rivers. N. Hoffmeyer2 describes a Fohn wind which blew down from Vatnajokull over south-east Iceland from September 18th to 26th, 1877, causing the temperature at Berufjord and on Papey to rise from 7° — 8° C. to as much as 18°— 20° C. ; otherwise, such a high temperature is very rare at these stations, even at midsummer. The oceanic character of the Icelandic climate is manifested in the low degree of heat experienced during summer and the mild- ness of the winter. The mean temperature of the coldest days along the coast of Iceland (Stykkisholm, Berufjord and Grimsey) was in 1 J. Hann: Die Anomalien der Witterung auf Island in dem Zeitraume 1851 bis 1900 und deren Beziehungen zu den gleichzeitigen Witterungsanomalien in Nordwestevropa (Sitzungsberichte d. Akademie d. Wissenschaften in Wien. Math.- naturw. Klasse. Band 113, Abteilung II a, Wien, 1904. S. 183—269). H. H. Hilde- brandsson: Quelqnes recherches sur les centres d'action de 1'atmosphere IV. Sur la compensation entre les types des saisons simultanes en differentes regions de la terre (Kungl. Svenska Vetenskapsakademiens Handlingar, Band 45, No 11. Stock- holm, 1910). N. Hoffmeyer: Vejrforholdene paa Island i Vinterhalvaaret 1877 — 78 (Tids- skrift for populaer Fremstilling af Naturvidenskaben, 5. Raekke, V Bind, 1878, pp. 161—172). PHYSICAL GEOGRAPHY 279 Fig. 25. The mean autumn temperatures for the period 1874 — 1901. (V. Willaume-Jantzen.) Fig. 26. The differences between the summer and winter temperatures of 28 years (1874—1901). (V. WiUaume-Jantzen.) 280 THORODDSEN February and in the first half of March - 2° to 4° C., which is only from 1° to 3° colder than the average temperature of the coldest days in Copenhagen; but, on the other hand, the warmest days in Iceland, which occur in the latter half of July and in v •- August have, on an average, a temperature of only from 7° to 10° C. There are, however, numerous oscillations from year to year, mostly during winter, less frequently during summer. The normal tempera- ture is below zero (Celsius) from the last days in November to the first days in April, or even May 1st on Grimsey. Therefore the Ice- landic summer is very unfavourable to plant-life, the temperature being so low during growth-period. The highest mean tempera- tures for any month fall in July and August and even these reach only to 11°— 12l/2° C. (Papey 8.5° and 9.6°; Vestmannaeyjar 12.7° and 12.3°) - the temperature of a mild May or cool September in Copenhagen. By exception, on some days, the heat may be very great; the greatest amount of heat registered at Berufjord was 26.3° C.,. on Grimsey 26.2° C., at Stykkisholm 22.9° C., and in the Vestmanna- eyjar 21.2° C. But in the interior of the island in the valleys — a higher temperature (27° — 28° C.) has occasionally been registered. The lowest temperature registered at Berufjord was 23.1°, in the Vestmannaeyjar 20.9; at Stykkisholm 26°, and on Grimsey — 30°. l At the coast in Iceland the average temperature of the day and night rarely exceeds 15° C., and at Berufjord there occur, as a rule, during the year, only four days with such a temperature, but in Modrudalur six to seven; a fact which shows among other things that the summer temperature is often higher in the interior of the country, while the cold there is also greater during the winter, On ascending to the interior districts the climate is found to be no- longer so decidedly oceanic as along the coasts, and the vegetation also increases in density the further one proceeds upwards, following the valleys. Coppice-woods often thrive at the head of valleys in places where birch -coppices cannot grow out at the coast. The mean temperature of the coldest day and night in Modrudalur was 29° C., at Berufjord 19° C. During the period March 6 — 12, 1892, when the cold was very severe, the mean tempera- 1 All these figures are those for the cold, drift-ice winter 1880—1881. But it should be mentioned here that the Grimsey station had no maximum thermo- meter, and that the temperature - 30° C. was one which was registered during the day. In reality the cold had probably been greater, because in the same winter I noted early in the morning of some days, at Modruvellir in Horgardal a tem- perature of • - 32° to — 36° C. PHYSICAL GEOGRAPHY 281 ture in the Vestmannaeyjar was 7° C., at Berufjord ll1/-'0 C., on Grimsey — lo1/?0 C. and in Modrudalur — 24V2° C. The mean temperature for the whole island , according to the observations to hand, is 21/2° C., but the real temperature is un- doubtedly considerably lower, because only one station on the im- mense plateau, was included. The mean temperature of the winter on the north and east coasts ranges between lc and - 31/*0 C., and of the summer between B1/^ and - o oo?o>O5i>toooccososco9OC5Cs oc oc c r- to ro o o «o co o ra ro o o o i-< o r- i-l lO - O >O r-< I> 00 CO LO Oi CO r-< I-H r-4 U J9CIIU3AOISJ '- co co co o500G3 osoooi>ososi>i>i>o«ooscs o o as oo o o o C5 l>- O G5 OO OC OS OO CO C5 OS O O i-< i— i r- i CiS os os « t> o i> >o o 0 os 0 o os cs oo oo o uo I-H co co c^ co co •* co r ^-H i- o o i- I II I I I I I rH»OrH OC CO 00 Meocieorjooeoi>r-i-«-Oi-coci-HTH I I M_ I I I II I Botany of Iceland. I. ( j r i m s e y From V? 1878 to 31/ia 190(>. ^ Mean temperature CM Highest mean temperature Lowest mean temperature . . . Absolute maximum temperature Absolute minimum temperature Mean maximum temperature . Mean minimum temperature. . * THORODDSKX !*[£^ ggS'S'lcwgii si'i'^l^li*^ S*g^S"|cE;gT3 'S^jQc*^ SC"'"""^''^^^ ^C— '^'"^"ssoc"^ C ^"'•-•"'•^SSOO?^* • • CD CD • • • • CD fD • • • CD CS ' • •• i c *- C! < • • S 4^- O O SO tc >— k 1C 1 1C k-k I-k 1C OCi-*C: 4*- »-k tC tC OC "— k i-k k-- tC ^J SO k-k tC i— k in C OS CO 00 O tC Jan. » •M oc in o in i— » c; o tc^iso H-^ii-itcincitc ococ4^-— ' i— k i— k tc in o to tc i— ' tnotcoosootc Feb. fl M •S OS tc O 00 ->J tC O tc o tc tc os tc tc •o so tc ocsotC4^4-**~Jtc in i— k tn so 01 co 4— (1 •• ^^ to i— k tc i— k to i— k oc to i— k 4^ >— k i— k ic so to i— k i— k 4-* c c: ic in i— • inoincotctcto March ft 1- c •• o cs o cs in 4^ cs be 4^^ sobscotO4^4*.in cs cc cc os o *>•*>« tcciocitcscto fl tC i-k 4* i— k tc O -J tc i-k i— k i— i k— k i_k tC i— k April p c o to oo oe I-1 cc to i— '-Jen ••o^ccsotcictc *— * cc tc 4^ 4^ in o i— ' 4^ o tc i— k oo -vj = — 1— k o 4^ to cs o in k-k .-* tO — i-k May M in •— k to o -J >— k -J toso-~j —*ococ4^~ji-kO oooctni-ko^Jtc ostotocoini-kcc « to tc oc cs >-k I-1 oc in i— k to i— k 1C i— k )—L>— i i— t tc *^-l in 4^ ^^ 4-* tn 4-* SO *^1 Oi 1C ^- H^ **J O SO in *"™k jc SO in O OC June s IH f*^ ^TI ^O ^^ *•*.! ^^i ^i ^^ in 4-*- '•^ tc 1C 4^ **J 4-^ tC OC •— k O CO OC tC 1C 1 ^ CO tC SO tC - -3 "— k 1C 1— k 1— k 1— 1 I— I ,_k H-k 1C I— k July m • OC "^ ^^^ t^^ ^j ^"^ ^^^ in in oc so oc oc tc 4^ tc "^j tc i"^ *-*j os )*™k »*j os tc »*j oc so i*^ 4-* so • » 1C __ — — 1C— _._-___. ^ | ^ _^ 4-sc~J in i-» tc tc cs o oo -^Jtcooosctco oiic>-kicotic:o Aug. f • P ^j cs o tc tc oc oc iccso inscsotni-k^itc csoctcinotcio -ootoo4^ctc tc ^a in oc tc oo f 1 1C — — — tc^ici 4^soinO4^0co: inotCOiOisooc 4^. cc 4^ cs tn co -^i Sept. H C H >-k so O tc •-k— k SO 4^ Oi O OCOSC^OIC-J c 1C 4* I-1 O CC CC 1C i-* 1 1C i— k t— k i— k tC i— k CilOO 1C i— k O i— k OS 1C ^ i— k 4* Ci O tC tC i— k 4*- i— k i— ' O -J i— k i— k Dec. - c •* >-k k-* 4» SO O — tc 4- — 4* inso4^in^]i-k~j CnoootoostO tcOk-k^socn^ r 1C O i— k i— k i— k bo *i— ' oo so Winter !• C! I— k o p *•* oo o -' tC CO SO Spring C • os CS OO SO CC ' . • • 4^ i— k SO i— i Summer 1C tc tc in tc 4^ OS 4*- 00 Autumn ! tc tc O CS C tC k-k 1 to tc to tc tc tc Otctc cc cs •-» *k oo o i— ' *- in in c: tc >— ' 4*- tc Average n o tc oc — in •— k^!4- i— k tc O 1C O SO 1C -^ SO i— k CTSOtCOC lor year PHYSICAL GEOGRAPHY 285 sequence. As the rainfall, however, is often accompanied by storms, the weather is frequently very unpleasant for man and beast. It is especially in South Iceland that complaints are heard of the storms of sleet during autumn, when everything becomes saturated with moisture and rain penetrates into the houses through every chink and crevice. The humidity is more considerable in South than in North Iceland, which is shown for instance bv the fact that houses with turf walls */ must be rebuilt more often in the southern districts. A cold, damp, whitish fog, accompanied by sleet, is very characteristic of the northern headlands, especially when the Greenland ice drifts back- wards and forwards along the coast ; then fog may shroud the coast for weeks and extend far into the valleys. The fog persists for a long time also in other places along the coast, especially in East Iceland. The amount of cloud also is generally great, especially in North Iceland.1 The amount of rainfall varies greatly in the different parts of Iceland, and in most places it is generally greatest during autumn and winter; on Grimsey, however, it is greatest during summer and autumn. The precipitation is greatest along the south and south- east coasts: in the Vestmannaeyjar 1320 mm., and at Berufjord 1166 mm. The greatest amount of annual rainfall in the Vestmanna- eyjar was 1587 mm., and at Berufjord 1737 mm. Along the west coast the rainfall is much less, and still less in North Iceland, being at Stvkkisholm 656 mm., but on Grimsev onlv 345 mm. At Stvkkis- I/ V */ •> holm there are, however, on an average 207 rainy days, but on Papey only 135, consequently, a much greater amount of rain falls there at a time. At Berufjord the greatest amount of rainfall within 24 hours was 1091/? mm. (July); at Stvkkisholm 51.9 mm. (January); and on Grimsey 34.3 mm. (November). Heavy showers and sudden torrents of rain do enormous damage to the soil-layer on the moun- tain-sides, which are usually woodless and unprotected, and occasion numerous rock-slips which have often proved very disastrous to property and human life. The frequency of fogs differs greatly at different parts of the coast, but it is most frequent along the southern part of the east coast where the warm and cold ocean currents meet. At Berufjord there are on an average 171 foggy days annually; here fogs are frequent at all the periods of the year, but they occur most frequently during summer. On Papey (in the neighbourhood 1 Amount of cloud (0 — 10): Vestmannaeyjar 6.2; Stvkkisholm 6.7; Berufjord 6.7; and Grimsey 8.3. 19* 286 THORODDSEN Table VII. Amount of Precipitation along the coas January «f February 1 CS S "E a < >> « S i Stykkisholmur From Ve 1878 to SI/i2 1906. Mean precipitation 78 2 65 5 45 0 34 6 38 6 39 4 Highest amount for the month 184 0 217 2 116 6 67 0 132 0 104 8 Lowest amount for the month 17 5 6 0 10 7 4 2 2 0 10 7 Greatest amount of precipitation during 24 hours Number of days with snow 51.9 16 50.1 14 19.2 13 17.6 8 19.8 3 19.8 0 4 fog 0 3 0 5 0 4 0 7 2 0 1 3 frost 26 25 26 19 9 1 storm 6 6 6 4 4 2 precipitation 21 19 18 16 15 15 V e s t m a n n a e y j a r From V? 1877 to 31/i2 1906. Mean precipitation 133 6 108 5 110 2 92 1 77 9 86 1 Highest amount for the month 225 6 195 i 239 0 198 1 149 1 145 6 ' Low'est amount for the month 34 5 ?9 6 26 1 15 0 11 5 30 0 Greatest amount of precipitation during 24 hours. Number of days with snow 41.8 9 54.8 g 62.6 8 33.6 4 46.9 1 34.4 0 2 — fog . 2 2 2 3 5 8 frost 20 18 17 10 3 0 1 storm 4 3 3 2 i 0 5 BerufjorSur From Vis 1872 to 3Vi2 1906. Mean precipitation 122 8 95 6 76 1 88 0 70 8 71 3 Highest amount for the month 258 7 214 9 231 6 222 9 237 6 168 2 1 ' Lowest amount for the month 26 6 4 6 2 6 15 0 0 0 17 9 Greatest amount of precipitation during 24 hours . Number of days with snow 57.2 8 59.0 8 62.6 9 44.8 5 52.6 3 108.1 04 : — fog . 11 11 11 14 17 19 "*O frost 24 22 25 18 10 2 P a p e y From Vs 1873 to »Vw 1906. Number of days with precipitation 14 13 12 12 9 8 snow 7 7 7 5 3 0 5 — fog 9 9 10 12 13 16 o G r i m s e y From V? 1873 to 3I/» 1906. Mean precipitation 21 5 22 2 22 5 159 19 3 21 8 Highest amount for the month ... 58 0 112 0 73 8 37 5 98 7 59.0 • Lowest amount for the month 1.2 0 6 2 4 1 2 3 0 3.2 . Greatest amount of precipitation during 24 hours. Number of days with snow 16.5 7 20.4 8 13.8 8 16.1 7 28.1 5 21.4 2 fog. 1 1 2 4 6 7 frost 25 24 27 24 17 5 storm 2 1 1 0 3 0 2 0.4 ^ PHYSICAL GEOGRAPHY 287 of Iceland in millimetres for the Period 1872—1906. >> j= -M X f4 _« 3f 7 < September October November December , Winter t£ c °£ C, C/3 Summer Autumn 0, i. II a, Arme- ria maritima, Ranunculus glacialis, Silene acaulis, all in flower July 22, 1886. Vattarfjall (401 metres), Papaver midicaule, Silene acaulis, Armeria maritima. Ba3Jarfell near Steingrimsfjord (316 metres), Silene acaulis, Armeria maritima, Cerastium alpinum, Polygonum vivi- parum, Empetrum nigrum, Saxifraga ccespitosa. Drangahals (315 metres), Salix herbacea, Polygonum viviparum, Oxyria digyna, Rho- diola rosea. Dalsheidi near Snaefjallastrond (660 metres), mosses, Silene acaulis, Salix herbacea, Gnaphalium supinum, Sibbaldia pro- cumbens. Klofningsheidi (606 metres), Grimmias, Salix herbacea, Ranunculus glacialis, R. pygmwus. Uxaskard near Latrar (Eyjafjord) (518 metres), Oxyria digyna, Salix herbacea, Armeria maritima, Ra- nunculus glacialis. Hafursey near Myrdalsjokull (587 metres), Grim- mias, Salix herbacea, Silene acaulis, Polygonum viviparum. Lod- mundur (1097 metres), NE. of Hekla, Grimmias, lichens, Salix herbacea, Silene acaulis, Cerastium alpinum, Saxifraga c&spitosa, Ca- rex rigida. Skalafell near 01fus (578 metres), Salix herbacea, Cassiope hypnoides, Armeria maritima. Arabis petrwa. On the plateau towards the west: Thristapafell near Eiriksjokull (714 metres), Salix glauca, Armeria maritima, Cerastium alpinum, Silene acaulis, Salix herbacea. Strutur (921 metres), on the summit no plants, but 130 metres further downwards, Arabis alpina, A. petrcea, Saxifraga hirculus, S. hypnoides, S. oppositifolia, Pedicularis flammea. Botnssulur (1108 metres), only some lichens and Grimmias and Salix herbacea and / " «, 306 THORODDSEN Empetrum nigrum - a miserable stunted specimen of each of these two species, 2 and 2l/2 cm. in height respectively. On the plateau towards the east: Kjarrdalsheidi near Lon (665 metres), Salix herbacea, Polygonum viviparum, Oxyna digyna, Ranunculus glacialis, all small and stunted. Markalda (961 metres) near the eastern edge of Vatnajokull, Polygonum viviparum, Armeria maritima, Salix herbacea, Saxifraga nivalis. Litla Snaefell (1133 metres), Salix her- bacea, Oxyria digyna, Arabis alpina, Ranunculus glacialis. On Hlid- arfjall near Myvatn (790 metres) where I was together with Gron- lund in 1876, we noted the following plants: - - Alsine biflora, Draba nivalis, Saxifraga cernua, Cassiope hypnoides, Pedicularis flammea, Oxyria digyna1. All that is situated outside the glacier-bearing mountains in the centre of Iceland at an altitude of 650 — 1100 metres may justly be regarded as a desert; seen both from a geological and geographical point of view the country here is desert-like in character, and in spite of considerable precipitation the plants suffer from drought, because the water disappears immediately over large areas owing to the porous nature of the rocky substratum - lava, tuff, volcanic gravel and sand. The few plants which grow in these wastes occur widely scattered; at a height of 900 — 1000 metres above sea-level a few7 lichens and mosses are seen only here and there, and at long intervals a few specimens of the hardy Armeria maritima, Silene maritima and Polygonum viviparum and in places where blown sand occurs a few7 tufts of Elymus arenarius; at a height of 1000 — 1100 metres one may ride for miles without coming across a single pha- nerogam. In addition to want of water, the frequent storms, often of sand and snow, check plant-growth during the short summer; moreover, a rather dry Form wind often blows across the wastes •/ north of the great Jokulls, having already deposited its moisture upon the great plateaus of the Jokulls. Somewhat lower down, at an altitude of 700 — 900 metres, a few more species are met with as scattered individuals, e. g. Silene acaulis, Arabis alpina and A. pelrcea, and here and there a few haulms of Luzula arcuala, Poa glauca or Fesluca ovina\ moreover, a few scattered cushions or tufts of lichens (Stereocaulon) or mosses (Grimmia) occur. Scattered here and there in these extensive wastes are sometimes 1 See also Chr. Gronlund's list of plants from Heljardalsheidi and Hrafn- tinnuhryggur. and the list of plants collected by Johnstrup on Dyngjufjoll (Karak- teristik af Plante-v;eksten paa Island, 1884, pp. 28 and 29). PHYSICAL GEOGRAPHY 307 found, at intervals of a whole day's journey, some small oases with a denser vegetation, in places where water is present, especially springs which issue from under the edge of the lava-streams. I propose to give here the names of some of these oases and their height above sea-level, and add some notes on their vegetation, which is generally very little known. As almost no notes are published on the vegetation of the plateau I shall avail myself of this oppor- tunity to give a list of the species I collected on some of these oases, as they have never been visited by botanists. Some of these plant-covered patches are only a few square metres in extent, and the largest of them are perhaps as much as a square km. in area1. These oases often originate around hot springs, the temperature of which need not be very high somewhat higher, however, than the mean temperature of the locality at which they rise. Thus NW. of Vatnajokull there are oases near Gaesavotn, 929 metres above sea-level, where the springs have a temperature of 5 — 7°C. , and near Marteinsflaeda (744 metres) with spring-temperature of 35V2° C. and near Hitalaug (672 metres) with a temperature of 33V20 C. Gaesavotn is the most highly situated of all the oases that I visited in 1884; plant -growth occurs here in connection with pools and springs; along the margins the vegetation is formed by mosses and Salix herbacea with scattered specimens of Polygonum viviparum, Saxifraga stellaris, Oxyria digyna, Armeria maritima, Cerastinm and Poa, while Carices and Eriophorum form a fringe along the water's edge. Here I collected the following plants: Equisetum arvense var. alpestre, Calamagrostis stricta var. borealis, Aira alpina, Poa pra- tensis, P. pratensis var. alpigena, Poa alpina var. vivipara, Carex in- curva, Eriophorum Scheuchzeri, E. angusti folium, Salix phylicifolia, S. herbacea, Polygonnm viviparum, Oxyria digyna f. pygm&a, Ar- meria sibirica, Saxifraga stellaris f. pygmcea, S. decipiens var. gron- landica, Ranunculus hyperboreus, Cerastium arcticum, C. trigynum2. From here I went to Hvannalindir near Kverkfjoll, the nearest oasis towards the east, but it took nevertheless two days (17, 18 Aug.) to reach it, travelling along the northern edge of Vatnajokull; along the whole of this stretch of land there were almost no plants, until at Jokulsa, 734 metres above sea-level, I found in the gravel a fairly large quantity of Cham&nerium latifolium and Oxyria digyna. 1 In my work "Island, Grundriss d. Geographic u. Geologic", 1906. there is a map showing the desert-boundaries and the oases known. 2 Determined by Prof. Joh. Lange. 308 THORODDSEN In Hvannalindir (656 metres) there occur along small streams a great many grasses and an abundance of tall specimens of Arch- angelica officinalis, but unfortunately I was prevented by a snow- storm from collecting plants there. At Vadalda at the springs of Svarta (674 metres) were found Archangelica officinalis and Juncus balticus. At Herdubreidarlindir (471 metres), north of the lofty mountain of Herdubreid, a rather rich vegetation occurs as a border along springs and branching streams while the surroundings consist of gravel quite bare of plant-life. Here I collected the following plants1: Equisetum uariegatum, Phleiim alpinum, Calamagrostis stricta var. borealis, Festuca rubra var. hirsuta, Eriophorum Scheuch- zeri, Juncus arcticus, J. triglumis, Luzula multiflora, Tofieldia borealis, Platanthera hyperborea var. major, Salix phylicifolia var. angasti- folia, S. lanata, S. herbacea, Achillea millefolium , Erigeron uuiflorus, Hieracium murorum, Galiiim verum var. aspera, Thymus serpyllum var. prostratus, Bartsia alpina, Pirola minor, Archangelica officinalis, Parnassia palusiris, Cerastmm uulgatum, Silene maritima, Epilobium alsinefolium , Chamcenerium latifolium, and Alchimilla alpina. On Odadahraun itself there are otherwise only extremely small plant- covered patches. In Hrutsrandir near Kollotta Dyngja (653 metres), there occurred on a bottom of Salix herbacea a few small individuals of Polygonnm vwiparum and Salix glauca and far out on a neigh- bouring lava-stream a single specimen of Taraxacum officinale — the only one for miles round. Along some mountain-streams in the south-eastern corner of Dyngjufjoll small patches of a similar vegetation were found, only Salix glauca was far more vigorous here. Otherwise no continuous oases are found in the higher parts of Odadahraun, but only a few very widely separated desert plants, and in some places Elymus arenarius on blown sand. Towards Bardardalur the vegetation increases gradually, and the sandy tracts, as far upwards as 500 metres above sea-level, are covered with Elymus arenarius, Salix lanata and S. glauca, and further down- wards till about 450 metres, Achillea millefolium grows in abundance on the sand. In Miklimor between Sudurarbotnar and Alftakill (about 450 metres) there is a considerable vegetation of different kinds of plants and even around Svartarvatn (409 metres) there is a rather rich vegetation like that on the sandy tracts of the lowlands. On the north-eastern part of Vatnajokull east of Jokulsa and close to the glaciers, oases occur separated by stretches of stony 1 Determined by Prof. Joh. Lange. PHYSICAL GEOGRAPHY 309 deserts which are as yet very little known, e. g. Thorlaksmyrar, Mariutungur and Eyjabakkar (672 metres). I visited the last lo- cality in 1894 and found there extensive areas covered with Cype- racese, cotton-grass and other plant-growth. In Fjallasveit, east of Jokulsa, a parish situated up on the plateau at an altitude of 400- 500 metres, the population of which is solely dependent on sheep- rearing for their sustenance, very large tracts are covered with blown sand which in many places is densely overgrown with Elymus arenarius, Salix lanata, S. glauca and Juncus balticus, while Carex incnrva grows abundantly on damp sandy flats. Near the farm- stead Vidirholl (415 metres) I collected in 1895 the following plants: - Junciis balticus, J. triglumis, J. trifidus, Elyna Bellardi, Carex incurva, C. capitata, C. capillaris, C.vulgaris, C.rigida, C.rariflora, Poa pra- tensis, P. annua, Phlenm alpinnm, Festuca rubra var. arenaria, Cala- magrostis stricta, Trisetum snbspicatum , Selaginella spinulosa, Salix lanata, S. glauca, Saxifraga aizoides, Gentiana tenella, Pleurogyne rotata. North of Hofsjokull, at an altitude of 600 — 800 metres, similar barren wastes occur, as around Odadahraun, consisting of ice-stri- ated doleritic lava, and here also the individual plants occur widely separated, but, in the neighbourhood of stream and lakes, the oases are larger both in number and size. The vegetation of these oases is generally confined to small swamps and pools and sometimes to rather extensive mountain-bogs; in the pool-vegetation Eriophorum is usually dominant, but sometimes Carices are in the majority. In mountain bogs numerous large knolls (dys) often occur which are usually dry at the top and covered with mosses and various rocky- flat plants, but wet below and overgrown with swamp-plants. Dry tracts between the swamps are generally covered with Grimmia- heaths wrhich sometimes pass into lichen-heaths. Of the oases oc- curring in these districts may be mentioned: Sydri Pollar, Nyrdri Pollar, Geldingaa and Laugafell (with hot springs), and southwards near Sprengisandur, on the eastern side there is Nyidalur, and on the western side Nauthagi and Arnarfell with an unusually luxuri- ant vegetation which has been described by St. Stefansson1, who also describes the plant-formations near Laugafell and Geldingaa. In Nydri Pollar (704 metres) I collected in 1896 the following plants: Poa alpina, P. flexuosa, Carex rigida, Eriophorum angusti folium, Salix lanata, S. glauca, S. herbacea, Pedicularis flammea, Tofteldia 1 Geografisk Tidsskrift, XVI, 1902. pp.230 and 231. 310 THORODDSEN borealis, Alsine biflora, Cerastium trigynum, Saxifraga Hirculus1. In Sydri Pol la r (729 metres) I collected in the same year the fol- lowing plants: Equisetum variegatiim, E. aruense, Carex rigida, C. lagopina, C. rariflora, C. pulla, Juncus arcticus, J. biglumis, Erio- phorum Scheuchzeri, E. angustifolium, Poa alpina, P. flexuosa, Phleum alpinum, Aira alpina, Festuca rubra var. arenaria, Calamagrostis stricta L borealis, Salix glauca, S. lanata, S.herbacea, S. phylici folia , Betula nana, Polygonum viviparum, Rumex acetosa, Arenaria ciliata, Cera- stium alpinum, Silene acaulis, Cardamine pratensis, Tofteldia borealis, Montia rivularis, Thalictrum alpinum, Pinguicula vulgaris, Armeria maritima, Saxifraga decipiens, S. Hirculus, Pedicularis flammea, Eri- geron uniflorus, Thymus serpyllum var. prostrata, Veronica alpina, Dryas octopetala, Vaccinium uliginosum, Empetrum nigrum, Barfsia alpina2. As already mentioned, on Tvidsegra there are extensive stretches of bog-land and a considerable vegetation of various spe- cies. In 1898 I collected here, near Ulfsvatn (453 metres), the fol- lowing plants: Juncus biglumis, J.triglumis, J. tri/idus, Luzula spicata, Eriophorum angustifolium, E. Scheuchzeri, Carex alpina, C. atrata, C. dioeca, C. lagopina, C. pulla, C. rigida, C. sparsiflora, Aira ccespitosa, A. flexuosa, Anthoxanthum odoratum, Festuca ovina, F. rubra, Phleum alpinum, Poa alpina. P. glauca, Trisetum subspicatum, Coelo- glossum viride, Salix glauca, S. herbacea, Polygonum viviparum, Alsine verna, Cerastium alpinum, C. trigynum, Silene acaulis, Viscaria alpina, Ranunculus acer, Thalictrum alpinum, Cardamine pratensis, Draba incana, Geranium silvaticum, Saxifraga ccespitosa, S. Hircnlus, S. stel- laris, Potentilla verna, Alchimilla vulgaris f. filicaulis, Sibbaldia pro- bens, Vaccinium uliginosum, Bartsia alpina, Pinguicula vulgaris, Thy- mus serpyllum var. prostrata, Gentiana nivalis, G. tenella, Galium sil- vestre, Hieracium alpinum, Taraxacum Iwvigatum. In the same year I collected near Hval vatn (378 metres) the following species: — Juncus triglumis, J. trifidus, Eriophorum angustifolium, Carex chordorrhiza, C. dioeca, C. Goodenoughii, C. lagopina, C. pulla, C. rariflora, C. rigida, Aira cwspitosa, A. flexuosa, Anthoxanthum odoratum, Festuca ovina, Phleum alpinum, Poa glauca, Tofieldia borealis, Loiseleuria procumbens. South of the great Jokulls the vegetation generally extends fur- ther upwards on the plateau than to the north of them, but here also there are vast deserts of gravel, lava and blown sand, with a few widely separated oases, but it is especially the tracts between 1 Determined by O. Gelert. Determined bv O. Gelert. PHYSICAL GEOGRAPHY 311 Thjorsa, Skafta and Myrdalsjokull which are desert-like in character, and here there are also several bare tun-ridges with numerous peaks. The sand is constantly drifting through the valleys and destroying all vegetation ; only upon the highest ridges and peaks, which cannot be reached by the coarser grains of the drifting sand are seen se- veral yellowish-green patches of mosses, and also along streams and around springs the moss-vegetation is sometimes fairly luxuriant and forms rather large green patches in places where very few or no phanerogams have been able to gain a foothold. Around Tjaldvatn (588 metres) in Veidivotn there is a considerable vegetation of differ- ent species; in 1889 I collected here the following plants: Carex ri- gida, Poa pratensis, Festuca ouina, Salix glaiica, Montia rwularis, Ranunculus acer, Batrachium paucistamineum , Thalidrum alpinum, Koenigia islandica, Empetrum nig rum, Rhodiola rosea, Parnassia pa- lustris, Chamcenermm latifolium, Hippuris vulgaris, Euphrasia latifolia. Near Thorisvatn (591 metres) which is situated in the centre of large sandy deserts quite bare of vegetation, I found only Salix glaiica, Chamcenerium latifolium and Carex rariflora. On Blesamyri (535 metres) near Tindfjallajokull I found Carex rigida and C. rari- flora. Near Hitalaug (650 metres) east of Torfajokull, in the neigh- bourhood of hot springs I found Coeloglossum viride, Saxifraga stel- laris, Sibbaldia procumbens, Pirola minor, Vaccinium uliginosum, Ve- ronica alpina, Gnaphalium Noruegicum, G. supinum, Hieracium alpi- num. Near Hvannabotnar (434 metres) in the neighbourhood of Skafta I collected: Equisetum palustre, Luzula campestris, Carex ri- gida, Calamagrostis stricla f. borealis, Anthoxanthum odoratum, Sib- baldia procumbens, Epilobium lactiflorum, E. Hornemanni, Gnapha- lium Norvegicum. Though the above list is naturally very incomplete owing to the author having had other work in hand (geographical survey and geological investigations) which left him no time for tho- rough botanical investigations or collections, yet these notes have been included here as these parts of the plateau are very difficult of access and are hardly ever visited by naturalists. Outside the deserts in central Iceland, and nearer to the sea, there are also many, large and small, high-situated rocky areas and broken groups of rocks, pieces of plateau, and isolated peaks in the numerous mountain-spurs which extend between the branching val- leys and fjords. The vegetation of these rocks is also very little known, but it resembles very much that of the plateau, only, it is usually richer in species. Highest up on the mountains, at an alti- 312 THORODDSEN lude of 600 — 800 metres, there are usually gravelly tracts with scat- tered individuals of rocky-flat plants, such as Silene acaulis, S. ma- ritima, Cerastinm alpinum, Luzula arcuata, Polygonum viviparum, Armeria maritima, Ranunculus glacialis, Saxifraga nivalis, S. oppo- sitifolia and others. In small damp hollows where the snow7 persists for a long time there is often a characteristic dense growth of Salix herbacea, which almost entirely conceals the moss-covering of the ground, together with Sibbaldia procumbens, Gnaphalium supinum, Oxyria digyna and Polygonum viviparum; in some places these are associated wTith several other plants. In other places there are small patches of Grimmia-heaths with scattered specimens of Pedicularis flammea or Cassiope hypnoides. Here and there streams and bogs occur writh Carex rigida, C. lagopina, C. rostrata, C. incurva and others, but most often with Eriophorum angustifolium and E. Scheuchzeri. As has been seen from the preceding notes on the vegetation of the plateau it is not easy to determine the altitudes or upper limits of the different plant-regions. Of the Scandinavian upper zones, the region of conifers is entirely absent, but, on the other hand, wre may be justified in speaking of a birch-region, of an osier-willow region, and perhaps a lichen- or moss-region, but these regions pass into one another in many ways, and overlap. During the period after the Ice Age (the Purpura-lapillus Period) when it was warmer than it is now, the birchigrew everywhere in the lowlands even on the northernmost headlands, but it had already retired from the latter at the time the first settlers came to the island, and since then, as we have already seen, its distribution has been considerably limited owing to the interference of man and sheep. The present Polar limit of Betula odorata in Iceland has not been fully investi- */ gated, but judging from what I sawr on my journeys it appears, on the east coast, to extend across Vopnafjordur to the wrest coast of Melrakkasljetta, across Axarfjordur and Skjalfandi to Eyjafjordur near the mouth of the valley of Fnjoskadalur. The stretch of land between Eyjafjordur and Hunafloi is now devoid of birch coppices, although these occurred there in olden times; how far out they ex- tended at that time upon the peninsulas between the fjords is not known. From Hunafloi the northern limit of the birch extends from Steingrimsfjordur to Isafjardardjup. This is, however, only quite a provisional limit; the subject requires to be investigated more closely. As regards the upper limit of the birch, it differs considerably in different parts of the island; it extends highest in Thingeyjarsysla PHYSICAL GEOGRAPHY 313 near Myvatn to 550 metres. Birch coppices occur on the moun- tains of Vindbelgur (540 metres) and Dalfjall (550 metres). In olden times there were in Kroksdalur and Yxnadalur a continuation of Bardardalur considerable woods up to 450 metres. On the south coast the birch coppices do not as a rule extend so far up- wards on the mountains; in Kollumuli near Lon birch coppices oc- Fig. 27. Woods in Hallormstad ; June 1909. (Phot. A. Hesselbo.) cur, however, at a height of about 500 metres, but, nowhere else along the south coast did I find birch coppices at that altitude. The forest of Nupsstadaskogur, south of Vatnajokull, reaches to an altitude of 400 metres, Bsejarstadaskogur in Orsefi to 320 metres, and coppices in Haukadalsheidi to 380 metres, but in the southern lowlands the limit for birch coppices usually occurs at a height of 200—300 metres; in some places at a slightly higher, in others at a slightly lower level. According to the Ordnance map the upper limits of the coppice-woods in South Iceland are as follows: Skaftafell in Orsefi 280 metres Jokulfell (Bcejarstadaskogur) 320 — Nupsstadaskogur 400 Thorsmork 320 — 314 THORODDSEN Markhlidar 250 metres Burfell near Thjorsardal 300 Skridufellsskogur 220 — Tungufellsskogur 260 — Haukadalsheidi 380 — Uthlidarhraun 320 — Laugardalsskogar 280 Thingvallaskogar 200 Botnsskogui 200 — Skorradalur 150 Svinadalur 200 — Husafellsskogur 260 Hvitarsida 300 — Although it is possible that birch coppices may be met with in some places at a somewhat higher level yet the limits will be approximately those given above. On the north-western pe- ninsula, birch coppices occur mostly on the southern side at the head of the valleys and fjords which extend upwards from Breidifjordur and face the sun ; but the birch coppices extend hardly anywhere higher than 200 — 300 metres, and usually occur at far lower levels. There is also a good deal of birch coppice at the branch-fjords of Arnarfjord, in Dyrafjord and at the southern fjords of Isafjardardjup, especially at Hestfjordur. North of Isafjardardjup I nowhere saw birch coppices proper, although a few individuals of Betula odorata occur in some places in Adalvik; nor are birch cop- pices known to occur on the east coast from Cape Nord to Stein- grim sfjord. To the birch region belongs also Sorbus aucuparia which occurs as scattered individuals both in the birch coppices and outside them; I do not think the mountain ash extends so far up as does the birch ; I did not observe it at higher altitudes than on Sluttnes in Myvatn, 290 metres above sea-level. Betula nana occurs now and then in birch woods, but grows most commonly on heather moors and in bog-lands; it rarely forms coppices proper. It extends higher on the mountains than Betula odorata; I found it, for instance, in Sydri Pollar, 729 metres above sea-level. Jnniperus commnnis is fairly common in birch coppices and on heather moors; the highest altitude at which I found it was in Yxnadalur near Odadahraun, 488 metres above sea-level. The heather moor is closely associated PHYSICAL GEOGRAPHY IH5 with the birch region , and it probably rarely extends higher than 300—400 metres above sea-level as an aggregate plant-formation ; but several of the different species of which the heather-moor is com- posed extend far higher up on the mountains without forming any heath-like associations, occurring as a few individuals only. 1 found Vaccininm uliginosum at the highest level (729 metres) near Sydri Pollar, and also Dryas octopetala at the same place; Empetrnm ni- grnm extends to a similar height, and I came across a stunted spe- cimen even on the top of Botnssulur at an altitude of 1108 metres. Cassiope hypnoides extends to a great height also; it is found, among other places, on Hlidarfjall at 790 metres above sea-level, and may perhaps extend even higher. Loiseleufia procumbens, on the other hand, did not occur at a higher level than 400 — 500 metres. Cal- luna viilgaris, Arctostaphylus uva ursi and Vaccinum Myrtillns were found near Myvatn at an altitude of about 400 metres, but I do not think they extend higher than the limits of the heather moor. Above the upper limit of the birch region an osier or a wil- low region may be said to occur — in the centre of the country at an altitude of 500 — 800 metres, in other places somewhat lower - where willows are dominant among the woody plants, although they have their greatest distribution at a far lower level, in the birch re- gion itself. At this altitude they do not form any coppice proper, but occur as flat expanses of low prostrate shrubs. Above the birch region it is especially Salix herbacea, S. lanata and S. glauca which are the dominants; Salix phylicifolia retires, although small speci- mens of the latter also are now and then met with even at this al- titude. Salix lanata, and to some extent S. glauca, occupy large areas of the lower parts of the plateau, e. g. in Fjallasveit, Myvatn- soraefi, Sudurarbotnar, and several other places; they are of great importance to sheep-breeding, and in some places quantities of wil- low leaves are gathered as winter- fodder for sheep and cattle; in the lowest part of this zone Betula nana occurs now and then. In the centre of the country above 800 — 900 metres and up to the snow-line at an altitude of 1000 — 1400 metres, mosses and lichens are undoubtedly the dominant plants, although a few widely scat- tered phanerogams also occur. Salix herbacea extends also through this region to the snow-line ; here and there, the most hardy of the previously mentioned rocky-flat plants occur as somewhat scattered individuals, but the main vegetation consists of mosses, although these do not occur in any great abundance compared with the vast The Botany of Iceland. I. 21 316 THORODDSEN areas. Mosses often form an edging or fringe around loose stones on gravelly flats or also cushions on heaps of stones or in rock- crevices. On the highest situated and most inhospitable rocky and gravelly tracts, where storms are constantly driving sand and gravel across the surface, not even mosses can thrive, and the ground is quite bare. A few species of lichens (Cladonia and Cetraria) are some- times found in the moss cushions, but usually they occur only as crusts on blocks of rock. Here and there on clayey soil in depres- sions crusts of liverworts occur, especially Anthelia nivalis, often associated with Grimmia hypnoides and Salix herbacea: round pools and along streams, cushions of bright green mosses are often seen, in which some flowering plants have sometimes found a home. In other parts of the country, the three regions mentioned here occur at a lower level, since they are correlated with the altitude of the snow-line in the different parts of the country. V. A SKETCH OF THE CHIEF PLANT-FORMATIONS OF ICELAND. WE will now give a brief account of the commonest and, from a geographical point of view, most important plant-formations of the land-vegetation of Iceland, without, however, entering into details or into local deviations of the different plant-associations. This account is based partly on my own observations and partly on those published by others; I have especially made use of Dr. Helgi Jonsson's numerous excellent descriptions of the vegeta- tion-forms of Iceland. Here I shall confine myself to the purely geographical distribution without entering more closely into rela- tions of causation, or into questions of general ecology, which will no doubt be exhaustively discussed later on. The division into plant-formations and -associations is as yet, in many respects, de- pendent upon individual opinion. In the following, with the ex- ception of a few deviations, I am adopting Dr. H. Jonsson's main divisions. Dr. Jonsson is the most experienced investigator of the vegetation of Iceland, and has described the plant-life from many more parts of Iceland than has anyone else. But much work re- mains to be done by future investigators ; the vegetation-conditions from more than one-half the total area of the island are still to be described, and so manv local variations, and such varied associa- •/ tions occur within the larger formations that it will be a long time before the details from everywhere are well-known. The following is only a brief account for the general orientation of the larger plant-formations and of the chief points regarding the distribution of the higher plants according to present knowledge1. 1 Besides the above-mentioned works on the flora of Iceland, there are the following descriptions of the vegetation: Chr. Gronlund: Karakteristik of Plantevsexten paa Island, sammenlignet med Floraen i Here andre Lande (Natur- historisk Forenings Festskrift, 1890 [printed 1884], pp. 107— 145 . C. H. Ostenfeld: Skildringer af Vegetationen paa Island (Botanisk Tidsskrift, XXII, 1899, pp. 227- 21* 318 THORODDSEN The Vegetation of the Coast-line. The halophilous plant- associations along the greatly extended coast-line of Iceland have, as in other places, a rather heterogeneous soil consisting of fine and coarse strand sand, large pebbles or boulders, blown sand and rocks of basalt and tuff. Very generally round the coast there is seen upon low rocks1, at the foot, a belt of Verrncaria maura; then come grey, yellow and green patches of several kinds of li- chens; above this level only a few plants occur, mostly scattered individuals of Cochlearia offlcinalis, Plantago maritima, Armeria ma- ritima and Glyceria distans. On steep, lofty coast- cliffs Cochlearia officinalis and Rhodiola rosea oflen occur in great abundance, also Silene maritima, Armeria maritima , Cerastinm alpinum and various species of Poa and Festuca ; to these should be added Haloscias sco- licnm in south-western Iceland and especially on the islands in Breidifjordur. As is well-known, there are several large sea-fowl cliffs along the coast of Iceland, but their vegetation has not yet been investigated; H. Jonsson has investigated only a few smaller sea-fowl cliffs in Dalasyssel and "sea-fowl-grass-slopes" (Fuglegraeslier) in Skaftafellssyssel. According to H. Jonsson (1905, p. 37) the grass- covered mountain-slopes in South Iceland frequented by sea-fowl differ from the common grass-slopes, among other things in the abundant occurrence of Poa pratensis and SteUaria media; the oc- currence of Festuca elatior and Avena elatior on "sea-fowTl-grass-slopes" is also characteristic of the latter. The vegetation of the lofty sea- fowl cliffs appears principally to consist of the same plants as are found on common coast-cliffs, for instance, Cochlearia, Rhodiola, Archangelica, as also Oxyria and Stellaria; all growing luxuriantly. Owing to the soil being manured there is an immigration of many other species, especially from the grass-slope and the grassland; these species grow as luxuriantly in rock crevices and on ledges as in the most well-manured home- fields. Collections of plants from these sea-fowl cliffs have never been made; the plants being extremelv difficult of access. ^ 253; XXVII, 1905. pp. 111 — 122). Th. Thoroddsen: Planteverdenen paa Island (Sal- monsens Leksikon. IX, 1899, pp. 606 607). Helgi Jonsson: Studier over Ost-Is- lands Vegetation (Bot. Tidsskr., XX, pp. 77—89). Vegetationen paa Snaefellsnes (Vi- densk. Medd. fra naturh. Foren. i Kobenhavn, 1900, pp. 15—97). Vegetationen i Syd- Island (Bot. Tidsskr.. XXVII, 1905, pp. 1 — 82) Vegetationen paa Island (Atlanten, 1904, pp. 41— 50). Gro9rar- og JarSvegsrannsoknir (Bunagarrit XX, 1906, pp. 146- 181 ; XXIII. 1909, pp. 41—54). Fyrirlestur urn groSur Islands (BtinaSarrit, XXI, 1907, pp. 6— 20). Bygging og Iff plantna, Kebenhavn, 1907, pp. 289— 300. 1 Cf. Eug. Warming: Dansk Planteva-kst, I, K0benhavn, 1906. PHYSICAL GEOGRAPHY 319 On the strand sand, especially in South Iceland, there is a fairly characteristic vegetation of Halianthus pephides, Cakile maritima, Atriplex patula, Stenhammaria maritima and Potentilla anserina; often, each of these species occurs separately and in abundance, but sometimes they are found intermixed in various ways. Cakile ma- ritima usually occurs nearest to the sea, and Potentilla anserina at the highest levels, occupying large areas, and occurring so abun- dantly that the ground is quite interwoven by its creeping shoots. On the strand sand are also seen in small numbers Cochlearia offici- nalis, Matricaria inodora, Silene maritima, Polygonnm aviculare, Stel- laria crassifolia, S. media, Capsella bursa pastoris, Glyccria distans, Festuca rubra, Elymus arenarius, Carex incurva, etc. On the vast sandy stretches along the south coast blown-sand formations are frequent nearest to the sea; here Halianthus peploides and Elymus arenarius occur in tufts; above this belt of sand dunes, tracts are found occupied by a vegetation richer in species and consisting of Potentilla anserina, Festuca rnbra var. arenaria, Thymus serpyllum, Galium uerum, Achillea millefolium, Juncus balticus, and several others. But below the glacier-bearing mountains (Jokulls) the strand sand quickh^ merges into glacier sand, which forms sandy wastes extremely poor in plant-life. At several places along the coast of Iceland salt-marshes are found, overflowed by salt wrater, for example in Borgarfjordur, Myrar, Hornafjordur and Lon. In Myrar the dominant species are Glyceria maritima, Agrostis alba, Plantago maritima, Stellaria crassifolia, more- over Heleocharis uniglumis, Triglochin maritima, Juncus bufonius, several species of Carex, etc. H. Jonsson found that in some places two-thirds of the area \vas occupied by Glyceria maritima and one-third by Agrostis alba, each species occurred separately in patches; some thick-leaved Plantago maritima occurred, however, in the Gly- ceria patches1. The Vegetation of the Fresh Water. Plant-life occurs very sparingly in running water, and where strong currents are felt it is usually absent. Nor do plants appear to thrive in glacier-rivers; this is probably due to the low temperature of the water and the current. In slowly flowing river-arms on level land, in rivulets and brooks there is often a considerable quantity of green algas (Zygnema and Spirogyra), both at the bottom and upon the surface; in South Iceland Enteromorpha intestinalis is common in streams. The fol- 1 Bunadarrit. XX. 1906. pp.150, 151. 320 THORODDSEN lowing mosses are found on stones in rivulets and brooks: Fonti- nalis antipyretica very frequently, also F. gracilis and F. thulensis, Amblystegium Kneiffii, A. ochraceum, Hypnum rusciforme var. altantica, and others1. The vegetation of lakes and pools is much richer and differs considerably according to the depth of the water and the nature of the bottom, etc. The plankton of the Icelandic lakes has as yet been verv little investigated: there are only a few notes K/ «j ^/ to hand from Myvatn and Thingvallavatn. In Myvatn zooplankton only was found; Thingvallavatn contained phytoplankton in which diatoms were dominant2. In deep lakes there is usually very little or no vegetation at greater depths, only in places where it is shal- lower does plant-life occur. In Thingvallavatn, however, there are large areas covered with Chara and Nitella, especially at a depth of 13 — 30 metres, and they extend even down to 38 metres3. Where ti the lakes are shallower various species of Potamogeton and Myrio- phyllum and also Batrachium paucistamineum are common. Near the margin and in smaller pools the most common, and usually dominant, species are the following: Heleocharis palustris, Equisetum limosum , Carex rostrata, Menyanthes trifoliata, and in the southern lowlands Glyceria fluitans is common; to these should be added Hippuris uulgaris, Eriophorum angustifolium, Sparganinm minimum, S. submuticum , Ranunculus hyperboreus and /?. reptans, Subularia aquatica, Callitriche hamulata and C. uerna, Limosella aquatica and several others, the occurrence of which varies somewhat according to the quantity of the water, the conditions at the bottom, etc. Where Equisetum limosum occurs in abundance it is cut annually and used for fodder for milch cows. In Myvaln and in other lakes in Thingeyjarsysla iVos/oc-lumps are found in abundance, often thrown up on the shore in very great quantities. Sometimes in warm sum- mers large areas of Myvatn become turbid; this phenomenon is known by the inhabitants as "leirlos" - it is said to be very injurious to salmon-trout, their gills becoming filled with the fine particles when this occurs they retreat in great numbers to the eastern shore of the lake, where the water is purer and clearer owing to the numer- ous springs which here issue from the lava4. 1 H.Jonsson. 1900, p. 17, 1905, p. 7. 2 C. H. Ostenfeld and C. Wesenberg-Lund: A Regular Fortnightly Explor- ation of the Plankton of the two Icelandic Lakes, Thingvallavatn and Myvatn. (Proc. R. Society Edinburgh, Vol.25. Part 12, 1906, pp. 1092— 1167). B. Sjemundsson in Andvari, 1904, p. 89. Dr. H.Jonsson informs me that the so-called "leirlos'1 is probably due to Blue-green Algae perhaps Aphanizomenon /Jos cifjuce. PHYSICAL GEOGRAPHY 321 Springs (kaldavermsl, dy1) are very common in Iceland, and are found in almost every valley below the mountain-sides or appear in rows upon the ledges of the basalt-layers. The peculiar vegeta- tion connected with these streams is widely distributed below7 the mountain-sides and often occurs in the midst of other formations or high up on the mountain-sides in narrow zones in places where other vegetation is wanting. This vegetation is characterized espe- cially by the fresh bright-green colour of the mosses; Philonotis f'on- lana is everywhere the dominant plant, but many other species of moss also occur, especially Pohlia albicaus var. glacialis, and various species of Brynm, Amblystegium, and several other genera. Several phanerogams occur among the mosses, the most common are Epi- lobium alsine folium, E. Hornemanni, Cerastium trigynum, Montia ri- vularis, Saxifraga stellaris, and Catabrosa aquatica, sometimes asso- ciated with Ranunculus hyperboreus and Caltha palustris, Marchantia polymorpha, etc. As this spring-vegetation usually occurs close to bogs and swampy tracts, there are often transitions to bog- and swamp-vegetation. Vegetation around Hot Springs. Peculiar to Iceland is the characteristic vegetation near hot springs. The heat of the soil and the hot water create exceptionally favourable conditions for plant-life, so that species which otherwise do not growr in Iceland can thrive here, and species from South Iceland which do not otherwise grow in North Iceland also occur here near hot springs. It is to be regretted that the vegetation connected with these springs has been more closely investigated in a few places only2. It is especially the rich algal flora, which for instance is of great importance as regards the separation of silica from the hot water, which requires investigation. The vegetation around the hot alkaline springs is usually very luxu- riant and may extend over fairly large areas, because the aqueous vapour floats above the surroundings and descends as a continuous, drizzling rain of tepid water. The plant-associations may differ some- what according to local conditions, as to whether the surroundings are damp or dry, gravelly or rich in humus. Where the surround- ings of the springs consists of a hard grass-bottom , the following 1 In Iceland the word "dy" is also applied to small pools in swampy tracts. 2 Plant-life near the hot springs has been investigated especially by C. Osten- feld, Bot. Tidsskr., vol.22. 1899, pp. 299—245; by Chr. Gronlund: Karakteristik af Plantevsexten paa Island, pp. 33—35 and by Japetus Steenstrup and F. M. Liebmann, Forhandl. v. d. skandin. Natnrforskeres 2. Mode 1840. Kebenhavn, 1841, pp. 336-340. 322 THORODDSEN plants are common: Potentilla anserina, Leontodon autumnalis, Tri- folium repens, Brnnella vulgar is, Gnaphalium uliginosum and G. sil- vaticum, Hydrocotyle uulgaris, Ranunculus acer, Sagina procumbens, Spergula arvensis, Plantago major, P. maritima f. pygmwa, Polygonum persicaria, P. aviculare and various Gramineae, and sometimes Vicia cracca. Where it is damper occur Epilobium palustre, E. alsini- folium, Cardamine multicaulis , Montia rivularis, Limosella aqnatica, Jnnciis lamprocarpus, J. bufonius and some Carices. The composition of these groups of species differs however considerably at the dif- ferent kinds of springs. Of vascular plants the following have been observed exclusively near alkaline hot springs: Hydrocotyle vulgare, Gnaphalium uliginosum, Veronica anagallis, Polygonum persicaria; rare outside the range of influence of hot springs are Galeopsis te- trahit, Plantago lanceolata, P. major and Blechnnm spicant. Moreover, C. Ostenfeld enumerates eight species of mosses which have been found only on warm soil, and three species which are especially connected with a warm ground. In North Iceland some species grow only near hot springs, which in South Iceland thrive also outside the area of the warm soil, for instance, Brnnella vulgaris, Plantago lanceolata, Cardamine mnlticanlis and Drosera rotnndi- folia. In the outlets from both the warm and hot springs an abun- dance of algae often occurs often in a great heat — which have not as yet been investigated. Near solfataras vegetation is ex- tremely sparse, and in the immediate neighbourhood of the sul- phurous-acid fumaroles no plant-life can thrive, therefore only bare clayey flats occur there. Ophioglossnm vulgatum grows only near solfataras; in 1882 I found at Bjarnarflag near Myvatn scattered, but numerous, specimens of it on clayey ground which had a tem- perature of 27° G. - it occurred together with Sagina procnmbens and Poa pratensis; and in 1883 I found it near Gunnuhver on Reykjanes also on warm clay in the neighbourhood of sulphur springs. In both localities it occurred together with the otherwise rare Riccia bifurca and Chomocarpon commutatus. C. Ostenfeld found it on Reykjanes in the neighbourhood of steaming holes, "first a crust was found held together by a Stigonema species, then a great many Muscineae of which the most characteristic were Riccia bifurca and Chomocarpon commutatus. Amongst them occurred some other mosses: Pohlia nntans v. filicaulis, Fossombroiua Dumortieri, Bryum ventrico- snm, Fissidens osmundioides and Philonotis fontana, moreover, the cha- racteristic Ophioglossnm vulgatum var. polyphylla" (loc. cit. p. 240). PHYSICAL GEOGRAPHY 323 Ostenl'eld also describes the vegetation near the boiling mud-pools, and writes: "Close to the mud-pools the ground is quite bare, and not until some distance from them do plants begin to appear; nearest to them occurred Agrostis alba forming a net-work with its long rhi- zomes, and beyond this came a dense, low carpet of Sagina procum- bens, Cerastium vulgatnm, Plantago major (dwarf-form), Stellaria media and a great abundance of Grimmia hypnoides: somewhat further off many other plants occurred." "The vegetation near the outlet of the large mud-pool Gunna was characteristic and peculiar; here the damp ground (about 30 °C.) was covered by a pure, green carpet of Nardia crenulata in which only one other plant occurred, viz. Jun- cus biifonius; within there was firm soil the particles of which were held together by moss-protonema. The moss-carpet became brownish- red where the ground was drier, but it was only Nardia which changed colour; some mosses occurred, however, along the outer edge but only as a subordinate component" Joe, cit. pp. 239, 240). Near numerous solfataras in the neighbourhood of Myvatn, on Odadahraun, Kerlingarfjoll, Torfajokull, and in several other places no vestige of plant-life occurs. Vegetation on Wet Soil (bogs, pools, swamps and wet mea- dows). Bog- and swamp-land (myrar) is very extensively distributed in Iceland, both on the plateau and in the lowlands and valleys; these tracts are also of great economic importance, as in the inha- bited districts the grass is cut for hav, and the hav (lithev) is used tJ V * »* \ V I as winter-fodder for sheep and ponies. In Iceland the swampy meadow-tracts are divided according to their water-content into two main divisions, myri (pi. myrar) and floi (pi. floar); the soil of the former is firm and tough owing to the interwoven roots and rhizomes; in the latter the soil is rotten, and more loosely connected, so that cattle thrust their legs through it and easily get stuck fast; in the former the surface is saturated with the ground water, but in the latter the water reaches to the surface or slightly above it, conse- quently here pools of all sizes abound. Upon the flat surface of the "myrar" small cone-shaped knolls usually occur, and in "floar" water- channels and swampy holes are often found between the knolls. The vegetation is far denser and more continuous in the former than in the latter. The dominant species in the swampy "floi" are Car ex chordorrhiza, Eriophorum angusti folium and Scirpus ccespitosns, and frequently occurring species are Carex rostrata, C. saxatilis, C. Goodenoughii, C. limosa, C. rariflora and several other Carices, as also 324 THORODDSEN Menyanthes trifoliata, Equisetnm limosum, Heleocharis palustris, etc. Upon the knolls, which are dry, many other plants occur, often Be- tula nana, Vaccinium uliginosum , Salix glaiica, etc. The ground- vegetation consists of several species of mosses, Amblystegium, Sphag- num, Hylocominm, and others. On the mountains Eriophorum-bogs are very frequent and Carex-hogs of rarer occurrence, but in the lowlands the reverse is the case. Moss bogs with different aquatic Fig. 28. Geitabergsvatn. Flooded meadow with Erioplutrnm and Care.v. (Phot. A. Hesselbo.) vegetation, as at the above-mentioned springs, often occur in these swampy tracts, and pools with Equisetum limosum, Hippuris uul- garis, Menyanthes and Sparganium. The Icelandic umyrar" are richer in species and have a denser vegetation than the "floar." The dominant plants are Carex cryptocarpa and C. Goodenoughii, but in addition many other species of Carex occur, vir. C. rariflora, C. canescens, C. microglochin and others, moreover, Eriophorum Scheuch- zeri, Equisetnm palustre, Comarum palnstre, which are characteristic of wet meadows, Caltha palustris, Parnassia palnstris, Cardamine pratensis, etc. Almost everywhere the above-mentioned genera and man}7 others are associated with a moss-bottom. Sometimes some Graminece, Polygonnm viviparum, Euphrasia offtcinalis, etc. occur upon the knolls, and in south-west Iceland, here and there on moss- covered knolls, Drosera rotundifolia. From the outer edge of the PHYSICAL GEOGRAPHY 325 water-saturated tracts, where the ground is becoming drier, are easy transitions to other plant-formations, such as grassland, heather- moor, etc. Rocky flats. Of all plant-formations, the rocky-flat-formation occupies the largest area in Iceland, and is the one which characteri- zes by far the greatest part of the island. In favourably situated loca- lities nearest the coast it passes into a "herb- flat" (Urtemark) with Fig. 29. Lake near Armuli. with Carex rostrata. (Phot. A. Hesselbo.) a dense vegetation of different plants, and with some mosses and lichens; but usually the plants are too scattered to have any influ- ence worth mentioning upon the appearance of the landscape. The vegetation of the rocky flat, which includes a great proportion of all the plant-species of the island, may be divided into many sec- tions according to soil-conditions. On climbing higher up on the plateau we find that species and individuals become fewer in num- ber and more scattered in growth, and as already stated, in the highest parts of the plateau only a few stunted , widely separated plants occur. On rocky flats situated at high levels mosses play an important part, especially Grimmia hypnoides, which gradually forms soil for higher plants; now and then some fruticose lichens are found intermixed with the mosses, and in many places on the 326 THORODDSEN plateau patches of Anthelia uiualis occur. The rocky-flat-formation appears to be an original, late Glacial formation from which a great many distinct formations have developed, the different species having become associated according to their conditions of life. The outer limits of the rocky flat are the Grimmia-heath and the "herb-flat," but transitional stages to heather-moor and grassland often occur. As sub-divisions or nearly related formations the following may be mentioned: - -gravelly flats (melar), stone-covered ridges (holt); fallen blocks and debris upon mountain slopes (urd, pi. urdir), steep cliffs (hamrar), gravelly river-plains and river-terraces (eyrar), sandy tracts of various kinds, clayey flats and lava-streams. Gravelly flats (melar) also occupy large areas in the lowlands; the soil-conditions differ somewhat, but generally the gravel is mixed with clay and then the surface often cracks into polygonal cakes and forms a "rudemark" (p. 257). These "rudemarks" greatly influence plant-distribution, as the plants generally resort to the gravel bands between the cakes, where they find shelter and protection. Some- times gravelly flats are so poor in plant-life that they appear quite bare and naked; sometimes they are so densely covered as almost to form a "herb-flat." The most common plants on gravelly flats in the lowlands are Cerastium alpinum, Arabis petrcea, Draba hirta, Silene acaulis, S. maritima, Armeria maritima, Salix herbacea, Sagina nodosa, Spergula arvensis, Arenaria ciliata, Alsine uerna, Thymus serpyllum, Dryas octopetala, Papaver nudicaule, Oxyria digyna, Rumex acetosella, Trisetum subspicatum, Poa glauca, Festnca ovina, Agrostis alba, Luzula multiflora, L. spicata, etc. Naturally all the above- mentioned species do not occur together; in some places a great many of them may occur, while in other places a very few, perhaps only three or four, may be found. The vegetation is also somewhat dependent upon neighbouring plant-formations. Usually mosses or lichens are very sparsely present upon these gravelly flats in the lowlands; only here and there small Grimmia-cushions occur. On stone- covered ridges (holt) there is usually a greater variety as regards soil and situation than on the gravelly flats, and the vegetation there is sometimes fairly luxuriant and conspicuous especially in early summer when Silene acaulis, Dryas octopetala and Thymus serphgllum are in bloom; these are very common there, and also the majority of the plants of gravelly flats. Moreover, the following are noteworthy: Alchimilla alpiua, Saxifraga ccespitosa and S. oppositi folia, Viscaria alpina, Empetnim uigrum, Sedum acre PHYSICAL GEOGRAPHY 327 and S. annuum. As the environment of the ridges differs greatly sometimes bogs, sometimes dry grassland, sometimes heather the vegetation on the ridges also differs somewhat in the different districts owing to immigration from these plant-associations. Mosses Fig. 30. Drycts octopetala (Vallanes; June 26. 1909). (Phot. A. Hesselbo.) are few in number, but there is often an abundance of crustaceous lichens, chiefly various species of Lecideas and Lecanoras, which often impart a strongly variegated appearance to the rocky boulders. River gravel has one characteristic plant all throughout Ice- land, viz. Chamcenerium latifolium , the splendid, purple flowers of which occur in large patches upon gravel-tongues between branching rivers, and can be seen from a distance. Besides plants common to gravelly flats several willows often occur here, viz. Salix glauca, S. lanata and S. phy lid folia, and also Saxifragacece , Galium verum, G. silvestris, and others. Clayey flats with a denser vegetation and a soil rich in humus frequently occur at the outer edges of gravelly 328 THORODDSEN flats; they have a characteristic vegetation consisting of Koenigia islandicctj Sedum villosnm , Juncus alpinns, J. biglumis and J. trigln- mis; moreover, Epilobium palustre, Spergula arvensis, Sagina pro- cumbens, S. nodosa, Stellaria crassifolia, Polygonum aviculare, Equise- tum palustre, Triglochin palustris, Agrostis alba, Luzula spicata and some other species occur frequently, which are distributed according to the water content, etc. of the clayey flats The vegetation of mountain- si opes is often only an extension of that of the rocky flats, with the difference that greater variations occur at the base of the mountains, the conditions there being more highly diversified: the stony tracts alternating with bogs, springs, grass-slopes, heather-moors, coppice-woods, etc. But frequently moun- tain slopes consists mainly of downward-gliding gravel-masses or an- gular rock-fragments, with little or no vegetation, the stone-covering being too unstable to permit plants to gain foothold; in other places are heaps of loose blocks of rock (urd) or solid rock-terraces or -faces; in many places mountain-streams excavate channels or deep ravines, and at the base of mountains they cause the formation of broad gravel-cones with branching streamlets with mosses and other plants connected with springs, or \vith transitions to bog-formations. On the rock terraces there is sometimes a soil-layer which, accord- ing to the conditions of moisture, supports either GramineaB or Cyperacese. Therefore, on mountain-slopes, many different plant- formations are found in patches close to one another in many transi- tional stages. In the rock-detritus on mountain slopes which are not too steep, plants common on rocky flats occur, but none that are really characteristic: the following have been noted: Sllene acanlis, S. maritima, Alchimilla alpina, Dryas octopetala, Thymns serpyllnm, Cerasthim alpinum, Armeria maritima, Saxifraga ccespitosa, S. hyp- noides, S. stellaris, PotentiUa macnlata, P. anserina, Sedum acre, Erige- ron alpinuSj Veronica saxatilis, Poa glauca, P. alpina, and several others. Nor are there many characteristic plants in the vegetation of the rock-faces. Tuff and breccia mountains are generally richer in plants than basalt mountains, their surfaces having many more crevices and hollows in which plants can gain foothold. The fol- lowing plants occur on steep mountain-sides: Archangelica offwinalis, Rhodiola rosea, Haloscias scoticnm, Polypodium uulgare and Woodsia ilvensiSj also Cochlearia officinalis, especially on sea-fowl cliffs; Saxi- fraga Cotyledon grows only on rocks in south-eastern Iceland. More- over, in rock-clefts various ferns occur, most frequently Cystopteris PHYSICAL GEOGRAPHY 329 fragilis, but also flowering plants, such as Sedum annuum, O.ryria digyna, Plantago maritima, Saxifraga ccespitosa, Poa glauca, Festuca ovina, and others. On damp rock-faces near large waterfalls these same species are met with, often as large, well-developed specimens (H. Jonsson, 1905, p. 30): also Poa alpina f. viuipara, Aira alpina, Saxifraga hypnoides, S. stellaris, S. n waits, S. ccespitosa and several Fig. 31. The river Thvera in Oxnadal. Epilobinm hiti folium and Aim on a small gravel-island in the river: July. 1909. (Phot. A. Hesselbo.) species of mosses. Nowhere is seen so mixed and variegated a plant- society as on extensive, fairly densely plant-covered mountain slopes where the majority of the plant-formations are met with, in patches, side by side. On a talus of fallen blocks and debris (urd) there is often a considerable vegetation of lichens, mosses and liverworts. In some places, especially far up on the mountains, the sloping heaps of rock-fragments are poor in plants: in other places they carry a rich vegetation of ferns or willows and birch shrubs and heather, with a variable admixture of herbaceous plants from dif- ferent formations, with no special character, the vegetation resembling closely that which occurs in ravines and lava-clefts. Where the conditions of life are specially favourable; where the 330 THORODDSEN situation is suited to plant-life with southern exposure, an adequate supply of water, shelter from sharp winds, and intense sunlight during spring a "herb-slope" (Urteli1) or "herb-flat" (Urtemark) is formed which on the one hand passes gradually into a rocky-flat formation, and on the other into a birch coppice; it chiefly contains the plants of these two formations, but the growth is dense and luxuriant, so that the ground often appears to be entirely covered by the closely-placed plants. Dicotyledonous flowering plants are the most important, grasses are absent or of subordinate importance. The soil consists of clay or gravel mixed with humus, upon which mosses sometimes occur. Such herb-slopes are found in patches on mountain-sides, on basalt-terraces or on the inclines below rocks, in large ravines and in sheltered, sunny hollows; they often form beautiful carpets, in which the various species usually occur inter- mixed with each other. As a rule, the dominant species are Gera- nium silvaticum , Spircea ulmaria, Archangelica officinalis, Angelica silvestris, Geum rivale, Bartschia alpina, Alchimilla viilgaris, Brunella uulgaris, Rnbus saxatilis, Vicia cracca, Myosotis arvensis, Leontodon autiimnalis, several species of Hieracium, Rnmex acetosa, Ranunculus acer, Poa, Agrostis and Aira; intermixed with these occur several other species, but less frequently. In some parts of the island other characteristic species are frequently noted in "herb-slopes," for in- stance in East Iceland, Campanula rotundifolia and Saxifraga aizoi- des, and in some places in South Iceland Valeriana officinalis and Lychnis flos cnculi. Sand-covered tracts (sandar). As already mentioned, sand}7 tracts occupy vast areas several thousand kilometres surface in Iceland, both in the lowlands and on the plateau. The physical conditions of these "sandar" differ somewhat, therefore their vege- tation, although usually homogeneous and poor in species, may now and then vary somewhat in details. The vast sandy wastes below the glacier-bearing mountains (Jokulls) of South Iceland are mainly formed of glacio-fluvial gravel and sand, but also partly of volcanic ashes and scoriae, while there are wide stretches upon which both the line and the coarse gravel is mixed with clay. Sometimes exten- sive stretches are occupied by alternating clayey flats and pebble- covered river-beds; there are also tracts strewn with ice-striated boulders, and extensive areas, especially on the plateau, covered Urteli ^herb-slope) and Gnesli (grass-slope; see p. 335) denote plant-covered slopes where dicotyledonous flowering plants and grasses are dominant respectively. PHYSICAL GEOGRAPHY 331 with blown sand. Where the sand-covered tracts reach the shore their outermost border supports a halophilous vegetation which, at a short distance from the coast, is replaced by the common plants of sandy soil and rockv flats. On gravelly tracts of sand in the low- \J *f V lands the following plants are the commonest: Sz7e/ie maritima, Ar- meria maritima, Festnca rnbra, v. arenaria, Carex incnrva, Agrostis alba, Jiincns bullions, Elymus arenarins and Polentillu unserina; also Gulinm vernm and T hymns serpijllnm, often occurring in patches. Although the vast stretches of glacial sand in South Iceland have a fairly variable surface yet they are extremely poor in plant-life; owing to "glacier-bursts," and to glacier-rivers constantly causing floods and changing their courses, the vegetation has rarely the chance of development. On Skeidararsandur (cf. H. Jonsson, 1905, pp. 20 — 22) the following plants occur widely scattered: Chamcenerium latifolinm, Arabis petrcea, Silene maritima, Saxifraga opposilifolia , S. ccespitosa and Poa glanca, also small patches of Grimmia hypnoides; but there are, in addition, large stretches quite naked and entirely destitute of plant-life. The main part of Myrdalsandur is a desert almost devoid of vegetation ; far apart occur a few specimens of Arabis petrcea, Silene marilima and Elymus arenarins; usually there is no vestige of plant-life, and one may ride for hours without seeing a single plant. In a few localities in Breidamerkursandur and Skei- dararsandur there is open grass-vegetation in small oases where the sandy gravel from some cause or another has for some length of time escaped inundation by the ice-cold glacier water; in such places, in addition to Chamcenerium latifolinm there occurs usually Agrostis alba, Poa alpina, P. glanca, Aira alpina, Calamagrostis stricta, Festnca ouina, F. rnbra v. arenaria, Carex incnrva, Jnncns bullions. J. triglnmis, Lnzula spicala, Salix lanata, Oxyria digynu and others. As may be seen, there is nothing specially characteristic in the vegetation of these tracts of glacial sand and if they were rescued from the destructive effect of the glacier-rivers, they would quickly become covered by the various plant-formations of the level country: we have already mentioned one such instance, when Brunasandur in 1783 was rescued from the inundations ot glacier-rivers by a lava-stream, which pushed a large rtver aside. Blown sand supports a somewhat more peculiar flora, the characteristic plant being Ely inns urenurins; but where it is very mobile, as e. g. on the plateau in the neighbourhood of Fiskivotn and between Tungna and Skafta, no plant-life can thrive on it. The Botany of Iceland. I. 22 332 THORODDSEN Where the sand becomes more stable, or there is shelter, Elymns arenarius appears, often associated with Festuca r libra v. areuaria; not until the sand becomes somewhat fixed do other species appear, e. g. Juncus balticus, Carex iucurua, Agrostis alba, Festuca oviua, Sileue maritima, Salix lauata, S. glauca and occasionally Haliauthus peploides in localities not too far away from the coast. Extensive Fig. 32. Sali.r lanata and. in the background, Betula odorata growing in black blown-sand (volcanic ashj near Jokulsa, between 'As and Svinadal ; July, 1909. (Phot. A. Hesselbo.) tracts are covered with different Salices, especially in Fjallasveit, where in some places on sandy flats Carex iucurua also occurs in abundance; moreover, in Myvatnsorsefi, Sudurarbotnar, Rangarvellir, and several other places. Where the soil is broken up and re- moved by sand-storms as deep down as to the underlying gravel, as in several places in the southern lowlands in the neigbourhood of Hekla, small tufts of Festuca rubra and F. ovina are the first to appear, then come Juncus balticus, Equisetum arueuse, Agrostis alba, Sileue maritima, Salix lauata and S. glauca; then gradually several species of grasses make their appearance until at last a grassy wil- low-flat is formed which makes an excellent sheep pasture. Here and there on the plateau are damp flats of old blown-sand inter- PHYSICAL GEOGRAPHY 333 sected with clefts, rent by frost, which are filled with- Grimmias and Salijc herbacea. The flora of the lava- streams cannot be referred to any single plant-formation, because according to the age and the progessive development of the vegetation, the lava may bear on it all possible kinds of plant-formations. Nevertheless in Iceland several plants Fig. 33. Lava-field in Nordrardal in the district of Borgarfjord (Aug. 3, 1909). The lava is covered to a depth of one foot with a layer of Grimmia hypnoides. Projecting parts of the lava are covered with crustaceous lichens. Here and there a tuft of Festnca ovina. Birch coppice in the background. (Phot. A. Hesselbo.) are more particularly associated with lava-streams and have there found shelter in the numerous clefts and depressions, where condi- tions of life are especially suitable for them. Paris qimdrifolia is found only on lava-streams, and also ferns are found, in abundance and often as very large specimens. Aspidinm filiv mas, A. spinnlo- sum, A.lonchitis and A. phegopteris very rarely occur in other habi- tats than lava-clefts; Athyrium filix femina, Woodsia ilvensis 9 Poly- podium vulgare, Aspidinm dryopteris and Cystopteris fragilis are also common in lava-clefts, although they are also met with fairly often in other localities, between blocks of rock and in rock -clefts. Milinm effiisum occurs also most frequently in lava-clefts. On lava- 22* 334 THORODDSEN streams there is a good opportunity of studying the development of the different plant-formations and plant-societies because, at the be- ginning, every lava stream is virgin soil, where plant-life must break entirely new ground. The first establishment of vegetation is due to mosses and lichens; on the 23-years-old Krakatindshraun Dr. H. Jonsson found twelve species of mosses, three of lichens and one alga. Two of the lichen -species, Stereocaulon alpinnm and Squamaria gelida, were widely distributed on lava-domes, the mosses usually occurred in clefts, but nowhere had a moss-carpet yet deve- loped (H. Jonsson, 1905, pp. 55 — 56). The next stage is the Grzmmza-heath, which occurs as a continuous covering over the low-lying parts of the lava , wrhile crustaceous lichens form crusts upon the protruding points; the Grimmias form the soil-layer which is primarily necessary for the growth of higher plants. In this moss- foundation occur several lichens and scattered specimens of flowering plants from different associations. After this the development on the lava-streams in the lowlands proceeds in various directions according to the surface-conditions of the lava, the nature of the rock and other circumstances, such as the greater or less amount of drifted sand or humus which has settled upon the surface, and whether a supply of water is available, etc. Consequently, in the course of time a lava-stream may support either a heather-moor, a coppice- wood, a "herb-flat", or grassland, or all these plant-formations may be simultaneously present on the same lava-stream. Rocky- flat forma- tions proper, do not occur on lava-streams, except very rarely, when the lava-streams become covered with gravel brought down by moun- tain streams or glacier rivers. On the plateau the vegetation on a great many of the lava-streams does not go beyond the lichen-stage, on others a considerable number of Grimmias are present, but al- most never as a continuous covering, such as they form in many places in the lowlands and especially on the peninsula of Reykjanes. On the other hand the lava-streams of the plateau are frequently covered with drifted sand and support a sand-vegetation which at higher levels consists of Elymus arenarias and at lower levels of different Salices. There are a fewr instances of old, partially blocked and sand-covered lava-streams in the valleys, in localities where water was abundantly present, forming the substratum of swampy grassland with peat. Thus, in the course of time, a lava-stream may give rise to almost any formation1. For further information regarding plant-life on lava-streams see H. Jons- PHYSICAL GEOGRAPHY 335 The G r i m m i a - h e a t h formation is co-ordinate with the rocky- flat formation, but is not so widely distributed by a great deal as is the latter. As already mentioned, it reaches its fullest development on the lava-streams in the lowlands, especially in the peninsula of Reykjanes, where it occupies vast areas; but it also occurs on many other lava-streams. Where there are sand-drifts Gri/nmia-carpets oc- cur very sparsely or are entirely absent, as Grimmia hypnoides, the most common species, cannot thrive in drifting sand; therefore ex- tensive areas around Hekla, Fiskivotn and several other places are practically bare of mosses. On the vast lava-fields on the plateau mosses occur very sparingly. In several places the Grz/n/nia-heaths on the lava-streams of the lowlands are comparatively quickly trans- formed into soil for higher plants, which are fairly numerous even in the Grimmia-carpet; generally the latter, in the course of time, passes into heather-moor; sometimes into patches of grassland. In several places the Grimmia-heath covers, with its characteristic grey carpet, stony mountain-slopes, and areas strewn with rock fragments, and. as is the case with the rocky flat, forms the foundation of a scattered vegetation of many different species, without any special character, but dependent upon the plant-formations of the neigh- bourhood. Usually the Grimmia-heath develops more quickly into heather-moor or grassland than does the rocky-flat formation, owing to the abundant material for soil-formation supplied by the mosses. Grimmia-heaths occupy large areas in the lower part of the pla- teau, but very few flowering plants are found there in them; on the other hand, lichens often occur numerously, especially Cetraria and Cladonia, not however so numerously that they form a lichen- heath proper, which occurs in Iceland only in patches, and is of no great importance. In the highest parts of the interior of Iceland the Grimmia-heath formation is of much less importance than in the lower part of the plateau. Grassland. Ground covered chiefly with grasses or grass-like plants may be classified under four heads: grass-slopes (Graesli); knolly grassland (Graesmo); flat uncultivated grassland; and home- fields, artificially manured soil. Grass-slopes (GraBsli). The lower son: Grodrarsaga hraunanna a Island! Skirnir, 1906. pp. 150— 163); Vegetation paa Smefellsnes (Vid. Medd. Nat. Foren.. 11)00, pp. 81—84), and Vegetation i Syd-Island (Bot. Tidsskr.. 27 Bd., 1905, pp. 53 — 61). C. H. Ostenfeld: Skildringer af Vegeta- tionen paa Island (Bot. Tidsskr., 22 Bd., 1899, pp. 245— 253). Chr. Gronlund: Karakteristik af Planteviexten paa Island, pp. 30 — 32. 33() THORODDSEN slopes of mountains are often grass-covered, especially when the rock is tuff or breccia; in South Iceland the tuff mountains are often en- tirely grass-covered, at least on the southern side, and sometimes they are covered with a thick layer of soil without knolls proper, which occur only on the clay ground of the lowlands. But some- times wavy rows of small knolls, or narrow ripple-like ledges occur in the lowest part of the grass-slope, and sometimes above these, for a great distance up, the surface of the soil is undulating and wave-like; this is undoubtedlv due to mud-flows in the clavey soil- */ */ «/ covering. On basalt mountains the grass-vegetation extends upwards in tongues or occurs in patches in depressions or on ledges, se- parated by considerable tracts of stones and gravel. On such a grass-slope, in addition to the grasses, many other kinds of herbaceous plants are more or less numerously represented. In South Iceland, according to H. Jonsson, the following are the dominant species: Agrostis vnlgaris, A. canina, Anthoxanthnm odoratnm, Festiica ouina, Poa alpina, P. nemoralis, Geranium silvaticum, Trifolinm repens, Brn- nella vulgaris and Leontodon autumnalis; less common, but often occurring locally in great abundance: Spircea nlmaria, Linnm ca- tharticum, Rubus saxatilis, Gentiana campestris, Myosotis arvensis, Parnassia palnstris, and many others: in Fljotshlid Carum carvi is very common and in Myrdalur and Sida Succisa pratensis. The vegetation is rich in species and is rather mixed, although grasses preponderate. In other parts of the country where basalt is domi- nant the grass vegetation of the mountain-slopes consists of similar species, but is not so luxuriant as in South Iceland. The following species are common: Agrostis vulgaris, A. alba, A. canina, Anthoxan- thum odoratum, Nardus stricta, Aira flexuosa, A. ccespitosa, Phleum pratense, Poa alpina, Hierochloa borealis, Festnca rnbra, etc. A spe- cial Nardzzs-association and an A/7//?oxa/?//?Z7m-association often occur. Knolly grassland (Graesmo). By this is understood dry, ex- tremely knolly stretches of clayey ground intermixed with humus, occurring on level land and in valleys with a mixed vegetation of Graminese, Juncaceae and Cyperaceae; it may therefore differ con- siderably in appearance, according to which of these families pre- dominates. When grasses predominate the "Grsesmo" resembles grassland, but sometimes Jnncns trifidns and Ely no. Bellardi are so dominant that large stretches attain a brownish tint like that of a heather-moor. Usually, the vegetation of the knolls differs from that of the depressions; in the depressions, mosses and some Carices PHYSICAL GEOGRAPHY 337 are often met with, but upon the knolls there are Elyna Bellardi and Jnncus trifidus: sometimes these are intermixed with mosses and lichens. In some places Jnncus balticus predominates, in others Luznla or Agrostis, Aira ccespitosa, Trisetum subspicatum and other grasses. The dry uncultivated grassland without knolls (hardvelli, vall- lendi) usually has for a substratum coarse sand, river-gravel, pebbles, etc., with a thin covering of humus, and a low and rather open vege- tation which chiefly consists of Gramineae (Festnca, Aira, Poa and Agrostis), but these are abundantly intermixed with Jnncus balticns, Luznla multiflora and L. spicata and Elyna Bellardi; Festnca r libra is sometimes a dominant species. Of dicotyledonous plants the fol- lowing are common: Leontodon antnmnalis, Thalictrnm alpinnm, Draba verna, Galinm verum, Euphrasia officinalis, Viola tricolor, Gentiana campestris, Achillea mille folium, Rhinanthns minor, and others. The ho me- fie Id (tun) is the manured grassland round the farm- buildings. It is usually enclosed and generally abounds in large knolls. In 1909 the area of the home-fields throughout the island was estimated at 188 square kilometres. The hay (tada) from these home-fields is kept chiefly for winter-fodder for cows. The culti- vation of the home-fields differs greatly, which again influences the vegetation, other plant-associations occurring in patches in the grass- covered area of the home-field; these are especially dependent upon the degree of moisture contained in the different parts of the field. In really well-cultivated home-fields throughout the country the vege- tation is everywhere homogeneous. The Graminea? are dominant, especially Aira ccespitosa, Poa pratensis and Festnca rnbra; moreover Festnca ouina, Poa trivialis, P. annua, Agrostis vulgaris, Alopecurus genicnlatns and Anthoxanthnm odoratiim are very common, and inter- mixed abundantly with these occur Ranunculus acer, Taraxacum of/ici- nale, Rumex acetosa and Polygonum viuiparnm. The home-fields are often quite yellow with Ranunculi (R. acer and R. repens); in many places Trifolium repens and Rhinanthus minor also occur abundantly. In some places Viola tricolor is abundant in home-fields, especially in Eyjafjordur, and Geranium siluaticum in some places in Myrdalur, and in other places Vicia cracca, etc. In addition, many other plants are found in patches, according to the dampness of the soil and the care given to the cultivation of the home-fields; in dry and sandy grassland there often occur an abundance of Galinm verum, G. siluestre, G. boreale, Achillea mille folium , Leontodon antnmnalis, 338 THORODDSEN Armeria marilima, as also Luzula, Elyna, etc.; in damp home-fields are found Caltha pahistris, Cardamine pratensis, Koenigia islandica, Montia rwularis and different Carices. Moreover, a great many spe- cies often immigrate from the plant-formations of the neighbourhood; in some places, in badly kept home-fields, even swamp vegetation and willow-coppices or heather may be met with in patches. The ^m^mmM:^^<, Fig. 34. Outer edge of the home-field (tun) of the farmstead Hnausar in Valnsdal Matricaria inodora and Capsella Bnrsa pastoris: in the background liiunex domesticus ; July 28, 1909. (Phot. A. Hesselbo.) farm-buildings are, as a rule, in the centre of the home-field, and around them there is almost always a characteristic vegetation con- sisting of Alopecurns geniculatus, Glyceria distans, Ranunculus repens, Poa annua and P. trwialis; quite near to the home and stables grow Stellaria media, Capsella bursa pastoris, Polygonum aviculare, Rumex domesticus. etc. Around farmsteads in the vicinity of the sea, and •) on islands, are often found in addition Cochlearia officinalis, Cakile marilima and Atriplex patula. In Iceland the walls and roofs of peasants' houses are generally built of turf and are therefore over- grown with various grasses, especially Glyceria distans. Flowering plants also often occur upon houses; they vary in the different dis- tricts; in south-west Iceland Matricaria inodora grows luxuriantly PHYSICAL GEOGRAPHY 339 on the roofs and in some places Achillea millefolium; in various districts Rhodiola rosea has been planted on the walls, and in the northermost districts Cochlearia officinalis often occurs in abundance upon the houses, and in some places Saxifraga rimilaris. At their base the house-walls are green with Prasiola crispa; various mosses also grow7 upon the houses, especially Bryiim argentenm l. Heather moors are extensively developed in Iceland both in the lowlands and valleys, on mountain-sides and on hills to a height of about 400 metres; they are not recorded from higher levels. On moun- tain-sides and on hill-slopes the heather moor is almost flat, on level land it is usually knolly; it is best developed upon old lava-streams and it often forms the ground vegetation of birch coppices. The dominant species are Empetrum nigrnm, Vaccimum uliginosum, V. myrtillus, Arctostaphylns uva ursi and Calluna vnlgaris; the last species is fairly common in many districts, but never occurs so abun- dantly as the others. Moreover, among the heather there is usually an abundance of Dry as odopetala, Juniper us communis, Betula nana, Rubus saxatilis and Salix herbacea. In heathy tracts at higher levels and also above these in scattered patches, are found Loiseleuria pro- cumbens, Cassiope hypnoides and Sibbaldia procumbens. On the penin- sulas on either side of Eyjafjordur patches of Phyllodoce coerculea which is not recorded from more southerly habitats - - are no\v and then found. There is a great variation in the distribution of the different character-plants in the heather moor; sometimes they all occur mixed with each other; sometimes each occurs 'separately in large or small patches, so that the different areas may be designated Vaccimum heaths, Empetrum heaths, Calluna heaths, Arctostaphylus heaths, etc. Between the heather, as ground vegetation, other plants occur in abundance, especially immigrants from the knolly grass- land from which the knolly heather-land often appears to have been derived for instance, Juncus tri/idus, Elyna Bellardi, Luzula multiflora, L. spicata, Nardus stricta, Agrostis canina, A. alba and se- veral other Gramineae, also Salix lanata, S. glauca, S. phylicifolia, Silene acaulis, Thymus serpyllum, Bartschia alpina, Alchimilla alpina, Thalictrum, Galium, Hieracium, and many others. In addition, there is very often a moss-carpet of Grimmia hypnoides beneath the heather. Willow coppices. Although the various species of willows lare widely distributed they rarely form coppices proper, and willow 1 As regards the moss-vegetation on old house-walls see Helgi Jonsson's above-mentioned paper on the vegetation of South Iceland. 1905, p. 54. 340 THORODDSEN coppices, as independent formations, have but a slight distribution in Iceland. As stated above, Salix lanata and S. glanca are distri- buted over extensive sandy areas, and occur there, together with some other plants, as dominants; in other places they occur on dry, flat, clayey tracts, as scattered shrubs of low growth (20 — 60 cm.) with an undergrowth of heather, Elyna Bellardi, various species of grasses, etc. It is chiefly Salix phylicifolia which forms coppices Fjg. 35. Knolly moor of Betula liana intermixed with Salix lanata and SalLv phylicifolia Ground vegetation: Empetrnm. Arctostaphylus iwa ursi, Anthoxanthum odoratum and Polygonum uiviparum. (Phot. A. Hesselbo.) (e. g. near Skaftafell, Fnjoskadalur, Hrafnkelsdalur, Kaldalon and Myvatn) in association with Salix lanata and Betula odorata and with a rich ground-vegetation of highly diverse plants common on "herb-flats." Now and then Salix phylicifolia forms the undergrowth of birch coppices, as for instance, in Bsejarstadaskogur near 0rsefi; the willows here have an average height of 2 metres; the highest individuals are 3 metres in height, but the stems are only 18 — 24cm. in circumference. Salix phylicifolia (often together with Salix lanata, which may attain a height of I--!1/? metres) is often found inter- spersed in birch coppices, in many places in different districts. In Several places near Myvatn Salix phylicifolia forms coppices, among other places in Sluttnes, where a stem which I measured in 1882, PHYSICAL GEOGRAPHY 341 had a circumference of 21 cm. and a length of 5 metres, but it could not stand erect. Birch coppices. An account has already been given of the distribution of birch coppices in Iceland, and in a subsequent part * ~ TV* Fig. 36. Hallormstadaskogur: 1894 (see text). (Phot. Th. Thoroddsen.) of this work there will no doubt be an exhaustive description of tree-distribution in Iceland, so here only a few very brief notes on the subject will be given. The greater part of the birch coppices in Iceland consists of stunted shrubs having, as a rule, a height of 1 — 3 metres only, sometimes even lower; a fact which is undoubtedly due to the destructive habits of sheep. Well-grown birches - - the re- mains of woods which have formerly been far more extensive occur, however, in a few localities. Of such woods the following 342 THORODDSEN are the most important: Hallormstadaskogur near Lagarfljot in East Iceland, Bsejarstadaskogur below Jokulfell in 0raefi in South Ice- land, and Thordarstadaskogur and Halsskogur in Fnjoskadal in North Iceland. In Hallormstadaskogur some erect birch trees have a height of 8 — 9 metres and a circumference of 70 — 80 cm., and many others have a height of 5 — 7 metres. In Thordarstadaskogur the highest tree is 8l/2 metres high, with a circumference of 32 cm.; several of the trees are 6 — 7 metres high, and the average height of the whole wood is 3 — 4 metres. Halsskogur is somewhat lower; some of the trees are, however, 6 — 7 metres high, and several 4—5 metres1. Ba?jarstadaskogur is somewhat lower, but the trees are well-grown and erect, and stunted birches are absent; the average height of the birch trees is 4 — 5 metres and may often be as much as 6 metres2. In a ravine near Skaftafell I measured in 1894 a birch tree which was 7 metres high and a mountain ash which had a height of 91/* metres. This tree occurs sometimes dispersed in birch coppices, and sometimes separately in ravines and on mountain slopes; it has often been allowed to stand on account of some superstition. In some places in North and South Iceland the mountain ash (Sorbus aiicnparia) has been planted around farmsteads and by houses in towns. It at- tains a height of 7- -10 metres, but in birch coppices it is gene- rally only 4 — 5 metres, or even less. In birch coppices are also found Betnla nana, Salix phylicifolia , 8. lanata and 5. glaiica and Jimiperns communis. The soil in coppice-woods consists often of "moar" knolly clay which rests sometimes on gravel and some- times upon rock. Coppices often occur also on a stony bottom, as in ravines, between rocky boulders, and often upon mountain slopes occasionally they are found on boggy soil. The wood-floor is very often occupied by heather moor; and birch coppices of lower growth often even pass into heather moor; in the latter case the same species are found in the woods as are found on ordinary heather moors, and they form similar associations3. 1 S. Sigurdsson: Skogarnir i Fnjoskadal (Andvari, XXV, 1900, pp. 144-175). H..J6nsson, 1905, pp. 46— 50. Th. Thoroddsen in Geograiisk Tidsskrift, XIII, 1895, pp. 16—17. During latter years man}- papers have been written on the woods of Ice- land. One of the most important is that by C. V. Pr\'tz: Skovdyrkning paa Island in Tidsskrift for Skovvaesen. vol. XVII, 1905, pp.20 — 89; it also contains interesting notes on the Icelandic soil. Moreover, works dealing with the woods of Iceland are enumerated in Lysing Islands, vol. 2, on pp. 443 — 445. PHYSICAL GEORRAPHV 343 Sometimes the trees grow in grassland, and sometimes on "herb- flats;" occasionally the wood-floor consists of mosses. In grassland the commonest species are Agrostis viilgaris, Air a flexuosa and An- thoxanthum odoratnm and other more scattered Gramineae. Where the herb-flat formation is dominant many different species occur, especially Spiraea ulmaria, Angelica siluestris, Rubus saxatilis, Leon- todon autamnalis , Geranium silvaticnm, Bartschia alpina, Myosotis arvensis, Alchimilla vnlgaris, Fragaria vesca, and others. Trientalis enropcea grows here and there in woods in East Iceland only. Where mosses are dominant on the wood-floor the vegetation consists chiefly of Hylocomium (H. Jonsson, 1900, p. 76). On the whole, in coppices, scattered individuals of the numerous species from the different form- ations in the neighbourhood may be found.