| \RCTIC ISLANDS OF NEVV ZEALAND. POLE MITD PRT TIS Oe aed ea atl Ce oho. NS Nis Oo (5 SUBANTARCTIC [ISLANDS OF NEW ZEALAND. ARTICLE XIX.—ON THE SYSTEMATIC BOTANY OF THE ISLANDS TO THE SOUTH OF NEW ZEALAND. By T. F. Cuerseman, F.L.S., F.Z.S8., Curator of the Auckland Museum. I. INTRODUCTORY. Tue islands to the south of New Zealand consist of six widely separated groups, situated between the parallels of 48° and 55° 8. lat. and 159° and 179° E. long. Their physiography and geology are fully described by Dr. Marshall and Mr. Speight in separate memoirs, so that all that need be done here is to briefly indicate their posi- tion and size. The most northerly group is that known by the name of the Snares, its nearest point being within sixty-five miles of the South Cape of Stewart Island. It consists of two main islands, one much larger than the other, and a number of storm-swept rocks of small size. The larger of the two islands is barely more than a mile and a half in greatest diameter, and its extreme altitude does not exceed 500 ft. It is surrounded by precipitous cliffs, landing being possible only in a little boat- harbour on the north-east side, and in one or two indentations on the coast-line. Next in distance to New Zealand is the well-known group to which the name of the Lord Auckland Islands has been applied. It is situated about 190 miles 8.S.W. from Stewart Island; Port Ross, at the northern end of the group, being in §. lat. 50° 32’. The entire group is under thirty-two miles in length, and has a greatest breadth of about fifteen miles. It consists of two large islands of unequal size— Auckland Island and Adams Island—separated by a narrow and somewhat tortuous strait; and a considerable number of smaller ones, of which Disappointment Island on the west, and Enderby, Rose, and Ewing Islands to the north, are the most im- portant. The western side of the two main islands forms an almost unbroken line of precipitous cliffs, which in the south of Adams Island attain a vertical height of 1,400 ft., but which sink to a much lower elevation in the north of the group. The eastern side is cut up into many deep inlets, forming excellent harbours, some of them advancing to within a few miles of the western coast. The whole island is hilly, but the hills nowhere exceed 2,000 ft., and are usually much less. The lower part of the island, especially on the eastern side, is mainly covered with forest ; the upper portion is composed of open meadow or tussock. Campbell Island les about 140 miles to the south-east of the Auckland Islands, or, in round numbers, 330 miles from Stewart Island, its exact position being §. lat. 390 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. 52° 22” and H. long. 169° 9’. It is irregularly rhomboidal in outline, and has a greatest length of about ten miles, with a breadth of nearly eight. With the exception of two or three inlets, the middle one of which almost divides the island into two, it is walled round by abrupt rocky cliffs, which Sir J. D. Hooker compared to those of St. Helena. The hills are steeper and more rugged than those of the Auckland Islands, although not quite so high, and there is no proper forest, its place being taken by a dense woody scrub, common at low levels. About 400 miles E.N.E. from Campbell Island, and about 490 miles from Stewart Island, lie the Antipodes Islands, consisting of one large island about four miles long by two broad, and seven small islands or rocks, of which Bollons Island is the only one oi any importance. Although the coast-line is precipitous and rocky, when the top of the island is reached there is a considerable extent of undulating country. The Bounty Islands (S. lat. 47° 43’, EH. long. 179°-5) are situated about ninety miles north of the Antipodes Islands, and about 490 miles east of Stewart Island. The group consists of twenty or more small rocks or islets, frequented by penguins or other sea-birds, but without any land vegetation whatever. The five groups just mentioned all stand on a comparatively shallow oceanic plateau, stretching far to the south and south-east of New Zealand, the depth on which, according to Sir J. Hector, nowhere exceeds 400 fathoms, and, so far as the Auckland Islands are concerned, 1s much less than that. The sixth and most remote eroup—that of Macquarie Island—tlies to the south-west of Stewart Island, from which it is distant about 570 miles. It is separated both from New Zealand and the plateau above mentioned by an oceanic depression which is said to have a depth of nearly 2,000 fathoms, a statement which must be kept clearly in mind when considering its peculiarities and the history of its flora and fauna. Its approximate position is 8. lat. 54° 30’ and H. long. 159° 3’. There is some uncertainty as to its size, but it is probably from fifteen to eighteen miles long by five or six in greatest breadth. Although hilly, the hills are of no great height, nowhere exceeding 600 ft. or 700 ft., and on many portions of the coast-lne there is a considerable area of flat or gently sloping land near the sea. ‘There are two outlying islets—one, known as the “ Bishop and Clerk,” is nearly thirty miles south of the main island ; the other is about seven miles to the north. According to Dr. Scott, who visited the island in 1880, its appearance is barren and ‘desolate in the extreme. There is not a single woody plant, and the facies of the vegetation is dreary and monotonous to a degree. Il. HISTORY OF BOTANICAL EXPLORATION. The botanical investigation of the islands to the south of New Zealand dates back to 1840, during which year they were visited by no less than three scientific expeditions of the first rank. Mentioning these in the order of time, the first was the well-known United States Exploring Expedition, under the command of Commodore Wilkes; the second the elaborately organized and equipped French Expedition, led by Admiral D’Urville; and the third the Antarctic Expedition of Sir James Clark Ross, to which Sir Joseph Hooker was attached as naturalist. As the collections obtained by Ross’s voyage far surpassed in extent and importance those made by the other two, and as, in addition, the results were the first published and made available to the scientific world, it will be best te commence this sketch with an account of its work. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 391 Ross’s expedition consisted of two vessels—the “ Erebus,” commanded by Ross himself, and the “ Terror,” under Captain Crozier. To the first-mentioned vessel, as already mentioned, Sir J. D. Hooker was attached as naturalist ; whilst Dr. Lyall served in a similar capacity on the “ Terror.” After calling at the Cape of Good Hope, Kerguelen Island, and Tasmania, the expedition arrived at the Auckland Islands on the 20th November, 1840, anchoring in Rendezvous Harbour (now known as Port Ross), at the northern end of the island. Here the ships remained until the 12th December, or twenty days in all. During this stay Hooker and Lyall were actively engaged in investigating the natural history of the group. Nothing was previously known of the vegetation, and the isolated situation of the group, far removed from any land except the Islands of New Zealand, promised many im- portant discoveries, and made its exploration a matter of peculiar interest. That these anticipations were fully realised the published records of the expedition amply prove; but it may be allowable to quote a few sentences from a short account of the vegetation contributed by Hooker to Ross’s “ Voyage to the Southern and Antarctic Regions” (vol. 1, p. 144): “ Possessing no mountains rising to the limits of perpetual snow, and few rocks or precipices, the whole land seemed covered with vegetation. A low forest skirts all the shores, succeeded by a broad belt of brush- wood, above which, to the summits of the hills, extend grassy slopes. On a closer inspection of the forest, it is found to be composed of a dense thicket of stag-headed trees, so gnarled and stunted by the violence of the gales as to afford an excellent shelter for a luxuriant undergrowth of bright-green feathery ferns and several gay- flowered herbs, With much to delight the eye, and an extraordinary amount of new species to occupy the mind, there is here a want of any of those trees or shrubs to which the voyager has been accustomed in the north.” “ The woods consist en- tirely of four or five species of trees, or large shrubs, which are here enumerated in the order of their relative abundance.” (1, Metrosideros lucida ; 2, Dracophyllum longifolium ; 3, Panax sumplex ; 4, Veronica elluptica ; 5, Coprosma foetidissima.) “ Under the shade of these, about fifteen different ferns grow in great abundance, the most remarkable of which is a species of Aspidiwm (A. venustum, of the French South Polar Voyage), with short trunks, 2-3 ft. high, crowned with a tuft of spread- ing feathery fronds, each 3-5 ft. long: this is one of the most graceful and ornamental productions of the group. The Aralia polaris, Homb. & Jacq., and the Pleuwrophyllum erumferum are two highly remarkable plants, very common near the sea: the former is allied to the ivy, but has clusters of green waxy flowers as large as a child’s head ; and its round and wrinkled leaves, of the darkest green, measure a foot and a half across. They form the favourite food of the hogs, which run wild on these islands.” “Tt is upon the hills, however, that the more beautiful plants abound; amongst which the most striking is a liliaceous one, allied to Anthericum (Chrysobactron Rossi), whose conspicuous racemes of golden flowers are often a span long, and many speci- mens have three or four such spikes, The Plewrophyllum speciosum resembles a large Aster, bearing numerous purple flowers, the size of a large marigold. The Celmisia vernicosa has linear glossy leaves, spread out on the ground like the spokes of a wheel, and pure-white flowers, with a purple eye, as large as those of the last-named plant. Finally, the Veronica Benthami may be mentioned ; it is of shrubby growth, with spikes of flowers of a deep ultramarine blue. Amongst those of humbler stature, several Kuropean genera occur, as species of Cardamine, Ranunculus, Plantago, 392 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. Geramum, and Epilobium, two beautiful white- and red-flowered gentians, and a forget-me-not with flowers much larger than those of any English species.” I have quoted these passages not only because they give a vivid picture of the salient fea- tures of the vegetation, but because they evidence the ardour and enthusiasm with which the two naturalists of the expedition pursued their investigations. Departing from the Auckland Islands on the 12th December, Campbell Island was reached on the morning of the 13th, the ships anchorimg in Perseverance Har- bour. A stay of four days only was made, the expedition leaving on the 17th Decem- ber for a cruise to the Antarctic Circle and the south polar regions ; but notwith- standing the shortness of the visit a good deal of botanical work was done. Respecting the character of the vegetation, Hooker remarks, in the publication already quoted from (p. 158), “ Although Campbell Island is situated 120 miles to the southward of Lord Auckland’s Group, and is of much smaller extent, 1t probably contains fully as many native plants. This arises from its more varied outline, and from its steep precipices and contracted ravines, affording situations more congenial to the growth of grasses, mosses, and lichens. Its ironbound coast and rocky mountains, whose summits appear to the eye bare of vegetation, give it the aspect of a very desolate and unproductive rock, and it is not until the quiet harbours are opened that any green hue, save a few grassy spots, is seen. In these narrow bays the scene suddenly changes: a belt of brushwood, composed of some of the trees mentioned as inhabitants of the last-visited island, but in a very stunted state, form a verdant line close to the beach. This is succeeded by bright-green slopes, so studded with the Chrysobactron as to give them a yellow tinge, visible a full mile from the shore. Most of the beauti- ful plants of Lord Auckland’s Group, including the elegant caulescent ferns, are equally abundant here, and from many of them growing in this higher latitude at a proportionally lower elevation, their beauty strikes every one on first landing.” In 1843 Ross returned to England. The important scientific results obtamed by the expedition, and the large collections amassed during the voyage, both pointed to the desirability of speedy publication ; and through the liberality of the Admiralty this was promptly arranged for. The botanical portion was intrusted to Hooker, and a special vote of £2,000 was granted to defray the cost of printing and illustrat- ing the work. The result is found in the two volumes of the superb “ Flora Ant- arctica,” a publication in every way worthy of the importance of the subject and the reputation of its distinguished author. The first volume, issued in 1844—45, is confined to the plants of the Auckland and Campbell Islands, and will always remain the foundation for investigations into the botany of the islands. A briet analysis of its contents (after deducting two or three naturalised plants, and including | some additional species mentioned in the second volume) shows that it records the occurrence of 105 phanerogams and eighteen vascular cryptogams, or 123 in all. In addition to this, 277 lower cryptogams are included, making a total flora of exactly 400 species. ‘The illustrations consist of eighty beautifully prepared plates, fitty-six of which are of phanerogams. The new species are described with the utmost care, and with an unusual amount of detail; and the whole work, from beginning to end, is full of special information upon the relationships, systematic position, and geographical distribution of the species treated upon. As I have elsewhere remarked, it is a splendid monument of painstaking exploration and research, and it seems almosthincredible that the observations and material upon which it is founded should have been collected in less than a month. Systematic Botany] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 393 Second only in importance to Ross’s expedition is that of D’Urville in the “ Astrolabe” and “ Zélée.” Leaving France in 1837, its main purpose was geo- eraphical discovery, and especially the exploration of the little-known antarctic regions. Returning northwards in the early part of 1840, after an adventurous voyage to the south polar regions, during which Adelie Land was discovered, D’Ur- ville called at the Auckland Islands, his ships anchoring in Rendezvous Harbour on the 11th March, 1840. A stay of nine days was made, the expedition leaving for New Zealand on the 20th March. During this period important collections were made, not only by Hombron and Jacquinot, the naturalists of the expedition, but also by D’Urville himself. The vessels returned to France about the end of 1840. The French Government, with its usual liberality im such matters, at once authorised the publication of the results of the voyage, in a style which can only be described as lavishly sumptuous. The history and cartography, together with memoirs in all departments of science, were arranged for on an elaborate scale, and publication was commenced early in 1841. But D’Urville’s tragic death in a railway accident, which occurred within two years of his return, delayed the progress of the work, and it was not until the end of 1854 that the last memoir of the series appeared. The work bears the title, “ Voyage au Pole sud et dans lOceanie sur les corvettes l Astrolabe et la Zélée, pendant les années 1837 4 1840,” and consists of eighteen octavo volumes of letterpress and seven folio volumes of plates. The botanical portion, with which we are alone concerned, was edited by Hombron and Jacquinot, and consists of a volume of plates and two volumes of letterpress. The plates, which were put in hand the first, were several years in preparation, and some of them must have been separately issued in advance of the rest, for Hooker quotes several in the “ Flora Antarctica,” published in 1844-45. Their appearance was further delayed by the death of Hombron, so that it was 1852 before they were actually issued as a whole. As representations of the species they are excellent —in fact, some of them are exquisite works of art; but the arrangement and nomenclature are both open to serious criticism. The total number of plates is sixty-five, illus- trating 170 species, and forty-eight of them are devoted to plants from the Auckland Islands. The first volume of the letterpress, containmg the cellular cryptogams, was intrusted to the well-known cryptogamist Montagne, and appeared in 1845. The second volume, contaiming the phanerogams and vascular cryptogams, was to have been written by Hombron. After his death a long delay ensued, but the work was eventually taken up by the eminent botanist Decaisne, and was published in 1853. So far as I can estimate, the total number of species collected by the French Expedition is about 100, of which seventy-five are lower cryptogams, and twenty- five phanerogams and vascular cryptogams. The United States Explormg Expedition, under the command of Commodore Wilkes, was engaged from 1838 to 1842 in a voyage of discovery in various parts of the world. On the 26th December, 1839, the expedition, which consisted of four vessels, left Sydney for a cruise to the south polar regions. During the outward voyage, the “ Peacock,” one of the ships comprising the expedition, called at Mac- quarie Island, and with great difficulty succeeded in landing two men in a small bay on the western side of the island. Some penguins and other birds were obtained ; but the only reference made to the vegetation in Wilkes’s “ Narrative ” (vol. 2, p. 290), is contained in the words, “ The island is apparently covered with verdure, although \ | q \t 394 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. a long tulted rank grass was the only plant seen by those who landed.” No doubt this grass was Poa foliosa, which has been recorded by all visitors to the island. On the return voyage of the vessels, the brig “ Porpoise,” which had become separated from the rest, paid a visit to the Auckland Islands, anchoring in Port Ross on the 7th March. A stay of three days only was made, a period altogether insufficient for any scientific work of importance, especially when it is borne in mind that the whole of the scientific staff remained in Australia during the cruise. However, the surgeon of the “‘ Porpoise,” Dr. Holmes, collected a few plants, which were the first obtained on the island. Later on, the expedition proceeded from Australia to the Bay ot Islands, where the scientific staff rejoined it, and made larger collections, with which Dr. Holmes’s Auckland Island plants were incorporated. Atiter Wilkes’s return to the United States, and after many delays, the botanical collections were intrusted to the eminent American botanist Asa Gray, and ultimately appeared (in 1854) in two volumes quarto, with a folio atlas of 100 plates. Not one of the Auckland Island plants is figured, and Ranunculus aucklandicus is the only species described which had not been previously recorded by Hooker. For many years after the visits of the three scientific expeditions mentioned above no addition of any importance was made to our knowledge of the botany of the southern islands. Now and then they were visited by a British man-ot-war, and on one of these occasions the late General Bolton made a small collection of plants, which included a few species not noticed by Hooker. These are recorded in the first part of the Handbook, published in 1864. In 1874 the French Govern- ment despatched an astronomical expedition to Campbell Island for the purpose of observing the Transit of Venus. Two of the party, Dr. Filhol and Lieutenant Rathouis, made collections of plants. Some of the specimens were retained in the colony, and were examined by Mr. Kirk, who contributed a list of the species to the “ Transactions of the New Zealand Institute ” (vol. xiv, p. 287). In 1880 Professor J. H. Scott visited Macquarie Island for the purpose of in- vestigating its flora and fauna. Before referring to his visit, however, it should be mentioned that prior to 1832 Mr. C. Fraser, then Superintendent of the Sydney Botanical Gardens, sent to Sir W. J. Hooker a few plants which he had obtained from the island, probably through the master of some vessel. These are Acaena Sanguisorbae, Acaena adscendens, Azorella Selago, Pleurophyllum Hookeri, Cotula plumosa, Luzula campestris var. ermmta, Poa foliosa, and Aspidium vestitum. The collection is interesting, as being the first recorded from any of the southern islands. Professor Scott has published an account of his visit in the “ Transactions of the New Zealand Institute ’ for 1882 (vol. xv, p. 484), from which an excellent idea of the physiography of the island and the general character of the vegetation can be obtained. His catalogue of the plants observed includes the names of sixteen phanero- gams and three vascular cryptogams, or nineteen in all. But an introduced grass (Poa annua) rust be deducted ; and it is now quite clear that a mistake was made in including Azorella lycopodioides, the specimens so identified being probably a form of Colobanthus muscoides. Deducting these two species, the total number becomes seventeen. In this place it will be convenient to mention the later visit of Mr. A. Hamilton, made in 1894. Although much hindered by illness and by excep- tionally severe weather, he succeeded in materially adding to our knowledge Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 395 of the vegetation. A list of the plants collected will be found in his “ Notes on a Visit to Macquarie Island” (Trans. N.Z. Inst., xxvu, 559). It includes eleven phanerogams additional to those recorded by Professor Scott, thus bringing up the total flora to twenty-eight species. At first sight, this seems a small number for so large an island ; but it is really only in keeping with what is known of the character of the vegetation in high southern latitudes. Kerguelen Island, which is not only much larger, but is 5° further north than Macquarie Island, has a flora of only thirty species. Further, South Georgia, which is in almost the same latitude as Macquarie Island, has so far only yielded eighteen phanerogams and ferns. About the end of 1883 the late Mr. John Buchanan visited Campbell Island in the Government s.s. “ Stella,” with the view of procuring living plants for various public gardens in New Zealand, and of further investigating the flora of the island. His paper on “Campbell Island and its Flora” (Trans. N.Z. Inst., xvi, 398), although short, contains several observations of value. The most important result of his visit was the proof which he obtained of the existence of a third species of Plewro- phyllum (P. Hookert, Buch.), which had been confounded with P. crinferum by all previous collectors. In 1887 the numerous disastrous shipwrecks which had occurred on the Auckland Islands, and the hardships experienced by the castaways, induced the New Zealand Government to provide depots of provisions, &c., and to arrange that a Government steamer should regularly visit the islands at least twice every year. Since the esta- blishment of this means of communication it has been possible for naturalists and others to reach the islands with much greater comfort and certainty. The most noteworthy visit made in this manner 1s undoubtedly that of Mr. Kirk in 1890. He was able to botanize not only on the Auckland and Campbell! Islands, but also on the Snares and Antipodes Island, the vegetation of which was previously altogether unknown. The results of his journey were embodied in a memoir contributed to the Report of the Australasian Association for 1891, portions of which were also separately published in the “ Transactions of the New Zealand Institute” and the “ Journal of Botany,” and which constituted the most important contribution made to the botany of the southern islands since Hooker’s visit exactly fifty years before. The additional information recorded may be briefly summarised as follows: First, the addition of at least twelve species to the known flora of the Auckland and Campbell Islands, and the collection of many fresh facts bearing on the vegetation generally. Second, the exploration of the Snares, which not only gave us a knowledge of the general character of the vegetation, and placed upon record a list of twenty-three indigenous phanerogams and ferns, but which brought to light some totally unex- pected facts of geographical distribution, such as the occurrence of Olearia Lyallu and Stiubocarpa Lyall var. robusta, together with the discovery of the endemic Lagusticum acutifolium. Third, a similar exploration of Antipodes Island, which led to a knowledge of the chief features of its florula, and to the recognition of the important fact that fifty-four species of phanerogams and ferns, two of them endemic, are found on this remote outlier of New Zealand. At the same time as Kirk, Mr. F. R. Chapman (now Mr. Justice Chapman) visited the islands, and made collections of some size, both of living and dried plants. His paper, entitled “ The Outlying Islands South of New Zealand” (Trans. N.Z. Inst., xxii, 491), is replete with interesting matter relating to the flora and fauna, and should be read by all persons interested in the subject. Among other gentlemen 396 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. who have visited the southern islands and have interested themselves in their vege- tation, either as observers or collectors, may be mentioned Captain F. W. Hutton (1901), Mr. H. J. Matthews (1895), and Sir J. Hector (1895). In June, 1903, Dr. L. Cockayne made his first visit to the islands. All previous explorers had reached them in spring or summer; and, as the winter aspect of the vegetation was absolutely unknown and undescribed, at any rate so far as the endemic species were concerned, the journey could not fail to be productive of important results, That this was the case is amply evidenced by his paper on “ A Botanical Excursion during Midwinter to the Southern Islands of New Zealand ” (Trans. N.Z. Inst., XXXVI, 225-330). In this he not only gives a detailed account of the “ plant formations ”’ which make up the flora of the islands visited, but also contributes a list of the flowering-plants and ferns, and a sketch of the physiography, geology, climate, &c. The greater portion of the paper is, of course, purely ecological, and will no doubt be amplified and extended in the memoir which I understand Dr. Cockayne is to contribute to the series constituting this volume ; but, quite apart from that portion of the subject, the paper is a most important and accurate contri- bution to our knowledge of the botany of the various groups of islands, and must be carefully studied by any one engaged in the investigation of their vegetation. In the above remarks I have dealt with the history of botanical exploration in the southern islands prior to the departure of the expedition to illustrate whose work this series of memoirs has been projected. I understand that the reasons for organ- izing the expedition, its composition, and the history of its proceedings will all be fully dealt with in another part of this volume. Nevertheless, I think it incumbent upon me to particularise the various collections placed in my hands. These are as follows :— 1. The collections made by Dr. Cockayne on the Auckland Islands during the visit of the expedition ; also a more extensive series obtained by him on the Auck- land, Campbell, and Antipodes Islands during his previous visit of June, 1903. 2. An extensive series of specimens collected by Mr. B. C. Aston; all, with the exception of a few plants from the Snares, obtained on the Auckland Islands. 3. A similar collection, also formed on the Auckland Islands, made by Mr. J. 8. Tennant, of Wellington. 4. A series of specimens, partly dried and partly preserved in formalin, obtained by Mr. R. M. Laing on Campbell Island. 5. A few specimens in formalin obtained on Campbell Island by Messrs. Chambers and Des Barres. 6. A packet of dried specimens collected on the Auckland Islands by Mr. F. R. Fieid. Tn addition to the above, I am indebted to Mr. Petrie for the loan of the southern- island specimens contained in his herbarium. Mr. Kirk’s collections, mostly formed in 1890, and some of Mr. Buchanan’s, collected in 1883, I had full opportunities of studying when engaged in the preparation of my “ Manual of the New Zealand Flora.” Finally, I have compared all the above with the specimens in my own herbarium, obtained from various sources during the previous thirty years. It is, perhaps, necessary to say that, with the view of making the following enumeration as complete as possible, I have included all the species that have been recorded on reliable authority from the islands, whether gathered by the expedition or not. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 397 Ill. ENUMERATION OF THE PHANEROGAMS, FERNS, AND LYCOPODIACEAE HITHERTO COLLECTED IN THE ISLANDS TO THE SOUTH OF NEW ZEALAND, INCLUDING THE SNARES, AUCKLAND AND CAMPBELL ISLANDS, ANTIPODES ISLAND, AND MACQUARIE ISLAND. RANUNCULACEAE. Ranunculus pinguis, Hook. f. Ranunculus pinguis, Hook. f., Fl. Antarct., 1, 3, t. 1 (1844); R. pingues, var. b, Hook. f., Handb. N.Z. FI., 5 (1864). Auckland Islands: Not uncommon throughout the whole of the group, from sea-level to an elevation of 1,400 ft.; Hooker, Kirk! Cockayne! Aston ! Tennant! Campbell Island: From sea-level to 1,200 ft.; Hooker, Filhol! Kirk! (Endemic. ) Varies greatly in size, attaiming its greatest luxuriance in sheltered hollows on peaty soil. Specimens collected by Dr. Cockayne on the hills above Camp Cove, Carnley Harbour, are quite 14in. high, with a stout 2-flowered scape, and the leaves are considerably more than 2 in. across. On the other hand, I possess two specimens gathered by Kirk in Campbell Island which barely exceed 14 in. in total height, and in which the blade of the leaf is not more than $in. diam., ovate-rhomboid or cuneate-rhomboid in outline, with 2-3 coarse teeth at the tip. This form is evi- dently identical with Hooker’s variety rhombifolius (“ Flora Antarctica,” 3), but it cannot be looked upon as being more than a depauperated state induced by exposure, &c. In the “ Handbook of the New Zealand Flora” Hooker united R. pinguis with the New Zealand R. Monroi; but that species differs altogether in the more con- stantly branched scapes, in the more numerous flowers, in the much larger petals, which are always considerably longer than the sepals, and in the more slender style. In the “ Manual of the New Zealand Flora” I have therefore followed Mr. Kirk in reaffirming the distinctness of the two plants. Ranunculus subscaposus, Hook. f. Ranunculus subscaposus, Hook. f., Fl. Antarct., i, 5 (1844). Campbell Island: Banks of streams, not common; Lyall, Filhol! Rathouis! Kirk! Laing! (Endemic.) This is very closely allied to the following species ; in fact, I am not at all sure that the two are really distinct. Its distinguishing characters, such as they are, lie mainly in the ovate-deltoid leaves, which are usually 3-partite to the base, the seg- ments or leaflets being acutely 3-5-lobed. In R. aucklandicus the leaves are broader and more cordate, with a reniform or rounded outline, and are seldom completely 3-partite, and the segments are more obtusely lobed. Fruiting specimens of R. subscaposus gathered by Kirk show the achenes to be ovate, compressed, and dis- tinctly margined, and the style is short and only very slightly hooked. According to Asa Gray, the achenes of R. aucklandicus are not margined ; but, as none of the specimens obtained of late years has ripe fruit, I cannot say whether this statement is strictly correct. 398 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. Ranunculus aucklandicus, A. Gray. Ranunculus aucklandicus, A. Gray, Bot. U.S. Expl. Exped., i, 8 (1854); Hook. {., Handb. N.Z. Fl., 723 (1867). R. Hectori, T. Kirk, Students’ Fl., 16 (1899). Auckland Islands: Usually in wet ground near the sea; U.S. Expl. Exped., Kirk! Cockayne! Aston! Tennant! (Hndemic.) Half-ripe fruiting specimens collected by Aston show that the achenes are villous in the young state. Further study of Kirk’s &. Hectori has convinced me that it can only be regarded as a trivial form of R. aucklandicus, and I have therefore rele- gated 1t to that species. Ranunculus acaulis, Banks & Sol. Ranunculus acaulis, Banks & Sol. ex D.C. Syst., 1, 270 (1818); Hook. f., Fl. Antarct., 1, 4, t. 2 (1844). Auckland Islands: Moist places near the sea; Hooker, Kirk, Cockayne, Aston! Tennant! (Muddy beaches and sandy or gravelly shores, from the North Cape to Stewart Island; also in the Chatham Islands. Very closely allied to the Chilan R. stenopetalus, Hook., of which it is probably only a recent derivate form.) The specimens submitted to me are by no means good, but are apparently identi- cal with the common New Zealand state. Hooker remarks that his specimens “ en- tirely agree with others gathered in the Bay of Islands.” &. acaulis is the only species of Ranunculus which stretches through the whole length of the Dominion, from the North Cape to the Auckland Islands; it has thus a latitudinal range of nearly a thousand miles. Ranunculus biternatus, Smith. Ranunculus biternatus, Smith in Rees’s Cyclop., xxix, n. 48 (1802) ; Wildem., Res. Voy. “ Belgica,” 89, t. 12, f. 12-2 (1906). &. erassipes, Hook. f., Fl. Antarct., 11, 224, t. 81 (1844); A. Hamilton in Trans. N.Z. Inst., XXvll, 566 (1895). Macquarie Island: Swampy ground near the sea, usually in the shelter of the tussocks of Poa foliosa; A. Hamilton! (Fuegia; Falkland Islands; South Georgia ; Kerguelen Island; Crozets; Marion Island; New Amsterdam.) I have followed Dr. Schenk in uniting the Macquarie and Kerguelen R&. crassipes with the Fuegian R. biternatus. R. crassvpes was originally distinguished on account of its greater size and less-divided leaves; but these are characters of small im- portance, and Mr. Hamilton’s Macquarie Island specimens have the leaves almost as much divided as those of a Fuegian specimen figured as R. biternatus in the “ Botany of the Voyage of the ‘ Belgica.’ ” CRUCIFERAE. Cardamine glacialis, D.C., var. subcarnosa, O. R. Schulz. Cardamine lirsuta, Linn., Sp. Plant, 655, var. subcarnosa, Hook. f., FI. Antarct., 1, 5 (1844). C. glacialis, D.C., Syst., 1, 268, var. subcarnosa, O. R. Schulz in Engl. Jahr., xxx, 542 (1903). Stout, succulent, glabrous or nearly so, variable in size, 15-12 cm. high. Rhizome perennial, stout, spongy, often branched at the top, each division crowned Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 399 with a rosette of densely crowded radical leaves and with a short and stout flowering- stem. Radical leaves 1-6 cm. long; petiole half the length or nearly so, broadly sheathing at the base; leaflets 1-5 pairs, with a terminal one ; terminal leaflet much the largest, 5-15 mm. diam., entire or angularly sinuate-repand or -lobed ; lateral similar but usually smaller, sessile or shortly petiolulate. Cauline leaves few, smaller and on shorter petioles; leaflets fewer and smaller. Racemes short and dense, corymbose, probably elongating in fruit. Flowers usually numerous, rather large, 3-5 mm. diam., white, pink, or purple. Sepals sometimes purplish at the tips. Stigma broad, almost sessile. Ripe fruit not seen; in an immature state crowded, very short and broad ; style almost wanting. Auckland and Campbell Islands: Not uncommon in stony débris or in moist grassy places; Hooker, Kirk, Cockayne! Aston! Tennant! Laing! Macquarie Island: Swampy places near the coast; A. Hamilton. (Southern Chili; Fuegia ; Falkland Islands.) But few specimens of this plant were available during the preparation of my Manual, and most of them were in poor condition. I therefore followed Sir J. D.. Hooker in placing it under C. hirsuta. A much more complete series, mainly col- lected by Mr. Aston, has now fully convinced me of its distinctness. It differs from C. hirsuta in the stout and fleshy habit, perennial rhizome, large often coloured flowers, short and broad often crowded pods, and almost sessile stigma. It is much more Closely allied to the South American C. glacialis, to which it was united by Schulz in his recent monograph of the genus, and it appears best to follow this arrangement until a leisurely comparison of specimens can be made. Cardamine hirsuta, Linn., var. corymbosa, Hook. f. Cardamine corymbosa, Hook. f., Ic. Plant., t. 686; Fl. Antarct., 1, 6 (1844) ; Schulz in Engl. Jahr., xxx, 561 (1903). C. horsuta, Linn., Sp. Plant., 655, var. corymbosa, Hook. f., Handb. N.Z. Fl., 12 (1864). Auckland Islands: Port Ross; Tennant! locality not stated; Cockayne. Camp- bell Island: Turfy ground near the sea, common; Hooker, Lamg! (New Zealand, from the Tararua Range northwards; Orange Harbour, Fuegia.) This was first treated by Hooker as a distinct species, but at a later period he reduced it to the position of a variety of the northern C. hirsuta, including with it some mountain forms from New Zealand. Schulz, in his monograph of the genus, restored it to specific rank, limiting it to the original Campbell Island plant. For the present I have followed Hooker’s arrangement, although I consider it probable that most, if not all, of the southern forms previously referred to C. hirsuta will ultimately be separated from that plant. Schulz describes the flowers of var. corymbosa as being apetalous. But this is clearly erroneous, for not only are the petals shown in the plate given in the “ Icones Plantarum,” and are described in the original diagnosis, but they are also present in all the flowering specimens I have seen. In addition to the ordinary inflorescence, var. corymbosa produces very curious cleistogamic flowers. These were overlooked until quite recently, when they were detected by Dr. Schulz during an examination of some of Hooker’s specimens, col- lected nearly seventy years ago. I failed to find any traces of them in Mr. Laing’s 400 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. specimens, but a single example obtained by Mr. Tennant at Port Ross showed two or three in excellent condition. They are developed singly at the tip of a short peduncle springing from the axil of one of the radical leaves, and are so small as to be only just visible to the unaided eye. ‘There are 4 minute scale-like sepals, no petals, and 4 small stamens, the filaments of which are barely as long as the ovate-cordate anthers. Schulz states that there is only one stamen, but in two flowers examined by me both had four. There is a short conical 2-celled ovary, each cell contaiming 2-6 ovules, of which (according to Schulz) only one ripens. Aitter fertilisation the peduncle elongates and curves downwards, ultimately burying the ripening pod in the soil. Similar cleistogamic flowers are known in the South American C. cheno- podifolia, and have been described with considerable detail by Grisebach and other authors. Mr. Hamilton, in his notes of a visit to Macquarie Island, records var. corymbosa as a native of that island. Ihave not seen his specimens, but I understand that they were subsequently identified by Mr. Kirk as C. hirsuta, var. subcarnosa, referred to C. glacialis in this memoir. For the Fuegian locality 1 am indebted to Wildemann’s “Results of the Voyage of the ‘ Belgica,’” p. 98. Cardamine depressa, Hook. f Cardamine depressa, Hook. f., Fl. Antarct., 1, 6, t. 3 and 4B (1844). The Snares: Kirk. Auckland Islands: From sea-level to 1,500 ft.; Hooker, Kirk ! Cockayne! Aston! Tennant! Campbell Island: Cockayne. (New Zealand : in mountain districts from Nelson southwards.) Cardamine depressa, Hook. f., var. stellata. Cardamine stellata, Hook. f., Fl. Antarct., 1, 7, t. 44 (1844); Schulz in Engl. Jahr., xxxii, 560 (1903). C. depressa, var. stellata, Hook. f., lemon Nea a2. Campbell Island: Usually on rocky débris ; Hooker, Kirk ! Cockayne ! Laing! (Endemic.) Schulz considers that both the forms of C. depressa are entitled to specific rank, and he also excludes the New Zealand mountain forms associated with it by Hooker; but specimens which I have gathered in the mountains of Nelson and Canterbury come so near to the type that I feel compelled to keep them together. Dr. Schulz omits all mention of these closely allied New Zealand plants in his monograph. Lepidium oleraceum, Forst Lepidium oleracewm, Forst., Prod., n. 248 (1786); Hook. f., Fl. Nov. Zel., i, 15 (1853). The Snares: Rocky places on the cliffs; Kirk, Aston! Auckland Islands: Exact locality not stated; General Bolton, Kirk! (ex Herb. Petrie), Tennant! (New Zealand; rocky places near the sea, from the Three Kings Islands south wards.) Mr. Aston’s specimens from the Snares have long narrow linear leaves, thus presenting a somewhat different appearance to the ordinary state of the species. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 401 CARYOPHYLLACEAE. Stellaria decipiens, Hook. f. Stellarva decipiens, Hook. f., Fl. Antarct., 1, 7 (1844); Ic. Plant., t. 680 (1845). Auckland and Campbell Islands: Not uncommon in the lower portions of the islands, especially in the woods and near the sea; Hooker, Kirk! Chapman! and many others. Macquarie Island: A. Hamilton. (Hndemic.) Stellaria decipiens, Hook. f., var. angustata, T. Kirk. Stellarva decuprens, Hook. f., var. angustata, T. Kirk, Students’ FI., 57 (1899). Antipodes Island: Tussock meadow, not uncommon; Kirk! Chapman! Cockayne! (Hndemic.) Hooker compares the typical state of S. decupiens with the naturalised S. media, which differs in having a pubescent line down each side of the branches, and in the more distinctly cymose inflorescence. Extreme states of the variety angustata have a very distinct appearance, but I have seen specimens collected by Kirk in which the leaves are broader and shorter, thus approaching the ordinary form. Colobanthus muscoides, Hook. f. Colobanthus muscoides, Hook. f., Fl. Antarct., 1, 14 (1844); Homb. & Jacq., “ Voyage au Pole sud,” Bot., t. 17 (1853). Snares: Confined to a patch of swampy ground; Kirk! Auckland, Camp- bell, and Antipodes Islands: Maritime rocks and cliffs, abundant; Hooker and all collectors. Macquarie Island: Rocky beaches and cliffs; A. Hamilton. (Endemic.) This 1s evidently an abundant plant on maritime rocks and cliffs, forming large dense rounded cushions 3-18 in. diam., or even more. Mr. Kirk remarks that these cushions consist of “the partially decomposed stems and leaves of old plants and the roots of young plants. The seeds often germinate in the capsule, and it is no uncommon thing to find capsules still attached to the stem and with apparently perfect seeds imbedded some 3 in. or 41n. below the surface of the mass, the old surface having been covered with a growth of young plants too quickly to allow of the germination of the buried seeds.” Colobanthus Billardieri, Fenzl Colobanthus Billardierr, Fenzl m Ann. Wien Mus., 1, 49 (1836); Hook. £., Fl. Antarct., i, 14 (1844). Auckland Islands: Carnley Harbour; Kirk! Campbell Island: Banks near the sea; Hooker, Kirk, Laing! Antipodes Island: Kirk. Macquarie Island: Hamilton. (New Zealand, from the East Cape southwards; Victoria; Tasmania.) Although this species has been recorded from the whole of the southern islands except the Snares, it does not seem to be at all abundant, and I have seen but few 27S. 402 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. specimens. They all belong to an unusually small state, barely more than Jin. high, but they have the grassy and almost flaccid leaves of the type. Colobanthus subulatus, Hook. f. Sagina subulata, D’Urv., Fl. Malouines in Mem. Soc. Linn. Paris, iv, 617 (1826). Colobanthus Benthamianus, Fenzl in Ann. Wien Mus., i, 49 (1836). C. subulatus, Hook. f., Fl. Antarct., 1, 13 (1844). Auckland Islands: Exact locality not stated; Aston! Campbell Island: Rocky places near the summits of the hills; Hooker, Kirk! (Victoria; Fuegia; Falkland Islands ; South Georgia.) Very closely allied to the New Zealand C. acicularis, which differs in the much longer acicular points to the leaves, and in the much longer and narrower sepals, which are lmear-subulate and considerably exceed the capsule, whereas in C. subulatus they are about the same length. The New Zealand plants referred by Hooker to C. subulatus are in all probability small forms of C. acicularis. The plate of C. subulatus given in the “ Flora Antarctica,” based on Falkland Island specimens, represents a plant with a much more lax habit than the Auckland Island or Campbell Island species, and the sepals are only four in number instead of five. It appears to me that the specific identity of the two plants is open to doubt; but, as I have not seen American specimens, I leave the question in abeyance for the present. In the Manual I adopted the specific name of Benthamianus, that being the oldest name under the genus Colobanthus. But now that the Vienna Congress has definitely ruled that the oldest name under any genus must take precedence it will be necessary to go back to the name subulatus, which was published by D’Urville (under Sagina) in 1826, whereas Benthamianus did not appear until 1836. PORTULACACEAE. Montia fontana, Linn. Montia fontana, Linn., Sp. Plant., 87 (1753); Hook. f., Fl. Antarct., 1, 13 (1844). Auckland and Campbell Islands, Antipodes Island, Macquarie Island: Moist places near the sea; abundant. (Generally distributed throughout the North and South Temperate Zones.) GERANIACEAE. Geranium microphyllum, Hook. f. Geranwum microphyllum, Hook. f., Fl. Antarct., 1, 8, t. 5 (1844). Auckland and Campbell Islands: Not uncommon in moist places; Hooker, Kirk! Aston! Laing! (Throughout New Zealand ; and very close to an Australian ee from Hooker’s plate and from the few specimens I have seen, the Auck- land and Campbell Islands plant is smaller in size and has a more slender habit than is usual in New Zealand. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 403 ROSACEAE. Geum albiflorum, Cheesem., n. comb. Sreversia albiflora, Hook. f., Fl. Antarct., 1, 9, t. 7 (1844). Geum parvi- florum, Hook. f., Handb. N.Z. Fl., 55 (1864), so far as the Auckland Island locality is concerned, but not of Smith. G. aucklandicum, Greene, Pittonia, iv, 225 (1895). G. sericeum, T. Kirk, Students’ FI., 128 (1899). Auckland Islands: Rocky places on the hills; Hooker: highest part of Adams Island, alt. 1,000-2,000 ft.; Kirk, Aston! Tennant! (Endemic.) I much regret that the few specimens I have seen of this interesting plant are in poor condition, and out of flower, so that 1 am unable to add anything to the descriptions given by Hooker and Kirk. The foliage looks very different to that of G. parviflorum, partly owing to the upper surface being much more densely villous with long soft hairs, and to the very prominent veins of the under-suriace, which give it a strongly rugose appearance. Kirk also mentions as a diagnostic character of importance that the style is much shorter than the ripe achene, whereas it is much longer in G. parviflorum. It is not at all clear why both Greene and Kirk, in recognising that the species was distinct from parviflorum, should have passed over Hooker’s original name of albiflorum. As it is obviously entitled to precedence, I have adopted it here. Acaena Sanguisorbae, Vahl. Acaena Sanguisorbae, Vahl., Hnum., 1, 294 (1804); Hook. f., Fl. Antarct., i, 9 (1844). Auckland and Campbell Islands, Antipodes Island, Macquarie Island. (Throughout New Zealand; Australia; Tristan d’Acunha; Amsterdam Island.) Although this species has been recorded from the whole of the southern islands, it is not quite clear that the typical state is indigenous in any one of them, the previous records probably referring to the var. antarctica. Dr. Cockayne informs me that he has collected it only at Port Ross, where it is probably introduced. Acaena Sanguisorbae, Vahl., var. antarctica, Cockayne. Acaena Sanguisorbae, var. antarctica, Cockayne in Trans. N.Z., Inst., xxxvi, 319 (1904). Auckland, Campbell, and Antipodes Islands: Abundant throughout. (Hn- demic.) This is the common Acaena of the southern islands, and the one to which most of the previous records refer. Acaena adscendens, Vahl. Acaena adscendens, Vahl., Enum., i, 297 (1804); Hook. f., Fl. Antarct., i, 10 (1844), and ii, 268, t. 96 (1847). Macquarie Island: Near the sea, Fraser, Scott, Hamilton. (New Zealand ; Fuegia; Falkland Islands; Kerguelen Island; Crozets; Marion Island; South Georgia.) 404 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. This 1s one of the few species found on nearly all the widely scattered islands— specks in a vast ocean—which girdle the globe between lat. 50° and 60° S., although, curiously enough, it does not occur on the Auckland and Campbell Islands. Hooker associated with it a plant not uncommon in subalpine localities in the South Island, but a comparison with specimens from Macquarie and Kerguelen Island show several important differences, and I suspect that the New Zealand plant will constitute a separate species. CRASSULACEAE. Tillaea moschata, D.C. Crassula moschata, Forst. in Comm. Gott., 1x, 26 (1789). Bulliarda mos- chata, D’Urv., Fl. Malouines in Mem. Soc. Linn. Paris, iv, 618 (1826) ; Hook. f., Fl. Antarct., 1, 15 (1844). Tllaea moschata, D.C., Prodr., iii, 382 (1828). The Snares: Kirk, Aston. Auckland and Campbell Islands: Maritime rocks, abundant ; Hooker, Kirk! Cockayne, Laing! Aston! Tennant! Antipodes Island : Kirk. Macquarie Island: Maritime rocks; Scott, Hamilton. (New Zealand, from Cook Strait southwards; Fuegia; Falkland Islands; Kerguelen Island; Marion Island.) One of the commonest plants of the islands, almost invariably present on mari- time rocks and cliffs. DROSERACEAE. Drosera stenopetala, Hook. f. Drosera sp., Hook. f., Fl. Antarct., 1, 8 (1844). D. stenopetala, Hook. f., Fl. Nov. Zel., 1, 19, t. 9 (1853). Auckland Islands: Hooker, Le Guillon (“ Flora Antarctica,” 1, 246), Kirk, Cockayne, Tennant! Campbell Island: Peat bogs; Laing! (Mountain districts in New Zealand, from the Ruahine Range southwards.) My specimens are all immature, and show no trace of flowers, but are apparently referable to this species. HALORAGACEAE. Haloragis micrantha, R. Br. ex Sieb. & Zucc., Fl. Jap. Haloragis micrantha, R. Br. ex Sieb. & Zucc., Fl. Jap. i. 25 (1815); Hook. f., Handb. N.Z., Fl., 66 (1864). Auckland Islands: Head of Norman’s Inlet; Laing! (Throughout New Zea- land ; eastern Australia, and Tasmania ; Malaya; China; Japan.) This 1s an addition to the florula of the islands. The specimens are mere scraps, but there seems to be no reason to doubt their identity with the well-known H. macrantha. Callitriche antarctica, Hngelm. Callitriche verna, Hook. f., Fl. Antarct., 1, 11, non Linn. ; also, var. terrestris, lc. (1844). C. antarctica, Engelm. ex Hegelm. in Verh. Bot. Ver. Brand,, ix, 20 (1867). Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 405 The Snares: Kirk! Auckland and Campbell Islands: Common on the ground and on wet rocks near the sea; Hooker, Kirk! Cockayne, Tennant! Laing! Anti- podes Island: Wet rocks; Kirk, Cockayne. Macquarie Island: A. Hamilton. (Fuegia; Falkland Islands; Kerguelen Island; Marion Island; Heard Island ; South Georgia.) Hooker remarks (Phil. Trans. Roy. Soc., 168, p. 20) that two forms of this species are found in Kerguelen Island and elsewhere—one aquatic, with long stems and pro- portionally large spathulate leaves; the other smaller, terrestrial, suberect, with obovate or oblong leaves. To this latter form he applies the name of var. terrestris in the “ Flora Antarctica.” Specimens of the larger form collected by Kirk on the Snares and on the Auckland Islands bear abundance of ripe fruit, which has the rounded angles characteristic of the species. The smaller state I have seen only in young flower. MYRTACEAE. Metrosideros lucida, A. Rich. Melaleuca lucida, Forst., Prodr., n. 216 (1786). Metrosideros lucida, A. Rich, Hi Nov. Zel. 333 (1832) > Hook i, KI. Antarct., 1, 12 (11844), Agalmanthus umbellatus, Homb. & Jacq., Voy. Astrol. et Zél., 78, t. 1 (1845). Auckland Islands: Abundant from sea-level up to 500 ft.; Hooker and all subsequent collectors. Campbell Island: Extremely rare; Kirk. (New Zealand, from Whangarei and the Great Barrier Island southwards.) The most abundant tree on the Auckland Islands, skirting almost the whole of the eastern shore with a broad belt of forest. According to Hooker, it reaches a height of from 20 ft. to 40 ft., with a trunk 2-3 ft. in diameter. It is seldom erect ; usually the trunks are inclined or almost prostrate at the base, giving off many naked more or less gnarled and distorted branches. Generally speaking, the branch- lets of the Auckland Islands plant are stouter and the leaves more coriaceous than is the case with specimens from the mainland of New Zealand. There is also a considerable range of variation in the size and shape of the leaves. I have quoted the Campbell Island locality on the authority of Mr. Kirk, who states that “‘ it is extremely rare and of very low stature.” It does not seem to have been observed by any subsequent botanist. Mr. Laing has expressly informed me that he searched for it In vain. Sir Joseph Hooker was the first to draw attention to the high latitudes reached by Myrtaceous plants in the Southern Hemisphere. In Auckland Island Metro- sideros lucida attains the latitude of 52° 8., while in South America Myrteola num- mularia stretches through Fuegia almost as far south as Cape Horn (S. lat. 56°), and is also found in the Falkland Islands. The monotypic genus Tepualia, first discovered in southern Chili, and so closely allied to Metrosideros as to be included in that genus by Hooker (Fl. Antarct., 1, 275), is now known to extend along the south-western coast of South America as far south as Desolation Island (S. lat. 53°), where it was collected by Dusén, the botanist of the Swedish Magellanic Expedition. 406 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Systematic Botany. ONAGRACEAE. Epilobium confertifolium, Hook. f. Epilobium confertrfolium, Hook. f., Ic. Plant., t. 685 (1844); Fl. Antarct., 1, 10 (1844); Haussk., Monog. Epilob., 295 (1884). Auckland and Campbell Islands: Meadows and open slopes from sea-level to 1,000 ft. alt.; Hooker, Kirk! Chapman! Aston! Tennant! Lamg! Chambers! Antipodes Island: Kirk, Cockayne. (Hndemic.) Although the specimens brought back by the expedition are not so complete as could be wished, not one of them showing perfectly ripe fruit, they are sufficient to prove that the species is quite distinct from the New Zealand mountain-plants associated with it by Hooker. Its distinguishing characters, m the limited sense in which I now understand it, are the creeping and rooting often densely matted stems, the young branches alone rising from the ground ; the densely crowded pale- green and almost fleshy leaves, which are almost sessile, obovate or obovate-oblong, entire or remotely and obscurely denticulate, the lower ones more or less distichously placed, the upper spirally arranged; the few almost sessile flowers, placed at the tips of the branchlets, and apparently always of a bright-pink colour; and the strict and erect perfectly glabrous capsules, the peduncles of which only slightly elongate after the flowering period. Epilobium alsinoides, A. Cunn. ? Epilobvum alsinoides, A. Cunn., Precur., n. 540 (1839); Hook. f., Fl. Nov. Zel., i, 59 (1853). Antipodes Island: Kirk, Cockayne! (In New Zealand the species ranges from the North Cape to Stewart Island.) This plant was originally discovered by Mr. Kirk, who referred it to H. alsinordes. I have seen only a few flowerless specimens collected by Dr. Cockayne. These agree with H. alsinoides in the fine even pubescence of the stems, but differ so much in habit and general appearance that I am compelled to regard Kirk’s identification as highly doubtful. Flowering specimens are required to safely determine the systematic position of the plant. Epilobium linnaeoides, Hook. f. Epilobium linnaeordes, Hook. f., Fl. Antarct., 1, 10, t. 6 (1844); Haussk., Monog. Epilob., 301 (1884). Auckland and Campbell Islands: Abundant in woods, in scrub, and in open tussock ; Hooker and subsequent collectors. Antipodes Island: Kirk, Cockayne. Macquarie Island: A. Hamilton. (New Zealand, in damp mountainous localities from the Ruahine Mountains southwards.) A well-marked and easily recognised species, excellently figured by Hooker in the “ Flora Antarctica.” Systematic Botany. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 407 Epilobium nummularifolium, R. Cunn., var. nerterioides, Hook. f. Epilobium nerteriordes, A. Cunn., Precur., n. 541 (1839); Hook. f., Fl. Antarct., 1, 11 (1844). #. nwnmularifolium, R. Cunn., var. nerteriordes, Hook. f., Fl. Nov. Zel., 1, 57 (1853). #. pedunculare, A. Cunn., Precur., n. 536; var. aprica, Haussk., Monog. Epilob., 303 (1884). Auckland Islands: Moist rocky places on the hills; Hooker, Homb. & Jacq., Kirk, Tennant! Macquarie Island: In swamps; A. Hamilton. (New Zealand, from the North Cape southwards.) The specimens gathered by the expedition entirely agree with the common New Zealand state. UMBELLIFERAE. Azorella Selago, Hook. f. Azorella Selago, Hook. f., Fl. Antarct., u, 284, t. 99 (1847). Macquarie Island: Sides of the hills, forming prominent globular masses often 4 ft. im diameter; Fraser, Professor Scott! A. Hamilton! (Fuegia; Kerguelen Island ; Crozets; Marion Island ; Heard Island.) Forms large convex masses several feet in diameter, composed of the closely packed stems and branches of the plant ; the young shoots, with their leaves, form- ing a bright-green smooth and even crust or surface, so densely compacted as to bear the weight of a man. Under this living surface, as Sir Joseph Hooker remarks, lies ““a vast mound of débris, the remains of former years’ growth, through which the roots descend to the ground.”’ According to Dr. Scott, it is comparatively rare in Macquarie Island; but in Kerguelen Island it is the most abundant plant, often forming sheets or hummocks investing the soil for acres in extent with an elastic green coating. Its apparent absence in the Auckland and Campbell Islands is a very curious fact, in no way easy to account for. Azorella reniformis, Hook. f. Pozoa reniformis, Hook. f., Fl. Antarct., 1, 15, t. 11 (1844). Azorella reni- formas, Benth. & Hook. f. in Gen. Plant., 1, 875 (1865). Auckland and Campbell Islands: Rocky places on the hills; Hooker, Kirk ! Cockayne, Aston! Tennant! (Hndemic.) The New Zealand A. Haastw is so close to this that it is highly doubtful if it can be maintained as a distinct species, for the characters of the ciliate stipules and long pedicels to the fruit, relied upon by Hooker, break down when large suites of Specimens are under examination. A. Haastw, however, attains a much larger size than A. reniformis. Apium prostratum, Lab. Apium prostratum, Lab., Relat., 1, 141 (1804). A. australe, Thouars, F). Tristan d’Acugn., 43 (1811); Hook. f., Fl. Nov. Zel., i, 86 (1853). Antipodes Island: Maritime rocks, abundant; Kirk, Cockayne. (Shores of New Zealand, from the Kermadec Islands to Stewart Island ; Australia, from Queens- 408 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. land to Tasmania, and westward to Swan River; south Chili; Fuegia and the Falkland Islands; Tristan d’Acunha; St. Paul’s; Amsterdam Island; South Africa.) So closely allied to the common celery (Apiwm graveolens, Linn.) that it was treated as a form of that species by Hooker (FI. Antarct., 11, 287), although he sub- sequently regarded it as distinct in the “ Flora Novae Zealandiae.” Wildemann, in the recently published “ Botany of the Voyage of the ‘ Belgica,” has again re- ferred it to A. graveolens (p. 134). It differs, however, as has been pointed out by Bentham and other botanists, in the prostrate or decumbent habit, and especially in the thick and prominent almost corky ribs of the fruit; and I think it is best kept as a distinct though closely allied species. Ligusticum latifolium, Hook. f. Amisotome latifolia, Hook. f., Fl. Antarct., i, 16, t. 8, 9, and 10, B (1844). Calosciadium latifolwm, Endl. ex Walp. Ann., 1, 702 (1850). Lagus- ticum latifoluum, Hook. f., Handb. N.Z. FI., 94 (1864). Auckland and Campbell Islands: From sea-level to the tops of the hills, abundant ; Hooker and all subsequent collectors. (Hndemic.) A noble species, excellently figured and described by Hooker in the “ Flera Antarctica.” He remarks that “it attains a height of 6 ft., bearmg several umbels of rose-coloured or purplish flowers, each compound umbel as large as the human head. The foliage is of a deep shining green, and the whole plant emits, when bruised, an aromatic smell.” In the “ Genera Plantarum ” Mr. Bentham referred to Acvphylla all the plants from the Southern Hemisphere placed by Hooker in Lagusticwm. No doubt he was influenced by the fact that the floral and carpological characters of the two groups are very similar. But, on the other hand, the habit of the true Acvphyllae, with their rigid and coriaceous spinous leaves, and paniculate inflorescence with spinous bracts, is most distinct, and to me certainly appears to justify their separation as a distinct genus, which would really be as well characterized as many genera of Umbelliferae accepted by most botanists. If it be necessary to limit Lagustiewm to the forms found in the Northern Hemisphere (and there is much to be said in favour of such a course), then it would be far more preferable to revive Hooker’s genus Amisotome for the southern species than to transfer them to Acrphylla. Ligusticum antipodum, Homb. & Jacq. Ligusticum antipodum, Homb. & Jacq., Bot. Voy. Astrol. et Zel., 63, t. 3, Dicot. (1853). Anzsotome antipoda, Hook. f., Fl. Antarct., 1, 17, t. 9, 10 (1844). Calosciadiwm antipodum, Endl. ex Walp. Ann., 11 (1850). Auckland and Campbell Islands, Antipodes Island: From sea-level to the tops of the hills, abundant ; Hooker and all subsequent collectors. (Hndemic.) Almost as striking a plant as the preceding, from which it is easily distimguished by the finely divided leaves with numerous linear segments. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 409 Ligusticum acutifolium, T. Kirk. Ligusticum acutifoluum, T. Kirk in Journ. Bot. (1891), 237; Students’ FI., 201 (1899). The Snares: Rare and local; Kirk! Cockayne! (Hndemic.) This must be considered as one of the rarest plants in the florula. Mr. Kirk states that it was seen only in one place, at an altitude of about 350 tt. ; and Dr. Cockayne informs me that, although special search was made, only one plant was noticed by the recent expedition. As a species, it is very close to L. intermedium, of which it must be considered a recent derivate form. Its chief differential charac- ters, according to Kirk, are the total absence of the viscid milky juice so evident in L. intermedium ; 1m the upper part of the sheath at the base of the petiole being free, forming a hooded ligule; in the more acute segments of the leaves; and in the smaller umbels and shorter fruits. I have seen specimens of the foliage only. ARALIACEAE. Stilbocarpa polaris, A. Gray. Aralia polaris, Homb. & Jacq. MS., ex Hook. f., Fl. Antarct., 1, 19 (1844) ; Hook. f., Ic. Plant., t. 701; Homb. & Jacq., Bot. Voy. Astrol. et Zel., 55, t. 2, Dicot. (1852). Stedbocarpa polaris, A. Gray, Bot. U.S. Expl. Exped., 714 (1854). Auckland and Campbell Islands: Abundant from sea-level to a considerable height on the hills; Hooker and all subsequent collectors. Antipodes Island: Plentiful over the greater portion of the island ; Kirk, Cockayne. Macquarie Island : Abundant over the whole of the island; Scott, Hamilton. (HKndemic.) I have nothing to add to the excellent account of this most striking plant pub- lished by Hooker in the “ Flora Antarctica,” and to the still more elaborate descrip- tion given by Decaisne in the “ Voyage au Pole sud.” Hooker well remarks that it is “ one of the most handsome and singular of the vegetable productions in the group of islands it inhabits, which certainly contains a greater proportion of large and beautiful plants, relatively to the whole vegetation, than any country with which IT am acquainted. Growing in large orbicular masses, on rocks and banks near the sea, or amongst the dense and gloomy vegetation of the woods, its copious bright- ereen foliage and large umbels of waxy flowers, often nearly a foot in diameter, have a most striking appearance.” Stilbocarpa Lyallii, Armst., var. robusta, T. Kirk. Stilbocarpa Lyall, Armst. in Trans. N.Z. Inst., xii, 336 (1881), in part. Aralia Lyallu, T. Kirk in Trans. N.Z. Inst., xvii (1885), 295, var. robusta, Journ. Bot., xxix, 237 (1895). The Snares: Not uncommon; Kirk! Cockayne, and other members of the expedition. (Endemic.) I very much regret that my acquaintance with this fine plant is confined to a single imperfect specimen collected by Kirk, no example having been brought back by the members of the expedition. Kirk states that it differs from the type in being 410 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. more robust and less hispid, and in wanting the curious arcuate stolons. The leaves are larger and have the teeth strongly mucronate, and the flowers are smaller, with yellowish petals. I suspect that it will constitute a separate species, but at present I have not the material from which to form a satisfactory diagnosis. The late Mr. Kirk removed Stilbocarpa Lyall to Aralia, mainly on account of the two-celled ovary and fruit not hollowed at the apex. But in other characters it approaches very closely to Stedbocarpa polaris, and | now think that Harms was right (“ Natirlichen Pflanzenfamilien,” teil mi, abt. 7, 57) m extending the characters of Stilbocarpa so as to receive both plants. From the point of view of the plant- geography of the southern portions of New Zealand, it is a decided gain to be able to treat the two species as constituting a single endemic genus. Panax simplex, Forst. Panax simplex, Forst., Prodr., n. 399 (1786); Hook. f., Fl. Antarct., 1, 18, t. 12 (1844). Nothopanax simplex, Seem. in Journ. Bot., iv, 296 (1866). Auckland Islands: Abundant in woods, from sea-level to an altitude of about 500 ft.; Hooker and all subsequent botanists. (New Zealand, in forests from Cape Colville southwards.) Auckland Island specimens agree fairly well with New Zealand examples of this well-known plant. RUBIACEAE. Coprosma parviflora, Hook. f. Coprosma myrtillifolia, Hook. f., Fl. Antarct., 1, 21 (1844). C. parviflora, Hook. f., Fl. Nov. Zel., 1, 107 (1853); Handb: N.Z. FI, 116 (1864). Auckland and Campbell Islands: Not uncommon, especially in ravines on the face of the hills; Hooker, Kirk! Cockayne! Aston! Lamg! (New Zealand, irom the North Cape to Stewart Island.) There are two states of this represented in the collections submitted to me. The first has comparatively large oblong-obovate leaves 4-4 in. long, and the young shoots are clothed with whitish shagey pubescence. The other has smaller and narrower linear or lanceolate more acute leaves, and the twigs are not nearly so pubescent. Specimens of the first-mentioned form, collected by Mr. Aston, show a few imperiect flowers, which in the main agree with those of C. parviflora ; but neither of the two exactly matches any of the New Zealand forms of C. parviflora, although the differences are not of a pronounced character. At the same time, it must be admitted that C. parviflora, as at present defined, is a somewhat elastic species, including several plants which further research may prove to be worthy of specific rank. For instance, the northern variety, abundant in the Bay of Islands and Mangonui Counties, and remarkable for its flat spreading branches, has a very different aspect to the form so common in the mountain districts of New Zealand. Should the species be divided in the future, the name parviflora should be retained for the northern plant, to which it was first applied, and Hooker’s original name of myrtllifolia, as the oldest in date, be kept for the Auckland Island plant and the allied forms found in the mountains of the South Island. Systematic Botany] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 411 Coprosma ciliata, Hook. f. Coprosma ciliata, Hook. f., Fl. Antarct., 1, 22 (1844). C. parviflora, var. pilosa, Cheesem. in Trans. N.Z. Inst., xix, 241 (1887). Auckland and Campbell Islands: Abundant, ascending to at least 1,000 ft., and often forming almost impenetrable thickets; Hooker, Kirk! Aston! Laing! Antipodes Island: Kirk! (Probably also, in a slightly different state, in the moun- tains of the South Island of New Zealand). Although the flowers of this species have still to be obtained, the specimens brought back by the expedition leave little doubt that it will prove identical with my C. parviflora var. pilosa, a plant found in many localities in the central and western portions of the South Island, and which differs from the Auckland Island plant only in being more slender and not so copiously branched, and in having less copiously ciliate leaves. Coprosma foetidissima, Forst. Coprosma foetidissima, Forst., Char. Gen. 138 (1776) ; Hook. f., Fl. Antarct., i, 20, t. 13 (1844). C. affinas, Hook. f., Fl. Antarct., 1, 21, t. 14 (1844). Auckland Islands: Abundant in woods near the sea, and ascending the valleys to an elevation of nearly 1,000 ft.; Hooker and all subsequent collectors. Camp- bell Island: Quoted in the Handbook by Hooker, and by Kirk in the “ Students’ Flora,” but I have seen no specimens, and it was not collected by Cockayne in his visit of 1903, nor by Laing during the recent expedition. (New Zealand, hilly forests from Cape Colville southwards. ) Varies greatly in the size and shape of the leaves, which in luxuriant specimens may exceed 2 in. in length, with a breadth of over ?1in., thus exceeding the usual size in New Zealand. Hooker’s C. affinis was separated mainly on account of the acute or even acuminate leaves, but was subsequently abandoned by its author. I have not seen specimens which exactly match the plate given in the “ Flora Ant- arctica,” although some show a sufficiently close approach to justify the merging of the two forms. Coprosma cuneata, Hook. f. Coprosma cuneata, Hook. f., Fl. Antarct., 1, 21, t. 15 (1844). Auckland, Campbell, and Antipodes Islands: Common from sea-level to a considerable height, forming much of the subalpine scrub in open gullies and depressions ; Hooker, Kirk! Cockayne! Aston! Tennant! Lamg! (New Zealand, mountain districts from the East Cape and Lake Taupo southwards.) In the “ Manual of the New Zealand Flora” I have followed Hooker and others in assuming that this is the same as the mountain species so common in New Zealand. It is well to state, however, that the Auckland Island plant differs in having its leaves always widest at the upper extremity, which is rounded or almost truncate, and decidedly retuse or even emarginate, whereas the leaves of the New Zealand plant are usually broader, subacute, or even acute. Unfortunately, all the speci- mens I have seen from the islands are out of flower; but if it should be proved that the above differences are accompanied by corresponding deviations in the flowers or fruit, then the New Zealand plant should be separated as a distinct species. 412 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. Coprosma repens, Hook. f. Coprosma repens, Hook. f., Fl. Antarct., 1, 22, t. 164 (1844). Auckland, Campbell, and Antipodes Islands: Abundant from sea-level to an elevation of at least 1,200 ft.; Hooker and all subsequent collectors. Macquarie Island: Not uncommon; Scott, Hamilton. (New Zealand, from the Hast Cape and Tongariro southwards ; mountains of Victoria and Tasmania.) The specimens I have examined show no points of difference from the usual state of the species in New Zealand. Nertera depressa, Banks & Sol. Nertera depressa, Banks & Sol. ex Gaertn. Fruct., 1, 124, t. 26 (1788) ; Hook. f., Fl. Antarct., 1, 23 (1844). Auckland Islands: Common amongst moss in the woods, and also on the open hillsides ; Hooker, Kirk! Cockayne, Aston! Tennant! Campbell Island: Appa- rently not uncommon; Buchanan, Cockayne. (New Zealand, mountain districts from the Ruahine Range southwards; mountains of Victoria and Tasmania ; America, from Mexico along the Andes to Peru, Chili, Fuegia, and the Falkland Islands ; Tristan d’Acunha.) The distribution of Nertera is most remarkable. In addition to N. depressa, with the exceptionally wide range quoted above, there are three other species in New Zealand, one of which is said to occur in the Philippine Is'ands; an endemic species in Australia (NV. reptans, F. Muell.); another in Tristan d’Acunha (N. assur- gens, Thouars) ; a species in Java and Borneo, very closely allied to N. depressa ; and one in China (N. sonensis, Hemsl.). It is worthy of remark that the Bornean and Javan species (which 1s considered by Stapf to be identical with N. depressa) occurs on the summit of Mount Kinibalu, in north Borneo, at an altitude of over 10,000 ft., associated with such southern genera—“ austral-antarctic types,” as Dr. Stapf calls them—as Drimys, Coprosma, Pratia, Gaultheria, Podocarpus, Dacry- dium, and Phyllocladus ; and, what is equally extraordinary, in company with alpine species of Vaccinium, Rhododendron, and other northern genera. For fuller particulars reference should be made to Dr. Stapfs memoir on the ~ Flora of Mount Kinibalu ” (Trans. Linn. Soe., ser. 1, vol. iv, p. 69 ef seq.). COMPOSITAE. Lagenophora pumila, Cheesem., n. comb. Calendula pumila, Forst., Prodr., n. 305 (1786). Mucrocalia australis, A. Rich., Fl. Nov. Zel., 231 (1832). Lagenophora Forsterv, D.C., Prodr., v., 307 (1836). Auckland Islands: Probably not uncommon: Port Ross and other places ; Kirk. Enderby Island, on sand-dunes; Cockayne, Tennant! Carnley Harbour, at sea-level; Aston. Campbell Island: Cockayne, Chambers! Antipodes Island : Kirk, Cockayne. (Common throughout New Zealand, from the Kermadec Islands to Stewart Island.) Mr. Tennant’s specimens, which are the only ones I have seen in flower, are in poor condition, but agree fairly well with the ordinary form found in New Zealand. Systematic Botany. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 413 The closely allied Lagenophora nudicaulis, Dusén (L. Commersonw, Cass.), represents the species in South America, ranging from Chil to the south of Fuegia and the Falkland Islands. It is also found in Tristan d’Acunha. I very much regret to have to change an old-established name like that of L. Forsteri, but in this case it is unavoidable, Forster’s name having fifty years’ priority over De Candolle’s. It is worth remarking that A. Richard’s name of australis also antedates that of Forsteri by four years. Olearia Lyallii, Hook. f. Olearia Lyall, Hook. f., Fl. Nov. Zel., 1, 116 (1853). The Snares: The principal tree on the island, and forming the greater portion ~ of the bush; Kirk! Cockayne, Aston, &c. Auckland Island: Rare and local, but forming a dense forest on Ewing Island, at the entrance to Port Ross; Lyall, Bolton, Kirk! Cockayne! Aston! Tennant! (Hndemic.) I have no new matter to add to the excellent accounts of this noble plant which have already appeared. I would specially refer the inquirer to Mr. Kirk’s paper on the “ Botany of the Snares” (Trans. N.Z. Inst., xxi, 426); to Mr. Justice Chap- man’s interesting “ Visit to the Outlying Islands of New Zealand ” (/.c., 491); and to Dr. Cockayne’s “ Botanical Excursion to the Southern Islands” (Trans. N.Z. Inst., xxxvi, 252). Pleurophyllum speciosum, Hook. f. Pleurophyllum speciosum, Hook. f., Fl. Antarct., 1, 31, t. 22, 23 (1844). Auckland and Campbell Islands: Abundant from sea-level to an altitude of 800 ft. or more; Hooker and all subsequent collectors. According to Hooker, it 1S most common in wet ground near the sea. (Hndemic.) Of the many handsome herbaceous plants found in the Auckland and Camp- bell Islands this is one of the finest. Its rosettes of huge ribbed leaves, pressed flat to the ground, are frequently more than 4 ft. in diameter, and from the centre rise from 4 to 6, or even as many as 8-10, tall flowering-stems 14-3 ft. high, each of them bearing numerous rayed flower-heads 1} in. diameter. The rays are usually lilac or whitish-purple; the disc is a deep rich purple. As the plant frequently covers extensive tracts, the effect, as Mr. Kirk has stated, “1s unique, and approaches the magnificent.” Both this and the followimg species are very fully described and beautifully figured in the “ Flora Antarctica” (plates 22-25). Mr. Kirk’s mono- graph (Trans. N.Z. Inst., xxi, 431) also contains a full account of the species constituting the genus. Pleurophyllum criniferum, Hook. f. Pleurophyllum criniferum, Hook. f., Fl. Antarct., 1, 32, t. 24, 25 (1844). Albinea oresigenesa, Homb. & Jacq., Bot. Voy. Astrol. et Zél., t. 4, Dicot. (1852). Plewrophyllum Hombronw, Decne. in Bot. Voy. Astrol. et Zél., 36 (1853). Auckland, Campbell, and Antipodes Islands: Abundant in wet peaty ground from sea-level to 1,100 {t.; Hooker and all subsequent collectors. Has been re- 414 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. corded from Macquarie Island, but apparently in mistake for P. Hooker. (Kn- demic.) Almost as fine a species as the preceding, from which it differs in its suberect petiolate leaves, taller scapes, which sometimes are over 4 ft. in height, and in the flower-heads being without rays. Hooker’s plate of the inflorescence (“ Flora Ant- arctica,’ t. 24, 25) is excellent, but unfortunately the artist has inadvertently figured the leaf of P. speciosum in mistake for that of this species. Pleurophyllum Hookeri, Buch. Pleurophyllum Hookeri, Buch. in Trans. N.Z. Inst., xvi, 395 (1884). P. Hookervanum, T. Kark in Trans. N.Z. Inst., xxi, 435 (1891). P. Gulresianum, T. Kirk in Trans. N.Z. Inst., xxi, 435 (1891). Auckland and Campbell Islands: Upland meadows, usually above 500 ft. and ascending to 1,400 ft., not uncommon; Buchanan! Kirk! Cockayne! Aston ! Tennant! Laing! Macquarie Island: Abundant in all suitable localities; Fraser, Scott, Hamilton! (Hndemic.) Easily separated from the preceding by the much smaller size, silvery tomentose acuminate leaves, and by the shorter rigid scapes with fewer and smaller heads. Celmisia vernicosa, Hook. f. Celmisia vernicosa, Hook. f., Fl. Antarct., 1, 34, t. 26, 27 (1844). Auckland and Campbell Islands: From sea-level to at least 1,500 ft., most abundant in moist grassy places above 500 ft. elevation ; Hooker and all subsequent collectors. (Kndemic.) An exceedingly handsome plant, but very variable in size. A large series of specimens collected by Cockayne at the head of the North Arm of Carnley Harbour, and a similar set obtained by Laing in Campbell Island, contain specimens varying from 14 in. diam., with a solitary scape barely 2 in. high, to others quite 6 in. across, with 8-10 scapes from 6in. to 8in. high. The peculiar varnished surface of the leaves is most remarkable, and unique in the genus. Hooker compares it with the smooth and shining surface of a shell, or looking as if covered with a thick coat of copal varnish ; while Mr. Justice Chapman describes 1t as “ gleaming like polished nephrite newly from the lapidary’s hands.” The usual colour of the flowers is as described by Hooker in the “ Flora Antarctica ’—that is, the ray-florets are pure white, some- times with a faint tinge of rose at the tips, and the disc-florets are dark purple; but I am informed that individual specimens were noticed by members of the expedi- tion in which the flowers were purple throughout. Celmisia Campbellensis, F. R. Chapm. Celmisia Campbellensis, F. R. Chapm. in Trans. N.Z. Inst., xxin, 407 (1891). C. Chapman, T. Kirk in Gard. Chron., 1x, 731, t. 146 (1891). Auckland Islands: North Arm of Carnley Harbour, alt. 600 ft. ; Captain Dor- rien-Smith and Dr. Cockayne, J. 8. Tennant! (only two plants observed). Camp- bell Island: Venus Cove, Perseverance Harbour; F. R. Chapman! T. Kirk! (not more than a dozen specimens observed). (Hndemic.) Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 415 The flower-heads and scapes hardly seem to differ from those of C. vermecosa, but the leaves are altogether different, being broader and not so coriaceous, longi- tudinally furrowed above and strongly ribbed beneath. (NorE.—Celnusia verbascifolia, Hook. f., has been recorded by Kirk as a native of Campbell Island, on the strength of specimens said to have been collected in 1874 by Dr. Filhol and Lieutenant Rathouis, of the French Transit of Venus Expedition. It has not been observed since, although repeatedly searched for, especially by Mr. Kirk in 1890, by Dr. Cockayne in 1903, and by Mr. R. M. Laing during the present expedition. I feel sure that some accidental admixture of specimens has been made, and I have consequently expunged the species from the list.) Gnaphalium luteo-album, Linn., Sp. Plant. Gnaphalium luteo-album, Linn., Sp. Plant., 851 (1753) ; Hook. f., Fl. Ant- arct., 1, 37, and 1, 543 (1844-47). Auckland Islands: On the sea-beach, exact locality not stated; D’Urville (Fl. Antarct.), Kirk, Cockayne. (Most warm-temperate or tropical countries.) I have seen no specimens, but there is no reason for doubting the locality. At the same time, it is singular that the species should extend so far to the south, seeing that 1t is mainly a tropical or warm-temperate plant, absent from the extreme north of Kurope and the polar regions, and also wanting in North America and the extreme north of Asia. In the Southern Hemisphere it is found in South Africa and the warmer portions of South America, but it does not reach Fuegia or the Falkland Islands. It occurs on Tristan d’Acunha and Inaccegssible Island, but not on Ker- guelen Island or the Crozets. Helichrysum bellidioides, Willd., var. prostratum, T. Kirk. Xeranthemum bellidioides, Forst., Prodr., n. 293 (1786). Helichrysum prostratum, Hook. f., Fl. Antarct., 1, 30, t. 21 (1844). Gnaphaliwm prostratum, Hook. f., Fl. Nov. Zel., 1, 137 (1853). Helichrysum belli- diodes, Willd., Sp. Plant., i, 1911; var. prostratum, T. Kirk, Students’ Fl. 307 (1899). Auckland, Campbell, and Antipodes Islands: Abundant throughout, ascend- ing to over 1,500 ft.; Hooker, Kirk! Cockayne, and others. (Mountains of New Zealand, from Mount Egmont southwards, but not nearly so abundant as the type.) Some botanists will prefer to treat this as distinct from the typical H. bellidioides, but there is little to separate the two plants, except that bellidioides has long brac- teate peduncles to the heads and prostratum none at all. But passage forms are not uncommon, and, on the whole, it is best to treat the plant as a well-marked variety rather than a separate species. Cassinia Vauvilliersii, Hook. f. Ozothamnus Vauvilliersu, Homb. & Jacq. ex Hook. f., Fl. Antarct., i, 29 (1844) ; Decn., Voy. Astrol. et Zel., 38, t. 5, Dicot. (1853). Cassinia Vauvilliersu, Hook. f., Fl. Nov. Zel., 1, 133 (1853). Auckland Islands: Abundant, often forming a dense scrub, but rarely ascending to any great height on the hills; Hooker, D’Urville, and all subsequent collectors, 416 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. Campbell Island: Recorded by Kirk, but not observed by any other botanist. (New Zealand, common in mountain districts from the Hast Cape and Taupo southwards.) There are no differences of importance between the Auckland Islands plant and that found in the mountain districts of New Zealand. Cotula australis, Hook. f. 3 Anacyclus australis, Sieb. ex Spreng, Syst., u, 497 (1826). Strongylosperma australe, Less., Syn. Comp., 261 (1832). Cotula australis, Hook. f., Fl. Nov. Zel., 1, 128 (1853). Campbell Island: Buchanan. See Trans. N.Z. Inst., xvi, 399. (Throughout New Zealand ; east and south Australia and Tasmania ; Tristan d’Acunha.) Mr. Buchanan’s paper on Campbell Island and its flora, quoted above, is the sole authority for including this species in the florula, no subsequent botanist having met with it. Its occurrence im Tristan d’Acunha, quite six thousand miles from its Australian or New Zealand habitats, is one of the most puzzling instances of disconnected distribution known. Cotula plumosa, Hook. f. Leptinella plumosa, Hook. f., Fl. Antarct., 1, 26, t. 20 (1844). Cotula plumosa, Hook. f., Handb. N.Z. Fl., 141 (1864); Phil. Trans. Roy. Soc., 168, p. 20 (1879). Auckland and Campbell Islands: Maritime rocks and meadows near the sea, abundant ; Hooker, Kirk! and all subsequent botanists. Antipodes Island: Not uncommon near the sea; Kirk, Cockayne. Macquarie Island: Plentiful close to the sea; Fraser, Scott, Hamilton. (Kerguelen Island and the Crozets.) This is one of the few New Zealand plants found in Kerguelen Is'!and which does not also extend to Fuegia or the Falkland Islands. In Kerguelen Island it is the only Composite, and forms immense blue-green patches where the soil is enriched by the dung of the birds. It 1s by far the finest species of the genus. Cotula lanata, Hook. f. Leptinella lanata, Hook. f., Fl. Antarct., 1, 25, t. 19 (1844). L. propinqua, Hook. f., Fl. Antarct., 1, 27 (1844). Cotula lanata, Hook. f., Handb. N.Z. Fl., 141 (1864). Auckland Islands: Abundant on rocks near the sea ; Hooker and all subsequent collectors. Campbell Island: Recorded by Hooker in the Handbook, but not men- tioned in the “‘ Flora Antarctica’: North-east Harbour; Cockayne. (Endemic.) Varies much in the size and breadth of the leaves and the extent to which they are divided, also in the degree to which the loose white cottony hairs of the young stems and leaves are developed. I have seen no specimens of C. propinqua, said to be intermediate between C. plumosa and OC. lanata. It was originally described by Hooker as a separate species, but in the Handbook he merged it with C. Janata. Judging from the description, 1t appears to differ from both species in several im- portant points, and quite possibly should be restored to specific rank, Systematic Botany] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 417 Cotula dioica, Hook. f. Leptinella dioica, Hook. f., Fl. Nov. Zel., 1, 129 (1853). Cotula dioica, Hook. f., Handb. N.Z. Fl, 143 (1864). Auckland Islands: Near the shore; Dr. R. Koettlitz. (New Zealand, from the Kaipara Harbour southwards.) T insert this on the authority of Dr. Cockayne. See Trans. N.Z. Inst., xxxvi, 321. Abrotanella spathulata, Hook. f. Trineuron spathulatum, Hook. f., Fl. Antarct., 1, 24, t. 17 (1844). Abrota- nella spathulata, Hook. t., Handb. N.Z. Fl., 139 (1864). Auckland and Campbell Islands: Peaty soil from sea-level to 1,500 it., but not common. below 500 ft. ; Hooker, Kirk! Cockayne, Aston ! Tennant ! Buchanan ! Laing! (Endemic.) Specimens collected by Kirk at Carnley Harbour are much larger than those figured by Hooker, and have the flower-heads raised on a peduncle sometimes over 2in. long. On the other hand, those gathered by Aston and Tennant are smaller, with the inflorescence hidden amongst the uppermost leaves. Abrotanella rosulata, Hook. f. Ceratella rosulata, Hook. f., Fl. Antarct., i, 25, t. 18 (1844). Abrotanella rosulata, Hook. t., Handb. N.Z. Fl., 139 (1864). Campbell Island: Rocky places on the hills, above 800 ft. elevation, ascending to the summit of Mount Honey, alt. 1,860 ft., not common ; Hooker, Kirk ! Cockayne, Lamg! (Endemic.) Mainly distinguished from the preceding by the smaller size and more compact habit, smaller and broader rosulate leaves, and 4-horned achenes. Mr. Kirk’s speci- mens are in full flower ; Mr. Laing’s are quite immature. Erechtites, sp. ? Hrechtites prenanthordes, D.C., Prodr., vi, 296 (1837); Hook. f., Fl. .Ant- arct., u, 544 (1847). Auckland Islands: Precise locality not stated; Le Guillon: Enderby Island ; Cockayne, Tennant! (New Zealand ?) A species of Hrechtites was gathered by D’Urville’s expedition in 1840, and was referred by Hooker to the widely distributed HL. prenanthoides. Probably the same plant was collected by Dr. Cockayne on Enderby Island in 1903, and again by the present expedition in 1907. The single specimen I have seen is immature and in very poor condition, and is quite insufficient for determination; but it differs from the usual state of H. prenanthoides in the toothing of the leaf-margin, and can hardly be referred to that species. Senecio antipodus, T. Kirk Senecio antipodus, T. Kirk in Trans. N.Z. Inst., xxii, 440 (1891), name only : Students’ Fl., 341 (1899). Antipodes Island: Abundant over the greater part of the island: Kirk! Cockayne. (Hndemic.) 28—S. 418 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. A remarkably distinct species, not closely allied to any of the New Zealand forms. Kirk compares it with the Fuegian S. candicans, and no doubt there is a resemblance in the inflorescence and discoid heads; but S. candicans has ovate leaves with crenate toothed margins, widely different in appearance to the deeply pinnatifid leaves of S. antupodus. The paucity of species of Senecio in the New Zealand subantarctic islands 1s a very remarkable fact, of which no adequate explanation has yet been offered. In Fuegia, a large portion of which lies in the same latitude as the Auckland Islands, no less than thirty-three species are recorded in the most recent enumeration, that given by Wildeman in the “ Voyage of the ‘ Belgica.’”” New Zealand contains thirty ; and Tasmania fourteen. With the genus so well developed in the nearest large land-masses, 1ts almost total absence in the islands to the south of New Zealand is one of the most striking peculiarities of their flora, both as compared with that of New Zealand and with that of subantarctic America. Senecio Stewartiae, Armstr. Senecio Stewartiae, Armstr. in Trans. N.Z. Inst., xii, 339 (1881). S. Muelleri, T. Kirk in Trans. N.Z. Inst., xv, 360 (1883). The Snares: Not uncommon; Kirk! Chapman! and all subsequent collectors. (Herekopere Island, in Foveaux Strait.) A very handsome species, Its nearest ally is doubtless the Chatham Island S. Huntiw, trom which it differs in the broader more acuminate leaves, with much white tomentum beneath, and in the narrow contorted ray-florets. The restricted geographical range of many species of Senecio has often been remarked ; but hardly any instance is more striking than that of the present species, which is confined to two small rocky islets not much more than a square mile in area. Taraxacum officinale, Wigg. Taraxacum officinale, Wigg., Prim. Fl. Holsat., 56. T. dens-leonis. Dest., Fl. Atlant., 1, 228. Campbell Island: Not plentiful; R. M. Laing! (Throughout the temperate regions of both hemispheres.) Mr. Laing is the first to collect this in the southern islands. It is possible that it may be naturalised; but his specimens belong to the small mountain form known to be indigenous in New Zealand, having been collected by Banks and Solander, and not to the larger state common in pastures and cultivated soil, which is certainly an importation. . . Sonchus asper, Hill. Sonchus asper, Hill, Herb. Brit., 1, 47. S. oleraceus, Linn., var. B, Hook. f., Fl. Nov. Zel., 1, 153 (1853). The Snares: Kirk. Auckland Islands: Dr. Koettlitz. Campbell Island: Near the shore; R. M. Laing! Antipodes Island: Kirk. (Cosmopolitan.) The only specimens which J have seen are Mr. Laing’s, which are so immature as to make their identification somewhat doubtéul. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 419 STYLIDIEAE. Phyllachne clavigera, F. Muell. Forstera clavigera, Hook. f., Fl. Antarct., 1, 38, t. 28 (1844). F. aretriastri- folia, Homb. & Jacq., Bot. Voy. Astrol. et Zél., t. 16c (1853). Helo- phyllum clavigerum, Hook. {., Handb. N.Z. Fl., 167 (1864). Phyllachne clavigera, F. Muell., Fragm., vi, 40 (1873). Auckland and Campbell Islands: Abundant in moist peaty localities, 500- 1,800 it. elevation ; Hooker, D’Urville, Kirk! Cockayne! Aston! Tennant! Laing ! (New Zealand, alpine localities from Canterbury southwards.) According to all travellers, the bright-green hard cushions of this plant are familiar objects on the hills of the Auckland and Campbell Islands, particularly above 500 it. elevation. It is closely allied to the Fuegian P. whiginosa, which represents it in that country. CAMPANULACEAE. Pratia arenaria, Hook. f Pratia arenaria, Hook. f., Fl. Antarct., 1, 41, t. 29 (1844). P. angulata, var. arenaria, Hook. f., Fl. Nov. Zel., 1, 157 (1853). Auckland Islands: Sandy soil on the shores of Enderby Island; Lieutenant Oakely (Hooker, Fl. Antarct.), Cockayne, Aston. Campbell Island: Locality not stated; Laing! Antipodes Island: Kirk! Cockayne! (South-eastern coast of Otago; Chatham Islands.) This appears to me to differ from all the forms of P. angwlata in the peculiar habit, larger, very obscurely toothed and nearly sessile leaves, and in the very short peduncles. J have therefore reinstated it as a species, Hooker’s description and plate are excellent. EPACRIDACEAE. Cyathodes empetrifolia, Hook. i. Androstoma empetrifolia, Hook. f., Fl. Antarct., 1, 44, t. 30 (1844). Cyathodes empetrifolia, Hook. f., Fl. Zel., 1, 164 (1853). Auckland Islands: Abundant above the bush-level, ascending to quite 1,500 ft. ; Hooker, Kirk ! Cockayne, Aston, Tennant! Campbell Island: Recorded by Hooker in the Handbook, but not mentioned in the “ Flora Antarctica.” (New Zealand, from Cape Colville southwards.) The Auckland Island specimens of this common plant do not differ in any im- portant character from the usual state of the species in New Zealand. Dracophyllum longifolium, R. Br. Epacris longifolia, Forst., Prodr., n. 68 (1786). Dracophyllum longifolium, R. Br., Prodr., 556; Hook. f., Fl. Antarct., 1, 45, t. 31, 32 (1844) ; Homb. & Jacq., Bot. Voy. Astrol. et Zél., 33, t. 27, Dicot. (1853). Auckland and Campbell Islands: Abundant from sea-level to about 700 it. elevation ; Hooker and all subsequent collectors. (New Zealand, common in moun- tain districts from the East Cape southwards.) 420 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. One of the chief components of the forest which clothes the lower slopes of the Auckland Islands, also of the scrub of Campbell Island. According to Hooker, it sometimes attains a height of 25 ft., with a trunk 14 ft. in diameter. The majority of the specimens from the Auckland Islands agree with those from Stewart Island and the sounds of the west coast of Otago, but others, with shorter and narrower leaves, answering to the var. retortum, figured by Hombron in the “ Voyage au Pole sud,’ come much nearer to the mountain form so common in the Southern Alps, and show a marked approach to D. Urvilleanum. It is curious that neither Gaultherva nor Pernettya occur in the Auckland Islands, seeing that in South America species of these genera advance as far south as Cape Horn, 8. lat. 56°. Dracophyllum scoparium, Hook. f. Dracophyllum scoparvum, Hook. f., Fl. Antarct., 1, 46, t. 33 (1844). D. Uroileanum, var. scoparvum, Hook. f., Handb. N.Z. FIl., 182, in part (1853). Campbell Jsland: Common near the sea; Hooker, Kirk! Cockayne! Laing ! Chambers! (Chatham Islands.) Two forms of the species, as I understand it, occur on Campbell Island: one is the plant figured by Hooker in the “ Flora Antarctica,’ which has the margins of the leaves conspicuously ciliate and the upper surface densely pubescent; the other has smaller and narrower less-pubescent leaves, Kirk restricted the species to the first (see his account of the botany of Campbell Js!and, publ:shed in the re- port of the Australasian Association for 1891); the second, which js much the more common, in many places constituting the greater portion of the scrub, he referred to D. Urvilleanum. This view can hardly be accepted, for they are certainly much more closely connected with one another than either of them is with D. Urvilleanum. That species has decidedly less-pubescent leaves, with less- evidently-ciliate margins, and the top of the leaf-sheath 1s much move truncate or even auricled. I regard 1% as sufficiently distinct from both of the Campbell Island plants. PRIMULACEAE. Samolus repens, Pers. Sheffieldia repens, Forst., Char. Gen., 18 (1776); Prodr., n. 67 (1786). Samolus repens, Pers., Syn. Plant., 1, 171 (1805); T. Kirk im Report Austr. Assoc., 1891, 222. S. littoralis, R. Br., Prodr., 428 (1810). Auckland Islands: Rocks on Adams Island, Carnley Harbour ; Kirk! Cockayne, Aston. (Shores of New Zealand, from the Kermadec Islands and the North Cape southwards ; Australia; Tasmania; New Caledonia.) Mr. Kirk’s specimens, which are the only ones I have seen, agree very well with the ordinary state in New Zealand. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 42) MYRSINACEAE. Myrsine divaricata, A. Cunn. . Myrsine divaricata, A. Cunn., Precur., n. 406 (1838). Suttonia dwwaricata, Hook. f., Fl. Antarct., 1, 51, t. 34 (1844). Auckland and Campbell Islands : Abundant in scrub from sea-level to a consider- able height on the hills, Hooker and all subsequent collectors. (New Zealand, from Mangonui southwards.) This, with several species of Coprosma and Cassinia Vauvilliersu, together with stunted Metrosideros and Dracophyllum, usually form a belt of almost impenetrable scrub above the level of the rata forest in the Auckland Islands, and with the Co- prosmas and Dracophyllum in Campbell Island, from sea-level to a height of 500 ft. Few plants occupy such a range of diverse habitats as M. dwaricata. It appears equally at home in the moist sheltered river-valleys of the North Island, in the moun- tain forests of the South Island, in the subalpine scrub of the Southern Alps, and on the wind-swept hills of the Auckland and Campbell Islands. With respect to its latitudinal range, Hooker has well drawn attention to the fact that it is the only shrubby plant in the New Zealand area which can be found at or near sea-level in the respective latitudes of 35° and 524° 8. GENTIANACEAE. (The early period of the season at which the expedition visited the islands made it impossible to obtain a really good series of the Gentians, G. cera and G. concunna were only just coming into bloom on the Auckland Islands, and no flowering speci- mens of any species were obtained on Campbell Island or Antipodes Island. The following enumeration is therefore mainly based on the collections made by Mr. Kirk in 1890. During this visit Mr. Kirk paid special attention to the genus, his observations being incorporated in his “ Revision of the New Zealand Gentians ” (Trans, N.Z. Inst. xxvul, pp. 330-341.) Gentiana cerina, Hook. f. Gentiana cerina, Hook. f., Fl. Antarct., 1, 54, t. 36 (1844) ; T. Kirk in Trans, N.Z. Inst., xxvu, 338 (1895). G. Campbell, Homb. & Jacq., Bot. Voy. Astrol. et Zél., 26, t. 31c. (1853). Auckland Island: Not uncommon from sea-level to ovér 1,000 ft. elevation ; Hooker, D’Urville, Kirk! Chapman! Cockayne! Aston! Tennant! Campbell Island: I have seen no specimens, but Hombron and Jacquinot’s G. Campbelli clearly be- longs to this species, judging from the plate. (Endemic.) Kirk remarks that this 1s “one of the most beautiful plants in the flora, the corolla varying in colour from a pure waxy white to white with a vertical red stripe, purple, reddish-purple, and violet.” He also states that “it varies much in habit and luxuriance, according to the nature of the habitat in which it is growing.” Ex- cluding G. concinna, he admits two forms—cerima proper, in which the stems are prostrate or trailing, and the leaves thick and fleshy, the flowers crowded at the ends of the branches; and suberecta, which has more slender suberect or decumbent 422 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. stems, with the leaves more remote and the flowers in rather lax corymbs. Judging from his own specimens and from those brought back by the expedition, 1t appears certain that these two forms are directly connected by intermediates, A remark- ably slender state collected by Mr. Aston on Disappointment Island, with long and narrow linear-spathulate leaves, looks more distinct, but it is, unfortunately, not in flower. Gentiana concinna, Hook. f. Gentiana concinna, Hook. f., Fl. Antarct., 1, 53, t. 35 (1844). G. ceria, var. concumna, T. Kirk in Trans. N.Z. Inst., xxvii, 329 (1895). Auckland Islands: Open places on the hills, abundant ; Hooker, Kirk! Chap- man! Cockayne! Aston! (Hndemic.) . A charming little plant, excellently figured by Hooker in the “ Flora Antarctica.” Kirk considered it to be a form of the preceding, and it is certainly a close ally ; but it appears to be sufficiently distinct in the annual duration, the more erect habit of growth, and the smaller size of all its parts. Kirk remarks that it is usually found on the surface of the masses of Trichocolea tomentella and other Hepaticae, which often cover the ground in sheltered localities in the Auckland Islands. Gentiana antarctica, T. Kirk. Gentiana antarctica, T. Kirk in Trans. N.Z. Inst., xxvii, 339 (1895). G. concinna, var. robusta, Hook. i., Fl. Antarct., 1, 53 (1844). Campbell Island: Not uncommon on the hills; Hooker, Kirk! Laing! Cham- . bers! (Hndemic.) Chiefly distinguished by the pale greenish-yellow hue of the entire plant, by the stout erect habit, by the minutely verrucose surface of all its parts, and by the much-branched slender inflorescence, the flowers being almost hidden among the crowded cauline leaves and bracts. Kirk’s var. imbricata is not verrucose, the leaves are more coriaceous, and the flowers are larger. Gentiana antipoda, T. Kirk. Gentiana antepoda, T. Kirk in Trans. N.Z. Inst., xxvu, 340 (1895). Antipodes Island: Abundant; Kirk! Cockayne! (Hndemic.) Closely allied to the preceding species, from which it mainly differs in the more numerous stems and narrower lingulate leaves. Mr. Kirk remarks that there are two trivial forms or states — one, pallida, with yellow stems and delicate white flowers; the other, rubra, with red stems and white flowers longitudinally veimed with red. BORAGINACEAE. Myosotis antarctica, Hook. f. Myosotis antarctica, Hook. f., Fl. Antarct., 1, 57, t. 38 (1844). Campbell Island: Rocky débris at the base of precipices and in clefts on the summits of the mountains; Hooker: Venus Cove, Laing! (specimens very imma- ture, and possibly not referable to this species). (New Zealand, from the Auckland Isthmus southwards.) Systematic Botany. SUBANTARCTIC ISLANDS OF NEW ZEALAND. 423 I much regret that a good series-of specimens of this species was not obtained by the expedition. A comparison could then have been made with the numerous New Zealand forms which are at present referred to the species. Hooker’s plate, which must be taken as representing the type, might stand for a common New Zea- land form, so far as habit and foliage are concerned; but the flowers are repre- sented as blue, and the anthers are placed lower down the corolla-tube than is usual in New Zealand. All the New Zealand specimens I have seen have white flowers, sometimes with a faint tinge of yellow, and blue or purple flowers are unknown. Myosotis capitata, Hook. f. Myosotis capitata, Hook. f., Fl. Antarct., 1, 56, t. 37 (1844). Auckland Islands: Gravelly banks close to high-water mark; Hooker: open hillsides and rocky places, ascending to over 1,000 ft. ; Kirk ! Chapman ! Cockayne ! Aston! Campbell Island: Filhol! Kirk, Laing. (Hndemic.) This must be a very beautiful plant when fresh, judging from Hooker’s plate and description. He gives the colour of the flowers as “deep violet-blue.” The late Mr. Kirk informed me that it was violet-blue fading to purple, and Mr. Laing states that on Campbell Island it varies from magenta to light purple. I now limit this species strictly to the Auckland and Campbell Island plant, for the excellent specimens brought back by the expedition have fully satisfied me that the variety albiflora of my Manual is a distinct species, which may be described as follows : — Myosotis albida, Cheesem., n. sp. Myosotis capitata, Hook. f., Fl. Nov. Zel., i, 200, so far as the New Zealand localities are concerned (1853). M. capitata, var. albiflora, Armstr. in Trans. N.Z. Inst., xii, 340 (1881). M. capitata, var. albida, T. Kirk in Trans. N.Z. Inst., xvitl, 224 (1885). Perennis, Rhizoma crassum, elongatum, horizontale vel obliquum, fibras crassas emittens, ad apicem reliquis foliorum emarcidorum vestitum. Caules vel rami florifer! e rhizomate plurimi, validi, ascendentes, basi plerumque decumbentes, 20-45 cm. alti, undique pilis brevibus mollibus et densis patentibus vestiti. Folia radicalia numerosa, 10-15 cm. longa, 15-3 cm. lata, crassa, plana, lineari-obovata vel lineari-spathulata vel lineari- oblonga, obtusa, in petiolum latum gradatim attenuata, supra et infra pilis brevibus albidis densissime obsita. Folia caulina pauca, minora, 2-4 cm. longa. Racemi terminales et m axillis foliorum superiorum, pedunculati, simplices vel furcati, scorpioidei, capitati. Flores numeros‘ssimi, con- ferti, breviter pedicellati. Calyx 4-5 mm. longus, pilis albidis densissime obsitus ; lobis elongatis, lmeari-oblongis, obtusis. Corolla hypocrateriiormis, alba; tubus elongatus, anguste cylindraceus, calyce sub } longior, fauce gibbis emarginatis in- structa; limbus sub 5mm. diam., explanatus lobis rotundatis. Stamina 5, fauci affixa ; filamenta antheris longitudine aequalia ; antherae fauce.n omnino superantes. Nuculae ovoidae, acutae, atrae, nitidae. The Snares: Not uncommon on the clifis; Kirk! (South-east coast of Otago, islets in Foveaux Strait, shores of Stewart Island.) 494 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. As Mr. Armstrong’s varietal name of albiflora has already been applied to a Fuegian plant (MW. albiflora, Banks and Solander), I have used Kirk’s later name of albida. As a species it differs from M. capitata in the larger size and stouter habit, and in being more copiously covered with soft white hairs. The leaves are larger, 3-6 in. long, with longer petioles, and are equally hairy (or nearly so) on both surtaces, whereas in M. capitata the under-surface of the leaves is less hairy than the upper side, often becoming nearly glabrous in age. The inflorescence is much more de- veloped, the lateral racemes being more numerous, and on longer peduncles. The flowers are always white, never blue. They are rather smaller m diameter than M. capitata, but have a proportionally longer tube. But the most important difference between the two species is in the stamens. In M. capitata they are inserted some distance below the corolla-scales, and the filaments are very short, so that the tips of the anthers just reach above the corolla-scales. In WM. albida the stamens are inserted between the corolla-scales, and the filaments are as long as the anthers, or a trifle longer, so that the anthers are altogether above the level of the scales. The position of the stamens is therefore about intermediate between Humyosotis and Haarrhena. SCROPHULARIACEAE. Veronica elliptica, Forst. Veronica elluptica, Forst., Prodr., n. 10 (1786); Hook. f., Fl. Antarct., 1, 58 (1844). V. decussata, Ait., Hort. Kew, i, 20, (1789). The Snares: Not uncommon; Kirk! Cockayne! Auckland Islands: Fairly plentiful, especially along the western coast and on Disappointment Island; also on the exposed islets of the eastern coast, as Ewing Island, Enderby Island, &c. ; rarer in the inlets of the eastern shore ; Hooker and all subsequent travellers. Camp- bell Island: Hooker, Kirk, Lamg. (Shores of New Zealand, from Cook Strait southwards; South America, from Magellan Strait to Cape Horn (8. lat. 56°); Falkland Islands.) Two states of this well-known plant are found in the Auckland Islands. One, with rather small leaves, varying from in. to lin. long, agrees fairly well with Forster’s drawing in the unpublished Banksian plates, which must be taken as repre- senting the type of the species. The other is much more robust, with larger and thicker leaves, 1-13 in. long, and with larger flowers and capsules. I understand that this state always has the flowers pure white; whereas the other (and especi- ally in some of the New Zealand localities) frequently has the upper petals finely pencilled with purple. Veronica odora, Hook. f. Veronica odora, Hook. i., Fl. Antarct., 1, 62, t. 41 (1844). V. buadfolca, var. odora, T. Kirk in Trans. N.Z. Inst., xxviii, 524 (1896). V. elliptica, var. odora, Cheesem., Man. N.Z. Fl., 516 (1906). Auckland Islands: “In woods near the sea, not uncommon, forming scattered bushes”: Hooker: Norman’s Inlet and Port Ross; Aston! (Hndemic.) Systematic Botany.) SUBANTAROCTIC ISLANDS OF NEW ZEALAND. 425 This is a plant which is very imperfectly understood, and the systematic position of which is consequently doubtful. In the “ Flora Antarctica ” Hooker figured and described it as a distinct species; but in the subsequently published Handbook he merged it with the widely spread V. buwifolia, Benth. Kirk, in his paper on Veronica, published in the “ Transactions of the New Zealand Institute ” (vol. xxvii, p. 524), also referred it to buaifolia, but as a separate variety, comprising Hooker’s Auckland Island plant and a patent-leaved form of buaxifolia common in mountain districts from Nelson to Stewart Island. ‘Looking at the matter solely from the point of view ot Hooker’s plate and description, this reduction to Veronica buaifolia has always appeared to me to be highly doubtful. According to Hooker’s description, odora differs from buaxzfolva in the delicate fragrance of the flowers, in the shorter corolla- tube and more obtuse corolla-lobes, in the absence of the large concave bracts so conspicuous in buaifolia, and in the regularly patent and more distantly placed leaves. In all these points it shows an affinity with V. elliptica, and in the Manual I consequently placed it under that species. In this view I was fortified by the opinion of Dr. Cockayne, who states (Trans. N.Z. Inst., xxxvi, 322) that it “is at most a form of ellaptica.” On the other hand, Mr. N. E. Brown, who examined for me the whole of the New Zealand Veronicas in the Kew Herbarium, and who had the advantage of inspecting Hooker’s original specimens, informed me that odora was distinct from buxifolva, and constituted a good species. Unfortunately, the collections submitted to me for determination do not include any specimens which I can refer to V. odora. But I am informed by Mr. Aston that a Veronica was observed at Norman’s Inlet and Port Ross which resembled V. buxzfolva, but which was not seen in a flowering state. A living plant is now in cultivation at Wellington, a scrap from which has been kindly forwarded by Mr. Aston. The material is not sufficient to form the basis of any opinion, but the foliage certaily comes nearer to the plate of V. odora than any other specimen I have seen. An approach to V. buazfolia is also evident, but there are differences which would seem to preclude the union of the two plants, and until flowering specimens are obtained it is best to treat the two as distinct. - It may be remarked that should future investigation prove the specific identity of V. odora and V. buaxzfolia, then odora must be the name retained. It was described by Hooker in 1844, whereas V. buaifolua was not published until 1846, two years later.* Veronica Benthami, Hook. f. Veronica Bentham, Hook. f., Fl. Antarct., 1, 60, t. 39, 40 (1844). Auckland and Campbell Islands: In open and rocky places from a short distance above sea-level to the summit of the hills, abundant; Hooker and all subsequent collectors. (Hndemic.) Excellently figured and described by Hooker, and a very distinct and beautiful plant, with flowers of the deepest blue. Like Hooker, I find that abnormal flowers frequently occur, in which the calyx and corolla are 6-lobed, with 3 stamens and a 3-celled ovary. * For further information respecting Veronica odora see page 447. 426 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. PLANTAGINACEAE. Plantago aucklandica, Hook. f. Plantago aucklandica, Hook. f., Fl. Antarct., 1, 64, t. 42 (1844). Auckland Islands : Not uncommon on the hills, usually above 1,000 ft. elevation ; Hooker, Kirk ! Chapman! Aston! Tennant! (Hndemic.) This is a very distinct species, of which Hooker remarks that “it 1s remarkable for its numerous leaves, which are generally quite smooth and fleshy, often forming a dense head, not unlike that of a small cabbage. In the smoothness, thick and succulent habit, and stout collum it bears some affinity to the caulescent species of Juan Fernandez and other insular situations.” Plantago Brownii, Rapin. Plantago Brownw, Rapin in Mem. Soc. Linn. Par., vi, 485 (1827); Hook. f., Handb. N.Z. Fl., 227 (1864). P. carnosa, R. Br., Prodr., 425 (1810) ; Hook. £., Fl. Antarct., 1, 65 (1844); non P. carnosa, Lam., Illust., 1, 372 (1791). Auckland Islands: Abundant on maritime rocks; Hooker, Kirk! Cockayne, Aston! Tennant! Campbell Island: A single specimen without locality; R. M. Laing! (As understood in Hooker’s Handbook, this is a common mountain-plant in New Zealand, from the East Cape southwards, and is also found in Victoria and Tasmania.) This differs from the New Zealand mountain-plant in having much more fleshy and more deeply toothed bright-green leaves, and in often being nearly glabrous. The spikes are also rather larger. I consider that it should be regarded as a distinct species, probably endemic in the Auckland Islands ; but, as I understand that the whole family is at the present time being fully worked up for the “ Pflanzenreich,” it is better to avoid proposing names, which would doubtless be anticipated before publication. POLYGONACEAE. Rumex neglectus, T. Kirk. Rumex cunerfolius, var. alismaefolius, Hook. f., Fl. Antarct., 1, 67 (1844), but not R. cunerfolius, Campd., Monog. Rum., 66, 95. AR. neglectus, T. Kirk in Trans, N.Z. Inst., 1x, 493 (1877). Auckland Islands: Sandy and pebbly beaches at Port Ross and Enderby Island ; Hooker, Kirk ! Cockayne, Tennant! (New Zealand, pebbly and sandy beaches from Cook Strait southwards.) The only specimens I have seen from the southern islands are mere scraps in poor condition ; but they seem to agree with the New Zealand plant. ; URTICACEAE. Urtica australis, Hook. f. Urtica australis, Hook. f., Fl. Antarct., 1, 68 (1844). Auckland Islands: Margins of woods near the s2a, not common; Hooker, Kirk ! Cockayne, Aston, Tennant! Campbell Island: Near the sea-shore, and also Systematic Botany. SUBANTARCTIC ISLANDS OF NEW ZEALAND. 427 in rocky clefts of the western mountains; Laing! Antipodes Island: Not un- common, especially on the eastern side; Kirk! (Chatham Islands; small islands in Foveaux Strait.) The apparent absence of this species from Stewart Island and the Snares is somewhat peculiar. The North Island locality quoted in the Handbook on the authority of Bidwill is no doubt erroneous. Urtica aucklandica, Hook. f. Urtica aucklandica, Hook. f., Fl. Antarct., 1, 69 (1844). This has not been gathered since its discovery by Sir J. D. Hooker in 1840. Judging from the original description, there is little to distinguish it from the pre- ceding, except the smaller size and copious greyish-white pubescence, but Hooker states that it is apparently quite distinct. ORCHIDACEAE. Thelymitra longifolia, Forst. Thelymatra longifolia, Forst., Char. Gen., 98, t. 49 (1776) ; Hook. £., Handb. N.Z. Fl., 270 (1864). TT. stenopetala, Hook. f., Fl. Antarct., 1, 69 (1844). T. Forsteri, Swz. in Vet. Akad. Handb. Stockh., xxi, 228 (1800) ; Hook. f., Fl. Nov. Zel., 1, 243 (1853). Auckland Islands: “In woods and on the bare ground in exposed places, not uncommon”; Hooker. (Throughout New Zealand; Australia; Tasmania.) I insert this solely on the authority of Sir J. D. Hooker, no subsequent botanist having collected it. Thelymitra uniflora, Hook. f. Thelymitra uniflora, Hook. f., Fl. Antarct., 1, 70 (1844); and u, 544 (1847). Auckland Islands: On the bare ground and in tufts of moss, Phyllachne, &c. ; Hooker, Le Guillon (D’Urville’s expedition). (New Zealand, upland districts from Rotorua southwards.) Like the preceding, this rests entirely on the record given in the “ Flora Ant- arctica.” It should be remarked, however, that most of the recent visits to the southern islands have been either too late or too early in the flowering season to secure good specimens of Orchidaceae. Prasophyllum Colensoi, Hook. f. Prasophyllum Colensor, Hook. f., Fl. Nov. Zel., 1, 241 (1853); T. Kirk in Trans. N.Z. Inst., xxi, 440 (1891). Antipodes Island: Exact station not named; Kirk. (Throughout New Zealand.) I have seen no specimens. Lyperanthus antarcticus, Hook. f. Lyperanthus antarcticus, Hook. f., Fl. Antarct., 1, 544 (1847). Auckland Islands: Hooker, Le Guilon (D’Urville’s expedition), General Bolton. Adams Island, alt. 1,200 ft.; Aston! (New Zealand, upland districts from the Tara- rua Range southwards ) 428 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. Two specimens collected by Mr. Aston are all that I have seen. These are very immature, but there is no reason for doubting the identification of the species. Caladenia Lyallii, Hook. f. Caladema sp. (4), Hook. f., FI. Antarct., i 1, 70 (1844). C. Lyall, Hook. f., RIE Nov. Zele, 1, 247 (1853). | Auckland Islands: “ Amongst moss in the woods near the sea” (New Zealand, subalpine localities from Nelson southwards.) I have seen no specimens. The colour of the flowers, as quoted in the “ Flora Antarctica,’ hardly corresponds with New Zealand specimens. ; Hooker. Caladenia bifolia, Hooker, f. Caladema sp. (5), Hook. f., Fl. Antarct., 1, 79 (1844). C. bsfolia, Hook. f., Fl. Nov. Zel., 1, 247 (1853). Chiloglottis Travers, F. Muell., Veg. Chath. Is., 51 (1864). Auckland and Campbell Islands: Amongst moss, &c., not uncommon ; Hooker, Laing! Antipodes Island: Kirk. (New Zealand, upland districts from Lake Taupo southwards.) Two immature specimens collected by Laing are all that I have seen. So far as a comparison can be made, they agree fairly well with New Zealand specimens. Chiloglottis cornuta, Hook. f. Chiloglottis cornuta, Hook. f., Fl. Antarct., 1, 69 (1844). Auckland and Campbell Islands: Mossy sheltered places, not uncommon; Hooker, Kirk! Aston! Laing! Antipodes Island: Kirk. (New Zealand, from Kai- tala southwards.) An examination of some flowers preserved in formalin, collected by Mr. Laing, proves that the calli on the labellum vary in number from 6 to 12, the shape and size also being very inconstant. Not one of the specimens exactly matches the description given in the “ Flora Antarctica.” I find that the calli are equally vari- able in New Zealand specimens, Corysanthes rivularis, Hook. f. Acanthus rvvularis, A. Cunn., Precur., n. 312 (1836) ; Hook. f., Fl. Antarct., 1, 71 (1844). Nematoceras rwwularis, Hook. f., Fl. Nov. Zel., 1, 251 (1853). Corysanthes rwularis, Hook. f., Handb. N.Z. Fl., 266 (1864). ‘Auckland Islands: Locality not stated; General Bolton: Norman’s Inlet; Aston! (New Zealand, wooded ravines from the North Cape southwards.) Mr. Aston sends numerous fine specimens of this, agreeing oa with the New Zealand specimens, | Corysanthes rotundifolia, Hook. f. Nematoceras rotundifolia, Hook. f., Fl. Nov. Zel., i, 251 (1853). Cory- santhes rotundifolia, Hook. f., Handb. N.Z. Fl., 266 (1864). Auckland and Campbell Islands: Hooker, Lamg? (Throughout New Zea- land.) Two specimens in formalin, collected by Laing, apparently belong to this species. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 429 Corysanthes triloba, Hook. f. Nematoceras triloba, Hook. f., Fl. Nov. Zel., i, 250 (1853). Corysanthes triloba, Hook. f., Handb. N.Z. Fl., 265 (1864). Auckland Islands: Camp Cove; Aston! Campbell Island: Laing! Chambers ! (New Zealand, shaded places from the North Cape southwards.) This is an addition to the florula of the southern islands. Mr. Laing’s speci- mens, preserved in formalin, clearly show the characteristic obtuse upper sepal. C. macrantha, which is sometimes difficult to separate from C. triloba, always has an acute or acuminate upper sepal. Corysanthes macrantha, Hook. f. Nematoceras macrantha, Hook. f., Fl. Nov. Zel., 1, 249 (1853). Corysanthes macrantha, Hook. f., Handb. N.Z. Fl., 266 (1864). Auckland and Campbell Islands: Shaded mossy places, not uncommon; Bolton, Aston! Tennant! Laing! Corresponds with New Zealand specimens. LILIACEAE. Astelia linearis, Hook. f. Astelva linearis, Hook. f., Fl. Antarct., i, 76 (1844). Auckland and Campbell Islands: Abundant in open places; Hooker and all subsequent collectors. (New Zealand, upland districts from the East Cape and Ruapehu southwards. ) This is represented in Fuegia by the closely allied A. pumila, R. Br., and in Victoria and Tasmania by the broader-leaved A. alpina, R. Br. Auckland Island specimens agree weil with the ordinary form of the species in New Zealand. Astelia subulata, Cheesem., n. sp. Astelia linearis, var. subulata, Hook. f., Fl. Antarct., i, 76 (1844). Herba pusilla, densissime caespitosa, habitu A. lineart sed multo minor. Folia minora, erecta vel paullo curvata, 1-5-2°5 cm. longa, 0°5-0:75 mm. lata, subulata, teretia vel superne obscure canaliculata, versus apicem cylindracea ; vaginae mem- branaceae, hyalinae, paleis argenteis obtectae. Scapus solitarius, terminalis, bre- visissimus. Flos solitarius, terminalis, simillimus ad A. lineari sed minor. Fructus lgnotus. Auckland and Campbell Islands: Wet open places on the hills, not uncommon ; Hooker, Cockayne, Aston! Tennant! Laing! (Kndemic.) This differs so much in foliage from all the states of A. linearis that I am acquainted with that I feel compelled to grant it the rank of a species. A. linearis has leaves varying from 1-8 in. long by ;4,-4+in. broad, channelled above and keeled below, more or less sulcate or grooved, and with the midrib and margins clothed with narrow hyaline scales. In A. subulata the leaves are seldom over 4in. long by 35 im. broad, terete or slightly flattened or canaliculate above, smooth, quite glabrous except the sheaths. I have seen only very imperfect flowering specimens, 430 SUBANTARCTIC ISLANDS OF NEW ZBHALAND. [Systematic Botany. all of them being males. The flowers closely resemble those of A. linearis, but are rather smaller, and are apparently always solitary. Mr. Aston mforms me that it is often found side by side with typical A. linearis, so that the difference between the two species can hardly be due to different environment. Bulbinella Rossii, Benth. & Hook. f Chrysobactron Rossu, Hook. f., Fl. Antarct., 1, 72, t. 44, 45 (1844). Vera- trum Dubouzetr, Homb. & Jacq., Voy. Astrol. et Zél., t. 4 (1853). Anthericum Rossu, Hook. f., Handb. N.Z. Fl, 285 (1864). Bulbinella Rossw, Benth. & Hook. f., Gen. Plant., 11, 784 (1883). Auckland and Campbell Islands: Abundant in open moist situations; Hooker and all subsequent collectors. (Hndemic.) I have nothing to add to the excellent account of this beautiful plant given in the “ Flora Antarctica.” JUNCACEAE. Rostkovia magellanica, Hook. f. Juncus magellanicus, Lam., Eneycl., ii, 266 (1789). Rostkovia sphaero- carpa, Desv., Journ. Bot., 1, 327 (1808). &. magellanica, Hook. f., Fl. Antarct., 1, 81 (1844). Auckland Islands: Exact locality not stated ; Aston! Campbell Island: Mossy and spongy places on the hills, especially at the sources of streams; Hooker. (Fuegia; Falkland Islands ; South Georgia.) Mr. Aston’s specimens are the first obtained on the Auckland Islands, Rostkovia gracilis, Hook. f. Rostkovia gracilis, Hook. f., Fl. Antarct., 1, 83, t. 47 (1844). Marsuppo- spermum gracile, Buchen. in Abh. Nat. Ver. Bremen, vi, 374 (1879) ; Monog. Junc., 68 (1890). Auckland and Campbell Islands: Not uncommon in marshy and rocky places ; Hooker and most subsequent collectors. (New Zealand, alpine localities from Nelson southwards.) The specimens submitted to me correspond fairly well with the ordinary state found in the mountains of the South Island. Juncus bufonius, Linn. Juncus bufonwus, Linn., Sp. Plant., 328 (1753); Hook. f., Fl. Nov. Zel., i, 264 (1853); T. Kirk, Report Austral. Ass., 111, 218, 222 (1891). The Snares: Kirk. Auckland Islands: Port Ross; Kirk! (Cosmopolitan.) Mr. Kirk’s specimens, gathered at Port Ross, are the only ones I have seen from the southern islands. Juncus effusus, Linn. Juncus effusus, Linn., Sp. Plant., 326 (1753). J. communes, H. Mey., Junc., 12 (1822); Hook. f., Handb. N.Z. FIl., 290 (1864). Campbell Island: In a watercourse not far from the sea-shore; Laing! (Cos- mopolitan, or nearly so.) Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 431 This is the first record of the presence of this species in the southern islands. Possibly it is introduced ; but I have no evidence of this, and consequently give it the benefit of the doubt. Juncus planifolius, R. Br. Juncus planifolius, R. Br., Prodr., 259 (1810): Hook. f., Fl. Antarct., u, 545 (1847); Handb. N.Z. Fl., 290 (1864). Auckland Islands: Le Guillon (D’Urville’s expedition, fide Hooker). (New Zealand ; Australia and Tasmania ; Chili.) The inclusion of this species in the florula of the southern islands rests solely on Hooker’s statement in the second part of the “Flora Antarctica” (p. 545), where he says, ““ Specimens very diminutive, but, I think, clearly referable to this species, which is not uncommon throughout the islands of New Zealand.” No subsequent botanist has met with it, and the doubt naturally arises as to whether a large state ot J. antarcticus (which is very closely allied to J. planifolius) has not been mis- taken for it. Juncus antarcticus, Hook. f Juncus antarcticus, Hook. f., Fl. Antarct., 1, 79, t. 46 (1844). J. brevifolius, T. Kirk in Trans. N.Z. Inst., xiv, 382 (1882). Auckland Islands: Adams Island; Kirk! Tennant! North Arm of Carnley Harbour, at sea-level; Cockayne! Campbell Island: Exposed summits of the mountains, alt. 1,000 ft., rare; Hooker. (New Zealand, mountain districts from Ruapehu southwards ) In New Zealand the ordinary state of the species has the culms considerably longer than the leaves, and this is also the case with Dr. Cockayne’s and Mr. Tennant’s Auckland Island specimens. The plate in the “ Flora Antarctica,” however, repre- sents a depauperated state (also found in New Zealand), in which the culms barely exceed the leaves. Mr. Kirk’s specimens are intermediate in character. Juncus scheuchzerioides, Gaud. Juncus scheuchzervoides, Gaud. in Ann. Sci. Nat., v, 100 (1825); Hook. f., Fl. Antarct., 1, 79 (1844). Auckland Islands: Marshy places, not uncommon; Hooker, Cockayne (with J. antarcticus), Aston! Campbell Island: Near the sea; Hooker. Antipodes Island: Kirk! (New Zealand, mountains of Central Otago ; South America, from southern Argentina to Fuegia and the Falkland Islands; South Georgia; Ker- guelen Island.) The few specimens I have seen of this species are in poor condition, and do not satisfactorily match the drawing and description given by Buchanan in the “Pflanzenreich.” It is possible that they may represent a distinct variety, or even species, but a better series must be obtained before the question can be finally decided, 432 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. Luzula campestris, D.C., var. crinita, Buchen. Luzula cromta, Hook. f., Fl. Antarct., 1, 84, t. 48 (1844); Handb. N.Z. FL, 293 (1864). L. campestris, D.C., var. crinita, Buchen. in Oesterr. Bot. Zeit., 215 (1898). Auckland and Campbell Islands: Not uncommon from sea-level to the sum- mits of the hills; Hooker and all subsequent collectors. Antipodes Island: Kirk! Macquarie Island: Fraser, Scott, Hamilton. (Hndemic.) The distinguishing characters of this variety are the very robust and firm, some- times almost rigid, habit ; leaves with the margins distinctly thickened and furnished with very long but somewhat distantly placed cilia; inflorescence often composed of a single conglobate head, but sometimes with 1-3 lateral heads; and very dark- chestnut bracts, which are usually most densely ciliate. Luzula campestris, D.C., var. floribunda, Buchen. Luzula campestris, D.C., var. floribunda, Buchen. in “ Pflanzenreich,” heft 25, 93 (1906). Auckland Islands: Port Ross; Aston! Tennant! (New Zealand.) Differs from any form of the preceding in the much less ngid habit, much broader and thinner leaves (sometimes 8 mm. diam.), and especially in the greatly developed inflorescence, which consists of very numerous (in one of Mr. Aston’s specimens over thirty) pedunculate heads arranged in an umbellate manner, and in the paler bracts, with almost white margins. The Auckland Island plant suits the description of var. floribunda better than any other, and comes very close to some South Island mountam zooms collected by myself which Buchenau includes in the Ween CENTROLEPIDACEAE. Gaimardia ciliata, Hook. f. Gamardia ciliata, Hook. f., Fl. Antarct., 1, 85 (1844). Centrolepis viridis, T. Kirk in Trans. N.Z. Inst., xxi, 441 (1891). Alepyrum ciliatum, Hieron., Pflanzent., teil 11, abt. 4, 13 (1895). Auckland and Campbell Islands: Abundant in boggy places, in open situations ; Hooker! Kirk! Cockayne, Aston! Laing! Chambers! (New Zealand, upland bogs from Ruapehu southwards.) I do not think that erther this or the foltowrinie species ought to be placed in Centrolepis, as has been done by Kirk and other authors. As pointed out in the Manual, they differ m the perennial densely pulvinate habit, the shape of the leaves, the flowers seldom more than one in each floral bract, and in the cells of the ovary or carpels) being frequently reduced to one. In the “ Pflanzenfamilien ” Hieronymus keeps up the genus Alepyrum for their reception, a course which is likely to lead to confusion, seeing that not one of the species which Robert Brown placed in the original genus Alepyrum is retained in it by Hieronymus. Until the whole family recelves a careful examination it 1s best to leave the species in Gavmardia. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 433 Gaimardia pallida, Hook. f. Gaimardia pallida, Hook. f., Fl. Antarct., 1, 86 (1844). Alepyrum pallidum, Hook. f., Fl. Nov. Zel., 1, 268 t. 62c (1854). Centrolepis pallida, Cheesm., Man. N.Z. Fl., 757 (1906). Campbell Island: Boggy places on the hills; Hooker, Kirk! Laing! (New Zealand, mountain-bogs from Ruapehu southwards.) This is doubtless found in the Auckland Islands as well, but I have seen no specimens from thence. CYPERACEAE. Scirpus aucklandicus, Boeck. Isolepis aucklandica, Hook. f., Fl. Antarct., 1, 88, t. 50 (1844); Handb. N.Z. FIl., 302 (1864). Scirpus aucklandicus, Boeck. in Linnaea, xxxvi, 491 (1869-70). The Snares, Auckland and Campbell Islands, Antipodes Island: Abundant in moist places from sea-level to a considerable height on the hills; Hooker, Kirk ! Cockayne, Aston! Tennant! Laing! (Mountain districts in New Zealand, from Lake Taupo southwards ; Tasmania; New Amsterdam.) This appears to be a most abundant plant all through the southern islands. Scirpus cernuus, Vahl. Scirpus cernuus, Vahl., Enum., 11, 245 (1806). S. riparius, Poir., Encycl., v, 103 (1813). Isolepis riparia, R. Br., Prodr., 222 (1810); Hook. f£., Handb. N.Z. Fl., 302 (1864). The Snares: Kirk. Auckland Islands: Kirk! Aston! Tennant! Antipodes Island: Kirk! (Cosmopolitan.) All the specimens I have seen from the southern islands are in an immature condition, and it is quite possible that some of them may prove to belong to S. auck- landicus, some forms of which come very near to S. cernuus. Scirpus antarcticus, Linn. Scurpus antarcticus, Linn., Mant., 11, 181 (1766). S. cartilagineus, Poir., Encycl., suppl., v, 103 (1813). Isolepis cartilaginea, R. Br., Prodr., 222 (1810) ; Hook. {., Handb. N.Z. Fl., 302 (1864). The Snares: Kirk. Reo HehiO ak New Tonanel but local; extra-tropical Aus- tralia ; South Africa; St. Helena.) I insert this solely on the authority of the late Mr. Kirk (Trans. N.Z. Inst., xxiii, 430), for no other botanist appears to have met with it in the southern islands. Carpha alpina, R. Br. Carpha alpina, R. Br., Prodr., 230 (1810); Hook. f., Handb. N.Z. Fl., 299 (1864). Auckland Islands: Carnley Harbour; Kirk! Cockayne! Adams Island: Aston. (New Zealand, mountain districts from Cape Colville southwards; Tasmania ; Victoria; New Guinea; Chili; Fuegia.) This does not seem to have been observed by Hooker, 29—S. 434 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. Oreobolus pumilio, R. Br., var. pectinatus, C. B. Clarke. Oreobolus pectinatus, Hook. f., Fl. Antarct., 1, 87, t. 49 (1844). O. pumilro, R. Br., var. pectinatus, C. B. Clarke ex Cheesem., Man. N.Z. Fl., 797 (1906). Auckland and Campbell Islands: Abundant in open situations, forming dense convex masses; Hooker and all subsequent collectors. (New Zealand, mountain districts from the Hast Cape and Taupo southwards.) The strongly distichous leaves give this variety a very different appearance from the type; so much so that it almost deserves recognition as a species. Uncinia riparia, R. Br. Unena riparia, R. Br., Prodr., 241 (1810). U. Hooker, Boott m Hook. f., Fl. Antarct., 291, t. 51 (1844). U-. rupestris, Hook. f., Handb. N.Z. FI., 310 (1864), in part, but not U. rupestris, Raoul. Auckland and Campbell Islands: Grassy places in the woods, and among rocks on the hills; Hooker, Kirk! Cockayne, Aston! Tennant! Laing! Macquarie Island: Hamilton. (Throughout New Zealand; Victoria; Tasmania.) The late Mr. C. B. Clarke referred the U. Hookeri of Boott to Raoul’s U. rupestris, and I adopted this determination in the Manual. But the specimens now placed in my hands for determination differ so much from Raoul’s figure and description, and at the same time approach so closely to U. rvparza, that I feel compelled to place them under that species, a course which was suggested to me some years ago by Dr. Kukenthal, who has made a special study of the genus. Carex appressa, R. Br. Carex appressa, R. Br., Prodr., 242 (1810); Hook. f., Fl. Antarct., 1, 90 (1844). C. paniculata, Linn., var. appressa, Cheesm. in Trans. N.Z. Inst., xvi, 427 (1884). Auckland, Campbell, and Antipodes Islands: Abundant, especially near the sea, often forming large tussocks; Hooker, Kirk! Cockayne! Aston, Laing! (New Zealand, from Dunedin southwards ; extra-tropical Australia ; Tasmania.) I have no additional information to record respecting this well-known species. Carex ternaria, Forst. Carex ternaria, Forst., Prodr., n. 549 (1786); Hook. f., Fl. Antarct., i, 89 (1844) ; Handb. N.Z. FI., 314 (1864). Auckland Islands: Margins of woods near the sea, not common; Hooker. Campbell Island: Exact locality not stated; Lamg. Antipodes Island: Boggy places, abundant; Kirk, Cockayne. (New Zealand.) No specimens of this were brought back by the expedition, probably on account of the early period of the season making it impossible to obtain flowering specimens. Carex trifida, Cav. Carex trifida, Cav., Ic., v, 41, t. 465 (1799); Hook. f., Fl. Antarct., i, 89 (1844). The Snares: Not uncommon; Kirk! Auckland and Campbell Islands: Moist places near the sea; Hooker, Kirk! Aston, Laing! Antipodes Island; Kirk. (New Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 435 Zealand, shores of the South Island, from Cook Strait southwards, but local; Chili ; Falkland Islands.) The specimens submitted to me agree perfectly well with others gathered from various localities in the South Island. The distribution of the species in South America is remarkable. It was gathered by Darwin at Cape Tres Montes, to the south of the Chonos Archipelago, and it is abundant im the Falkland Islands, but so far has not been found in the intervening districts of southern Patagonia and Fuegia. GRAMINEAE. (The grasses collected by the expedition form the subject of a separate memoir, by Mr. D. Petrie. In all, twenty-five species are recorded, which, following the enumeration adopted in this article, brings the total number of phanerogams up to 147.* A list of the species will be found in the tabular arrangement of the florula given towards the close of this article; for further particulars reference should be made to Mr. Petrie’s memoir, p. 472.) FILICEs. Hymenophyllum rarum, R. Br. Hymenophyllum rarum, R. Br., Prodr., 159 (1810); Hook. f., Fl. Antarct., 1, 105 (1844). H. semibwalve, Hook. & Grev., Ic. Fil., t. 83, (1829). H. wmbricatum, Col. in Tasm. Journ. Nat. Sc., u, 187 (1844). Auckland Islands: Woods near the sea, not common; Hooker, Tennant! Campbell Island: A single scrap without locality; Laing! (Throughout New Zea- land; Victoria; Tasmania; extra-tropical South America; South Africa; Mauritius ; Polynesia; Japan.) Mr. Tennant sends a dwarf state with fronds less than an inch in height, and with much imbricated pinnae, evidently answering to Colenso’s H. wmbricatum. The range of the species quoted above is that given in the “‘ Synopsis Filicum,” but many authors now separate the American and African plants. Hymenophyllum polyanthos, Swartz, var. sanguinolentum, Hook. Hymenophyllum polyanthos, Swartz, Gen. et Sp. Fil., 102 (1801), var. sanguinolentum, Hook., Sp. Fil., 1, 107 (1846). Trichomanes sanguino- lentum, Forst., Prodr., n. 465 (1786). Hymenophyllum sanguinolentum, Swartz, Gen. et Sp. Fil., 101 (1801). Auckland Islands: Amongst moss, &c.; Kirk! (The type of the species is widely distributed ; var. sanguinolentum extends throughout New Zealand.) Mr. Kirk’s specimens are the only ones I have seen from the southern islands. Hymenophyllum villosum, Col. Hymenophyllum villosum, Col. in Tasm. Journ. Nat. Sci. 11, 185 (1844) ; T. Kirk in Trans. N.Z. Inst., x, 395 (1878). Auckland and Campbell Islands: Not uncommon amongst moss, &c.; Kirk, Tennant ! Laing! (New Zealand, mountain districts from Cape Colville southwards.) * A few additions to these numbers have been made in the appendices to the papers by Mr. Petrie and Mr. Cheeseman (see pp. 479 and 446). These additions are included in the Tabular View on p- 448.—Ep. 456 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. Hymenophyllum dilatatum, Swartz. Trichomanes dilatatum, Forst., Prodr., n. 467 (1786). Hymenophyllum dila- tatum, Swartz, Gen. et Sp. Fil, 100 (1801); Hook. f., Handb. N.Z. Fl., 354 (1864). Auckland Islands: Hooker. (Throughout New Zealand ; Polynesian islands ; Java.) I have seen no specimens from the southern islands. Hooker does not quote it in the “‘ Flora Antarctica,” but includes it as a member of the florula in the Handbook. Hymenophyllum demissum, Swartz. Trichomanes denussum, Forst., Prodr., n. 468 (1786). Hymenophyllum denussum, Swartz, Gen. et Sp. Fil., 180 (1801); Hook. f., Fl. Antarct., i, 104 (1844). Auckland Islands: Woods near the sea, often covering the ground in large patches; Hooker. (Throughout New Zealand ; Polynesian islands ; Java.) Sir J. D. Hooker remarks that his specimens are smaller than those collected in the North Island of New Zealand, but do not otherwise differ. Curiously enough, no specimens exist in any of the collections formed by the members of the recent expedition. Hymenophyllum flabellatum, Lab. Hymenophyllum flabellatum, Lab., Pl. Nov. Holl., u, 101, t. 250 (1806) ; Hook. f., Fl. Antarct., 1, 105 (1844). Auckland Islands: Not uncommon on trunks of trees and rocks, both near the sea and on the hills; Hooker, Tennant! (Throughout New Zealand ; extra-tropical Australia; Tasmania; Polynesian islands.) Mr. Tennant’s specimens, which are all I have seen, belong to a curious form intermediate between H. flabellatum and H. rufescens, in which the stipes, rhachis, and costae are furnished with long flexuose hairs exactly as in H. rufescens, but in which the fronds are not nearly so delicate nor the stipes so long and slender as in typical rufescens. It could almost be referred to either species, but, on the whole, it seems best to keep it with H. flabellatum. Hymenophyllum minimum, A. Rich. Hymenophyllum mimmum, A. Rich., Fl. Nov. Zel., 91, t. 14, f. H (1832) ; Hook. f., Fl. Antarct., 1, 103 (1844). Auckland Islands: On trunks of trees in the woods, rare; Hooker. (New Zealand, coastal districts from Tasman’s Bay southwards; rare and local.) This was not collected during the recent expedition. As I have remarked in the Manual, it is a much misunderstood species, many collectors confusing it with small states of H. Tunbridgense, from which it differs in the smaller size, in the uni- formly terminal and solitary sori, and in the indusium being spinose on the faces as well as on the margins. Hooker’s description and remarks, given in the “ Flora Antarctica,” amply prove that he was dealing with the typical plant, Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 437 Hymenophyllum Tunbridgense, Smith. Trichomanes Tunbridgense, Linn., Sp. Plant., 1098 (1753). Hymenophyllum Tunbridgense, Smith, Engl. Bot., t. 162 (1794); Hook. f., Handb. N.Z. Fl., 352 (1864). Auckland Islands: Hooker. (Most temperate and subtropical countries.) This is another fern of which I have seen no specimens from the southern islands. It has been included in the lists on the authority of Hooker’s Handbook, where “ Lord Auckland’s Group ” is quoted as a locality ; but if this is correct it is singular that it is not included in the enumeration of species given in the “ Flora Antarctica.” Hymenophyllum multifidum, Swartz. Trichomanes multifidum, Forst., Prodr., n. 473 (1786). Hymenophyllum multifidum, Swartz, Syn. Fil., 146, 378 (1806) ; Hook. f., Fl. Antarct., i, 104 (1844). Auckland and Campbell Islands: “In all situations, from the level of the sea to the tops of the mountains, growing on the ground, on trunks of trees, and on rocks” ; Hooker and all subsequent collectors. Antipodes Island: Kirk ! Cockayne. (Throughout New Zealand ; New South Wales; Polynesia; Malaysia.) — This is the most abundant species of the genus in the southern islands, and has a wonderful range of habitats. When growing in open places, amongst rocks or erass, the fronds are often curved downwards, and the pinnae rolled inwards, so that the appearance of the species is very different from that which it assumes when erowing in woods. Hymenophyllum bivalve, Swartz. Trichomanes bivalve, Forst., Prodr., n. 466 (1786) ; Hymenophyllum bivalve, Swartz, Syn. Fil., 146, 372 (1806); Hook. f£., Handb. N.Z. Fl., 353 (1864) ; T. Kirk in Report Austral. Assoc., 11, 222 (1891). Auckland Islands: Port Ross; Kirk! (New Zealand, from Cape Colville south- wards ; New South Wales.) I have seen no specimens but Mr. Kirk’s. Hemitelia Smithii, Hook. Cyathea Snuthu, Hook. f., Fl. Nov. Zel., u, 8, t. 72 (1853). Hematelia Smithu, Hook., Syn. Fil., 31 (1874); Cockayne in Trans. N.Z. Inst., Xxxvi, 246 (1904). Auckland Islands: Rata forest, Norman’s Inlet; Joss and Cockayne! (Throughout New Zealand.) The discovery of this species in the Auckland Islands, first recorded by Dr. Cockayne in his paper on “ A Botanical Excursion to the Southern Islands ” (Trans. N.Z. Inst., xxxvi, 246), 1s of considerable interest from the point of view of botanical geography, as it advances the known southern limit of tree-ferns by rather more than 2 degrees. 438 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. Hypolepis millefolium, Hook. Hypolepis millefoliwm, Hook., Sp. Fil., , 68, t. 958 (1858); T. Kirk in Trans. N.Z. Inst., xxii, 441 (1891). Campbell Island: Kirk, Laing! Chambers! Antipodes Island: Kirk, Cockayne. (New Zealand, mountain districts from the Hast Cape southwards.) The apparent absence of this species from the Auckland Islands is remarkable. Pteris aquilina, var. esculenta, Hook. f. Pteris esculenta, Forst., Prodr., n. 418 (1786). P. aquilina, Linn., Sp. Plant., 1075 (1758), var. esculenta, Hook. f., Fl. Nov. Zel., m, 35 (1855). Piterrdium aquilinum, Kuhn in Deck. Reis. Bot., 11 (1879). Campbell Island: Hooker? (Pteris aquilina in some of its forms is almost cosmopolitan ; var. esculenta is widely spread in the Southern Hemisphere.) I do not find any specimen of this in the collections made by the expedition ; nor was it gathered by Kirk in 1890 or Cockayne in 1903. It is inserted in the enumeration on the strength of Hooker’s statement in the Handbook, where he says, ‘“ Abundant throughout the islands, from the Kermadecs to Campbell Island.” But he does not mention it i the “ Flora Antarctica,” and it is quite possible that the locality found its way into the Handbook by an accidental slip of the pen. Pteris incisa, Thunb. Pteris incisa, Thunb., Fl. Cap., 171 (1800); Hook. f., Handb. N.Z. FI., 364 (1864). P. vespertilionis, Lab., Plant. Nov. Holl., 1, 96, t. 245 (1806); Hook. f., Fl. Antarct., 1, 110 (1844). Hustiopteris incisa, J. Sm., Hist. Fil., 295 (1875). Auckland, Campbell, and Antipodes Islands: Not uncommon, usually not far from the sea; Hooker, Kirk, Cockayne, Aston, Tennant, Laing! Chambers! (Uni1- versal through the tropics and South Temperate Zone.) This reaches its southern limit in Campbell Island, for although it stretches as far south as southern Chili, Juan Fernandez, and Tristan d’Acunha, it is not found in Fuegia, the Falkland Islands, or Kerguelen Island. Lomaria Patersoni, Spreng. Stegama Patersom, R. Br., Prodr., 152 (1810). Lomaria Patersoni, Spreng., Syst. Veg., iv, 62 (1827). L. elongata, Blume, Enum. Fil. Jav., 1, 201 (1828); Hook. f., Fl. Nov. Zel., 11, 29 (1855). Blechnum Patersona, Mett., Fil. Lips., 64, t. 4 (1856). Auckland Island: Scrubby forest at the head of the North Arm of Carnley Harbour; Cockayne! Aston! Tennant! (Philippine Islands; India and Malaysia southwards to Polynesia ; Australia and Tasmania ; New Zealand.) Now recorded for the first time from the southern islands. L. Patersoni was formerly separated into two varieties, or even species, according to the different views of authors—first, the typical Paterson, which was believed to always have simple Systematic Botany.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 439 fronds, and, secondly, elongata, in which the fronds are pinnatifid. But it was long ago proved that elongata commonly produces simple fronds, and Patersoni has also been found with pinnatifid ones, so that the supposed distinction falls to the eround. The specimens brought from the Auckland Islands show both forms of fronds, Diels, in the “ Pflanzenfamilien,” and Christensen in the more recent “‘ Index Filicum,” have both followed Mettenius in reducing Lomaria to Blechnum, and it must be confessed that the two genera are much too closely allied. Lomaria discolor, Willd. Osmunda discolor, Forst., Prodr., n. 413 (1786). Lomaria discolor, Willd., Sp. Plant., v, 293 (1810). Hook. f., Fl. Nov. Zel., 1, 30 (1855). Blech- num discolor, Mett., Fil. Lips., 64 (1856). Auckland and Campbell Islands: Hooker. (Throughout New Zealand ; Nor- folk Island ; Australia and Tasmania.) This is not mentioned in the “ Flora Antarctica,” but both Sir J. D. Hooker, in the Handbook, and Sir W. J. Hooker, in the “ Species Filicum,” give it as a native of the Auckland and Campbell Islands. So far as I am aware, it has not been found by any recent collector, and the supposition arises that large forms of L. dura (a species which was not recognised at the time of Hooker’s visit) were taken for it. Lomaria dura, Moore. Lomaria lanceolata, Hook. f., Fl. Antarct., 1, 110 (1844), not of Spreng. LI. dura, Moore in Gard. Chron., 1866, 290; Hook. f., Handb. N.Z. FL., 748 (1867). Blechnum durum, C. Chr., Ind. Fil., 153 (1905). The Snares, Auckland, Campbell, and Antipodes Islands: Rocks and woods near the sea, abundant; Hooker, Kirk! and all subsequent collectors. (New Zealand, coastal localities from Banks Peninsula and the Chatham Islands southwards.) I think there can be no reasonable doubt that the Lomaria lanceolata of the “ Flora Antarctica” is identical with Lomaria dura. No subsequent traveller has met with L. lanceolata ; and L. dura is much too abundant in the islands to have escaped the notice of such a keen observer as Hooker. Lomaria penna-marina, Trev. Polypodium penna-marma, Poir., Encycl., v, 529 (1804). Stegania alpina, R. Br., Prodr., 152 (1810). Lomaria alpina, Spreng., Syst. Veg., iv, 62 (1827); Hook. f£., Fl. Antarct., n, 393, t. 150 (1847). Blechnum alpinum, Mett., Fil. Lips., 64 (1856). B. penna-marina, Kuhn, Fil. Afr., 92 (1869). Lomaria penna-marima, Trev., Att. Inst. Ven., 14, 570 (1869). Antipodes Island: Abundant; Kirk, Cockayne. Macquarie Island: Scott, Hamilton. (New Zealand, from the upper Thames and Rotorua southwards; extra- 440 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. tropical Australia, and Tasmania; Juan Fernandez; South America, from Chili southwards to Fuegia and the Falkland Islands; Kerguelen Island and the Crozets ; Tristan d’Acunha ; Amsterdam and St. Paul Islands.) From the above sketch of the distribution of this species it will be seen that it occurs in almost all lands between S. lat. 35° and 55°. This makes it all the more remarkable that it has not yet been detected in Auckland or Campbell Islands. Lomaria capensis, Willd. Osmunda capensis, Linn., Mant., 306 (1771). O. procera, Forst., Prodr., n. 414 (1786). Stegania procera, R. Br., Prodr., 153 (1810). Lomaria capensis, Willd., Sp. Plant., v, 291 (1810). L. procera, Spreng., Syst. Veg., iv, 65 (1827); Hook. f., Fl. Antarct., i, 110 (1844). Blechnum capense, Schlecht, Adumb. Fil., 34, t. 18 (1832). Auckland, Campbell, and Antipodes Islands: Not uncommon, especially on the banks of streams, and in a stunted form ascending to a considerable height on the hills; Hooker, Kirk, Cockayne, Aston, Chambers! (New Zealand, from the Kermadec Islands southwards; Australia and Tasmania; Malaysia; Polynesia ; South America, from Chili to Mexico ; South Africa.) Attains its extreme southern limit in Campbell Island, for in America it does not advance further south than Juan Fernandez and southern Chili. In Fuegia, the Falkland Islands, and Tristan d’Acunha its place it taken by the closely related L. Boryana, Willd. (L. tabularis, Mett.). . Lomaria fluviatilis, Spreng. Stegama fluviatilis, R. Br., Prodr., 152 (1810). Lomaria fluviatilis, Spreng., Syst. Veg., iv, 65 (1827); Hook. f., Fl. Nov. Zel., u, 28 (1855). Blech- num fluriatile, Salom., Nom., 115 (1883). Auckland Islands: Rata forest near Camp Cove, Carnley Harbour; Cockayne ! (Throughout New Zealand ; extra-tropical Australia ; Tasmania.) This was first gathered in Dr. Ceckayne’s visit of 1903. Asplenium obtusatum, Forst. Asplenvum obtusatum, Forst., Prodr., n. 430 (1786) ; Hook. f., Fl. Antarct., 1, 108 (1844), in part; Homb. & Jacq., Voy. au Pole sud, Crypt., t. 1, f. B., small specimen (1853). The Snares, Auckland and Campbell Islands, Antipodes Island: Not uncommon on rocks near the sea ; Hooker, D’Urville, Kirk, and others. (Coasts of New Zealand, from the Kermadec Islands southwards; Australia and Tasmania; Polynesia ; Juan Fernandez ; South America, from Peru to southern Chil1.) A. obtusatum is here restricted to Forster’s original plant, with short obtuse pinnae. Hombron and Jacquinot’s plate, cited above, exactly represents a small state of this, and I have seen specimens collected by Mr. Chambers on Campbell Island which also match the type. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 44] Asplenium lucidum, Forst., var. obliquum, Moore. Asplenvum obliquum, Forst., Prodr., n. 429 (1786). A. obtusatum, Forst., var. oblaquum, Hook. f., Fl. Antarct., 1, 108 (1844). A. apice-dentatum, Homb. & Jacq., Voy. au Pédle sud, Crypt., t. 1, D (1853). A. obtu- satum, Homb. & Jacq., l.c., t. 1, B., large specimen (1853). A. lucidum, Forst., var. oblaguum, Moore, Index Fil., 142 (1858). Auckland and Campbell Islands: Rocks and woods near the sea, not uncommon ; Hooker, D’Urville, and subsequent collectors. (Throughout New Zealand; Aus- tralia ; Polynesia ; extra-tropical South America; Tristan d’Acunha.) All botanists agree in considering that Forster’s A. obliquum is not entitled to rank as a separate species; but, as it stands just half-way between A. obtusatum and A. lucidum, it becomes very much a matter of taste to decide under which of the two it should be placed. Personally, as stated in the Manual, I think it falls more naturally into A. lucedum. Asplenium lucidum, Forst., var. scleroprium, Moore. Asplenium scleroprium, Homb. & Jacq. ex Hook. f., Fl. Antarct., 1, 109 (1844); Voy. au Pole sud, Crypt., t. 1, D (1853). A. flaccedum, Forst., var. aucklandicum, Hook. f., Fl. Antarct., 1, 109 (1844). A. lucidum, Forst., var. scleropriwm, Moore, Index Fil., 142 (1858). Auckland Islands: Woods near the sea, abundant; Hooker, D’Urville, Kirk ! Cockayne ! and others. (Herekopere Island, in Foveaux Strait.) Hombron and Jacquinot’s figure shows the pinnae less-deeply divided than in the majority of the specimens I have seen. The bright shining green very coriaceous fronds give it a handsome appearance when fresh. Asplenium bulbiferum, Forst. Asplenwum bulbiferum, Forst., Prodr., n. 433 (1786); Hook. f., Fl. Nov. Zel., ii, 34 (1855). Antipodes Islands: Not uncommon ; Kirk! Cockayne! (Throughout New Zea- land ; Australia and Tasmania, and northwards to Malaysia and India ; Polynesia.) The Antipodes Island plant is an erect, rather fleshy form, not at all like the usual state of the species in New Zealand, and in some respects showing an approach to A. flaccodum. Aspidium vestitum, Swartz. Polypodium vestitum, Forst., Prodr., n. 445 (1756). Call | Be “3 Panax simplex, Forst. xe New Zealand. RUBIACEAE. Coprosma parviflora, Hook. f. xe fe ac New Zealand. 35 cmtaTA, Hook. f. | Sm ll axl) ee Endemic (?). Pa foetidissima, Forst. | Me kG) a New Zealand. A cuneata, Hook. f. x 3K 5 ee a aa repens, Hook. f. | x | x | x | New Zealand, Victoria, ‘Tasmania. Nertera depressa, Banks & Sol. {heer a [peeexe New Zealand, America from Mexico to Fuegia and Falkland Islands, . Tristan d’Acunha. CoMPOSITAE. Lagenophora pumila, Cheesem. eg exis lime ax New Zealand. Olearia Lyatiit, Hook. f. x >) eae Endemic. PLEUROPHYLLUM SPECIOSUM, Hook. it |) 22 S82 oe s 7 CRINIFERUM, Hook. f | x x xu - cf Hooxert, Buch. essex x FA Celmisia VERNICOSA, Hook. f. Neeson (fax ap a CAMPBELLENSIS, F. R. Chapm. Brsel|| sexs Il! =x: cs Gnaphalium luteo-album, Linn. der ile Rs Most warm-temperate or tropical regions. Helichrysum bellidioides, Willd., var. eae T. Kirk] .. | || oe libs New Zealand. Cassinia Vauwwilliersi, Hook. i ; i / x xx (2) 3 Craspedia uniflora, Forst. suuiimciee New Zealand, Australia, Tasmania. Cotula australis, Hook. f. ieee «lex New Zealand, Australia, Tasmania, | Tristan d’Acunha. plumosa, Hook. f. lex x x | Kerguelen, Crozets. », LANATA, Hook. f. xe || x Endemic. ;, dioica, Hook. f. Pesca bitte New Zealand. Abrotanella sPATHULATA, Hook. A | x x Endemic. a ROSULATA, Hook. f. j-- | x fe Erechtites prenanthoides, D.C. (?) .. | a New Zealand. Senecio ANTIPODUS, T. Kirk ne x Endemic. = Stewartiae, Armstr. xe New Zealand (Herekopere Island, in Foveaux Strait). Taraxacum officinale, Wigg x Temperate regions of both hemi- tole} I t=) spheres. Sonchus asper, Hill la. pelt Se II Sz Cosmopolitan. | | STYLIDIACEAE. | Phyllachne clavigera, F. Muell. | me | xe New Zealand. | CAMPANULACEAE. Pratia arenaria, Hook. f. xg x x South-eastern New Zealand, Chatham Islands. EPACRIDACEAE. Cyathodes empetrifolia, Hook. f. x4) |< (@) New Zealand. Dracophyllum longifolium, R. Br. xe || 3x 3 5 scoparium, Hook. f. x Chatham Islands, Pp : 30—S, 450 SUBANTARCTIC ISLANDS OF NEW ZHALAND. [Systematic Botany. TABULAR VIEW OF THE DISTRIBUTION oF THE PHANEROGAMS, ETC.—continued. 4\2¢ |e lou | es eae Sales | ee | 88] 28 Distribution. &)Sq| 83 | 2a | ge BI O7 | ay™ | PRIMULACEAE. Samolus repens. Pers. x New Zealand, Australia, Tasmania. MYRsSINACEAR. Myrsine divaricata, A. Cunn. xe | ae New Zealand. GENTIANACEAE. Gentiana CERINA, Hook. f. eX eX: Endemic. me CoNcINNA, Hook. f. ails a sk ANTARCTICA, T. Kirk Kelle es 5 ANTIPODA, T. Kirk x “A BORAGINACEAE. Myosotis antarctica, Hook. f. ea line < New Zealand. a CAPITATA, Hook. f. Weel Endemic. c albida, Cheesem. x New Zealand. ScROPHULARIACEAE. Veronica elliptica, Forst.. . Xa Soa ES New Zealand, Fuegia, Falkland Is- lands. 5 oporA, Hook. f. a) ieee Endemic. 6 BentuHami, Hook. f. x x NS PLANTAGINACEARH. Plantago AUCKLANDICA, Hook. f. .. P| ne Endemic. 3 Brown, Rap. .. Ky u|| Bk New Zealand, Victoria, Tasmania. POLYGONACEAE. Rumex neglectus, TV. Kirk x New Zealand. URTICACEAE. | Urtica australis, Hook. f. x x IK New Zealand (Foveaux Strait), Chat- ham Islands. > AUCKLANDICA, Hook f. x Endemic. ORCHIDACEAE. Thelymitra longifolia, Forst. 3K New Zealand, Australia, Tasmania. uniflora, Hook. f. | x oe New Zealand. Prasophyllum Colensoi, Hook. f. | ak x 5 Lyperanthus antarcticus, Hook. f. | K 3 Caladenia Lyall, Hook. f. XA) eel es 3 me bifolia, Hook. f. xe |) se | xs an Chiloglottis cornuta, Hools. f. x x XK a Corysanthes oblonga, Hook. f. x 55 35 rivularis, Hook.f. .. elie ; Ap rotundifolia, Hook. f.. . | x x 3 a triloba, Hook. f. 5 x x 5 on macrantha, Hook. f. .. x x * LILIACEAE. Astelia linearis, Hook. f. me il (ose New Zealand. >, SUBULATA, Cheesem. aie x x Endemic. Bulbinella Rosstr, Benth. & Hook. f. x x BB JUNCACEAR. Rostkovia magellanica, Hook. f. x x Fuegia, Falkland Islands, South Georgia. aD gracilis, Hook. f. ll eae ROX New Zealand. Juncus bufonius, Linn. eed Wi Nae hiner Cosmopolitan. » effusus, Linn. Pete line: 35 planifolius, R. Br. x New Zealand, Australia and Tas- mania, Chili. » anitarcticus, Hook, f. ar x || 38 New Zealand. Systematic Botany. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 451 TaBULAR VIEW OF THE DISTRIBUTION OF THE PHANEROGAMS, ETC.—continud. g Ee a. on 3 ats 22/35/28) 88] 38 Distribution. | Sh ST pe | JUNCACEAE—~ontinued. | Juncus scheuchzerioides, Gaud. a = a ilo | x x | .. | New Zealand, Fuegia, Falkland Is- lands, South Georgia, Kerguelen. Inzula campestris, D.C.. var. cRIntTA, Buchen. sey lla ll Seok See x | Endemic. 3 a », var. floribunda, Buchen. ..|..| x |../|.. | .. | New Zealand. CENTROLEPIDACEAER. Gaimardia ciliata, Hook. f. am Ki po | oo dt ee | ae ll Se loo 1) New Zenlbuack 3, pallida, Hook. f. B0 30 Saul eevee d|seis sy iia Soe liars Waa ” CYPERACEAE. Scirpus aucklandicus, Boeck. 8 ae -.| x | x | x | x | .. | New Zealand, Tasmania, New Am- sterdam. me cennwuus, Vahl: a ae a) x | x 2. | x | a) | Cosmopolitan: 3 antarcticus, Linn. BO ae .. | x |..|..|.. | .. | New Zealand, Australia, South Africa, St. Helena. Carpha alpina, R. Br... aa on ..{|..|xif..i]..|].. |New Zealand, Australia and Tas- | mania, New Guinea, Chili, Fuegia. Oreobolus pumilio, R. Br., var. ae ee C. B. Clarke | .. | x | x | New Zealand. Uncinia australis, Pers. oe | eetedee< .. | New Zealand, Sandwich Islands. » reparia, R. Br. .. fe be ..|.. |x | x |.. | x | New Zealand, Victoria, Tasmania. Carex appressa, R. Br. ../| x | x | x |.. | New Zealand, Australia, Tasmania. ;, ternaria, Forst. .. Be aN Sc oall waniece al ce al es x New Zealand. » trifida, Cav. eg ae as Soe fi e< a lmee- Sl tos lc New Zealand, Chili, Falkland Islands. GRAMINEAE. Microlaena Thomsoni, Petrie Ss als co | ao |) S| os I oo Neo | Ne@w Zeallamel, avenacea, Hook. f. a re caee | eset lpoasad |i: steals . H jefechioe redolens, R. Br. cae Ups Se le eical lta < x | ..|.. | New Zealand, Australia and Tas- mania, Fuegia. te Brunonts, Hook. f. .. oe olen ILS -< x | .. | .. | Endemic. Agrostis magellanica, Lam. oe 3 .|.. | x | x | x | x | New Zealand, Chili, Fuegia, Falkland Islands, Kerguelen, Marion Island, Heard Island. 3 subulata, Hook. f. (A. Muelleri, Benth.) x || sx New Zealand, Australia. Deyeuxia filiformis, Petrie, var. Lyallii, Hack. so se ll xs || os | oo th on | Wen Zepleyact = = Ps var. MICRATHERA, Hack... | .. | .. | x x | .. | Endemic. oO setifolia, Hook. f. x New Zealand. Deschampsia caespitosa, Beauv. Se lleae Cosmopolitan. Rs Chapmani, Petrie se || xe |) xe New Zealand. 8 GRACILILIMA, T. Kirk sian eon Oa Kalle .. | Endemic. 55 PPNICILLATA, T. Kirk =e pStack Ne cartarl taka, o|naoees dls 5 Trisetum subspicatum, Beauv. x x New Zealand, Tasmania, Fuegia, Arctic regions. Danthonia antarctica, Hook. f. ac is ee lee exs el excliee alien lNews Zeallands Poa foliosa, Hook. f. me oi6 & .. | x | x | x | x | x | New Zealand (Herekopere Island, in Foveaux Strait). ,, TENNANTIANA, Petrie a w ee lex: Ol es Shee exe Endemic. ,, ANTIPODA, Petrie XK x x is ;> Livorosa, Cheesem. sig ee Sf aeatl| ea ae BE ,, Astoni, Petrie 5 oe wh a Stenee< x New Zealand. ,, RAMOSISSIMA, Hook. f. He Be so | bo | && | 32 Ion |) oo. |) ldinokemnie, >, Hamitont, T. Kirk ae se BOR ee ccah Moral ey alltea ull abc Fe s>» AUCKLANDICA, Petrie x xe 35 s5 INCRASSATA, Petrie .. x xe 55 >, BREVIGLUMIS, Hook. f.* x XE 3 Atropis stricta, Hack. iy: ae We oe lp || se een! eeeieNew Zealand, , ANTIPODA, Petric a he 50 | oo bo los |) Se |) oo |} Mthackaomie, Festuca conTRACTA, T, Kirk os ao Seritl Prous. llegae) ul oaaecedl eae tae $5 » rubra, Linn. as ws eee eee lee| eee Cosmopolitan: a ovina, Linn. a ae 23 SONA eee Meee (ote lie. Seal ois “ * Mr. Petrie considers this to be identical with the New Zealand P. imbecilla, Forst. 452 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. TABULAR VIEW OF THE DISTRIBUTION OF THE PHANEROGAMS, ETC.—continued. \/89/eg|tdlad — 25 a8 a5 Ee as Distribution. i) an an Bie Sr FILICEs. Hymenophyllum rarum, BR. Br. x x New Zealand, Victoria, Tasmania, : temperate South America, South Africa, Mauritius, Polynesia, Japan. oe polyanthos, Swartz, var. sanguinolen- Bee New Zealand. tum, Hook. Ee villosum, Col. x x 5 3 dilatatum, Swartz x New Zealand, Polynesia, Java. a demissum, Swartz x aS 99 99 ae flabellatum, Labill. x New Zealand, Australia, Polynesia. a minimum, A. Rich. S| igs New Zealand. A Tunbridgense, Smith KR l[ecsareilh cae Almost cosmopolitan. 3 multifidum, Swartz PX |e Nex New Zealand, Australia, Polynesia, Malaya. 3 bivalve, Swartz x New Zealand, Australia. Hemitelia Smuthu, Hook. axle evr [are New Zealand. Hypolepis millefoluwm, Hook. 6 | 5K 2 Pteris aquilina, var. esculenta, Forst. x (?) Throughout the Southern Hemi- sphere. > ~mncisa, Thunb. K K | Tropics and Southern Hemisphere. Lomaria Paterson, Spreng. x New Zealand, and widely spread in the Southern Hemisphere. ss discolor, Willd. . . x x New Zealand, Australia and Tas- mania, Norfolk Island. a4 dura, Moore x || 3x x x | .. | New Zealand. ne penna-marina, Trey. x x | Throughout the South Temperate Zone. f capensis, Willd. x x ox Ditto. 5 fluviatilis, Spreng. oe Exch || sea ee New Zealand, Australia, Tasmania. Asplenium obtusatum, Forst. Ke (vex | exch esx Throughout the South Temperate Zone. 33 lucidum, Forst., var. obliquum, Moore .. iK x Ditto. ES om » var. scleropriwm, Moore .. x ah Stewart Island. i bulbiferum, Forst. x New Zealand, Australia, Tasmania, Polynesia, Malaya, India. Aspidium vestitum, Swartz xX x | x | x | x | Throughout the South Temperate Zone. 9 mohrioides, Bory. x (?) Chili, Fuegia, Falkland Islands, South Georgia, Marion Island, Amsterdam Island. 5 cystostegia, Hook. Sx) ge New Zealand. Polypodium punctatwum, Thunb. x || x Throughout the tropics and the South Temperate Zone. 53 australe, Mett. x i x x | Throughout the South Temperate Zone. sp grammitidis, R. Br. x x New Zealand, Australia, Tasmania. a Billardiert, BR. Br. x | x New Zealand, Norfolk Island and Lord Howe Island, Australia, Tas- mania. Schizaea fistulosa, Labill., var. australis x | & New Zealand, Fuegia, Falkland Island. Todea superba, Col. x New Zealand. LycoropracHAL. Lycopodium varium, R. Br. x x x x | New Zealand, Victoria, Tasmania. ay fastiguatum, R. Br. x x x New Zealand, Australia, Tasmania. as scartosum, Forst. x New Zealand, Australia, Polynesia. Tmesipteris tannensis, Bernh. i =n 3 50 Totals 23 | 158) 118) 57 | 28 | Total number of species, 194, of which 53 are endemic: percentage, 27-3. Systematic Botany.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 453 SUMMARY. F . | | 3 Zs | 32 ga g | | 5 5m 38 BA a 3 | | & = aS OE ie a n ae n a R ° 6 i) oa oe o 8 ors 2 Paes en merce Secu ee ek oe |g 5 S Sue ee dl coealiegs § SAS 2S ae cies nae A q x oo = oS = yas = aes) € es 5 3 2A Se | 7 él S 255 g | 25a 8 S| ko | Se ERE Bo Bee e | Baie) og Beer aS a isa] Ay A ay A AY A AY A aw Snares .. af Na ee 23 2 8-7 17'| 73:9 7 | 30-4 10 | 43-4 5 | 21-7 Auckland Islands .. se it 158 6 3°7 | 117 | 74:0 47 | 29:7 35 | 22-1 16 | 10-1 Campbell Island .. Ea ax 118 + 3:3 79 | 66-9 33 | 27-9 30 | 25-4 13 | 11-0 Antipodes Island .. tha oe 57 4 7-0 40 | 70-1 19 | 33-3 18 | 31-5 13 | 22:8 Macquarie Island .. a om 28 3 | 10-7 15 | 53-6 9 | 32-1 11 | 39-0 10 | 35-7 Southern islands as a whole ae 194 53 | 27-3 133 | 68-5 53 | 27-3 41 | 21-1 20 | 10-3 New Zealand as a whole .. .. | 1,571 | 1,143 | 72:1 ath oe 366 | 23-3 108 6:8 23 1-4 VY. AFFINITIES OF THE FLORA, WITH SUGGESTIONS AS TO ITS PREVIOUS HISTORY. In the enumeration of the species and in the foregoing “tabular view ” I have endeavoured to briefly state the geographical distribution, so far as it is known at present, of the species constituting the florula of the islands to the south of New Zealand. Before proceeding further, it will be well to say a few words respecting the position, physical features, and vegetation of the various land-masses situated within similar parallels of latitude, or placed contiguous to those parallels, either to the north or the south. And it will be convenient to commence with the ice-clad lands surrounding the South Pole, to which the name Antarctic Continent is now com- monly and appropriately applied. The great activity with which antarctic exploration has been pursued during the last ten or fifteen years has resulted in furnishing us with more precise ideas regarding the extent of the Antarctic Continent, and in disclosing many of its more remarkable physical features. It is now tolerably certain that the South Pole stands almost in the centre of a vast land area stretching uninterruptedly from Sabrina Land, far to the south of Australia, to Graham Land, immediately to the south of Fuegia, a distance of over 3,000 miles. The smallest diameter of this land-mass cannot be much less than the distance (about 1,800 miles) separating the head of Ross Sea from the furthest point reached by Weddell in 1823. These boundaries include an area considerably larger than that of the great Australian Continent. Although the major part is yet unexplored, we know that it contains mountain-ranges with peaks rising from 12,000 ft. to 16,000 ft. in height; active volcanoes; and elevated plateaux of immense extent, in one case reaching a height of at least 10,000 ft. We also know that it contains the largest snowfields and glaciers in the world—that, in fact, the whole continent, except a few outlying islands, the faces of cliffs, and the steep escarpments of the mountain ranges, is covered with an everlasting mantle of snow and ice. Under such physical conditions a highly organized flora cannot exist, and it is not at all surprising to learn that a single grass (Deschanvpsia antarctica), col- lected on Graham Land and the adjacent islands, is the sole phanerogam yet observed, the remainder of the land vegetation consisting of mosses, a small number of Hepaticae, some lichens, and other lowly organized cryptogams. 454 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Systematic Botany. But although a phanerogamic flora does not exist at the present time, geologists have long considered it probable that in the south polar regions, as has long been known to be the case in the Arctic Zone, a much milder climate once existed, with its natural accompaniment of a luxuriant vegetation. Proof of this has now been obtained on opposite sides of the Antarctic Continent—near the head of Ross Sea, in the veins of carbonaceous matter discovered by Ferrar during the “ Discovery ” Ex- pedition, and in the beds of coal observed by Shackleton in his recent adventurous journey towards the South Pole ; and in the more important, because more complete, discovery of two fossil floras made by Dr. Anderson on the islands flanking Graham Land, during the Swedish Antarctic Expedition led by Nordenskiold in the years 1902-3. According to the well-known palaeo-botanist Nathorst, in whose hands the collections made by Anderson have been placed, and who has contributed a short preliminary account to the Paris Academy of Sciences, they prove the existence, in the first place, of a Jurassic flora remarkably rich in species, including Hquisetuneae, ferns, and conifers, its general character resembling that of the Jurassic flora of Europe, and particularly that of the Upper Gondwana in India; secondly, a Tertiary flora, including specimens of a Sequoza or of some allied genus, an Araucarza allied to A. brazliensis, leaves of Fagus, and various small-leaved Dicotyledons. The im- portance of this discovery can hardly be overestimated, and a full exploration of the localities will probably throw a flood of light on the composition of the former vege- tation of the Antarctic Content, and give great aid in working out the origin of southern floras generally. A vast sea, formed by the confluence of the southern portions of the Pacific, Atlantic, and Indian Oceans, everywhere surrounds the Antarctic Continent, com- pletely cutting it off from the land-masses to the north. The average depth of the Southern Ocean at a distance of about 500 miles from the continent bas been esti- mated by Sir John Murray at not less than 1,700 fathoms, or nearly two miles. So far as 1s known, this great depth is wonderfully regular and constant; and, as Sir John Murray remarks, “ There is no trustworthy evidence of ridges, barriers, or banks extending far northwards from Antarctica.” The distances which separate the Antarctic Continent from the land-areas to the north are also very great, as may be gatnered from the following table :— Miles. Cape Horn to Graham Land . os = 2 550 Macquarie Island to Adelie Land s9 O70 New Zealand (Stewart Island) to Victoria Land so (LLC2S Tasmania to Adelie Land .. a”) i .. 1,600 Kerguelen Island to Enderby Land .. ne Pel 225 Cape of Good Hope to Tihudlevee Land KS > ALSO At the present time, and with our present knowledge, the on and width of this ocean are the chief difficulties in the way of accepting any theory of previous land connection with the Antarctic Continent. Proceeding northwards from Antarctica, we next arrive at a ring of scattered islands, all of small size, and placed at vast distances apart, mere dots on an immense expanse of ocean, which encircle the globe between lat. 45° and 60° S. The islands to the south of New Zealand, whose vegetation it is the business of this memoir to Systematic Botany] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 455 elucidate, may be considered as forming a portion of this ring or zone; and we shall find that the whole of the islands composing the ring possess many curious and remarkable points of agreement, so far as their flora is concerned. For information respecting the relative position of the islands reference should be made to the map of the antarctic and subantarctic regions accompanying this volume; but it is advisable to give a few details respecting each group. Considerably more than 4,000 miles to the westward of Macquarie Island and about 2,200 miles to the south-east of the Cape of Good Hope is Kerguelen Island, situated between 48° 39’ and 49° 44’ S. lat., and in long. 70° E., and which thus lies about 1,250 miles from the nearest part of the Antarctic Continent. Its extreme length is about eighty-five miles and greatest breadth about seventy-nine, but, as the coast-line on all sides is much cut up by deep sounds or fiords, no part of the interior being more than ten or twelve miles from the sea, the total area does not much exceed 2,000 square miles.. Its surface is everywhere rough and moun- tainous, the highest peak (Mount Ross) attaining a height of 6,120 ft. In the centre of the island a district of considerable extent 1s covered by snowfields, from which small glaciers descend on all sides; but evidences of more complete recent glaci- ation are everywhere present. The geological structure is purely volcanic, with the exception of certain thin beds of coal, interstratified with the volcanic rocks. The occurrence of these is of considerable interest, from the proof which they give of the former existence of a much milder climate and more copious vegetation. Kerguelen lies in the zone of westerly winds, and its climate is tempestuous, chilly, and wet. At the same time, the range of temperature is not excessive, the thermometer seldom falling much below 32° F. in winter, or rising much above 60° in summer, About 950 miles to the north-west of Kerguelen is a group of several contiguous islands of small size, known collectively as the Crozets, of which the principal are Possession, Hog, and Hast Islands, to which may be added Inaccessible or Penguin Island. Possession Island, which is situated in 47° S. lat. and 50° 30’ EH. long., is the largest of the group, being fifteen or twenty miles in length by about half the breadth. It is high, bare, and rocky, and attains a height of about 5,000 it. Nearly 500 miles still further to the west is the Prince Edward Group, consisting of Marion Island and Prince Edward Island. Marion Island, which is much the larger of the two, is in S. lat. 46° 52’ and H. long. 37° 45’. It is about eleven miles in length by eight in extreme breadth, its area being about eighty square miles. Like Kerguelen and the Crozets, it is purely volcanic, and attains a height of 4,250 ft. Rather more than 300 miles to the south of Kerguelen lies the isolated Macdonald Group, of which Heard or Young Island is the only one which has been examined. It has an elongated shape, being about twenty-five miles in extreme length by six in breadth. It is exceedingly rough and rugged, Big Ben, the highest mountain, attaining a height of 7,000 it. A considerable part of the island is snow-clad, and numerous glaciers come down to the shore. The next island to be considered is South Georgia, which is situated about 3,500 miles south-west of Marion Island and about 1,000 miles to the eastward of Cape Horn. From the nearest poimt of the Antarctic Continent it is barely more than 750 miles distant. New Zealand lies fully 5,300 miles to the eastward. Its average latitude may be given as 54° 8’ S., and the longitude as 37° W. It is much 456 SUBANTARCTIC ISLANDS OF NEW ZEALAND. __ [Systematic Botany. larger than any of the above, being about ninety-five miles long Ly twenty miles in greatest breadth. It forms a high mountain-chain, rising very steeply out of the sea, and many of the peaks attain a height of over 6,000 it. A large portion of the in- terior is occupied by extensive snowfields, and glaciers discharge into all the fiords. As there is little low-lying land or level soil, there are few suitable stations for a phanerogamic flora, even if the cold, wet, and exceptionally boisterous climate did not necessarily limit plant-growth. The island has been twice carefully explored— first by the German Expedition of 1882-83, and then again by Nordenskiold’s Swedish Antarctic Expedition of 1902-3—but only fourteen phanerogams and four pteridophytes have been observed, not one of them being endemic. The Sandwich Group, situated to the south-west of South Georgia, and consist- ing of about sixteen small islands, has never been botanically explored. Cook, who discovered the group, mentions that he saw two hills free of snow, apparently covered with green turf, so that probably there is a scanty phanerogamic vegetation, doubt- less composed of some of the grasses found in South Georgia. The small group of the Bouvet Islands, situated in the same latitude as South Georgia, and about midway between it and Marion Island, is also very imperfectly known. For many years its existence was looked upon as doubtful, but in 1898 the German expeditionary ship “ Valdivia” rediscovered it, fixing its position as 54° 26° S. lat. and 3° 24’ W. long. Unfortunately, it was impossible to effect a landing, but no sign of vegetation could be discerned from the ship. Far to the westward of Cape Horn—ain fact, almost a third of the distance between it and the meridian of New Zealand—is Dougherty or Keates Island, one of the most isolated spots in the Southern Ocean. Captain Keates gives its position as 59° 20’ S. lat. and 120° 20° W. long., which would make it nearly 1,700 miles from Cape Horn and about 2,500 from Campbell Island. It has never been landed upon, and nothing whatever is known of its physical features and natural productions. Its position with regard to New Zealand and South America is exceptionally imteresting, and it is much to be wished that some of the expeditions now regularly visiting the Ant- arctic Continent would give it a careful exploration. A group called the Nimrod Islands is sometimes marked on our charts about 1,000 miles south-east from Antipodes Island, but it 1s very doubtful if it has any real existence.* The general character of the vegetation of the islands described above has been vividly portrayed by Sir J. D. Hooker in his memoir on the vegetation of Kerguelen Island (Phil. Trans. Roy. Soc., vol. 168, p. 10), where he says, “ The three small archipelagoes of Kerguelen Island (including Heard Islands), Marion and Prince Kdward Islands, and the Crozets are individually and collectively the most barren tracts on the globe, whether in their own latitude or in any higher one, except such as lie within the Antarctic Circle itself; for no land, even within the north polar area, presents so impoverished a vegetation.” All students of botanical geography know that this statement is by no means overdrawn. To mention a single proof out of many that could be quoted, the Island of Spitzbergen, which is situated 30° nearer the Pole (in N. lat. 77° 80’) possesses 120 vascular plants, whereas Kerguelen has * Since this was written, both Dougherty Island and the Nimrod Group have been unsuccessfully searched for by Captain Davies, of the Antarctic exploring vessel ‘ Nimrod.” Systematic Botany.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 457 only twenty-eight. But, although the vegetation is poor and scanty, the few species which do exist have a distribution so remarkable, and are so closely connected with those found in the islands to the south of New Zealand, that it is advisable to give full details respecting them. ‘This is best presented in the following shape :— CATALOGUE OF THE PHANEROGAMIC PLANTS AND FERNS FOUND IN THE KERGUELEN AND Sout GroRGIA GRovuers OF ISLANDS, WITH THEIR DISTRIBUTION IN DETAIL. ; ; = SI & cs 8 a as) : —— rs a é = a ad il Se = E os e 8 a, | Be bee 5 ee 2 S as = Ss | 8S | so 2 o = co 2 =o SN S ee a} ‘S) a wa & 4 E: Fam. ULVACHAE. Utva (L.), Wittrock, 1866. Distribution.—All seas. Ulva lactuca, var. latissima (L.), Le Jolis. Ulva latissoma, Hook. f. & Harv., Fl. Nov. Zel., u, p. 265, 1855. U. lactuca, B latissima, Le Jolis, Alg. Mar. de Cherb., p. 39, 1880. A plant which has many forms and states; but all these are probably reducible to one species (which is often, indeed, considered identical with the following) and a few varieties. (Identified by A. Gepp.) 494 SUBANTARCTIO ISLANDS OF NEW ZEALAND. [Marine Algae. Auckland Islands; Hooker, R.M.L. Campbell Island; Hooker, R. M. L. (All seas.) Ulva rigida, C. Agardh. Ulva rigida, C. Agardh, Spec. Alg., p. 410, 1823-28; Hook. f. & Harv., Fl. Nov. Zel., 1, p. 265. 1855. U. lactuca, var. rigida, Le Jolis, Alg. Mar. de Cherb., p. 38, 1880. * Auckland Islands; J.C.S.! (All seas.) The form U. lobata also occurs in the specimens from Enderby Island collected by Mr. Crosby Smith. ENTEROMORPHA (Link), J. Ag., Till. Alg. Syst., pt. vi, 1890. Distribution.—All seas. Enteromorpha compressa (L.), Greville. Enteromorpha compressa, Greville, Alg. Brit., p. 180, 1830; Hook. f. & Harv., Fl. Nov. Zel., ii, p. 264, 1855. Auckland Islands; R.M.L. (All seas.) (Identified by A. Gepp.) Amongst Mr. Smith’s Campbell Island specimens there are fragments of an Enteromorpha, possibly belonging to a different species. Enteromorpha clathrata, Greville. Enteromorpha clathrata (Grev.), var. erecta, Mont., Voy. au Pole sud., Bot., 1, p. 31, 1845. Lord Auckland Group; d’Urville. I know nothing of this variety, and it would probably be impossible to re-identify it unless some of the original specimens were obtainable. The species recorded by Hooker from the mainland as EL. clathrata, var. ramulosa, Harv., has sometimes been re-identified as H. acanthophora, Kiitz.; but there is apparently no doubt that H. clathrata, var. ramulosa, does occur among the New Zealand specimens col- lected by Hooker(1). This plant is now known as E. ramulosa (Engl.), J. Ag.(?). Fam. CLADOPHORACEAE. CLADOPHORA, 1843. Distribution.—All seas. Cladophora verticillata, Harvey. Cladophora verticillata, Hook. f. & Harv., Fl. Antarct., vol. i, p. 193, 1847. * Campbell Island, * Snares; R.M.L. Auckland Islands; Hooker. (Stewart Island.) (*) A. D. Cotton, “ Marine Algae from the Chatham Islands,” Kew Bulletin No. 2, 1907. (7) J. Ag., Till. Alg. Syst., mu, p. 154. Marine Algae.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 495 Cladophora pacifica, Kiitzing. Conjerva pacifica, Mont., Voy. au Pole sud., Bot., i, p. 7, 1845. Cladophora pacifica, Kiitz., Sp. Alg., p. 419, 1849. Auckland Islands ; d’Urville, R. M. L. (Kerguelen, Falklands.) (Identified by A. Gepp.) The South African C. virgata, Kiitz., Sp. Alg., p. 388, is also recorded by d’Urville from Auckland Islands, but has not since been collected. RuIzocLonium, Kiitzing, 1843. Distribution.—All seas. Rhizoclonium Hookeri, Kiitzing. Rhizoclonium Hookeri, Kiitz., Sp. Alg., p. 383, 1849. R. ambiquum (?), Kiitz., Sp. Alg., p. 387, 1849. Conferva ambigua (?), Harv., Lond. Journ., p. 295, 1845; Hook. f. & Harv., Fl. Antarct., vol. u, p. 494, 1847. *The Snares; J. C. 8. *Hnderby Island; J. ©. S. Monument Harbour, * Campbell Island; J.C.S. (New Zealand, Kerguelen Land, Tierra del Fuego.) This is the same plant as one that I collected at the mouth of the Kaikorai, near Dunedin, some years ago, and forwarded to Major Reinbold for identification. He returned it as Rhizoclonum Hooker, Kg. (= R. africanum, Kg., = (?) R. am- biguum, Hooker). Rhizoclonium ambiquum was collected by Sir James Ross’s expedition at Kerguelen Land, and is figured in the “ Flora Antarctica,” tab. exci, and scarcely seems to be different from the plant under discussion. A Rhizoclonium from Tierra del Fuego has also been identified as R. ambiguum('). Recently the numerous and badly discriminated species of Rhizoclonium have been subjected to a careful revision by 8. Stockmayer(?). Mr. Stockmayer reduces the large number of species described to five. &. Hookeri is retained, but not R. ambiquum. Un- fortunately, I have not yet a copy of Stockmayer’s paper, but presume that R. ambiguum has become a synonym for R. Hooker. The only other species yet recorded from New Zealand is Rhizoclonium hierogly- phicum, Kitz. f. warkitensis, F. Hauck., from the Waikato River. A few notes on the form obtained during the “ Hinemoa ” trip may be useful. Rhizocloniwm Hookert, Kiitz.: A small tufted filamentous alga, sending up long unbranched filaments from a creeping rhizome, which gives off a few usually colourless unicellular rhizoids, narrower than the filaments. Occasionally, however, a slight stream of endochrome is seen passing into the rhizoids.. The chromatophores in each cell become waisted like an hour-glass before division. They are of a deep olive-green. Dissepiments between the cells are only faintly visible in the glycerime-prepared specimens, and not visible at all where the cells are young, or have only recently divided. In the centre of the branches the cells are about one and a halt to twice as long as broad. (Plate XXIV, figs. 5, 6.) (*) “ Mission scientifique du Cap Horn, Alques,” Hariot, p. 18, 1886. (*) “ Ueber die Algengattung Rhizoclonium,’ S. Stockmayer. K. K. Zoologisch-botanischen Gesellschaft, Wien, 1890. 496 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marie Algae. Fam. CODIACEAE. Copium (Stackhouse), C. Ag., 1823. Mstribution.—All seas. Codium tomentosum, Stackhouse. Codvum tomentosum, Stackhouse, Nereid. Brit. (1795-1801); C. Ag., Sp. Alg., p. 452, 118235 look. fs Glan. HI Nove Zelean) p..261e shor ee Till. Alg. Syst., vi, p. 41, 1886. Auckland Islands ; Hooker, R. M. L. (All seas.) The cosmopolitan species C. tomentosum has been divided into several sub- species by various writers, especially Agardh (loc. cit.). It 1s impossible to say to which of these the Auckland Island form should be assigned, as no specimens of it were brought home by us. It doubtless occurs at Campbell Island and the Snares, though not observed by us at these places. The New Zealand species is termed by Agardh (loc. cit., p. 43) Codiwm mucromatum, var. y novae-zelandiae. Fam. CAULERPACEAKE. CAULERPA, Lamouroux, 1809. Caulerpa Brownii, Endlicher. Caulerpa Brownu, Endlicher, Gen. Plant. Suppl., 1837-45 ; Hook. f. & Harv., Fl. Nov. Zel., 11, p. 260, tab. 1214, 1855. * The Snares; R.M. L. (South-western Australia, New Zealand, Tasmania.) Clags IP IEE 18) @) 1 Tel WC) Ja) Ay Jae Fam. SPHACELARIACEAE. STYPOCAULON, Kiitzing, 1843. Distribution.—North America, Mediterranean, Southern Ocean. Stypocaulon funiculare, Kiitzing. Sphacelaria jumculare, Mont., Voy. au Pole sud., Bot., i, p. 38, tab. 14, 1845. Stypocaulon funiculare, Kiitz., Sp. Alg., p. 467, 1849. Sphacelaria jumculare, Hook. f. & Harv., Handb. N.Z. Flora, p. 661, 1864. Auckland Islands; Hooker, R. M. L. (New Zealand.) (Identified by A. Gepp.) Fam. DESMARESTIACEAE. DESMARESTIA, Lamouroux, 1813. Distribution.—Most temperate seas. Desmarestia ligulata, Lamouroux. Fucus ligulatus, Turn., Hist. Fuc., u, p. 74, 1809. Desmarestia liqulata, Lamx., Hss. s. Thallassophytes, p. 25, 1813; Hook. f. & Harv., Fl. Nov. Zel., ui, p. 217, 1855. Lord Auckland Group; Hooker. Enderby Island; R.M.L. (Australasia, Chili, Fuegia, South Africa, North Atlantic, North Pacific). Marine Algae.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 497 Desmarestia Willii, Reinsch. Desmarestia viridis (Lamx.), Hook. f. & Harv., Fl. Antarct., vol. 1, p. 178, 1847. D. Will, Reinsch, Meeresalgen Flora von Sud-Georgien, p. 401, 1890; Skottsberg, Subant. und Antark. Meeresalgen, p. 16, 1907. Auckland Islands. (Tierra del Fuego, Falkland Islands, South Georgia, Ker- guelen Land, Victoria Land, Franklin Island.) This species, obtained by Hooker at the Auckland Islands, was not collected by us. By the earlier investigators it was considered to be identical with D. viridis of the Northern Hemisphere; but Skottsberg, following Reimsch, considers that, though the two plants closely resemble each other, the southern form is worthy of specific rank. He has examined some of the southern forms collected by Hooker, and considers they must be assigned to this species. Fam. DICTYOSIPHONACEAE. DictyosiPHon, Greville, 1830. Distribution.—Widely distributed in northern and southern seas. Dictyosiphon fasciculatus, Hook. f. & Harv. Dictyosiphon fasciculatus, Hook. f. & Harv., Fl. Antarct., vol. i, p. 178, tab. 69, 1847. Auckland Islands; Hooker, R.M.L. (Kerguelen Land, Falkland Islands.) (Identified by A. Gepp.) This species appears in Hariot’s list of the seaweeds of Cape Horn, but does not seem to have been recently collected in Kerguelen Land, as it is included neither in Dickie’s nor Reinbold’s(*) list of the algae of that region. Moreover, Skotts- berg, who has examined some of Hooker’s type specimens, considers that this species is either Scytothamnus australis or must at least be brought within the limits of the genus Scytothamnus. Untortunately, the specimens of D. jasciculatus named by Mr. Gepp, and collected by myself, have not yet been returned to me, when the point might be definitely decided. Scytothamnus australis is found in Tierra del Fuego, Falkland Islands, South Africa, New Zealand, Chatham Islands. Fam. CHORDARIACEAE. Cuorparia, C. Agardh, 1823. Distribution.—All extra-tropical seas. Chordaria flagelliformis, C. Agardh. Chordaria flagellijormis, C. Ag., Syst. Alg., 1, p. 166, 1823; Hook. f. & Harv., Fl. Antarct., vol. 1, p. 180, 1847. Campbell Island ; Hooker (not recorded from New Zealand or Auckland Islands). (North Atlantic, Kamschatka, Japan, Cape of Good Hope, Chili.) The above distribution shows that the Campbell Island habitat requires con- firmation. (*) “ Die Meeresalgen der Deutschen Tiefsee Expedition, 1907.” 33—S. 498 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. Fam. ENCOELIACEAE. ASspERococcus, Lamouroux. Distribution.—Probably all temperate seas. Asperococcus echinatus, Greville. Asperococcus echinatus, Grev., Alg. Brit., p. 50, 1830; Hook. i. & Harv., Fl. Antarct., vol. i, p. 180, 1847. Lord Auckland Group; Hooker. (Kurope, King George’s Sound, Western Australia.) The specimens from habitats in the Southern Hemisphere are much in need of re-identification. It is quite possible that the plant obtained is distinct from the northern form. Such a species as Corycus prolifer, Kjellman, might even pos- sibly have been confused with the northern Asperococcus echinatus. Fam. LAMINARIACEHAE. CHorDA, Stackhouse, 1795-1801. Distributvon.—All temperate seas. Chorda lomentaria, Lyngbye. Chorda lomentaria, Lyngb., Tent. Hydroph. Danic., p. 74, tab. 18, 1819; Hook. £. & Harv., Fl. Nov. Zel., 1, 218, 1855. Auckland Islands; Hooker, J. C. S. (Atlantic, Mediterranean, Southern and Subantarctic Oceans.) ApEnocystis, Hook. f. & Harv., 1845. Distribution.—Subantarctic South America, South Georgia, Kerguelen Land, Tasmania, New Zealand, Auckland and Campbell Islands, South Orkney Islands, Graham Land. Adenocystis utricularis (Bory), Hook. f. & Harv. Asperococcus utricularis, Bory in d’Urville, “ Flore des Malouines,” p. 594, 1826. A. Lessonu, Bory, “ Voyage de la Coquille,” Botanique, p. 199, 1828. Adenocystis Lessonw, Hook. f. & Harv., Fl. Antarct., vol. i, p. 179, 1847. Auckland Islands; Hooker, R. M. L. Campbell Island; Hooker, R. M. L. (Distribution as for the genus.) Specimens were not brought back, but the species was observed both at Camp- bell Island and at the Aucklands. Lessonta, Bory, 1828. Distribution.Sea of Ochotsk, Tahiti (? growing), Patagonia, Falkland Islands, South Georgia, Kerguelen Land, Heard Island, Graham Land, Marine Algae.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 499 Lessonia variegata (?), J. Ag. Lessonia fuscescens (partim), Bory, ““ Voyage de la Coquille,” tab. 2, fig. 2, 1828 ; Hook. f. & Harv., Fl. Nov. Zel., 1, p. 217, 1855. L. variegata, J. Ag., Alg. Mar. Nov. Zel., p. 6, 1877; Laing, Trans. N.Z. Inst., vol. xxvi, p. 304, 1894. * Snares; R. M. L. Auckland Islands; R. M. L. Campbell Islands; J. C. 8. (New Zealand, Chil.) A species of Lessonia was observed at Enderby Islands. Fragments of a Lessonia were also collected at Monument Harbour, Campbell Island, by Mr. J. Crosby Smith. They are, unfortunately, in a poor state of preservation, but seem to belong to the species L. variegata. I also noted the presence of L. variegata at the Snares, but brought back no specimens. Macrocystis, C. Agardh, 1823. Distribution.—West coast of South America to the Galapagos, South California to Sitka, Alaska, Ochotsk Sea, east coast of South America to 45° S. lat., Magellan Straits, Falkland Islands, South Georgia, Cape of Good Hope (? growing), Prince Edward Islands, Crozets, St. Paul, Kerguelen Land, south coast of Australia and Tasmania, New Zealand, Tahiti (? growing). Macrocystis pyrifera (L.), C. Ag. Fucus pyriferus, L., Mant. Plant., p. 311, 1771; Turn., Hist. Fuc., u, p. 104, 1809. Macrocystis pyrifera, C. Ag., Rev. Algengattung Macrocystis, 1839 ; Hook. f. & Harv.; Fl. Nov. Zel., u, p. 217, 1855. Auckland Islands; Hooker, R. M. L. Campbell Islands ; IDOLE and Harvey, vee: (Distribution as for the genus.) The immense length formerly assigned to this well-known species was un- doubtedly a mistake. Specimens of more than 100 ft. in length are very uncommon. A variety with almost spherical bladders was observed at Monument Harbour, Campbell Island, but unfortunately no specimens of it were brought to New Zealand. Fam. FUCACEAE. DurviILLaAEA, Bory, 1826. Distribution.—South America—Cape Horn to Valparaiso and the east coast as far north as 50° S. lat.; Falkland Islands, South Georgia, Kerguelen Land, Tahiti (? growing), New Zealand, Chatham Islands, Auckland and Campbell Islands. Durvillaea antarctica (Cham.), Skottsberg. Fucus antarcticus, Chamisso. (Choris, Voy. Pitt., pp. 7-8, 1822). Durvillaea utilis, Bory, “ Flore des Malouimes,” p. 588, 1826; Hook. f. & Harv., FI. Nov. Zel., u, p. 216, 1855. D. antarctica, Skottsberg, Subant. und Antark. Meeresalgen, p. 140, 1907. 500 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. * The Snares; R. M. L. Auckland and Campbell Islands ; Hooker and Harvey, R. M. L. (External distribution as for genus.) As Skottsberg points. out (Joc. cit.), Bory incorrectly replaced Chamisso’s specific name by one of his own. XIPHOPHORA, Montagne, 1845. Distribution.—Australia, Tasmania, New Zealand, Chatham, and subantarctic islands of New Zealand. Xiphophora gladiata (Labill.), Mont. Fucus gladiatus, Labill., Fl. Nov. Holl., u, p. 111; Turn., Hist. Fuc., iv, p. 102, 1819. Xzphophora Billardierr, Mont., Voy. au Pole sud, tab. 1, 1845. Fucodvum gladiatum, J. Ag., Spec. Alg., i, p. 202, 1844. F. gladiatus, Hook. =. & Harv., Handb. N.Z. Flora, p. 652, 1864. X«phophora gladiata, Kjellman, Engler und Prantl’s “ Pflanzenfamilien,” Algae, p. 281, 1897. *The Snares; R. M. L, J. C. S. Auckland Islands; Montagne, Hooker, R. M. L. * Campbell Island; R. M. L., J. C. 8. * Antipodes; Dr. Cockayne ! (External distribution as for the genus.) Two species have been described, X. gladiata and X. chondrophylla ; but they seem to have been so completely confused as to defy separation (vide J. Ag., Ale. Noy. Zel. mar, p. 7). The first-recorded species, X. gladvata, was originally described in a publication (Labill., Joc. cit.) not available in New Zealand. X. chondrophylla is described by R. Brown in Turner’s Hist. Fuc., vol. iv, p. 64, where X. gladiata is also described (p. 102); but the descriptions given there are naturally very crude, and the plates obviously from very imperfect specimens. Agardh (loc. cit.) expresses doubt as to the authenticity of the specimen of X. gladiata depicted by Turner, and also of the figure given by Montagne (Voy. Pole sud, tab. vu) ; and further states that a specimen from the Auckland Islands sent to him by Harvey as Fucodium gladiatum should, in his opinion, be referred to F. chondro- phyllum. Under these circumstances, it is clearly impossible for a worker in New Zealand, without type specimens, and without access to the earliest description, to state definitely which species we have here; but I feel assured, after the examination of hundreds of specimens from all parts of New Zealand, that, though there may be a good many states of the plant dependent upon the season and environment, there is only one species in New Zealand, and this appears to approach more closely to Turner’s figure of F’. gladiatus than to that of F. chondrophyllus. 1 therefore feel inclined, with Harvey and Montagne, to call the plant of our southern islands X. gladiata rather than X. chondrophylla. It is, indeed, possible that both these species are the same. Mrs. A. Gepp (Miss E. S. Barton) (1) states, “ There are, however, many intermediate forms, showing Fucus gladiatus with comparatively short receptacles, approaching more nearly (+) “ Phycological Memoirs of the British Museum,” pt. 1, p. 35, Marine Algae. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 501 we vs pA - B. Broadhead del. PLATE XXII. 502 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. to the short dichotomous branches of F. chondrophyllus.” The Antipodes and Snares specimens that I have possess the long narrow sword-like receptacles that are supposed to be characteristic of F. gladvatus, but which I think are merely typical of a state of the species. All things considered, it seems better to use the first specific name given to the original plant of the genus. Mareinaria, A. Richard, 1832. Distribution.—New Zealand and its outlying islands. Marginaria Urvilliana, A. Richard. Margimaria Urvilliana, A. Rich., “ Voyage de l Astrolabe,” Botanique, p. 9, 1832. Sargassum Urvillianum, A. Rich., “ Voyage de |’ Astrolabe,” Botanique, Sert. Astrolab., p. 138, 1834. Margunaria Urvilleana, Hook. f. & Harv., Fl. Nowe Zel is) ps 2s saa: Auckland Islands; d’Urville, R. M. L. (New Zealand.) It will be noticed that Hooker and Harvey misspell the specific name, and insert ” in the original description. an ‘‘e’’ where there is an “1 Marginaria Boryana (A. Rich.), Mont. Sargassum Boryanum, A. Rich., “ Voyage de | Astrolabe,” Botanique, Sert. Astrolab., p. 138, 1834. Marginaria Boryana, Mont., Voy. au Pole sud, tab. 1, 1845; Hook. & & Harv.,-Kl. Nov. Zel., 1, p. 213, 1855. * The Snares; R. M. L. (New Zealand.) CARPOPHYLLUM, Greville, 1830. Distribution.—South Atlantic, Indian Ocean, and South Pacific. Carpophyllum macrophyllum, Montagne. Carpophyllum macrophyllum, Mont., Voy. au Pole sud, Bot., 1, p. 76, 1845. Auckland Islands ; d’Urville. A doubtful species, not collected since d’Urville’s time. CystopHora, J. Agardh, 1841. Distribution.—Australasia. Cystophora retroflexa (Labill.), J. Agardh. Fucus retroflecus, Labill., Fl. Nov. Holl., p. 113. Cystophora retroflexa, J. Ag., Spec. Alg., 1, p. 242, 1848; Hook. f. & Harv., Fl. Nov. Zel., u, p. 214, 1855. Auckland Islands; Hooker. (Australasia.) In the « Pflanzenfamilien” of Engler and Prantl (Algae, p. 286) the genus Lands- burgia is recorded from the Auckland Islands, on what authority I do not know. As I have not found L. quercifolia (Harvey) south of Banks Peninsula, it seems scarcely probable that it should be found so far south as the Auckland Islands. Marine Algae.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 503 Clas RHODOPHYCEAE. Fam. BANGIACHAE. PorpHyra, C. Agardh, 1824. Distribution.—All seas. Porphyra perforata, J. Agardh. Porphyra perforata, J. Ag., Till. Alg. Syst., i, p. 69, 1882. * The Snares; R.M.L. Campbell Island; R.M.L. (California, Valparaiso 7) The New Zealand species of Porphyra are inadequately defined, and until the species and subspecies of the mainland are properly delimited in the light of modern research it will be impossible to do more than state more or less crudely the species of the outlying islands. As is well known, the older investigators were content to define by means of their external shape and other superficial characteristics the species of Porphyra obtained by them; but the form, and sometimes even the con- sistency, of the plant frequently depend upon the environment and upon the accidents of its lite, so that in many cases they are quite unreliable for diagnostic purposes ; and until the type specimens of the older writers have been examined with modern thoroughness it is impossible to give anything like an accurate synonymy of the species. What follows is therefore only to be regarded as an endeavour to get a little closer to a correct appreciation of the New Zealand forms than has hitherto been attempted. Harvey attributed three species of Porphyra to New Zealand waters—P. lacmata, C. Ag.; P. vulgaris, C. Ag.; and P. capensis, Kiitzing. Of these, P. lacomata and P. vulgaris are probably the New Zealand forms or subspecies of the common cosmopolitan species, called by many names. P. capensis does not occur in New Zealand, but was reported because Harvey wrongly identified P. colwmbina, Mont., with the South African P. capensis, Kiitz. These two species are undoubtedly quite distinct. An authentic specimen of P. capensis which I have, collected by Becker, differs widely from any local form, and therefore P. columbina, Mont., must be restored to specific rank. In J. G. Agardh’s(') list of New Zealand seaweeds, P. Jaciniata and P. vulgaris are again recognised as local species; but subsequently(2) P. Jaciniata is defined as a variety of P. umbilicalis, L., and its habitat is restricted to the North Atlantic waters ; whilst P. vulgaris appears as a synonym for the variety a purpurea, Roth., of P. umbilicalis. In his classification of the genus, Agardh divides the species into two divisions—those that are monostromatic and those that are distromatic. This distinction, however, is admittedly to be used with circumspection, as a species at one stage monostromatic may at another stage or in another part be distromatic. On page 62, Agardh (loc. cit.) creates for New Zealand a new monostromatic species P. nobilis, and revives P. columbina, Mont., for certain distromatic forms from aie coast. (*) De Algis Nov. Zel. marinis, p. 1, 1877. . (7) Till. Alg. Syst. (vol. iii), vi, p. 63, 1882, 504 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Marine Algae. Now, the most careful and elaborate discussion of any of the species of Porphyra yet put forth is that of H. T. Hus.(1) Under the direction of Professor W. T. Setchell, he has described the forms of Porphyra of the north-west Pacific coast. Taking advantage of the kind offer of Professor Setchell, I forwarded to him specimens of one of the common forms of Porphyra from our coast. He defined it as P. perforata, f. lanceolata, Setchell and Hus, and stated, “ The specimens agree perfectly with a form on our Californian coast which we call P. perforata, f. lanceolata. It 1s probably only a subspecies of P. umbilicalis (P. lacomata, P. vulgaris, &c.), the common, and as I believe thoroughly cosmopolitan, species.” We are thus apparently left with three species of Porphyra in New Zealand waters—P. nobilis, J. Ag.; P. perforata, J. Ag.; and P. columbina, Mont. The first and third of these species will be found marked as “common” in my revised list of New Zealand seaweeds. I now consider that I was at least premature in identifying certain forms of Porphyra on the New Zealand coast as P. nobilis(?). I had examined them, and found them to be mono- stromatic, and otherwise agreeing with Agardh’s description of P. nobilis. A fuller investigation of them has shown me that some at least of those I thought to be mono- stromatic are monostromatic only at the edges of the frond and in the younger por- tions, and in identifying these with Agardh’s P. nobilis I was doubtless in error. (It may perhaps be mentioned here that a very distinct and still undescribed species of Porphyra is to be found in most parts of New Zealand, growing on Durvillaea antarctica.) Specimens of Porphyra were collected both by Mr. Crosby Smith and myself during the ““ Hinemoa” expedition. Those collected by myself at the Snares and at Campbell Island have been identified by Mr. A. Gepp as Porphyra perforata, J. Ag., and he gives as synonyms P. capensis, Hook. f. & Harv., and P. Kunthiana, Rahenh. As I have already stated, I believe P. capensis, Kiitzing, is a distinct species. P. Kunthiana, Kitz., is a form from Valparaiso of which I know nothing, but which Agardh thinks may possibly be identical with P. perforata. Porphyra nobilis, J. Agardh. Porphyra nobilis, J. Ag., Till. Alg. Syst. (vol. ii), vi, p. 62, 1882. Campbell Island; J.C.S. Antipodes; Dr. Cockayne! (New Zealand, St. Paul Island ?) Amongst the specimens collected at Perseverance Harbour, Campbell Island, by Mr. Crosby Smith are some that I think must be referred to this species. The mature plants are 35-40 cm. in length, broadly elliptical in outline, and from dull lilac to purple in colour. They are monostromatic, and the tetrads are separated from each other by a broad hyaline margin, exceeding in breadth the endochrome itself. The conspicuous cell-walls mentioned by Agardh I have not seen; but this is a point on which little stress can be laid, as it depends largely upon the method of preparation of the slide. The microscopic appearance of the cells on the suriace of the frond closely resembles that of the plant shown in tab. u, 49, / (Agardh, Joe. cit.) ; and the plant agrees well in other respects with Agardh’s description. (+) “ An Account of the Species of Porphyra found on the Pacific Coast of North America.” Cali- fornia Academy of Science, Botany, vol. 1, No. 6, 1902. (2) Trans. N.Z. Inst., vol. xxxu, p. 61. Marine Algae.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 505 I have also some specimens from the Antipodes collected by Dr. Cockayne, and identified by Major Reinbold as P. nobilis (?), which may, I think, be well included in the species. Porphyra columbina, Mont. Porphyra columbina, Mont., Voy. au Pole sud, Bot., 1, p. 33, tab. 9, fig. 2, 1845. P. laconata, Hook. f. & Harv., Fl. Nov. Zel., ti, p. 264, 1855. P. capensis, Hook. f. & Harv., Handb. N.Z. Flora, p. 715, 1864. Auckland and Campbell Islands ; d’Urville, Hooker. (New Zealand.) This species, originally described from the Auckland and Campbell Islands, is, I believe, common on the New Zealand coast. It, however, does not appear to have been collected either by Mr. Smith or myself during the “ Hinemoa ” expedition. Fam. CHAETANGIACEAE. CHAETANGIUM, Kiitzing, 1843. Distribution.._Cape of Good Hope, southern America, Kerguelen Land, Falk- land Islands. Chaetangium variolosum (Mont.), J. Ag. Nothogena variolosa, Mont., Voy. au Pole sud, Bot., 1, p. 109, tab. 10, fig. 3, 1845; Hook. f. & Harv., Fl. Antarct., vol. 1, p. 188, 1847. Chaetangium variolosum, J. Ag., Sp. Alg., u, pt. 1, p. 461, 1851. Chondrus variolosus, Hook. f. & Harv., Handb. N.Z. Flora, p. 702, 1864. Auckland Islands; d’Urville, Hooker, R.M.L. “Campbell Island; R. M. L. (Kerguelen, Falkland Islands.) (Identified by A. Gepp.) Fam. GIGARTINACEAE. GIGARTINA, Stackhouse (Mem. Soc. Mosce., u, p. 54-74). Distribution.—All seas. Gigartina tuberculosa, Grunow. Chondrus tuberculosus, Hook. f. & Harv., Fl. Antarct., vol. i, p. 188, 1847. Gigartina tuberculosa, Grunow, “ Reise der Novara,” Bot., teil 1, 1870. Auckland Islands; Hooker, R. M. L. (Identified by A. Gepp.) Gigartina divaricata, Hook. f. & Harv. Gigartina divaricata, Hook. f. & Harv., Fl. Antarct., vol. i, p. 187, 1847. G. pistillata (partim), Hook. f. & Harv., Handb. N.Z. Flora, p. 699, 1864. G. dwvaricata, J. Ag., Epicr. Florid., p. 195, 1876; J. Ag., Anal. Algol., cont. v, p. 10, 1899. Campbell Island; Hooker, R.M.L. Auckland Islands; J.C.8.! (New Zea- land.) 506 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Marine Algae. Gigartina Kroneana, Rabenh. Drift-weed and rocks, Monument Harbour, Campbell Island; R. M. L. Mr. Gepp has thus named specimens Nos. 5 and 7 sent by me, and adds the note, “ Apparently. No authentic specimen seen by me.—A. Gepp.” The following is Rabenhorst’s description, kindly extracted by Mr. Gepp from a paper entitled “ Beitrage zur Meersalgen-Flora der Auckland Inseln,”’ Hedwigia, Xvi (1878), p. 70. This paper I have, unfortunately, not seen :— “« Gagartina Kroneana, Rabenh., nov. sp.—Vage ramosissima, durissima cornea. dilute vel obscure brunnea; stipite e basi dilatata ramis ramulisque 2-4 millim, latis, tereti-compressis, ramulis ultimis (ordinis tertiae vel quartae) subulatis nunc brevibus nunc elongatis (ad 3cm. longis) rectis vel plus minus curvatis, axillis obtuse angularibus; cystocarpis subgloboso-capituli-formibus, singulis in ramellis ultimis. «« Hinige, gute mstructive Exemplare. Ihre nachste Verwandschaft dirite sie wohl in G. flabellata (Ktz., Tabul, phycol., xvi, Tab. 5) und in Sphaerococcus vagus (Ktz., l.c., Tab. 76, fig. a) finden, zumal entspricht sie habituell letzterer (aus der Mittelmeere) unserer Pflanzen ausserordentlich.” Gigartina radula, J. Ag. ! Gigartina radula, J. Ag., Sp. Alg., 1, p. 278, 1851; Hook. f. & Harv., Handb. N.Z. Flora, p. 700, 1864. Auckland Islands; Hooker, R. M. L. Campbell Islands; Hooker, R. M. L. (Antarctic, North and South Pacific Oceans.) Gigartina macrocarpa, J. Ag. Gigartina pistdlata, Hook. f. & Harv., Fl. Nov. Zel., 1, p. 252, 1855. G. macro. carpa, J. Ag., Epicr. Florid., p. 683, 1876. * The Snares; R. M.L. (New Zealand.) (Identified by A. Gepp.) TripakA, Bory. Distribution.—Colder seas of both hemispheres. Iridaea micans, Bory. Iridaea micans, Bory, “ Voyage de la Coquille,” Botanique, p. 110, tab. 13, 1828 >) Wooke i) & Hare. HieNove Zely iu.) ps 252, slsoor ira Aes eiiorer Florid., p. 181, 1876. Auckland Islands; d’Urville. (Falkland Islands, Cape Horn, Valparaiso, McMurdo Strait.) Iridaea laminarioides, Bory. Iridaea laminarioides, Bory, “ Voyage de la Coquille,” Botanique, p. 105, tab. 11, fig. 1, 1828; Hook. f. & Harv., Handb. N.Z. Flora, p. 701, 1864. Auckland Islands; d’Urville. (California to Alaska and southern America ; apparently not found in New Zealand.) Marine Algae.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 507 CALLOPHYLLIS, Kiitzing, 1843. Distribution.—Chiefly in the southern seas, but a few species in the North Atlantic and North Pacific. Callophyllis variegata (Bory), Kiitzing. Halymenia variegata, Bory, “ Voyage de la Coquille,” Botanique, p. 179, 1828. Callophyllis variegata, Kiitz., Phyc. Gen., p. 400, 1843; Hook. f. & Harv., Fl. Nov. Zel., 11, p. 250, 1855. Auckland Islands; Hooker, R. M.L., J.C.S.! (New Zealand, New Guinea, Chili, Tierra del Fuego, Kerguelen Land, St. Paul Island.) Callophyllis calliblepharoides, J. Ag. Rhodomenia Hombroniana, Hook. f. & Harv., Fl. Antarct., vol. 1, p. 186, tab. 72, fig. 2, 1847. Callophyllis calliblepharoides, J. Ag., Epicr. Florid., p. 231, 1876. * Auckland Islands; J.C.S8.! (New Zealand, Chatham Islands.) Harvey confused this form with the following. Callophyllis Hombroniana (Mont.), Kiitzing. Rhodymenia Hombroniana, Mont., Voy. au Pole sud, Bot., i, p. 156, tab. 1, fig. 2, 1845. Callophyllis Hombroniana, Kiitz., Spec. Alg., 746, 1849; J. Ag. Hpicr. Florid , p. 232, 1876. Auckland Islands; Hombron, R. M. L. (New Zealand.) C. Hombromana of Hook. f. and Harv. is C. callablepharordes. Callophyllis ornata (Mont.), Kiitzing. Rhodymenia ornata, Mont., Voy. au Pole sud, Bot., 1, p. 160, 1845. Calophyllis ornata, Kiitz., Spec. Alg., p. 746, 1849. Frond stipitate, subpalmately pinnate, with the segments linear-cuneate, margin often proliferous, terminal segments obovate, conceptacles immersed in the disc of the frond. Auckland Islands ; d’Urville. This species does not appear to have been collected since the time of d’Urville, and does not appear to have been known to Harvey. Agardh (Epicr. Florid.) sug- gests that it may have been a form of Hymenocladia lanceolata, but this scarcely seems probable. CaLLOCOLAX, Schmitz, 1895. Distribution.—British Isles, New Zealand, Chatham Islands, Auckland Telande. Callocolax neglectus, Schmitz. Callocolax neglectus, Schmitz ex Batters in Annals Bot., vol. ix, p. 316, 1895. * Auckland Islands; J.C.8.! On Callophyllis calliblepharoides. (Distribution as for the genus.) This remarkable little parasite is now known on C. laciniata, C. Hombroniana, and C. callablepharoides. 508 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. Ruopociossum, J. Agardh, 1876. Distribution.—Australasia, Kerguelen Land, California, and south-west America. Rhodoglossum latissimum (Harvey), J. Agardh (?). Halymenia latissima, Hook. f. & Harv., Fl. Antarct., vol. 1, p. 189, 1847. Rhodoglossum latissumum, J. Ag., Hipicr. Florid., p. 187, 1876 (?). Enderby Island; J.C. 8.! Hooker. Campbell Island; Hooker. In Engler and Prantl’s “* Pflanzenfamilien,” this genus is submerged in Iridaea ; but, as Setchell and Gardner(') point out, it is sufficiently distinct in the structure and position of the sorus to be kept apart. Unfortunately, it is doubtful whether the plant now known from New Zealand and California as R. latissumum, J. Ag., is that originally described by Harvey as Halymenza latissema. Amongst the plants collected by Mr. J. Crosby Smith at Enderby Island are two specimens which seem to me to agree well with Harvey’s original description and with the figure given in the “ Flora Antarctica,” tab. 73, fig. 1; but this plant, though a Rhodoglossum, must, I think, be regarded as distinct from Rhodoglossum latissumum, J. Ag. In R. latissemum, J. Ag., of the mainland the sori are minute, almost punctiform (“Sorisque sphaerosporarum punctiformibus per totam fere superficiem sparis”’: J. Ag., Epicr., loc. cit.). In the Auckland Island specimen the sori, though superficial, and containing the cruciate tetraspores of the genus, are elliptical in outline, from 2mm. to 3 mm. in length, and sometimes become confluent. The tetraspores them- selves are long and narrow. ‘The plant also in its length (50-60 cm.), cuneate base, and lanceolate segments, agrees better with Harvey’s Halymema latissima than with the R. latissuma of J. Ag., in which the segments are generally shorter and broader, and often ovate in outline. It is quite possible that it may be found on a com- parison of the New Zealand and Californian forms with that of the southern islands that R. latessema, J. Ag., is distinct from H. latissema, Harvey, in which case, of course, a new specific name will have to be found for the former. RHODOPHYLLIS, Kitzing, 1846. Distribution.—Most temperate seas. Rhodophyllis bifida (Turn.), Kiitzing. Fucus bifidus, Turn., Hist. Fuc., i, p. 45, 1811. Rhodophyllis bifida, Kiitz., Spec. Alg., p. 786, 1849; Hook. f. & Harv., Handb. N.Z, Flora, p. 693, 1864. Lord Auckland Group; d’Urville. (North Atlantic and Mediterranean.) It may be doubted whether this northern species has been found in the southern islands. It was reported by d’Urville, but has not since been collected, and is not, I believe, definitely known from New Zealand or from the southern seas. Rhodophyllis acanthocarpa (Harv.), J. Ag. Callophyllis acanthocarpa, Harv., Fl. Noy. Zel., 1, p. 251, 1855; J. Ag., Epicr. Florid., p. 364, 1876. * Auckland Islands; R. M. L. (New Zealand.) (Identified by A. Gepp.) (*) “ Algae of North-western America,” p. 300, 1902. Marine Algae.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. B. Broadhead del. PLATE XXII 510 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. Fam. RHODYMENIACEAE. RHODYMENIA, Greville, 1830 (?). Distribution.—Almost all seas. Rhodymenia linearis, J. Ag. Rhodymenia linearis, J. Ag., Spec. Alg., u, p. 379, 1851; Hook. f. & Harv., FL. Nov. Zel., 1, p. 248, 1855. Auckland and Campbell Islands; Hooker. (Australasia.) Rhodymenia corallina (Bory), J. Ag. (?) Sphaerococcus corallina, Bory, “ Voyage de la Coquille,’ No. 66, 1828. Rhody- mema corallina, J. Ag., Spec. Alg., 1, p. 379, 1851; Hook. f. & Harv., FI. Nov. Zel., 1, p. 248, 1855. Auckland Islands; Hombron. (New Zealand, Tasmania, Kerguelen Land, Chili.) J. Ag. (Epicr. Florid., p. 331) states that he is unable to decide whether this plant occurs on the shores of New Zealand, as the specimens he has seen which perhaps belong to this species are all sterile. It may therefore be doubted whether our species is the same as the Chilan one. It has not been recently collected in New Zealand ; and this adds to one’s hesitation in identifying the New Zealand and Chilian forms. Rhodymenia dichotoma, Hook. f. & Harv. Rhodymena dichotoma, Hook. f. & Harv., Fl. Antarct., vol. i, p. 186, tab. 72, fig. 1, 1847. Callophyllis dichotoma, Kiitz., Spec. Alg., p. 746, 1849. * Auckland Islands; R. M. L. Campbell Island; Hooker, R. M. L., J. C. 8.! (Queen Charlotte’s Sound; Lyall.) Kiitzing has transferred this species to the genus Callophyllis ; but specimens of the plant collected by Mr. Crosby Smith at Campbell Island, and examined by me, show the structure of a Rhodymenia, and not of a Callophyllis, in cross-sections of the frond (wde Plate XXIV, fig. 3). Specimens collected by myself have been com- pared with type specimens in the British Museum by Mr. Gepp, and also identified as Rhodymenia dichotoma. I have seen no New Zealand examples of this species. HpyMeEntA, Kiitz., 1849. Distribution.—The Subantarctic Ocean, South Orkneys. Epymenia Wilsonis, Sonder. : Epymema obtusa, Kiitz., Sp. Alg., p. 787, 1849; Hook. f. & Harv., Fl. Nov. Zel., 1, p. 249, 1855. EH. Wilsonis, Sond. in Linnaea, p. 516, 1853. Auckland Islands; Hooker. (Australasia.) Epymenia obtusa is a South African species, with which Harvey considered the New Zealand form identical. Agardh, however, identifies the New Zealand species with H. Wilsonis, Sonder (Anal. Algol., p. 92), (see also Laing, “ List of New Zealand ee Marine Algae.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 511 Seaweeds,” No. 213, Trans. N.Z. Inst., xxxiv, p. 340). We did not collect this plant on the “ Hinemoa” trip; but it is more likely that the Auckland Island species will be found to be the same as the New Zealand form than to be the same as the South African variety. It may therefore for the present be regarded as E. Wilsonis. Procamium, Lamouroux. Distribution.—All seas. Plocamium cruciferum, Hook. f. & Harv. Plocamium cruciferum, Hook. f. & Harv., Fl. Nov. Zel., i, p. 246, 1855; J. Ag., Epicr. Florid., p. 345, 1876. * Hnderby Island; J.C. 8.! (New Zealand, Chatham Islands.) Plocamium brachiocarpum, Kiitzing. Plocamivum coccineum (partim), Hook. f. & Harv., Fl. Nov. Zel., u, p. 246, 1855. P. brachiocarpum, Kiitz., Spec. Alg., p. 885, 1849; J. Ag., Epicr. Florid., p. 341, 1876, * Enderby Island; J. C. S.! *Campbell Island; J. C. S.! (New Zealand, Chatham Islands.) Plocamium coccineum, Lyngbye. Plocamiaum coccineum, Lyngb., Tent. Hydroph. Danic., p. 39, tab. 9, 1819; (partim), Hook. f. & Harv., Fl. Nov. Zel., i, p. 246, 1855. Auckland Islands; Hooker, R. M. L. Campbell Island; Hooker. (North and South Atlantic, Mediterranean, North Pacific, Cape of Good Hope, Australia, Tas- mania, Antarctic Ocean, Coulman Island.) Mr. A. Gepp reports this species amongst those collected by myself at Auckland Islands. It may possibly turn out to be the preceding species. Plocamium Hookeri, Hook. f. & Harv. Plocamium Hookeri, Harv., Fl. Antarct., vol. nu, p. 474, 1847; J. Ag., Epicr. Florid., p. 337, 1876. Macquarie Island; A. Hamilton! (Kerguelen’s Land, Heard Island, South Georgia, South Orkneys.) Fam. DELESSERIACEHAE. Puatyciinia, J. Agardh, 1898. Distribution.—Australasia. Platyclinia crispatum (Hook. f. & Harvy.), J. Ag. Nitophyllum crispatum, Hook. f. & Harv., Fl. Antarct., vol. i, p. 185, tab. 71, fig. 1, 1845. Platyclima crispatum, J. Agardh, Mant. Deless , p. 110, 1898. * Auckland Islands; R. M. L. (identified by A. Gepp). Campbell Island ; Hooker. 512 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. HyPpocGiossum, 1843. Hypoglossum crassinervium (?) (Mont.), Kitz. Delesseria crassinervia, Mont., Voy. au Pole sud, Bot., i, p. 164, tab. 8, fig. 1, 1845; Hook. f. & Harv., Fl. Antarct., vol. i, p. 184, 1847. Hypoglossum crassinervium, Kiitz., Sp. Alg., p. 876, 1849. Delesseria crassinervia, Laing, Trans. N.Z. Inst., vol. xxix, p. 447, 1896, and vol. xxxiv, p. 347, 1902. Campbell Island ; Hooker, J.C.S.! Auckland Islands; Hooker. New Zealand ; R. M. L. (The plants from Fuegia and Kerguelen Land described under this name are probably distinct species—vde J. Ag., Hpicr. Florid., p. 486: Delesseria phyllo- phora and note thereon.) A fragment of a plant very similar to the one described by me (Trans. N.Z. Inst., vol. xxix, p. 447) was brought by Mr. J. Crosby Smith from Campbell Island. It differs in certain points, which may be of varietal or even specific importance, but, for the present at least, both plants may be considered as one species. The costa 1s much broader than in the New Zealand form of the plant, being 4-5 mm. in breadth, while in the local form it is only 2-3 mm. ; the wings of the pinnae are also much better developed in the Campbell Island form, whilst its ultimate pinnules are much shorter and less numerous than in the mainland variety. These may possibly be differences due to the different ages of the plants and to varying local con- ditions. My previous description (loc. cit.) of the sori 1s inaccurate, and I would like to amend it now. The tetraspores are arranged in narrow bands along both sides of the costae of the pinnae and pinnules in both the Campbell Island and New Zealand forms. I am not at all satisfied that my plant is the same as Montagne’s species, which will perhaps never be definitely re-identified, but, as I have already described it under the specific name crassinervia, it will be well to retain this until the various forms and stages have been more fully collected and more closely ex- amined, and the genus redetermined. (Vide Plate X XI, figs. 1, 2.) ScuizonnurA, J. Agardh, 1898. Distribution.—Subantarctic Ocean. Schizoneura dichotoma. Delesseria dichotoma, Hook. f. & Harv., Fl. Antarct., vol. i, p. 184, tab. 71, fig. 2, 1847; J. Ag., Epicr. Florid., p. 480, 1876. Schizoneura dichotoma, J. Ag., Sp. Alg., vol. i, pt. i, p. 168, 1898. Auckland and Campbell Islands; Hooker. (Foveaux Strait, Kerguelen Land, St. Paul Island.) Fam. RHODOMELACEAE. HerrrosipHoniA, Montagne, 1842. Distribution.—Almost all seas. . Heterosiphonia Berkleyi, Montagne. Heterosiphonia Berkleyr, Mont., Voy. au Pédle sud, Bot., i, p. 137, tab. v, fig. 1, 1845. Dasya Berkleyr, J. Ag., Sp. Alg., u, p. 1179, 1852; Hook. f. & Marine Algae.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 518 Harv., Handb. N.Z. Flora, p. 673. Heterosiphona Berkleyi, Falk, ‘‘ Die Rhodomelaceen,”’ 1901. Auckland Islands; Hooker, Hombron. (Magellan Straits, Falkland Islands, Marion Island, Kerguelen Land, South Georgia.) LAvRENCcIA, Lamouroux, 1813. Distribution.—Nearly all seas. Laurencia pinnatifida, Lamouroux. Laurencia pinnatifida, Lam., Ess. sur Thalassophytes, p. 48, 1813; Hook, f. & Harv., Fl. Antarct., vol. 1, p. 184, 1847. Auckland Islands; Hooker. (North Atlantic, North Pacific, South Africa, Magellan Straits.) Laurencia perforata, f. tenuissima (?), Montagne. Laurencia perforata, Mont., Fl. Can., p. 155; J. Ag., Spec. Alg., vol. ii, pt. iii, p. 748, 1852. * Antipodes Island ; Dr. Cockayne. (Identified by Major Reinbold.) A doubtful species, of which Major Reinbold writes, “ Laurencia/ resembles in habit and ramification L. perforata, Mont. (syn., L. vaga, decumbens, radicans, Kg. Tab. Phyc., xv, tab. 49-51). I should regard it [as] a very thin form of this species (200-100 p thick), L. perforata, f. tenwissumia (if not a new species ??).” Agardh’s description of L. perforata (loc. cit.) is as follows: Genus Laurencia, section 3—‘‘ Fronde tereti subvage ramosa; sterile caespites depressos intricatos- pannosos formante, fertile pmnato-ramosa, ramis suboppositis. L. perforata pur- purea, fronde tereti filiformi quoquoversum et vage intricato-ramosa, ramis recurvis radicantibus sursum ranulos seriatos secundos emittentibus, ramulis clavates, fruc- tiferis.”” Laurencia caespitosa, Lamouroux, is also reported by Hooker from Auckland Islands, but has not been recently collected, and its occurrence there may be con- sidered doubtiul, as it seems to be confined to the warmer waters of the Atlantic. PotysipHontA, Greville. Distribution.—All seas. Polysiphonia decipiens, Montagne. Polysiphoma decipiens, Mont., Voy. au Pole sud, Bot., i, p. 131, 1845; Hook. f. & Harv., Fl. Antarct., vol. 1, p. 184, 1847. Auckland Islands ; Hombron, Hooker. (New Zealand.) This is a little-known species. The plant described by Agardh as P. decipiens is P. aterrvma, Hook. f. & Harv. (vide under P. aterrima Falkenberg, “ Die Rhodo- melaceen,” p. 126). 34S. 514 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. Polysiphonia dumosa, Hook. f. & Harv. Polysiphonia dumosa, Hook. f. & Harv., Fl. Antarct., vol. i, p. 182, tab. 75, fig. 1, 1847; J. Ag., Sp. Alg., vol. 1, p. 950, 1852. Campbell Island ; Hooker. A little-known and doubtful species. Polysiphonia Lyallii, Hook. f. & Harv. Polysiphona Lyallw, Hook. f. & Harv., Fl. Nov. Zel., ii, p. 230. Auckland Islands; Hooker. Enderby Island; J.C.S.! Agardh(!) seems to have been deceived as to the identity of this plant by erroneously named specimens distributed by Haivey. He describes the plant as having seven siphons, whereas it has only four. (See for a full discussion of its affinities Falkenberg, ““ Die Rhodomelaceen,” p. 143.) BROGNIARTELLA, Bory, 1822. Distribution.—Chiefly in Australasian seas. Brogniartella australis (C. Ag.), Schmitz. Cladostephus australis, C. Ag., Syst. Alg., p. 169, 1824. Polysiphonia clado- stephus, Mont., Ann. Sci. Nat., 1843; Mont., Voy. au Pole sud, Bot., 1, p. 182, tab. 13, i. 4, 1845. P. australis, J. Ag., Spec. Alg., p. 1044, 1863. Brogmartella australis, Fall., “« Die Rhodomelaceen,” p. 546. Auckland Islands; Hombron, Hooker. * The Snares; R.M.L. (New Zea- land, Australia, Tasmania.) Euzonrevia, Falkenberg, 1901. Distribution.—Australasian seas. Euzoniella cuneifolia (Mont.), Falkbg. - Polyzonia cuneifolia, Mont., Voy. au Pdle sud, Bot., i, p. 143, 1845; Hook. £. & Harv., Fl. Nov. Zel., u, p. 226, 1855. Huzomella cunerfolia, Falk., “* Die Rhodomelaceen,” p. 368, 1901. * The Snares; R. M. L. Auckland Islands; Hooker, R. M. L. Campbell Island; Hooker, J.C.S.! (Southern New Zealand.) - The specimens from Perseverance Harbour collected by Mr. Crosby Smith are unusually well developed, from 8 cm. to 10 cm. in length, not epiphytic, but attached to the rocks by holdfasts. (Plate XXII, fig. 2; Plate XXIII, fig. 3.) Euzoniella flabellifera (J. Ag.), R. M. L. Polyzoma flabellifera, J. Ag., Till. Alg. Syst., vi, p. 74. * The Snares; R.M.L., J.C.S.! Campbell Island; J.C.8.! (*) J. G. Agardh, “ Species Algarum,” p. 1021. Marine Algae.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 515 The following is Agardh’s description of the species: “ P. flabellofera foliis ad geniculum quodque secundum alterne egredientibus, ambitu flabelliformibus, margine inferiore et superiore rectis integerrimis, petiolum versus cuneatim adproximatis, antico margine arcuato saepius argute dentato cellulis paginalibus rectangularibus, primaris a petiolo radiantibus, radiis novis superne formatis bifurcationem folii properantibus.” The genus Huzoniella was established by Falkenberg(') for the insertion of various species previously included under the genus Polyzonia. He seems to have overlooked the species Polyzonia flabellifera, J. Ag.,(?) which, belonging to the same section as P. cuneifolia, should also have been transferred to Huzoniella. Further, E. flabellifera is probably to be identified with Polyzonia cuneifolia, var. bifida, from Port Preservation, of Hooker and Harvey(*), but, as no type specimens of this variety are available in New Zealand, a certain opinion on this point cannot be given. I have, however, New Zealand specimens of LH. flabellifera, identified by Agardh, and the plants from the Snares and Monument Harbour agree with these in all details. The leaflets are double throughout ; the divisions vary much in relative size—sometimes they are equal or nearly so, sometimes the inner leaflet is the smaller, though it is occasionally the outer one which is the less. The upper margin of the outer leaflet is usually much more erose than the upper margin of the inner leaflet, which is dentate, the teeth running out into short hyaline points. But there seems to be a good deal of variation in the matter. I have a form from Dunedin, also collected by Mr. Crosby Smith, in which most of the leaflets are trifid. The plant is, on the whole, rather smaller than E. cuneijolia, and usually the adjacent leaflets are more widely apart in FL. flabellifera than in EF. cuneijolia. However, the large non-epiphytic forms of H. cuneifolia from Perseverance Harbour have in parts the upper margin of one leaflet separated from the lower margin of the one above by a space equal to the breadth of the leaflet. Falkenberg considers the plant E. cwnerjolia, var. bifida, as not even a variety of EL. cuneifolia. He states that bifid leaflets are chiefly characteristic of the fertile and certain other portions of the shoots of H. cuneijolia, whereas the leaflets on the creeping stems are unsplit. He further points out that, as bifid leaflets and un- divided ones are sometimes to be found on the same shoot, they cannot be considered to mark even a variety. He thus identifies #. cwnevfolia, var. bifida, with the normal form of £. cuneifolia. It is, to my mind, much more likely that the variety bifida is really H. flabellifera, which is not unlike #. cunerfolia in many of its characteristics, and has its leaflets normally bifid. It is true that an occasionally bifid leaf may be found on &. cuneifolia ; on the other hand, an occasionally entire leaflet may be found on £. flabellijera ; but the two forms—(qa) with single leaflets, (b) with double leaflets—are typically quite distinct. It will, however, be seen from the figures given that forms of FE. flabellifera from Campbell Island, the Snares, and the Bluff differ somewhat. (Vide Plate XXII, figs. 3, 4,5; and Plate XXIII, figs. 4, 5, 6, 7, 8.) (+) “ Die Rhodomelaceen,” p. 360. (2) Till. Alg. Syst., vi, p. 74. (*) Fl. Nov. Zel, 11, p. 226, 516 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. MicrocoLax, Schmitz, 1897. Distribution.—As for species. Microcolax botryocarpa, Schmitz. Microcolax botryocarpa, Falkenberg and Schmitz, Engler and Prantl’s “ Pflan- zenfamilien,’ Algae, p. 458, 1897; Falk., ““ Die Rhodomelaceen,” p. 357, 1901. The plant described by Hooker and Harvey in the “ Flora Antarctica ” as Poly- siphonia botryocarpa was really a combination of a plant resembling a Polysvphonia and a little parasite. The host plant has been described by Schmitz as Streblocladia neglecta, and the parasite as Microcolax botryocarpa. The parasite appears as little bundles of filaments about $mm. high on all parts of the host. In the “ Pflanzen- familien ” the distribution of the genus Microcolax is given as the Antarctic Ocean, but there is no definite evidence that it has yet been collected elsewhere, except the Auckland Islands. Its distribution, however, will probably be found to be the same as the host. DrereRrosipHonta, Schmitz and Falkenberg, 1897. Distribution.—New Zealand, Australia, Brazil, Mediterranean Sea. Dipterosiphonia heteroclada (J. Ag.), Falkenberg. Polysiphonia dendritica, Hook. f. & Harv., Fl. Nov. Zel., u, p. 232, 1855. P.. heteroclada, J. Agardh, Till. Ale. Syst., iv, p. 98. Dupterosiphonia hetero- clada, Fall., ““ Die Rhodomelaceen,” p. 320, 1901. * Antipodes Island; Cockayne (identified by Major Reinbold). (Also known from New Zealand and (?) Australia.) As the specific name dendritica is preoccupied for the Brazilian Polysiphoma (Dipterosiphonia) dendritica, C. Ag., it must be replaced by the specific name hetero- clada, proposed by J. Agardh. HeERPOSIPHONIA, Naegeli, 1846. Distribution.—Widely distributed, especially in warmer seas. Herposiphonia ceratoclada (Mont.), Falkenberg. Polysuphonia ceratoclada, Mont., Voy. au Pole sud, Bot., 1, p. 130, tab. 5, fig. 2, 1845 ; Hook. f. & Harv., Fl. Antarct., vol. 1, p. 183, tab. 85, fig. 2, 1847. Herposiphona ceratoclada, Falk., “* Die Rhodomelaceen,” p. 313, 1901. Auckland Islands ; d’Urville, Hooker. STREBLOCLADIA, Schmitz, 1897. Distribution.—As for species.(*) Streblocladia neglecta (Mont.), Schmitz. Rhodomela (?) botryocarpa (partim), Mont., Voy. au Pole sud, Bot., 1, p. 140, 1845. Polysiphonia botryocarpa (partim), Hook. f. & Hary., Fl. Antarct., (+) There is another doubtful species from the Mediterranean Sea. Marine Algae.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. ili vol. i, p. 69, tab. 70, 1847. Rhodomela glomerulata (partim), Hook. f. & Harv., Handb. N.Z. Flora, p. 665, 1855. Streblocladia neglecta, Schmitz and Falkenberg, Engler and Prantl’s “ Pflanzenfamilien,’ Algae, p. 457, 1897; Falk., “ Die Rhodomelaceen,” p. 354, 1901. Auckland Islands; d’Urville, Hooker, R. M. L. (New Zealand.) (Identified by A. Gepp.) Fam. CERAMIACEHAE. GRIFFITHSIA, C. Agardh, 1817. Distribution.—All seas. A fragment of an alga probably belonging to this genus was collected by Mr. Crosby Smith at Enderby Island. (Vide Plate XXIII, figs. 1, 2.) CALLITHAMNION, Lyngbye. Distribution.—All seas. Callithamnion cryptopterum (Hook. f. & Harv.), Kiitz. Callithamnion micropterum, Hook. f. & Harv., Fl. Antarct., vol. 1, p. 192, 1847. C. cryptopterum, Kiitz., Sp. Alg., p. 646, 1849. Auckland Islands ; Hooker. A little-known and not recently collected species. As Montagne had previously described a distinct species from the Canaries as C. micropterum, Kiitzing altered the specific name of this plant to OC. cryptopterum. Callithamnion gracile, Hook. f. & Harv. Callithammon gracile, Hook. f. & Hary., Fl. Antarct., vol. 1, p. 191, tab. 78, fig. 1, 1847. Campbell Island ; Hooker. A little-known and not recently collected species. PTILOTHAMNION, Thuret, 1877. Distribution.—Probably in many seas. Ptilothamnion pectinatum (Mont.), R. M. L. Callithamnion pectinatum (2), Mont., Voy. au Pole sud, Bot., i, p. 90, 1845 ; Hook. f. & Harv., Fl. Antarct., vol. 1, p. 191, 1847. Ptilothamnion pectin- atum, Laing, ““ New Zealand Species of Ceramiaceae,” Trans. N.Z. Inst., vol. xxxvil, p. 388, 1905. Auckland Islands; d’Urville. The Snares; J.C.S.! (Stewart Island.) I previously described (loc. cit.) this plant from specimens obtained from Half- moon Bay, Stewart Island, and identified it somewhat doubtiully with Montagne’s C. pectinatum. My previous specimens were somewhat immature ; and the present ones, collected at the Snares by Mr. Crosby Smith, are much larger, and present various differences in branching from the Stewart Island specimens. 518 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. They are densely tufted, 1-2 cm. high, and much branched. In dried specimens the thallus is dull brown in colour. The pinnae are themselves irregularly pinnately and occasionally dichotomously branched. In the Stewart Island specimens, from towards the upper ends of each cell of the pinnae spring two ultimate distichous pinnules of limited growth. In the Snares specimens the ultimate pinnules are less regularly arranged. Perhaps the most common and characteristic arrangement 1s for them to be in alternate pairs. ‘Two joint-cells in succession give rise to a pinnule on the same side; then the two cells above them give rise to a pair of pinnules on the opposite side. This arrangement, however, is by no means the only one, though perhaps the most common. Occasionally joint-cells may be found without pinnules ; and at other times three pinnules will spring out in succession from one side of the pinna ; or sometimes a cell will give rise to a pair of pinnae, as in the Stewart Island specimens. I have seen as many as four pinnules in succession on one side without corresponding ones on the other side. The tetraspores are the same as those of the Stewart Island specimens, and similarly placed. PLEONOSPORIUM, Naegeli, 1861. Distribution.—Northern and southern extra-tropical seas. Pleonosporium hirtum (Hook. f. & Harv.), R. M. L. Callithamnion hirtum (partim), Hook. f. & Harv., Fl. Antarct., vol. 1, p. 192, tab. 78, fig. 2, 1847. Pleonosporium lurtum, Laing, “ New Zealand Species of Ceramiaceae,’ Trans. N.Z. Inst., vol. xxxvu, p. 393, 1905. Auckland Islands; Hooker. (New Zealand.) If, as I think probable, Harvey confused several species under the name C. rtum, tt may be doubted which of these is represented by the plant collected by Hooker at the Auckland Islands. As, however, in New Zealand at least, P. lartum is much more common than the others (Spongoclonium sp.), the Auckland Island plant is perhaps this species. Indeed, the figure in the “ Flora Antarctica ”’ almost certainly represents P. hirtum, or the very closely allied P. Brownanum, rather than a Spongoclonium. Kuprinota, Kiitzing, 1847. Distribution.—South temperate and subantarctic seas. Euptilota formosissima (Mont.), Kiitz. Ptilota formosissima, Mont., Voy. au Péle sud, Bot., i, p. 98, tab. 9, fig. 3, 1845 ; Hook. f. & Harv., Fl. Antarct., vol. 1, p. 190, tab. 77, 1847. Huptilota formosissima, Kiitz, Sp. Alg., p. 671, 1849. Ptidota formosissima, Hook. f. & Harv., Handb. N.Z. Flora, p. 708, 1864. * The Snares; R. M. L. Auckland Islands; d’Urville, Hooker, J. C. S.! R. M. L. | (New Zealand.) Marme Algae.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. By RS) B. Broadhead del. PLATE XXIII. 520 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. Bauuia, Harvey, 1843. Distribution.—All subantarctic seas. Ballia callitricha (C. Ag.), Kiitz. Sphacelaria callitricha, C. Ag., Syst. Alg., p. 166, 1824. Ballia Brunonia, Harv. in Hook. Journ. of Bot., 1, p. 191, 1848. B. Hombroniana, Mont., Voy. au Pole sud, Bot., 1, p. 94, 1845. B. Brunona, Hook. f. & Harv., Fl. Antarct., vol. 1, pp. 78, 182, 1847. B. callitricha, Kitz., Sp. Alg., p. 663, 1849; Hook. f. & Harv., Handb. N.Z, Flora, p. 710, 1864. * The Snares; J.C. 8.! Auckland Islands; Hooker and Harvey, J. C. S.! Campbell Islands; Hooker. (Australia, Tasmania, New Zealand, Cape Horn, Falk- land Islands, Kerguelen Land.) CrRAmium, Roth. Distribution.—Abundant in all seas. Ceramium rubrum, C. Ag. Ceramium rubrum, C. Ag., Spec. Alg., nu, p. 146, 1820-27; Hook. f. & Harv., Fl. Nov. Zel., u, p. 256, 1855. Auckland Islands; Hooker, J. C. S.! Campbell Island; Hooker, R. M. L. (Common in all temperate seas.) Ceramium diaphanum, Lightfoot. Ceramaum diaphanum, Lightfoot, Scot., p. 996 ; Hook. f. & Harv., Fl. Nov. Zel., i, p. 256, 1855; J. Ag., Eipicr. Florid., p. 98, 1876. So many ill-defined species have been included under this caption that it may well be doubted whether amongst the diaphanous forms occurring on the New Zealand coast we have any which belong to C. diaphanum in the stricter sense. The distribu- tion given below can only be regarded as true of C. diaphanum, as understood in a wide sense. Auckland Islands; Hooker. (Atlantic Ocean, Cape of Good Hope, New Zea- land, Australia.) Ceramium aucklandicum, Kiitzing. Ceranuum aucklandicum, Kiitz., Sp. Alg., p. 686, 1849; J. Ag., Anal. Algol., cont. 1, p. 36, 1894. Auckland Islands ; Hooker, J. C. S.! “ Ceramium pollicare vage et crebre dichotomum ramulis superioribus gra- cilibus forcipatis, forcipibus apertis; articulis diametro aequalibus. Tetrocharpia in ramis mediis numerosa nodiformia, semi-erumpentia.” (Kiutz., loc. cit.) Amongst the specimens of Ceramium collected by Mr. Crosby Smith at Enderby Island is a fragment of a plant which agrees well with Agardh’s description of Ceramium aucklandicum (loc. cit.). The specimen bears cystocarps. I have not seen the tetraspores. The cystocarps generally replace one of the terminal segments, are often in pairs, and are enclosed by one to four involucral branchlets, or by none. The joint-cells are from two-thirds as long as broad to as long as broad. Marine Algae.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 52] Ceramium virgatum, Hook. f. & Harv. Ceramium virgatum (?), Hook. f. & Harv., Fl. Nov. Zel., ii, p. 256; J. Ag., Anal. Algol., cont. u, p. 45, 1894. * Auckland Islands; J.C.S.! (New Zealand.) A fragment apparently belonging to this species occurs amongst those sent me by Mr. Crosby Smith. It bears tetraspores, but is almost too small for certain identification. Ceramium apiculatum (Hook. f. & Harv.), J. Ag Ceramium cancellatum (partim), Hook. f. & Harv., Fl. Nov. Zel., ti, p. 256, 1855. C. apiculatum, J. Ag., Epicr. Florid., p. 105, 1876; J. Ag., Anal. Algol., cont. 1, p. 20, 1894. Auckland Islands; Hooker. (New Zealand; J. Ag.) Ceramvum cancellatum is a species from the Cape of Good Hope, considered by Agardh to be quite distinct from the New Zealand plant, which he describes as fol- lows (Epicr. Florid., loc. cit.): “‘C. apiculatum tfronde elatinre compressa distiche pinnata pinnis dichotomo-pinnulatis subcorymbosis, segmentis patentibus, terminali- bus eximie divaricatis apiculatis, articulis inferioribus diametro subbrevioribus ubique corticatis, sphaerosporis strato corticali immersis in segmentis antepenultimis cir- cumeirea verticillatis, in verticillo pluribus, parum prominulis favellis.” Ceramium vestitum (?), Hook. f. & Harv. Ceramium vestitum, Hook. f. & Harv., Fl. Nov. Zel., 11, p. 256, 1855. Antipodes ; Cockayne ! I forwarded a species of Cerannum collected by Dr. Cockayne at Antipodes, and it was returned as Ceramium vestitum, Harv.(?) (C. Stechidiosum affine), (vide also J. Ag., Hpicr. Florid., p. 106). Agardh has also identified doubtfully some sterile specimens collected by Berggren at Dunedin and Banks Peninsula as belonging to this species (Anal. Algol., cont. 1, p. 40). His plants are minute, more slender than C. rubrum, but completely corticated, the lower internodes apparently showing the characteristic reticulated appearance of C. rubrum. The joint-cells are a little longer than broad. The primary pinnae branch outwards in all directions, and the thallus is densely clothed on all sides with prolifications. This characteristic he thinks may have led to the specific name vestitwm. Harvey’s original specimens came from Stewart Island; his diagnosis, however, is not sufficiently discriminative to enable the species to be identified with certainty without comparison with his type speci- mens. He considered his species as perhaps a state of C. rubrum. Ceramium stichidiosum (?), J. Ag. Ceramium stichidiosum, J. Ag., Epicr. Florid., p. 105, 1876; J. Ag., Anal. Algol., cont. il, p. 21, 1876; C. stichidiosum, J. Ag., var. Smithii, R. M. L. (var. nov.), 1909. C. stechidiosum, J. Ag., var. scopulorum, R. M. L. (var. nov.), 1909. I had originally imtended to describe certain specimens of Ceramiwm from Knderby Island and the Snares, collected by Mr. J. Crosby Smith, as a new species, 522 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. under the name C. Smith; however, after much consideration, it seems to me better to retain them under Agardh’s specific name stechidiosum, and to call the Enderby Island specimens C. stichidiosum, var. Smithii, and those from the Snares C. stichidiosum, var. scopulorum. If, however, they should hereafter be considered to constitute a distinct species I would recommend that the specific name Smuathii be adopted, in honour of Mr. J. Crosby Smith, who collected them, and who has also collected the algae of castern and southern Otago with great care and much enthusiasm. Unfortunately, I have seen no authentic specimens of C. stichidiosum, and am therefore not certain whether these plants can be retained under this specific name. I can, however, find in them no characteristic to separate them distinctly from C. stichidiosum as described by Argardh. However, in both the specimens from the Snares and those from Enderby Island the tetraspores are more or less confined to the ultimate and penultimate segments of the pinnules, and do not form the complex stichidiose-like organs described by Agardh (Anal. Algol., loc. cit.). It is possible, if they were more fully developed, or growing in warmer seas, the fertile segments would be more luxuriantly developed, when they would apparently agree in all respects with those described by Agardh. As, however, there may be some doubt of the identity of my plants with those described by Agardh, I give a fairly full description of them. Ceramium stichidiosum, var. Smithii. A small densely tufted filiform plant 2-4 cm. high, nearly black (Klinksieck’s “Code des Couleurs,’ No. 65) im dried specimens, irregularly alternately subdis- tichously pinnately branched in the lower portion of the plant. Main pinnae almost squarrose, with simple more or less branched prolifications. Terminal and sub- terminal pinnules more or less dichotomous, and somewhat apiculate; ultimate pinnules usually divaricating, at other times, when very young, sometimes slightly incurved. Branching of pimnae takes place at somewhat irregular intervals, but generally a ramulus is given off from about every fifth joint-cell, the forks beimg strict at the axils. The whole plant is completely and nearly regularly corticated by a layer of minute, irregularly shaped, angular, rounded, or elongated cells, rather more densely arranged over the nodes than over the internodes. A transverse section of the stem shows a few somewhat larger internal cortical cells, quite distinct, however, from the large internal cortical cells of a Microcladia. The tetraspores are large and numerously developed in irregular lines on both sides of the terminal and subterminal pinnules. They are first immersed, but aiter- wards protruded by about half their extent from the cortex, giving the pinnules a swollen and somewhat stichidiose appearance. They are occasionally developed on the face as well as on the edge of the pinnule. Cystocarps not seen. The species C. stichidiosum comes nearest apiculatum, from which, however, it may be distinguished by tts abundant prolifications, by its much more slender and less-flattened stems, by its long, narrow, and almost acuminate terminal seg- ments. (In C. apiculatum the ultimate pinnules are almost truncated. (Vide Plate XXII, fig. 1; Plate XXIV, figs. 1, 4.) * Knderby Island; J.C.S.! Marine Algae.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 523 Ceramium stichidiosum, var. scopulorum. General description as for C. stichidiosum, var. Smathii, but differing in the following points: A larger plant, 4-6cm. high, dark red in colour (Klinksieck’s “Code des Couleurs,” No. 578), much more luxuriant, denser in growth, and differing much in general appearance from the preceding, but scarcely to be distinguished from it by any important diagnostic character. The tips of the pimnae end in a pencil of acuminate pinnules, owing to the repeated dichotomous branching of the terminal and subterminal segments, the forks remaining very narrow, not so densely clothed with prolifications as the preceding. The cortication seems to present certain differences from the preceding, but these are perhaps due to the more luxuriant growth of the plant. The whole plant, except the young pinnules and prolifications, is completely corticated by threads consisting of minute cells which are developed upwards and downwards from the larger cells covering the nodes. The interspaces between the cells give to the inter- nodes a more or less longitudinally veined appearance. On the internodes of the young pinnules transverse hyaline lines are sometimes seen. Tetraspores as in C. stichidiosum, var. Snuthu. Cystocarps solitary or in pairs in the axils of the sub- terminal pinnules. (Plate XXIV, fig. 2, represents a fragment of frond of C. stichi- diosum, var. scopulorum, magnified to show cystocarps.) * The Snares; J.C€.S8.! MicrocuaptA, Greville. Distribution.—Most seas. Microcladia pinnata, J. Ag., Anal. Algol., cont. iv, p. 34, 1897. Enderby Island; J.C.S.! (New Zealand; R. M. L.) The species is scarcely to be distinguished from the northern M. Coulteri; but, as no description of it has been given in any publication dealing with New Zealand plants, I append one. Thallus 15-20 mm. long, filiform, compressed, the whole distichously, rather laxly, alternately decompound. Lower pinnae rather distant, 1-2 cm. apart, alternate elongate, generally not much branched, and with the secondary pinnae and pinnules about lanceolate in outline. Ultimate pinnules acute, and coming off at an angle of about 45° or more from the pinnae. The lower pinnae are set at about right angles to the main stem, but the upper ones become less and less divergent. The main stem and pinnae are sometimes provided with two series of distichous simple or subsimple prolifications 1-3 mm. in length. In some forms these are wanting, and the main stem is completely glabrous. The tetraspores (not hitherto described) are in tetrads, subterminal in the ultimate pinnules, at first completely immersed, but afterwards partially extruded, and more or less irregularly arranged, generally in irregular longitudinal series along both margins of the thallus, sometimes appearing to be pseudo-verticillate and at other times scattered, the arrangement varying in different parts of the same frond. The cystocarps are terminal or subterminal, sessile or shortly stalked, black in dried specimens, and surrounded by a more or less completely developed involucre of ramull. 524 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. Fam. GRATELOUPIACEAE. PacHyMEnIA, J. Agardh, 1876. Distribution.—Cape of Good Hope, New Zealand. Pachymenia lusoria (Hook. f. & Harv.), J. Agardh. Iridaea lusoria, Hook. f. & Harv., Fl. Nov. Zel., u, p. 252, 1855. Pachymenia lusorva, J. Agardh, Epicr. Florid., p. 146, 1876. * The Snares; R.M.L., J.C.S. * Antipodes; Cockayne! (Abundant in the southern parts of New Zealand.) Grateloupia (?) aucklandica, Mont. (Voy. au Péle sud, Bot. i, p. 115, tab. 110, fig. 1) is a very doubtiul and not recently collected species. Fam. DUMONTIACEAE. DumontrA, Lamouroux. ?? Dumontia filiformis, Grev. This plant is recorded from Campbell Island by Hooker, but is a distinctly northern species. Agardh (Hpicr. Florid., p. 150), writing under Gratelowpia pro- lijera, thinks it possible that this may have been the plant identified by Hooker as D. filrformis. Dumontia cornuta, Hook. f. & Harv. (Fl. Antarct., vol. i, p. 189) is another not recently collected plant of very doubtful identity. It is reported only from Camp- bell Island. Fam. CORALLINACHAE. Literature for a proper description of the Corallinaceae, including Foslie’s recent papers, is not available in New Zealand, and no attempt has yet been made to identity the species collected during the “ Hinemoa” trip. The following species were ob- tained by previous collectors :— CoraALuina (Tournefort), Lamouroux. Distribution.—All seas. Corallina Hombronii (Mont.), J. Ag. Jania Hombroni, Mont., Voy. au Pole sud, Bot., vol. i, p. 145, 1845. Coral- lina Hombronu, J. Ag., Sp. Alg., u, p. 574, 1852. Auckland Islands ; Hombron, Hooker. A not recently collected species closely allied to the better known C. Cuvieri of the South Temperate Ocean. Marine Algae.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 525 Corallina virgata, Zanardini. Corallina virgata, Zan., ““ Rivista Sulle Corallineae,” 1844. C. granijfera, J. Ag., Sp. Alg., ii, p. 569, 1852. Antipodes ; Cockayne (identified by Reinbold). (Mediterranean, Africa, Queens- land.) As there is no published description accessible in New Zealand, the following from Areschoug (J. Ag., Sp. Alg., loc. cit.) may be quoted :— “(C. granifera, fronde pollicari flaccida filiformi capillari, primaria inferne ramosa, ramis bipinnatis, pinnis patentibus irregulariter pinnatis, pinnulis simplicibus 1, dichotomo-ramosis, articulis ramiferis cuneatis, 1 subclavatis, eramiferis subcylin- draceis utrisque diametro 3 plo. —4 plo. longioribus, Keramidis apicalibus suburnae formibus poro producto biquadricorniculatis, cornibus attenuatis.” MetopestA, Lamouroux. Distribution.—In nearly all seas. Melobesia antarctica, Hook. f. & Harv. Melobesia antarctica, Harvey, “‘ Nereis Australis,” p. 111, 1847; Hook. f. & Harv., Handb. N.Z. Flora, p. 681, 1864. Auckland Islands; Hooker. (Cape Horn, Falkland Islands, Kergulen Land.) INCERTAE SEDIS. APoPHLOEA, Harvey, 1855. Distribution.—New Zealand. A genus of uncertain position, perhaps belonging to the Sphaerococcaceae. Apophloea Sinclairii, Harvey. Apophloea Sinclairu, Hook. f. & Harv., Fl. Nov. Zel., 1, p. 244, tab. 116B, 1855. * The Snares and * Campbell Island; R.M.L. (Otago, Stewart Island.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Marine Algae. 526 EXPLANATION OF PLATES XXI-XXIV. Prats XX. Fig. 1. Portion of specimen of Hypoglossum crassinervium (2) from Campbell Island. (Natural size.) Fig. 2. Portion of specimen of Hypoglossum crassinervium, St. Clair, Dunedin. (Natural size.) Puate XXII. Fig. 1. Portion of a frond of Ceramium stichidiosum, var. Smith, showing divaricating pinnules. (Magnified.) Fig. 2. Fragment of frond of Huzontella cuneifolia. (x 20.) Fig. 3. Fragment of frond of Huzoniella flabellifera. (x 20.) (Form from St. Clair, Dunedin.) Fig. 4. Fragment of same from Snares. (x 20.) Fig. 5. Fragment of same from Campbell Island. (x 20.) Prare XXIII. Fig. 1. Portion of frond of species of Griffithsia (?) from Enderby Island. (x 15.) Fig. 2. Node of same. (x 150.) Fig. 3. Portion of stem and base of pinnule of Huzomella cuneifolia, to show shape and arrangement of cells. (x 100.) Fig. 4. Portion of stem and pinnule of Huzoniella flabellifera, to show shape and arrangement of cells. Specimen from Snares. (x 100.) Fig. 5. Fragment of frond of Huzoniella adiantiformis, from the Bluff (collected by Berggren, type specimen of J. Agardh), given for comparison with H#. cunedfolia and HL. flabellafera. ( x 20.) Fig. 6. Portion of stem and pinnule of above. (x 100.) . Fig. 7. Portion of stem and pinnule of H. flabellifera. Specimen from St. Clair, Dunedin. (x 100.) Fig. 8. Portion of stem and pinnule of #. flabellifera. Specimen from Campbell Island. (x 100.) PLate XXIV. Fig. 1. Fragment of Ceramium stichidiosum, var. Smithit, showing joint-cells and cortication. ( x 500.) Fig. 2. C. stichidiosum, var. scopulorum, showing cystocarps. (x 50.) Fig. 3. Rhodymenia dichotoma : Transverse section of frond. (x 240.) Fig. 4. C. stichidiosum, var. Smithu, showing tetraspores. (x 100.) Fig. 5. Rhizoclonium Hookert: Portion of creeping rhizome, showing rhizoid and young filaments. (x 50.) Fig. 6. R. Hookeri : Tip of an older filament. (x 50.) Marine Algae.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 527 Sate 6S BSD SO 00 B. Broadhead del. PLATE XXIV. ARTICLE XXII.—LIST OF CRYPTOGAMS (EXCLUDING ALGAE AND PTERIDOPHYTA) COLLECTED BY MEMBERS OF THE EXPEDITION TO THE SUBANTARCTIC ISLANDS OF NEW ZEALAND. THE numbers correspond to those of the specimens in the collectors’ herbaria and in the Royal Botanical Museum of Berlin. The Antipodes Island cryptogams are here recorded for the first time ; they were collected by Cockayne in July, 1903. It may be noted that no Snares species are in this list, this group, strange to say, being almost without bryophytes, as pointed out by Kirk in 1891 (Rep. A.A.A.S., 1m, p- 217). The absence of these plants probably depends upon the close covering of the ground by tussocks, the constant destruction of the vegetation by penguins, and the absence of rata forest, with its hygrophytic interior. Jt (UF IN) Gell, Ibe (Identified by Mr. G. Massee, ¥.R.S., Royal Gardens, Kew.) PYRENOMYCETES, Fries. Lacunea, Fries, 1823. Distrib.—Temperate and subtemperate zones of both hemispheres. Lachnea scutellata, Guillet. Campbell Island; rare (J. B. Mayne). No. 2. Distrib.Kurope, Java, Tasmania, United States, Fuegia. BASIDIOMYCETES, Fries. OmpHALIA, Fries, 1821. Distrib.—Temperate and subarctic regions of both hemispheres. Omphalia umbellifera, Fries. Campbell Island ; plentiful in all the meadows (J. B. Mayne). No. 1. Distrib.Kurope, Greenland, Ceylon, Tasmania, and United States. LEentTINuS, Fries, 1836. Distrib.—Cosmopolitan. Lentinus strigosus, Fries. Campbell Island ; on Dracophyllum (J. B. Mayne). Distrib.—Ceylon, Tasmania, and United States. a ae Cryptogams. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 529 Fomes, Fries, 1851. Fomes lucidus, Fries. Campbell Island ; on dead Dracophyllum (J. B. Mayne). No. 4. Distrib.—Common everywhere, except in the Arctic Zone. Fomes annosus, Fries. Auckland Island (J. B. Mayne). No. 7. Distrib.Europe, United States, and Cuba. A very destructive wound-parasite to conifers in Europe. STEREUM, Persoon, 1796. Stereum vellereum, Berk. Campbell Island; on dead wood of Dracophyllum (J. B. Mayne). No. 3. Distrib. —Australia and New Zealand. LICHEN KS. (Identified by Mr. G. Massee, F.R.S., Royal Gardens, Kew.) CottEMA, Wigg, 1780. Collema flaccidum, Ach. Campbell Island ; common on Dracophyllum (Chas. Chilton). No. 10. Distrib.—Kurope, Australia, New Zealand, and United States. MY XOGASTRES, Fries. Tricu14, Haller, 1768. Trichia affinis, De Bary. Auckland Island (J. B. Mayne, Chas. Chilton). Nos. 8 and 9. Distrib.—Kurope, Tasmania, Australia, United States, and Cuba. 1 10 Nell (Ident fied by Professor Dr. G. Lindau, Botanisches Museum, Berlin.) PERISPORIACEAE. ANTENNARIA, Link., 1809. Distrib.—Subcosmopolitan. Antennaria scoriadea, Berk. _ Auckland Islands (J..S. Tennant). Nos. 65 and 73. 30—S. 530 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Cryptogams. THELEPHORACEAE. STEREUM, Pers., 1796. Distrib.—Cosmopolitan. Stereum hirsutum (Willd.), Pers. Auckland Islands (J. S. Tennant). No. 72. CAE SENSES: (Identified by Professor Dr. G. Lindau, Botanisches Museum, Berlin.) SPHAHROPHORACEAE. SPHAEROPHORUS, Pers., 1794. Distrib.—Nearly cosmopolitan, but most species on the Southern Hemisphere. Sphaerophorus fragilis (l.), Pers. Auckland Islands (J. S. Tennant); No. 63. Campbell Island (Dr. Cockayne) : No. 841. Sphaerophorus argus, Hook. f. et Tayl. Auckland Islands (J. S. Tennant). No. 68. Sphaerophorus tener, Laur. Auckland Islands (J. S. Tennant). No. 60. Sphaerophorus stereocauloides, Nylander. Campbell Island (R. M. Laing). Sphaerophorus australis, Laur. Auckland Islands (J. 8. Tennant). Nos. 55 and 62. LECIDEACEAE, Lopapium, Korb., 1855. Distrib.Lopadium is a nearly cosmopolitan genus. Lopadium pezizoideum (Ach.), Kérb., var. muscicola (Smf.), Korb. Campbell Island (Dr. Cockayne). No. 843. PANNARIACEAE. PannariA, Delis., 1827. Distrvb.—Cosmopolitan. Pannaria mariana (H. Fr.), Mull. Arg. Auckland Islands (J. 8, Tennant). No. 82, ns —s 2 Cryptogams. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 531 STICTACEAKE. Strcta, Schreb., 1791. Distrib.—Cosmopolitan ; most abundant in warmer countries. Sticta Freycinetii, Del. Antipodes Island (Dr. Cockayne). Nos. 828 and 832. Sticta filicina (Ach.), Nyl. Antipodes Island (Dr. Cockayne) ; No. 822. Auckland Island; No. 824. Sticta orygmaea, Ach. Auckland Islands (J. S. Tennant); Nos. 69 and 70. Antipodes Island (Dr. Cockayne) ; No. 837. Campbell Islands (R. M. Laing). Sticta fragillima. Auckland Islands (J. 8. Tennant); No. 67. Campbell Island (R. M. Laing). Of both localities the material is defective and the identification not certain. PARMELIACEAE. Usnea (Dill.), Pers., 1794. Distrib.—A_ subcosmopolitan genus. Usnea articulata (Hofim.), Nyl. Antipodes Island (Dr. Cockayne). Nos. 839 and 840. CLADONIACEAE. Ciaponia, Wiggers, 1780. Distrib.—Cosmopolitan. Cladonia furcata (Huds.), Schrad., var. a racemosa (Hoffm.), Fr. Habitat and collector uncertain. Cladonia gracilis (L.), Willd., var. « Campbelliana, Wun. Antipodes Island (Dr. Cockayne). No. 834. Cladonia pycnoclada (Gaud.), Nyl. Antipodes Island (Dr. Cockayne). No. 830. Cladonia firma, Laur. Campbell Island (Dr. Chilton). No. 27. Cladonia aggregata (Sw.), Ach. Auckland Islands (J. 8. Tennant); No. 83. Campbell Island (R. M. Laing). Antipodes Island (Dr. Cockayne) ; Nos. 819, 820, and 838, 532 SUBANTARCTIC ISLANDS OF NEW ZEALAND. Cladonia retipora (Lab.), Fr. Auckland Islands (J. S. Tennant); Nos. 64 and 70. Laing). Cladonia verticillata, Hoftm. Antipodes Island (Dr. Cockayne). No. 844. Cladonia fimbriata (L.) Fr., var. a simplex, Flot. Auckland Islands (J. S. Tennant). No. 66. Cladonia cornuta (L.), Schrad. Campbell Island (R. M. Laing). STEREOCAULON, Schreb., 1791. Distrib.— Cosmopolitan. Stereocaulon ramulosum, Ach. Antipodes Island (Dr. Cockayne). Nos. 821 and 842. Stereocaulon argodes, Nyl. Antipodes Island (Dr. Cockayne). No. 835. HEHEPATICA #. (Identified by F. Stephani, Leipzig, Germany.) MARCHANTIACHAE. MarcuHantiA (L.), Raddi, 1818. Distrib.—Cosmopolitan. Marchantia cephaloscypha, Steph. Auckland Island (Dr. Cockayne). Nos. 84 and 820. JUNGERMANNIACEAK. MertzcertiA, Raddi, 1820. Distrib.Almost cosmopolitan. Metzgeria glaberrima. Antipodes Island (Dr. Cockayne). Nos. 827 and 831. Metzgeria pinnata, Steph. Auckland Islands (J. 8. Tennant). No, 46, Campbell Island (R. M. ht ee de ee Thal sahil Cryptogams. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 533 Patuavicinia (S. F. Gray, 1821), Steph., 1892. Distrib.—Subcosmopolitan. Pallavicinia connivens (Col.), Steph. Antipodes Island (Dr. Cockayne). No. 823, partly. TyLIMANTHUS, Mitt., 1867. Distrib.—On the Southern Hemisphere. Tylimanthus homomallus, Steph. Antipodes Island (Dr. Cockayne). No. 826. Tylimanthus saccatus (Tayl.), Mitt. Auckland Islands (J. 8. Tennant); Nos. 35, 44, and 56. Campbell Island (R. M. Laing). PLAGIOCHILA, Spruce, 1885. Distvib.Subcosmopolitan, but more abundant in the tropics and the Southern Hemisphere. Plagiochila Howeana, Steph. Auckland Islands (J. 8. Tennant). No. 52, partly. Plagiochila ramosissima, Lindb. Auckland Islands (J. 8. Tennant). No. 58. Plagiochila strombifolia (Tayl.), Lehm. Auckland Islands (J. 8. Tennant). Nos. 48 and 53. Letoscypuus, Mitt., 1855. Distrib.—On the Southern Hemisphere and in the mountainous parts of the tropics. Leioscyphus chiloscyphoides, Mitt. Auckland Islands (J. 8. Tennant). No. 57, partly. LopHocoLEeA, Dum., 1835. Distrib.—Cosmopolitan, but more abundant in the temperate Southern Hemi- sphere. Lophocolea pallida, Mitt. Auckland Islands (J. 8. Tennant); No. 2. Auckland Island (Dr. Cockayne) : Nos. 814 and 822. Antipodes Island ; No. 823, partly. Lophocolea fulva, Steph. Auckland Island (Dr. Cockayne). Nos. 813 and 815. 534 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Cryptogams. CHEILOSCYPHUS, Corda, 1829. Distrib.Most species on the Southern Hemisphere, few on the Northern. Cheiloscyphus ciliatus, Steph. Campbell Island (R. M. Laing). Mastigopryum, Nees, 1838. Distrib.—Almost cosmopolitan. Mastigobryum involutum, Lindb. Auckland Islands (J. S. Tennant); Nos. 1, 144, 154, 50. Auckland Island (Dr. Cockayne); Nos. 84 and 817. Leprpoz1A, Dum., 1831. Distrib.—Almost cosmopolitan. Lepidozia asperifolia, Steph. Campbell Island (R. M. Laing). LEPIDOLAENA, Dum., 1835. Distrib.—In the temperate and cold parts of the Southern Hemisphere. Lepidolaena Menziesii (Hook.), Dean. Antipodes Island (Dr. Cockayne). No. 836. TRICHOCOLEA, Dum., 1822. Distrib.— Cosmopolitan. Trichocolea tomentella (Huds.), Nees. Auckland Islands (J. 8. Tennant); No. 52, partly. Auckland Island (Dr. Cockayne); No. 86. SCHISTOCHILA, Dum., 1835. : Distrib.—On the Southern Hemisphere and in the mountains of the tropics. Schistochila Lehmanniana, Nees. Auckland Island (Dr. Cockayne). No. 818. Schistochila pinnatifida, Hook. Auckland Islands (J. 8. Tennant). Nos. 42 and 45. MapotHeca, Dum., 1821. Distrib.—In all parts of the world; abundant in the tropics. Madotheca Stangerii, L. et G. Auckland Islands (J. S. Tennant). No. 57, partly. Mixed with Leioscyphus chiloscyphordes. ———— Cryptogams. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. MUSCL. (Identified by Dr. W. F. Brotherus, Helsingfors, Finland.) DICRANACEAE. DicRANOLOMA, Ren., 1897. Distrib.—Most species on the Southern Hemisphere. Dicranoloma Billardieri (Schw.), Brotherus. Auckland Islands (J. 5. Tennant); Nos. 3, 9, 11, 18, 23, and 34. 535 Campbell Island (R. M. Laing). Auckland Island (Dr. Cockayne); Nos. 81, 812, and 816. Dicranoloma Menziesii (Tayl.), Brotherus. Auckland Islands (J. 8. Tennant). Nos. 12 and 25. Dicranoloma Menziesii (Tayl.), Brotherus, forma nana. Auckland Islands (J. 8. Tennant). Nos. 14, 15, and 30. Dicranoloma setosum (Hook. f. et Wilson), Brotherus. Auckland Islands (J. 8S. Tennant). No. 28. CampyLopus, Brid., 1819. Distrib.—Nearly cosmopolitan. Campylopus introflexus (Hedw.), Mitt. Auckland Islands (J. 8. Tennant); Nos. 17 and 36. Forma brevipila : Auckland Islands (J. 8. Tennant) ; Nos, 224 and 29. DITRICHACEAE. DirricHum, Timm., 1788. Distrib.Subcosmopolitan. Ditrichum affine, C. Mill. Auckland Island (Dr. Cockayne). No. 819. Ditrichum strictum, Hook. f. et Wils. Auckland Islands (J. 8. Tennant). Nos. 26 and 27. SPLACHNACEHAE. Taytorta, Hook., 1816. Distrib.Scattered on both hemispheres. Tayloria purpurascens, Hook. f. et Wilson. Auckland Island (Dr. Cockayne). No. 811. 536 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Cryptogams. BRYACEAE. Leprostomum, R. Br., 1811. Distrib.South Asia, Polynesia, east Australia, New Zealand; one species im Chili. Leptostomum inclinans, R. Br. Antipodes Island (Dr. Cockayne). No. 833. Leptostomum gracile, R. Br. Auckland Islands (J. 8. Tennant). No. 5. Ruizoconium, Brid., 1827. Distrib.rWarmer countries ; not represented in Hurope. Rhizogonium bifarium, Hook. Auckland Islands (J.S.Tennant). No. 8. Bartramia, Hedw., 1789. Distrib. Cosmopolitan. Bartramia robusta, Hook.f. et Wils. Auckland Islands (J. 8. Tennant). No. 7. BREUTELIA, Schimp., 1856. Distrib.rRare on the Northern Hemisphere; more common on the tropical mountains and in the Southern Hemisphere. Breutelia pendula, Hook. | Auckland Islands (J. 8. Tennant); No. 10. Campbell Island (R. M. Laing). Breutelia comosa, Mitt. Auckland Islands (J. 8. Tennant). Nos. 19 and 33. PrycHomnium, Hook. f. et Wils. Distrib.— Ptychomnium aciculare, Brid. Auckland Islands (J. 8. Tennant); No. 20. Campbell Island (R. M. Laing). Auckland Island (Dr. Cockayne) ; No. 85. . = Cryptogams. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 537 LEMBOPHYLLACEHAE. CAMPTOCHAETE, Reichardt, 1870. Distrib.—Only on the Southern Hemisphere—Melanesia, eastern Australia, Tas- mania, New Zealand, Chili. Camptochaete deflexa (Wils.), Jacq. Campbell Island (R. M. Laing). LEMBOPHYLLUM, Lindb., 1872. Distrib.Kastern Australia, Tasmania, New Zealand. Lembophyllum cochlearifolium (Schwaegr.), Lindb. Auckland Island (Dr. Cockayne). No. 810. HOOKERIACEAE. DIsTICHOPHYLLUM, Doz. et Molk., 1846. DIstrib.—Most species in south-eastern Asia, Polynesia, eastern Australia, New Zealand, Fuegia. Distichophyllum pulchellum, Hook. f. et Wils. Auckland Islands (J. S. Tennant). No. 21. HY POPTERYGIACEAE. CYATHOPHORUM, Palis., 1805. Distrib.—From eastern and southern Asia, through Malaysia, to eastern Australia and New Zealand. Cyathophorum bulbosum (Hedw.), C. Miill. Auckland Islands (J. 8. Tennant). No. 51. HYPNACEAE. CaMPYLIUM (Sull.), Bryha., 1893. Distrib. More temperate regions; the nearest relatives of the New Zealand species in eastern Australia. Campylium relaxum, Hook. f. et Wils. Auckland Jslands (J. 8. Tennant). No. 4. 538 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Cryptogams. SrEREODON (Brid.), Mitt., 1869. Distrib.Through the more temperate parts of the globe. Stereodon chrysogaster, C. Mill. Auckland Islands (J. S. Tennant); No. 41. Auckland Island (Dr. Cockayne) ; No. 87. AcrocLapium, Mitt., 1869. Mstrib.— Acrocladium chlamydophyllum, Hook f. et Wils. Auckland Islands (J. S. Tennant); No. 22. Campbell Island (R. M. Laing). MnriopEnprRoNn, Lindb. Dstrib.— Mniodendron comosum, Labill. Auckland Islands (J. S. Tennant). No. 6. Mniodendron Sieberi, C. Mill. Auckland Islands (J. 8. Tennant). No. 32. Mniodendron Kroneanum, C. Mill. Auckland Islands (Dr. Cockayne). No. 89. ARTICLE XXIV.—TWO SPONGES FROM CAMPBELL ISLAND. By H. B. Kirx, M.A., Professor of Biology, Victoria College, Wellington. PLATE XXY. LEUCOSOLENIA, Bowerbank, 1862. Leucosolenia echinata. Leucosolenia echinata, Kirk, Trans. N.Z. Inst., xxvi, 177. This sponge was found between the tide-marks in Perseverance Harbour, Campbell Island. The sponge has not been reported outside New Zealand. STYLOHALINA, Nn. gen. Halichondrine sponges, in which the skeleton consists entirely of styll. Stylohalina conica, n. sp. (Plate XXV.) Sponge forming flattened or conical masses, each cone having at its apex a single large oscule. The surface of the sponge shows numerous minute elevations, between which lie definite pore-areas, irregular in outline. In each area there are from 5 to 20, or more, small round or oval pores. At times several pore-areas are confluent. The dermal membrane is glossy, and of close texture, but quite thin. The surface is echinated by projecting spicules. Diameter of one of the two speci- mens, 2°5cm.; height from base of sponge to summit of oscule-bearing cone, i cm. The other specimen has a diameter of 4cm., and is less regular in shape. Colour, yellowish-white. Skeleton entirely of styli (figs. 2, 3). The basal end of the spicule is usually evenly rounded, and from this end for about two-thirds of the length of the spicule there is little or no tapering; then the spicule tapers shghtly, and presently more noticeably, terminating in a sharp point. Occasionally there is a slight approach to the tylostylote condition, a slight swelling developing close to the base. This swelling does not always extend right round the spicule, but sometimes suggests a short, blunt, feebly developed spinule. All, or nearly all, the spicules are very slightly curved, and usually, near the apex, there is a still slighter curve in the opposite direction. Thus the curvature of the spicule is, on the whole, slightly sigmoid. Average size of spicules 0°36 mm. by 0-01 mm. In the dermal membrane the spicules are numerous on the papillae, where many spicules have the apical portion projecting at almost any angle. A few spicules lie parallel to the surface. In the pore-areas there are fewer spicules, the membrane being very thin. In the body of the sponge spicules are not numerous, and are usually scattered irregularly ; sometimes arranged in small irregular groups, 540 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Sponges. or making some approach to forming spiculo-fibres. In some of the larger inhalent canals, particularly those that terminate under the papillae, the wall is strengthened by spicules, with the apical ray directed towards the surface of the sponge. Spicules seldom project into the canals or subdermal spaces; never into the small spherical flagellated chambers. From these cavities run branching inhalent canals—sometimes one, sometimes many—starting from a single cavity. marks. appear to me to be almost entirely with Halichondria ; but Ridley and Dendy’s classification (1), which I prefer to follow as nearly as may be, expressly limits the family Homorrhaphidae to sponges whose spicules are oxea or strongyla. It seems to me impossible to refer this sponge to the Suberitedae. In von Lendenield’s classi- fication also (2) the sponge would have to be placed in the Suberitidae. i the genus Stylohalina becomes recognised, it will involve the founding of a new family, Stylo- halinidae, for Halichondrine sponges without spongin and with a skeleton consisting of styl only. men. I cannot be sure that the Campbell Island sponge is not identical with it. me wh — . “* Challenger’ Reports.” Monaxonida. Ridley and Dendy. . “Sponges in the Australian Museum.” Von Lendenfeld. 1888. . “ Monograph of the British Spongiadae.” Bowerbank. 1864. . Contribution to a Knowledge of the New Zealand Sponges. Kirk. ‘Trans. N.Z. Under the pore-areas there lie subdermal cavities varying in size and shape. Sponge found in Perseverance Harbour, Campbell Island, between the tide- It is with reluctance that I propose a new genus for this sponge. Its affinities Bowerbank’s Halichondria simplex(3) was described from a single dried speci- WORKS REFERRED TO IN TEXT. Inst., vol. xxvi. DESCRIPTION OF PLATE XXV. Stylohalina conca, un. sp. Fig. 1. Section at right angles to surface. Figs. 2, 3. Spicules. 54] Sponges. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. Ni N S s) ao ‘A eal NG SG ele ——S=—_ Y ® 5S | Se) i) NGS) eG 0-05mm.” PLATE XXV. ARTICLE XXV.—VERTEBRATA OF THE SUBANTARCTIC ISLANDS OF NEW ZKALAND. By Epear R. Warts, F.L.S., Curator, Canterbury Museum, Christchurch. TuoucH the southern islands of New Zealand have been occasionally visited by members of scientific expeditions, and frequently by persons interested in animal- life, it cannot be said that the vertebrate fauna is even yet adequately known. The mammalia, represented by species of cetaceans, by two kinds of resident seals, and occasional visitors or stragglers of the order, are tolerably known, but their seasonal or accidental occurrences need further investigation. It would appear at first sight that the birds have received a full share of atten- tion: this may be somewhat true of certain of the land-birds, but even a casual visit will convince the ornithologist that very much remains to be done to elucidate ‘| the avifauna of the several groups of islands. The species and habits of the penguins ; are easily studied, and these birds are tolerably well known, as are also, though m : a less degree, the albatroses. The smaller species—mollymawks, or mollyhawks—are ; still little known, and, though these birds breed in immense colonies, we are not aware of the nesting-grounds of all the different species, for many of the islets have not yet been visited by the zoologist. The same is even more true of the various petrels which surround the islands in myriads and resort to the land for breeding purposes. The number of species which nest on these southern lands must be very ereat indeed. There are no reptiles on the islands. I have note of thirteen species of fishes recorded from the islands, two of which are fresh-water forms. It may confidently be predicted that, when the deeper waters off the islands are netted, additional species will be found, though it is improbable that the fish fauna 1s anything but a poor one. I am indebted to Captain J. Bollons, of the N.Z. Government steamer ‘““ Hinemoa,” for the several notes associated with his name. SR ee ee ee MAMMALIA. Order CARNIVORA. Fam. OTARIIDAE. ARCTOCEPHALUS, Cuvier, 1824. Arctocephalus hookeri, Gray. (Sea-lion, sea-bear.) Arctocephalus hookeri, Gray, Brit. Mus. Cat. Seals & Whales, 1866, p. 53. These seals are quite numerous at the Snares, Auckland, and Campbell Islands, . and at certain seasons are to be seen at every beach or other part affording a : Mammalia.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 543 landing. This circumstance is due to the fact that the under-fur as found in the fur- seals 1s not developed, and the species is therefore a “ hair-seal,” and, being regarded of small commercial value, has not been subjected to the reducing influences which have well-nigh exterminated the fur-seal. As a colony was approached by a boat-party the seals shuffled about uneasily, but, as the men landed, the animals showed more forcible objection. The females, or sea-bears, generally made off, but the larger shaggy males stood their ground until actually beaten off, grunting, barking, and biting in a very ferocious manner. Their Fic. 1.—S8s-nions (dretocephalus hookeri), KNDERBY IstaNp, AucKLAND IsLANpDs. onslaught consists of a rapid shuffle for a few yards, accompanied by a roaring or barking, which is sufficiently alarming; but ordinarily there is small difficulty in avoiding them, and there is no evidence on the part of the seals to give any lengthened chase. Most of them took to the sea, but others retreated beneath the Olearia trees, where the ground was so soft as to make further pursuit after an angry bull too dangerous to be entertained, to say nothing of the filthy and odorous condition of these tracks, used by generations of seals. At the Snares we came across a blind bull, and it was pitiable to see it wildly charging in all directions; at one time it came into violent contact with a tree, and finally rushed over a bank falling heavily on to the ground beneath. Every one wondered how it was able to exist. 544 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Mammalia. On the occasion of my former visit, in February, 1907, the females had retired far into the bush, often to considerable heights, and we frequently came across them long distances from the shore. The breeding season was evidently close at hand, and we found a mother on the beach with her cub. The sailors of the “ Tutanekai ” attempted to secure the young one, but it was so determinedly defended by the mother that only the overwhelming number of men, who by diverting her attention first in this direction, then in that, enabled them to carry off the baby. The mother was driven into the waves, but kept close to the shore. On Fie. 2.—Swa-nion (drctocephalus hookeri), THE SNARES. being liberated, the young one attempted to hide itself under a rock, and, as we moved away, the mother came ashore to her offspring. During my later visit, in November, the females had not secluded themselves; the young are evidently, therefore, born early in the year. Hutton states that the seal breeds in February. The male, or sea-lion, is a large bulky animal of dark-brown colour, and the old bulls have a thick mass of long hair, disposed mane-like over the neck and shoulders. The female, or sea-bear, is smaller and sleeker, and grey in colour. She is not so ferocious as the male, unless she has a young one by her side, when she ae eh a OO LL <—“«—-< a Mammalia.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 545 defends it with true maternal instinct. The young of both sexes are similar to the female in colour. On land the seal is clumsy and slow, and the hind limbs appear to be unneces- sarily joined together, so that little independent movement is possible ; a waddling gait is therefore produced. In the water, however, the seal is active and graceful, and will glide past one just beneath the surface of the water for a long distance without the shghtest movement of its limbs after its initial impetus. Fic. 3.—Youne Sra-nion (dretocephalus hookeri), CannnEy Harpour, AucKLAND ISLANDS. The seals are very inquisitive, and will closely follow a boat until a blow from an oar speeds them away without protest. As my quest for fishes led me to spend much time alone on the shore, I had especial opportunities of watching the seals. I have seen them, unaware of my presence, systematically searching the crannies of rocks, much after the manner of a dog hunting for food. The stomach of the animal examined at Enderby Island contained remains of octopi and fish.* They were possibly taken among the rocks as above indicated. As mentioned in my article on fishes, the rock-living fishes are extremely wary, doubtless owing to this habit of the seals. * Captain Bollons tells me that he nearly always finds remains of the octopus in their stomachs. 36—S. 546 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Mammatia.— Two seals rose together in Carnley Harbour at a considerable distance from the beach. One held what appeared to be a large red-coloured fish. A lively struggle ensued for half a minute, when the possessor of the prize managed to swallow its prey. “Tate, when in a hurry, a seal swims under the water, it can throw its huge body quite into the air, but more frequently rolls over, precisely after the manner of a porpoise; and it is interesting to note that the penguins behave in a similar manner when making rapid progress. Fié. 4.—SEA-BEAR AND Cus (dretocephalus hookeri), KNprRBY IsuANnD, AUCKLAND IsnaNnps. The presence of a seal is usually indicated by the noise made when it reaches the surface for breath, but often I have been startled by one raising its head out of the water close beside me. If the seal has been similarly surprised at one’s presence its curiosity will cause it to linger and take stock for some time; occasionally a bull will give vocal demonstration disapproving of an intruder’s presence. Fishing from a rock surrounded with deep water, I was once considerably scared by a bull seal rismg within a few yards of me, and barking in a very savage manner. He lessened the distance to 6ft., then, jumping half out of the water, made a dash towards me, and I nearly stepped backwards off the rock into deep water. I am Mammalia.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 547 confident, however, that the seal merely intended to frighten me, and had no desire to take a piece out of my leg. However, I wound up my line, and moved to a less- exposed situation. It is only necessary to throw a stick or a stone at a seal when in the water, and, though he may not be hit, he at once makes off. eS le poe’ Fie. 5.—Srau-Track, Maskep Isntanp, AUCKLAND ISLANDS. Seals appear to haunt particular areas of the coast, and one old bull lived in a quiet bay near our camp, where he could be seen at almost any time. He was never disturbed by our party, and I am confident that he would have become quite tame in time, for he was much more tolerant of our presence after an acquaintance of a few days. 548 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Mammalia. Beddard* has published an interesting paper on this seal, and I have followed him in retaining the name Arctocephalus for the genus. Filholy records isolated individuals of Otaria cinerea, Peron, from Campbell Island, but this species is not certainly known from New Zealand seas. Hab.—Snares, Auckland, and Campbell Islands. Arctocephalus forsteri, Lesson. (Fur-seal.) Otaria forsteri, Lesson, Dict. Class. Hist. Nat., xim, 1828, p. 421. This animal was at one time hunted almost out of existence. “It has been calculated that in 1824 ten vessels touched on the New Zealand coast and at the islands, and took away seventy or eighty thousand skins, forty or fifty thousand going to Sydney, and the remainder to England. In those days a seal-skin in Sydney was worth about 15s. It is recorded that the industry was carried on so assiduously, and the south-west portions of the coast were hunted so industriously by sealers, who killed the females and their young for food, that there was fear of the seal in these parts becoming extinct. Two years later a vessel spent six months in a cruise searching for new sealing-grounds, and took back only 449 skins. Stewart Island was a specially favoured spot for sealers, and many of them also established stations in Dusky Sound and in bays and inlets on the coast.’ It is pleasing to be able to report that during recent years the numbers of the fur-seal have increased, and if the poaching, which undoubtedly takes place, can be stopped, there 1s no reason why the animal should not again people its old haunts. When steaming past the Bounties in 1907 one or two seals were noticed on the rocks, and Captain Bollons tells me that there is now quite a small colony of them. A smaller number are to be found at the Snares, and a few individuals occur at places on the west coast of the South Island, notably at Steeple Rock and The Cascades, while a few others are occasionally met with in the Sounds. | | Hab.—Australian and New Zealand seas. Pele eee hrs, Fam. PHOCIDAE. MacrRoRHINUS, Cuvier, 1824. Macrorhinus leoninus, Linnaeus. (Sea-elephant.) Phoca leonina, Linnaeus, Syst. Nat., ed. x, 1758, p. 37. Writing on the mammals of the National Antarctic (“‘ Discovery ”) Expedition, Dr. Edward A. Wilson makes some interesting and valuable observations on the sea-elephant, and points out that in its general bearing and life-history it approaches the Otarudae, but that its osteological characters, as described by Flower, ally it with the Phocidae. Respecting its distribution, Wilson§ writes, “ The Macquarie Islands have long been known as a stronghold of this seal, but it has from time to time been reported as equally abundant in the Kerguelen, Marion, Heard, and Crozet * Beddard, Trans. Zool. Soc., xu, 1890, p. 369. + Filhol, ‘‘ Mission de I’ Ile Campbell,” Zool., 1885, p. 28. + Hutton and Drummond, “ Animals of New Zealand,” 1904, p. 39. § Wilson, Nat. Ant. Exped. Zool., i, 1907, p. 53. Mammalia.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 549 Islands. It is said not to come so far north as the Campbell and Auckland Islands in the direction of New Zealand, although in other directions it has been taken at Tristan da Cunha, Juan Fernandez, the Falkland Islands, South Georgia, and Inac- cessible Islands; also from the Cape of Good Hope and the ‘ Antarctic seas.’ It is therefore a seal of wide distribution, occurring abundantly in South Indian and South Pacific Oceans, and wandering as much to the north as to the south.” The sea-elephant has occurred at Campbell Island, for Mr. Nicholson, who re- sides there, told members of our expedition how he had secured one at Perse- verance Harbour, and exhibited the skull, which he had preserved. Filhol* also records having obtained two individuals there, Lonnberet considers that Mirownga, Gray, 1827, is the correct name of the genus. Hab.—Campbell Island, and as above noted. Hyprurea, Gistel, 1848. Hydrurga leptonyx, Blainville. (Sea-leopard, leopard seal.) Phoca leptonyx, Blainville, Journ. Phys., xci, 1820, p. 288. Though, properly speaking, this seal is a southern form, it wanders as far north as the shores of Australia, generally in pairs. In 1907 I picked up the skull of a leopard seal in Venus Cove, at Campbell Island, and members of our expedition saw an example near the same place, and photographed it. Trouessart{ points out that Hydrurga, Gistel, 1848, has priority over Ogmo- rhinus, Peters, 1875. Hab.—Southern seas. Order CE TACEA. Fam. BALAENIDAE. BaLAENA, Linnaeus, 1758. Balaena australis, Desmoulins. (Southern right whale.) Balaena australis, Desmoulins, Dict. Class. Hist. Nat., 11, 1822, p. 161. The whaling industry, at one time prosecuted off the coasts of New Zealand from the Chatham to the Campbell Islands with some vigour, has entirely ceased, but I understand that, as whales are becoming more plentiful again, the industry may in a slighter degree be renewed. A party recently visited Campbell Island for the purpose of prospecting the waters of the neighbourhood, as a result of which “a syndicate is being formed to establish a depot at Campbell Island to hunt the right whale, and it is hoped to be able to commence operations at the beginning of next year’s season. The intention is to fit out a whaling-steamer with a heavy bow gun, firing an explosive shell, as at the Norwegian stations.”§ Though whales * Filhol, loc. cit., p. 30. t Lonnberg, K. Svensk. Vet. Akad. Handl., xl, 1906, p. 9. { Trouessart, Mammalia, Expéd. antarct. frangaise, 1907, p. 20. § News item, 16th April, 1909. 550 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Mammalia. are nearly always seen during the southern cruises of the Government vessels, it is doubtful if they are sufficiently numerous to pay for the hunting of them. Hab.—Southern seas. BaLaAENOPTERA, Lacépéde, 1804. Balaenoptera musculus, Linnaeus. (Blue whale.) Balaena musculus, Linnaeus, Syst. Nat., ed. x, 1758, p. 76. Though more commonly known under the name of B. sebbaldw, Gray, I have followed Mr. F. W. True* in considering this as a synonym of B. musculus. Wilsont regards the blue whale as one of the commonest species in Antarctica, and writes, ‘“‘ We saw a pair of these whales in Table Bay on our voyage out; again, a pair off the coast of New Zealand, but nowhere did we see them in numbers till we reached the ice.” I may here refer to the large example stranded near Okarito, on the west coast of the South Island of New Zealand, in February, 1908, the skeleton of which is now in the Canterbury Museum. A short notice, together with an illustration of the tail-flukes, appeared in Nature,{ but the following particulars may be of interest: The total length of the whale was 87 {ft., that of the skull 21 ft.; the direct length of the mandible is 20 it. 8in., and that of the hand 12 ft. The width of the tail- flukes together was 21 it. The total weight of the skeleton is estimated at 9 tons, while the ascertained weight of the cranium, exclusive of the maxillae, &c., is 1 ton 9cwt. The mandibles together weigh 1 ton 5 cwt. When I first arrived’ at the stranded whale I found that some men had com- menced to remove the blubber, and the exuding oil had saturated the sand all around. The effect of this was very noticeable on the would-be breakers—far from the vicinity of the whale they were smoothed down ; and an attempt to wash the hands in the sea-water only resulted in adding a little more oil to them. The whale was a cow, and the characteristic furrowed throat and chest pre- sented an appearance not unlike corrugated iron, the folds having become some- what lax, and having lost their sharp edges. A noticeable feature of the dead whale was the number of pits in certain portions of the skin; these were each the size of a crown-piece and $in. indepth. They marked the positions of the host of barnacles, which had dropped off ere the animal had been washed ashore. With the exception of the right maxilla, which it is hoped may still be recovered, the whole of the bones were secured, including the vestiges of the pelvis, &c., the chevrons, the digital and the terminal tail-bones. There are sixty-four vertebrae, as usual with the species; but there has been some difference of opinion as to the number of ribs possessed by the blue whale, some authors giving fifteen, others sixteen pairs. While our example has the lower number, it is significant that the transverse process of the sixteenth dorsal vertebra has an articular surface, so that if we did not know that all the ribs were secured and preserved we should, by examination of the vertebrae alone, pro- nounce the number of ribs to be sixteen pairs, Hab.—Axctic, temperate, and Antarctic seas. * True, Proc. U.S. Nat. Mus., xxi, 1898, p. 629. + Wilson, loc. cit., p. 4. + Waite, Nature, 26th November, 1908, p. 98. Aves.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 551 Fam. PHYSETERIDAE. PHYSETER, Linnaeus, 1758. Physeter macrocephalus, Linnaeus. (Sperm whale.) Physeter macrocephalus, Linnaeus, Syst. Nat., ed. x, 1758, p. 76. This whale is an intertropical species, and, though it ranges into temperate seas, 1t is unknown in colder areas. It is known to occur off the coast of New Zealand, and Captain Bollons informs me that he has seen it as far south as Campbell Island. He has never met with the killer (Orca) nor any of the smaller dolphins in the neighbourhood of the southern islands. Hab.—As above noted. Fam. DELPHINIDAE. GLOBIOCEPHALUS, Lesson, 1842. ? Globiocephalus melas, Traill. (Blackfish.) Delphinus melas, Traill, Nicholson’s Journ., xxii, 1809, p. 21. When members of the Transit of Venus Expedition arrived at Perseverance Harbour, Campbell Island, in 1874, they found the remains of two cetaceans ; these were identified by Wilhol* as Physeter macrocephalus, and Globiocephalus macro- rhynchus, Gray. The latter was more probably G. melas, well known from New Zealand waters. In February, 1907, I saw hundreds of bodies of blackfishes lying on the beach at Petrie Bay, Chatham Islands, where they had been stranded some months pre- viously. AEDS: The following account of the birds is a very inadequate one, principally for the following reasons: While most of the subjects to be studied had been allotted to different members of the expedition, the arrangements made in respect to the birds had fallen through; it was not, therefore, until after our return that I was asked to write the account. While on the islands I did not pay the special attention to the birds I would have done had I known what would later have been required of me, nor—and what was almost equally important—had I acquainted myself with the literature of the subject beforehand. It was, further, one of the rules of the expedition that neither birds nor their eggs were to be taken; consequently, I have little material for purposes of identification or description. The report, therefore, consists of an enumeration of the species recorded from the islands, together with some notes made on birds which attracted my attention. Some of these notes, more especially those dealing with the avifauna of islands other than the Aucklands, were made in February, 1907, when, by courtesy of His Excellency the Governor, J had the privilege of accompanying him on a tour through the southern islands, * Wilhol, loc. cit., p. 34. 552 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Aves. embracing the Snares, Aucklands, Campbell, Antipodes, and Chathams. In June, 1907, I visited Stewart Island; and the followmg somewhat uneven accounts are compiled from my notes made on the several occasions referred to. As Stewart Island was regarded as outside the scope of the expedition, such birds as Octhodromus obscurus, Gmelin; Herodias tumoriensis, Lesson; Catharacta maccormicki, Saunders ; and Phalacrocorax stewart, Buller (all recorded from Stewart Island), are not included in the following list. Order PASS HIR[ES. Fam. SYLVIIDAE. PETROICA, Swainson, 1832. Petroica macrocephala, Gmelin. (Yellow-breasted tit.) Parus macrocephalus, Gmelin, Syst. Nat., ed. xiii, 1, 1788, p. 1013. , Though this bird was everywhere met with in the Auckland Island bush, it can scarcely be said to be common. Found in the South Island and the Chathams, it has not been recognised from the Snares. Hab.—South Island of New Zealand, Chatham Islands, and Auckland Islands. Miro, Lesson, 1831. Miro dannefordi, Rothschild. (Snares robin.) Miro dannefordi, Rothschild, Nov. Zool., 1, 1894, p. 688. This species was quite common at the Snares, and was met with everywhere among scrub and timber. A nest of the species was found in the scrub by a member of the ship’s crew. It contained four eggs; : : > Fic. 6.—Haee or Miro dannefordi, they are pale cream in colour, thickly spotted with Towiaecinclal. small purplish markings, principally at the larger end. Tye dimensions taken by Mr. G. R. Marriner were—length, 0-9 in.; breadth, 0-65 in. Hab.—Snares. Fam. TIMELIIDAE. Bown LeriA, Rothschild, 1896. Bowdleria fulva, Gray. (Hern-bird.) Sphenoeacus fulvus, Gray, Ibis, 1862, p. 221. This bird is common at the Snares. It runs along the ground, and rapidly crosses open spaces to the cover of tussock or other shelter. It is comparatively fearless of man, and will pass close to one’s feet; if alarmed, it flies heavily for a short distance, and settles either on the ground or on a low bush. The fern-bird builds in the heart of a fern, and Mr. G. A. Buddle, a passenger on the “ Hinemoa,” found a nest in such situation at the Snares. Captain Bollons recently found a Aves.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 553 number of the birds on Taumaki Islets, off the Haast River, on the west coast of the South Island of New Zealand. Buller* distinguishes the Snares Island bird under the name Sphenoeacus caudatus. Hab.—Snares ; South Island of New Zealand. Fam. MOTACILLIDAE. Antuus, Bechstein, 1802. Anthus steindachneri, Reischek. (Antipodes Island pipit.) Anthus steondachneri, Reischek, Trans. N.Z. Inst., xxi, 1889, p. 388. The pipits are extremely tame, and run almost at one’s feet. I found them to be most numerous on the beach, feeding probably upon Amphipoda and other small creatures peopling the broad Durvillaea which is washed ever to and fro by the waves. They also ran among the penguins, apparently for the sake of the flies which rise in clouds from the guano of these birds. The pipits paid such small heed to my presence that I had no difficulty in catching them in a hand-net. Hab.—Antipodes Island. Anthus novae-zealandiae, Gmelin. (Auckland Island pipit.) Alauda novae-zealandiae, Gmelin, Syst. Nat., ed. xiii, 1, 1788, p. 799. It is unlikely that the habrts of this bird differ from those of A. stecndachneri, but I certainly noticed its absence from the beaches, while 1t was numerously met with on the higher lands. It frequented the deserted camps of the survivors of the “ Dundonald,” at Disappomtment Island, in large numbers, being attracted by the insect-life which gathered around the remains of dead birds, &c., left by the sailors. Mr. F. R. Chapman, writing in 1891, records this bird from the Snares; a pipit, presumably this species, was also seen there by members of our party. Bullert would regard A. aucklandicus, Gray, as distinct from A. novae-zealandiae. Hab.—New Zealand, Snares, Chatham and Auckland Islands. Fam. MELIPHAGIDAE. ZostERopS, Vigors and Horsfield, 1825. Zosterops caerulescens, Latham. (Blight-bird, white-eye, silver-eye.) Certhia caerulescens, Latham, Ind. Orn., Supp., 1801, p. 38. Originally self-introduced into New Zealand from Australia, this bird has not only spread throughout the greater part of the Dominion, but has also extended to the outlying islands. Hab.—Australia, New Zealand, Snares, Auckland, Campbell, and Chatham Islands. * Buller, Trans. N.Z. Inst., xxvu, 1895, p. 128. 7 Buller, Supp. Birds of N.Z., u, 1905, p. 153. 554 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Aves. PROSTHEMADERA, Gray,’ 1840. Prosthemadera novae-zealandiae, Gmelin. (Tui, parson-bird.) Merops novae-zealandiae, Gmelin, Syst. Nat., ed. xin, i, 1788, p. 799. Tuis are really common at the Auckland Islands, and much more confiding than on the mainland ; as they fly within a few yards of one’s head their wings make a distinctly harsh noise. Many times I watched individuals feeding, after the manner of a fly-catcher. The bird would choose the dead limb of tree, make a short erratic flight, catch an insect, and return to the same branch time after time. Hab.—New Zealand, Chatham and Auckland Islands. ANTHORNIS, Gray, 1840. Anthornis melanurus, Sparrman. (Bell-bird.) Certhia melanura, Sparrman, Mus. Carls., 1786, pl. v. Considering how rare this bird is becoming on the mainland, it is most pleasurable to find it really common at the Auckland Islands. Its tameness there is equal to that of the other birds unmolested by man and other enemies. Its wings make a harsh sound as it flits from bush to bush, and at the time of our visit 1t was feeding largely on the flowers of Dracophyllum, then in bloom. The birds also searched the bark of trees for insects, ascending much after the manner of a creeper, but they pick off the insects with the sticky tongue. The ordinary note is a single toll, as on a high-pitched sweet-toned gong. The bird also makes a series of sounds like the whirr of a miniature watchman’s rattle. Another common call is :— I can give the intervals ‘ y ka Tic. 22.—Grovur or SHacs (Phalacrocorax colensoi), AucKLAND IsLANDS. broad black band across the chest. Both species when immature are said to agree in this respect. One of the most striking features of the voyage was presented during our pas- sage down the Auckland Island coast. In the offing the sea was black in patches with shags, and by aid of glasses or when sufficiently near to such a body we noticed that considerable commotion was taking place. The birds were evidently feeding ; their food, whatever its nature, being scattered over wide areas, but evidently closely 582 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Aves. packed where it occurred. At the time I considered the birds were catching fish, but later changed my views. It is quite impossible to convey any idea of the num- ber of birds thus engaged. In passing from the shore to their feeding-grounds, or in the reverse direction, their flight behind, over, and in front of the vessel was a con- stant stream, and this continued for many miles. ‘The birds returning to shore were noticed to be heavy with food; their bodies were fully distended, and their flight, in consequence, distinctly laboured. The addition of several ounces of food to the weight of so delicately adjusted an organism as a flying-bird must have an appreciable effect on its flight, as, indeed, we had ample evidence in the case of the gorged shags. As the “Hinemoa”’ drew towards the shore we fell in with the satiated birds sitting on the water. A flat stretch of rock below our camp at the Auckland Islands proved to be a favourite assembling-ground for shags, and we had no trouble in thus okserving them. The accompanying photographs were taken there. The birds were always accompanied by gulls. We secured several on different occa- sions, and caged three individuals for study, two of which laid eggs. Being anxious to ascertain what the birds were feeding upon, I examined the stomachs of others, and found them to be crowded with small crustaceans resembling Mu- mda: one contained small fish-bones in addition. Rowing up North Arm, we landed on Figure of Hight Island, and came across a very large mob of shags, and charged them like schoolboys. While scores got past us, we forced others across a tussocky patch to the rocks Fig. 23.—Aucxranp Isutanp SHAG AT CLOSE RANGE. beyond, whence they took to wing. The birds scarcely attempted to fly until they reached the edge of the rocks, but ran at considerable speed. All, with the exception of six specimens, suc- ceeded in escaping. I examined the stomachs of our captures, and in every case found them to contain the reddish mass which I have above attributed to >. paral digested Munda. Since the foregoing was written, Professor Benham: writes to me, “The shags that were opened had the stomach filled val the crustacean Munida sub- PUGOSH. Hab.—Auckland Islands. @ Aves.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 583 Phalacrocorax ranfurlyi, Ogilvie-Grant. (Bounty Island shag.) Phalacrocorax ranfurlyi, Ogilvie-Grant, Bull. Brit. Orn. Club, xi, 1901, p. 66. I sailed past the Bounty Islands in February, 1907, without landing. This is the latest addition to the avifauna of the New Zealand region. Hab.—Bounty Islands. Order LAMELLIROSTRES. Fam. ANATIDAE. Anas, Linnaeus, 1758. Anas superciliosa, Gmelin. (Grey duck.) Anas superciliosa, Gmelin, Syst. Nat., ed. xi, 1, 1788, p. 537. Though recorded from both the Auckland and Campbell Islands, the bird was not, so far as I learned, seen by any member of our expedition. Hab.—Java, Australia, Polynesia, New Zealand; Auckland, Campbell, and Chatham Islands. Nesonetta, Gray, 1844. Nesonetta aucklandica, Gray. (Flightless duck.) Nesonetta aucklandica, Gray, Gen. Birds, ii, 1844, p. 627, pl. clxix, fig. 4. Dr. L. Cockayne having been commissioned by the Government to obtain living examples of the flightless duck for transference to the bird-sanctuary on Kapiti Island, a special excursion was made to Hwing Island (Auckland Group) for the pur- pose of securing examples. We found these birds also on Disappointment Island (Auckland Group). They frequent the kelp on the coast, but were also found high up the hillsides in the watercourses. The cages into which the ducks were placed chanced to have been furnished with perches, and it was found that the birds readily availed themselves of these, perching almost as well as a true passerene. The birds were safely conveyed to the sanctuary in Cook Strait, and at latest advices were doing well. Though usually spoken of as “ flightless,” Captain Bollons tells me that these ducks are able to fly for short distances, and, as a matter of fact, they reach their nesting-sites by this means. The eggs are laid in holes 15 ft. or 20 ft. above sea- level, excavations made by petrels being possibly utilised. As these holes are some- times made in the face of a cliff, a mere walking-bird would be unable to reach them. Captain Bollons informs me that he has often seen the ducks rise at the foot of the cliff and by aid of the wings alone plump straight into the holes. Even by means of a ladder he himself has been unable to reach the nests. Hab.—Auckland Islands. 584 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ dves. Ham. MERGANETTIDAE. HYMENOLAIMUS, Gray, 1843. Hymenolaimus malacorhynchus, Gmelin. (Blue duck.) Anas malacorhynchos, Gmelin, Syst. Nat., ed. xi, 1, 1788, p. 526. The authority for the inclusion of this species in the Auckland Island fauna is Captain J. Bollons, of the “ Hinemoa,” a keen and careful observer. In February, 1883, he saw five blue ducks together at Hanfield Inlet, on the east coast, but has not seen a single example since, and no member of our expedition was more for- tunate. _ Hab.—New Zealand and Auckland Islands. Fam. MERGIDAE. MERGANSER, Brisson, 1760. Merganser australis, Hombron and Jacquinot. (Southern merganser.) Mergus australis, Hombron and Jacquinot, Ann. Sci. Nat., 1841, p. 320. A sharp look-out was kept along the shores of the Auckland Islands for this bird, but it was not recorded as having been seen. This is less remarkable when, as Captain Bollons later told me, he had not seen the bird on the coast, but had found it occasionally at the head of the estuaries, and especially on the inland watercourses. As mentioned in my report on the fishes, the supposed littoral habit of the bird led the late Captain F. W. Hutton to infer that Galaazias bollansi was a marine species, a supposition I have shown to be erroneous. Hab.—Auckland Islands. BRITISH BIRDS. I saw thrushes (Turdus musicus) and blackbirds (1. merula) at the Snares ; also a nest of the latter containing eggs, Nests and eggs of the blackbird were found at Terror Cove, in the Auckland Islands, A momentary glance of two birds at the Snares led me to think they were linnets (Linota cannabina), and a description supplied by Mr. G. R. Marriner, who watched three birds, supports my belief. j Mr. Marriner was informed that Mr. Nicholson had seen the following birds on Campbell Island: House-sparrow (Passer domesticus), hedge-sparrow (? Accentor modularis), linet, and that flocks of starlings (Sturnus vulgaris) were to be met with on Mount Honey. Mr. Marriner saw birds which he believed to be linnets, but did not see the others mentioned. Dr. C. C. Farr reported having seen two goldfinches (Carduelis elegans) at North Arm, Carnley Harbour, Auckland Islands, and stated that there could be no possible doubt on the matter. I may mention that goldfinches are extremely common throughout New Zealand, and, though I doubt the occurrence of the hedge-accentor on Campbell Island, they, in common with a host of other British birds, are quite numerous in Canterbury, and possibly in other parts also. i | Pisces.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 585 PISCES. Examples of all species of fishes obtained were preserved, and form the basis of the following account. Several forms, however, not secured have been recorded as a result of previous expeditions or of private collectors ; these are also included, but are merely mentioned as constituents of the fauna. Being myself stationed at the Auckland Group, the collection made there is naturally larger than that made at Campbell Island; but there is little doubt, as might be expected, that the Auckland Islands, with their varied coast-line, large harbours, and sheltered bays, support a much richer marine fauna than the com- paratively exposed and little-mdented shores of Campbell Island. The species found at the respective groups are as follows, those marked with an asterisk being obtained : — SNARES. * Notothenia nucrolepidota, Hutton. * Bovichtus varvegatus, Richardson. * Trupterygion varvum, Forster. AUCKLAND ISLANDS. Anguilla aucklandw, Richardson. * Galaxias brevipinnis, Giinther. Galaxias fasciatus, Gray. * Syngnathus blainvillianus, Kydoux and Gervais. Rhombosolea tapirina, Giinther. * Notothenva mucrolepidota, Hutton. Notothenia coruceps, Richardson. Notothenia macrocephala, Ginther. * Notothenia colbecki, Boulenger. * Bovichtus variegatus, Richardson. * Trupterygion varvum, Forster. CAMPBELL ISLAND. Euprotonucrus bispinatus, Quoy and Gaimard. Galaxias attenuatus, Jenyns (= G. campbelli, Sauvage). * Rhombosolea tapirina, Gunther. Notothenia microlepidota, Hutton. Notothenia macrocephala, Gunther. * Notothenia colbecki, Boulenger. Notothenia filholi, Sauvage. Bovichtus variegatus, Richardson. ANTIPODES ISLAND. * Notothenia colbecki, Boulenger. 586 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Pisces. Fam. GALAXTIDAE. GALAXIAS, Cuvier, Régne Anim., ui, 1817, p. 183. Galaxias brevipinnis, Gimnther Galaxias brevipinms, Ginther, Cat. Fish. Brit. Mus., vi, 1866, p. 213. G. grandis, Haast, Trans. N.Z. Inst., v, 1873, p. 278. G. robinsoni, Clarke, Trans. N.Z. Inst., xxxi, 1899, p. 89, pl. v. G. bollansi, Hutton, Trans. N.Z. Inst., xxxiv, 1902, p. 198. In his review of the family, Mr. C. Tate Regan* admits two species as occurring at the Auckland Islands—namely, G. brevipinnis, Giinther, and G. fascratus, Gray. Members of the genus were taken at two places in Carnley Harbour by the expedition. Our Maori crew, at the instance of Dr. Farr, caught a large number at the magnetic station on North Arm, and Professor Benham obtained others near the main camp. In view of the opinion below recorded, I may mention that, though I did not actually discover the haunts of the fishes, I saw them in the positions ind1- cated to me. All the specimens obtained are referable to G. brevipinnis. They were taken in pools in fresh-water streams quite close to the sea, that at North Arm being but a few feet above high-water mark. The specimen of G. bollansi (which, according to Mr. Regan, is but an example of this species) was taken out of the mouth of a specimen of Merganser australis, at the Auckland Islands. From this circumstance Captain Hutton (“On a Marine Galaxias from the Auckland Islands’’)} concluded that the fish was a marine one, and this no doubt largely influenced him in regarding it as distinct. This circumstance induced Dr. G. A. Boulenger to write an article with the following heading: “The explanation of a remarkable case of geographical distribution among fishes.’ He mentions that “most text-books and papers discussing geographical distribution have made much of the range of a genus of small fishes, somewhat resembling trout, the Galaxias, commonly described as true fresh-water forms, which have long been known from the extreme south of South America, New Zealand, Tasmania and Southern Australia.” Dr. Boulenger then refers to the representation of the genus (G. zebratus and G. punctifer) in South Africa, and mentions the fact recorded by Clarke in New Zealand and Vallentin in the Falk- land Islands, that G. attenwatus lives in the sea as well as in fresh water, descending periodically to the sea (in New Zealand) for the purpose of spawning. He adds, “In accordance with these marine habits, this species has a much wider range than any of the others, beg known from Chili, Patagonia, Tierra del Fuego, the Falkland Islands, New Zealand, Tasmania and Southern Australia.” The peculiar distribution of Galaxias is frequently advanced as a strong link in the evidence of land connection with the Antarctic Continent, a link, however, somewhat weakened by the discovery that G. attenuatus breeds in the sea ; its cohesion is apparently still further reduced by the record of a second marine species, as before mentioned ; and Dr. Boulenger concludes his article by remarking, “ It is hoped * Regan, Proc. Zool. Soc., 1905, u, p. 363. t+ Hutton, Trans. N.Z. Inst., xxxiv, 1902, p. 198. ~£ Boulenger, Nature, Ixvii, 1902, p. 84. ~I Pisces.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 58 that, by giving greater publicity to these discoveries, the family Galaxiidae will no longer be included among those strictly confined to fresh waters and that students of the geographical distribution of animals will be furnished with a clue to a problem that has so often been discussed on insufficient data.” There is little fear that anything of importance respecting such an interesting subject as the distribution of the Galaxidae will be overlooked, and, in addition to the article above referred to and quoted, several other allusions to the marine habit of G. bollansi: have already been published. Dr. David 8. Jordan* has given the oreatest possible publicity to the matter by reprinting Dr. Boulenger’s article in his “Study of Fishes,” and in his paper above mentioned Mr. Regan has also referred to Hutton’s statement. I am very far from satisfied that G. brevivinms is to be classed as a marine species, and consider that the evidence is of the weakest nature, and I am surprised that it should have been so generally accepted. The pool above described was quite accessible to gulls and shags, and had I found Galaxias in their digestive tracts. I should certainly not have been surprised. Hutton seems to have taken it for granted that the merganser is essentially a marine bird ; 1f this is so, 1t differs from its Huropean ales. I have seen the goosander (Mergus merganser) in England seventy miles from the sea, and Yarrellt states that these birds “ frequent fresh-water lakes as well as the sea-shore and estuaries.” Little appears to be known of the habits of the southern merganser. Hutton{ states that “it does not frequent the coast and open waters, but only the sheltered harbours.” I have, of course, no evidence to prove that the Galaxias does not occur in the sea; but, on the other hand, I maintain that there is no evidence to show that it does. I am penning my conclusions, though based on negative evidence, to draw attention to what is probably a misreading of a simple incident connected with the natural history of the species. I cannot claim that such conclusions will throw much light on the question of antarctic land connections, for so many species of Galaxias are now known to enter the sea. In addition to G. attenuatus in New Zealand, we have records of G. maculatus in Chili, and this species and G. gracillimus in the Falkland Islands. Some time after the foregoing was written I met Captain Bollons, and told him that I had grave doubts as to the Auckland Island Galazias living in the sea. I asked him if he did not consider it likely that the merganser had obtained the fish from some fresh-water stream close to the sea. ‘‘ But why close to the sea?” said Captain Bollons. “I have never yet seen the bird feeding close to the sea; it is always picking about in the creeks.” I was naturally gratified to find my surmise so unexpectedly supported, and we may now safely dismiss the alleged marine habit of Galaxias brevipinnis (bollansi) as incorrect. The following is a description of one of the examples secured :— eeulieD) Vac DV. O-5 Ve7 P12. Oo ib 42 Vert., 36 25 = 61. _ Length of head, 5:2; height of body, 7-5 ; and length of caudal, 6-0 in the length ; diameter of eye, 5-8; length of snout, 3:4, and interorbital space 2°5 in the head. * Jordan, “ Guide to the Study of Fishes,” ii, 1905, p. 205. 7 Yarrell, “ British Birds,” 3rd ed., i, 1856, p. 399. { Hutton and Drummond, “ Animals of New Zealand,” 1904, p.-321. 588 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Pisces. The head is depressed and broad, its breadth one-fifth greater than its height. The eye lies wholly in its anterior half, ‘and the maxilla extends to beneath the middle of the orbit. The upper jaw overhangs the lower ; the teeth are in a single series, that in the lower jaws with canines. The anterior nostril is in a tube which lies nearer to the eye than to the tip of the snout, the posterior nostril is a little in advance and above anterior margin of the eye. Hight gill-rakers on the lower limb of the anterior arch. The distance between the origin of the dorsal and the end of the caudal is one- half that between the former point and the tip of the snout. The anal arises beneath the middle of the dorsal, and when laid back extends to the base of the hypural. The ventrals are inserted midway between the tip of the snout and the base of the caudal rays. The length of the pectoral is half the distance from its origin to that of the ventral; caudal slightly emarginate ; the length of the peduncle is one- half greater than its depth. Colour.—The colour of the ground is pale olive, which becomes yellowish in fluid. A young one of 60 mm. is colourless, with the exception of some pigmentation along the dorsal and ventral surfaces. Larger examples are strikingly banded with dark grey, the markings being, however, somewhat irregular. The bands break up with increasing size, and a general reticulation results; in old examples a general clouding takes place, but there is always some definition in the hinder part of the body. A dark blotch above the pectoral, partly concealed by the gill-cover, is generally present. The under-suriace is without markings, the pigmentation seen in the young not being maintained. Length of specimen described, 140 mm. As above noted, the branchiostegal rays are eight in number in our examples. Mr. Regan counts them as seven only. Hab.—New Zealand and Auckland Islands. Fam. SYNGNATHIDAE. Synenatuus, Linnaeus, Syst. Nat., ed. x, 1758, p. 336. Syngnathus blainvillianus, Hydoux and Gervais. Syngnathus blaanvillianus, Kydoux and Gervais in Guerin. Mag. Zool., vu, 1837, pl. xvu. Leptonotus blainvillu, Kaup, Cat. Lophob. Brit. Mus., 1856, p. 46. This species has generally been admitted as a member of the New Zealand fauna on the identification of Dr. Ginther,* who wrote, “Of this fish, which hitherto has been found on the west coast of South America only, we have received an example from Cook Strait, New Zealand, through Dr. Hector.” Dr. Kaupt had, however, identified the species in our waters twenty years previously, and had also recorded it from the Auckland Islands, writing, “It is a native of South America, Peru, Chili, Auckland Islands, and New Zealand.” * Giinther, Ann. Mag. Nat. Hist. (4), xvi, 1876, p. 402. { Kaup, swpra, p. 47. Pisces.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 589 In the absence of South American examples for direct comparison, I also pro- visionally associate specimens taken on the expedition with this species. I have not access to the original description and illustration, and though Kaup states that he has “ figured ” the species, and also that he has given a “ portrait ” of it, I find it neither on the plates nor mentioned in the list thereof. The following table shows the comparative measurements of six examples. The great relative depth of the body of the female is most conspicuous, and on this account Kaup made the species the type of his genus Leptonotus, including also the South Australian and Tasmania L. semistiatus, Kaup. —— A B C, D E F | Perec: Come Stee Aline. 8 $ g Total length (in millimetres)... a eee a Al 113 119 119 106 116 Head in total length .. ce be Se) 8-4 8-5 8-8 8-8 9-0 Head in head-and-trunk tie = ee 3-4 3-0 3-1 3-2 3-4 3:3 Head-and-trunk in total ai ie a 2-7 2 ii EM ALG 2:7 Head-and-trunk in tail tne Ae a Si 1-8 er et) 1-6 1-7, Height in head 4 a a Seok Pe 15 2:3 P38 Aer Ny. IES) Snout in head i? Ps be ee eee 2-7 2:3 2-2 2326 Eye in head .. = se ne a 5:8 6-7 (OM NGA 6:0 | 6-5 Ege-pouch in tail a iss me bs 5-0 ee 4-4 be Pee) ve taes Body rings .. ae o is eel 19 19 19 190 20 1s Caudal rings .. ah os .. | 47 47 46 46 | 46 | 46 Dorsal begins on ring No. ues sl ae ah 10 Pa r= Sh 17 17 Dorsal Sends on (number of rings) =f ee al, 1 10) ed 11 Pouch stands on (number of rings) as oe ae he ae 12 Phey eile Be Dorsal rays .. Se a ale At iene 35 36 Bist | B10 39 Pectoral rays. . He - ug Fame tare) NORE seal: TUL gol 4 eae} 13 Ganda nays +: ao he ay a ee) ue | 10 10 Oe 5. 20 Venom, 93 4b 4 = 67. The dorsal keels are separated by a space equal to two-thirds the diameter of the eye, but widen somewhat behind the nape. These keels are lost below the hinder part of the dorsal fin. The upper lateral keels follow the curvature of the back, become sinuous below the fin, thence form the upper keels of the tail. The lower lateral keels are similarly continuous with the caudal keels. There is also a very prominent ventral keel, which, however, is not continued on to the tail; the latter is therefore square in section. None of our females exhibit the “ knife-like acute- ness of the back”? mentioned by Kaup, nor does the height of the body measure more than three times its thickness. It appears, therefore, that we have not speci- mens of such “ advanced age ”’ as those examined by this writer. The general colour is olive-brown above and pale-grey beneath. The markings vary considerably, and in some instances are so extensive as to occupy almost the whole area. This condition is commoner in the female, in which the body is usually of even coloration. In a male, in which the markings are well defined, there are four light bands between the pectoral and the pouch, each with a margin darker than the ground-colour. There are two bands below the dorsal fin, and seven on the tail. The whole of the body, the lower surface excepted, is ornamented with Ou (ds) (=) SUBANTARCTIC ISLANDS OF NEW ZHALAND. [ Pisces. minute white spots, each encircled with a dark rmg. A dark-brown line passes along each side of the snout to the eye, and the ventral keel is dark. ‘Forty-eight eggs were counted in the pouch of one male, each egg having a diameter of 2mm. Another male yielded young which, though not ripe for extru- slon, measured 12 mm. in length. This species was taken only at the Auckland Islands. I obtained specimens in the dredge off Masked Island, and also took them freely among seaweed, by means of the seme, in Musgrave Harbour. The largest example obtained is a male, and measures 161 mm. in length. The female mentioned by Kaup measured 220 mm. Hab.—South America, New Zealand, Auckland Islands. Fam. PLEURONECTIDAE. RHOMBOSOLEA, Gunther, 1862. Rhombosolea tapirina, Ginther. Rhombosolea tapirina, Giinther, Cat. Fish. Brit. Mus., iv, 1862, p. 459. A young example, too small for satisfactory descriptive purposes, was obtained at Campbell Island by Messrs. Chambers and Des Barres, and kindly handed to members of the expedition. Many specimens (from the Auckland Islands) were examined by Giinther when originally describing the species, and the members of the “Southern Cross” Ex- pedition obtained examples at Campbell Island. Hab.—Australa, Tasmania, New Zealand, Auckland and Campbell Islands. Fam. NOTOTHENIIDAE. NototTHEntA, Richardson, 1844. Notothenia microlepidota, Hutton. Notothenia nucrolepidota, Hutton, Trans. N.Z. Inst., vii, 1876, p. 213. JN. parva, Hutton, abid., x1, 1879, p. 339. I obtained this species readily at the Snares by means of a hook and line. It was very common in a deep gulch, and took a fish bait ravenously. Smaller examples were common in rock-pools, the only other fishes there taken being Bovwichtus varie- gatus. My early efforts at the Auckland Islands were unsuccessful, and, though I threw my bait into likely and unlikely places, I failed to secure a single fish. The attempts of the others members of the party being similarly unavailing, we concluded that the fishes had been driven away by the seals, which were very numerous in the vicinity of our camp at Carnley Harbour. These conclusions, however, proved to be incor- rect, for one day a member of our Maori crew brought me three examples which he had caught. He took me to a tumbled mass of sunken rocks, and by means of 2 ft. of line attached to a short stick dropped the bait into a crevice and drew out a Notothenia. Adopting this plan, I obtained as many fishes as I required. It is doubt- less the presence of seals that has driven these fishes to such seclusive habit, for on two occasions, while fishing from the rocks, seals have glided close to me, and, unaware Pisces. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 591 of my presence, have searched all the crannies large enough to admit their heads. Many times thereafter I attempted to entice the fish outside a chink; it was, how- ever, only when the bait was actually dropped into a crevice that it was taken. I found Bovichtus variegatus also living in the crannies. One I had hooked was immediately swal- lowed by a large Notothenia, and I thus ob- tained both specimens. I found small Notothenias living in masses of seaweed in shallow water; they are of sluggish habit, bite freely, but do not show fight when hooked. When young they are quite fearless, and will follow a bait to the suriace of the water. Wherever sunken masses of rocks were found Notothenias were procured. The sur- vivors of the “ Dundonald” wreck told me that they had not been able to catch fish, and they came to the conclusion that this was to be accounted for by the presence of seals, The followmg is a description of a medium-sized individual :-— Bama De Wer 3O) sell. 226 2 Ve ois P., 18; ©.,11+6. Vert., 18+26=44. Length of head, 3:2; height of body, 4:0; length of caudal, 4:8 in the total length. Diameter of eye, 5-6; interorbital width, 3-2 in the length of the head. The maxillary extends to below the third fifth of the eye; the lower jaw is somewhat longer than the upper ; the gill-membranes are free from the isthmus for a short space. The gill-rakers are very short, 17 in number on the first arch, 13 being on the lower limb. The head is much depressed, one-half wider than deep; a broad rounded ridge arises on each side of the snout, and passes backwards over the eye towards the upper angle of the opercle; interorbital space other- wise flat. A cross-ridge connects the two branches in advance of the eyes, The eyes are sublateral, and the upper border of each is formed by the ridge above mentioned. The opercle bears two flat spines. The gape Fic. 24.—Notothenia microlepidota, Hutton. is almost horizontal, the mouth wide, and the lps thick and reflexed. The"single nostril is in a wide and long tube, provided del, Edgar R. Waite Wi ey Tee S$ Vf, l he tiff y/ SS - ~~ ‘ 592 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Pisces. with a posterior flap, and situated midway between the eye and the front of the head. The villiform teeth stand in a wide band above and below; the anterior teeth in both jaws form a single row of widely spaced canines; tongue and palatines smooth ; a broad frenum behind the teeth in both jaws. Fins.—The dorsal arises in advance of the opercular edge, and has a rounded margin; the second spine is longest, 3:3 in the length of the head. The highest rays of the second dorsal are somewhat longer, and its base measures half the distance between the anterior border of the eye and the base of the caudal rays. The anal commences beneath the fifth dorsal ray, and terminates evenly with that fin; its rays are thickened and free at the margin. The pectoral is evenly rounded, and extends to above the base of the third anal ray, its length being a little more than half that of the head. The ventral is placed beneath the margin of the preopercle, and its length equals that of the pectoral. The caudal is rounded, and the depth of its peduncle is twice the diameter of the eye. The skin of the head is roughened, and scales are developed on the interorbital space and occiput, also on the upper parts of the cheeks and opercles. Both divisions of the lateral line are quite straight; the upper one terminates beneath the last dorsal ray, and the lower one commences ten scales in advance of this oint. Colour.—Bluish-black above, grey on the sides, and yellowish beneath. The body is boldly mottled with black and grey; and the lips, cheeks, and opercles, and under-parts of the head bear small closely placed spots or irregular lines of erey. All the fins are black, but the pectoral, ventral, anal, and caudal bear grey spots. Length of specimen, 330 mm. Boulenger states that the species grows to a length of 400 mm., and that the very massive adults have quite a cottoid physiognomy. We have examples 440 mm. in length, and specimens show some diversity of coloration, mainly in the grey colour being replaced by yellow. I obtained this species at the Snares, both by line and in rock-pools; and in Carnley Harbour, at the Auckland Islands. It is also known from Campbell Island. In his report on the fishes of the voyage of the “ Southern Cross,” Dr. Boulenger* reviews the family Notothenudae, giving a synopsis of the species. The following table is derived from an examination of fourteen specimens of N. microlepidota taken on our expedition, and is supplied for comparison with details of other species dealt with by Dr. Boulenger :— * Boulenger, Rep. Nat. Hist. Coll. “ Southern Cross,” 1902, p. 174. a 593 OF NEW ZEALAND. SUBANTARCTIC ISLANDS | ot + 2F| 91 + $9| Ti + 6F| qi+9 |er +9 joc +8 IL + 0¢| 91 + g¢ | | eae) IHL ae 9 ST + 1g LL +8 SL +8 | 1k Sr S| Oise Sip | ell ar @ on) Ne) iter) tt G&G EGG FL + 7G Air |GL + 6G OME sae Y) aul] [B104R] UI SeTBOS TepNQny, SOLOS OSIOA -SuURI] B UL SeTBog eur [etoyryT oAoqe Sollos eurpny -19UO] B UL SayVag peoy Jo YyouUsy Ut YAP [B41q.10.109 0] proy Jo yyQsu0] ut add JO ToJOUVICT YQouey UI pRoy Fo YA cueT YQouoy ut Apoq jo yydoq “* jeue ul SARI [esiop puosos ur sAvy jesiop | IoLeyue ut sway epneo qnoyyIa yyouery (sorqouury [run Ul) Youe, [RIO], 39 594 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Pisces. Boulenger has indicated by means of footnotes his conception of the synonymy of the many species of Notothenia. According to these notes, the New Zealand species are as follows :— 1. N. corwceps, Rich., Voy. Hreb. & Terr., Fishes, 1844, p. 5, pl. ii. 2. N. colbecki, Boul., Rep. “Southern Cross,” 1902, p. 185, pl. xvi. 3. N. microlepidota, Hutt., Trans. N.Z. Inst., vii, 1876, p. 213. 4. N. macrocephala, Ginth., Cat. Fish. Brit. Mus., 11, 1860, p. 263. 5. N. filhoh, Sauv., Bull. Soc. Philom. (7), iv, 1880, p. 228. Notothenia colbecki, Boulenger. Notothenia colbecki, Boul., Rep. Nat. Hist. Coll. “Southern Cross,” 1902, p. 185, pl. xvi. The species was first described from examples collected at Campbell Island. In February, 1907, while on a tour of the southern islands by kind invitation of His Excellency the Governor, I found it to be quite common at Antipodes Island. While at anchor on the north-east side of the island I put lines over and caught this “ cod ” very readily. It affords no sport, and is hauled unresistingly to the surface. Many of the fishes were infested with fluke-like parasites, which glided over the surface of the body; these and also internal worms have been examined by Professor W. B. Benham, who identifies the former as species of Tristoma. The gills harboured crustacean parasites which Dr. Chilton identifies as Lovoneca novae-zealandiae, Miers, a common parasite of New Zealand fishes. The presence of these varied parasites renders the fish most uninviting, and no one on board thought of asking the cook to supply fresh fish. All the specimens caught were larger than the Campbell Island examples taken by the “‘Southern Cross” Expedition, the maximum being 515mm. Hither this or an allied species attains gigantic dimensions, for a headless and otherwise mutilated fish “apparently closely alhed to N. colbecki” was obtained by members of the National Antarctic HExpe- dition, 1901-4, at the winter quarters. The remains of this specimen measured a torn The only specimen obtaimed by members of our Campbell Island party was preserved by Professor Kirk. It was secured at Perseverance Harbour, among stones, at low water. I netted a small specimen in the seine at Musgrave Harbour, Auckland Islands, these two being the only ones taken, durimg the expedition. The following is a description of one of the Antipodes Island specimens :— B. Vi 3D. Vil} 285"A 23% Vol bs P21. C1345 17. Scales—Wonorali2e tr., 13+38; lin. lat., 72433. Vert., 19+26=45. Length of head, 3:4; height of body, 4:6; length of caudal, 7:0 in the total length ; diameter of eye 5-6; interorbital width 3-1 in the length ofthe head. The maxillary extends to below the first third of the eye; the lower jaw is longer than the upper. The gill-membranes are wholly attached to the isthmus, and the gill- rakers are rather short, 22 in number on the first arch, 15 of which are on the lower limb. Pseudobranchiae well developed. * Boul., Nat. Ant. Exp., 1, 1907, Fishes, p. 1. Pisces.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 595 The head is not markedly broad, its width being equal to its depth ; it is smooth and naked, with the exception of the upper part of the cheeks and opercles, which bear small imbedded scales. The upper lateral line arises close behind the opercle, rises shghtly, and then runs almost straight to beneath the last dorsal ray; the lower line occupies the middle of the caudal peduncle, arising posterior to the middle of the anal. Teeth.—A row of widely spaced canines stand in front of each jaw; behind these a band of villiform teeth, broad in front and narrow at the sides; no teeth on the vomer, palatines, or tongue. Fins.—The dorsal commences slightly behind the opercle ; its second spine is the longest, 3-9 in the length of the head ; the outline of the fin is rounded, and the length of its base is twice the diameter of the eye, that of the second dorsal 2°3 in the total. The anal arises beneath the sixth dorsal ray, and extends further back than the dorsal; the pectoral is feebly rounded, and does not reach to the vent ; the ventral extends to the middle of the pectoral rays, and is equal to the caudal in leneth, or one-half that of the head. The caudal is slightly forked, and the depth ‘of the peduncle is one-half longer than the diameter of the eye. Colowrs.—Olive above and pale yellow beneath ; fins dark brown, the median ones margined with pale yellow. Length of specimen described, 475 mm. The following table gives the relative dimensions of some other examples from the islands where indicated :— see A B | C 1D) E FE G Total length (in millimetres) an 515 475 | 509 484 484 129 125, Length without caudal .. i 415 391 | 410 389 400 108 102 Rays im anterior dorsal .. es VI VIE 2 svi VII VII VII VII Rays in second dorsal .. at 28 28 26 28 26 21 28 Rays in anal .. a 23 23 23 23 22 24 22 Depth of body in length — 28 4-7 4-6 5:3 5-0 4-8 4-7 4-0 Length of head in length | 3°3 3-4 33 3-4 3-4 3°3 3-1 Diameter of eye in length of head 5-0 5-6 5-5 5-0 5-3 4-0 4-25 Interorbital width in length of head | 3-1 all: | 3:0 3-1 3:2 3:5 3-7 Scales in a longitudinal series above | 100 102 110 116 111 111 115 lateral line | Scales in a transverse series | 122P 36) 13-F 38) 12-236) 134-39) 124-36) 13-436 | 13-436 Tubular scales i in lateral line 5. || 72" 37 | 724-33 | 72 31 | 71+32 | 64431 | 73434 | 67+31 i B, oO iD} a E, Lcnihene Island ; zr Coren iene: and G, Auckland Islands. Hab.—Auckland, Campbell, and Antipodes Islands. Bovicutus, Cuvier and Valenciennes, 1831. Bovichtus variegatus, Richardson. Bovwichtus variegatus, Richardson, Voy. Hreb. & Terr., 1846, p. 56, pl. xxxiv, figs. 1-4. B. roseo-pictus, Hutton, Trans. N.Z. Inst., xxxvi, 1904, p. 148. 596 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Pisces. The type of B. roseo-pictus 1s a small beach-dried specimen, but, as already noted,* I find this to be but a young example of B. variegatus. The species is very common in the rock-pools around the coast in the neighbourhood of Christchurch, and proves to be equally so in similar situations at the Snares. Owing to their mottled markings, these fishes are quite inconspicuous; they lie very still until disturbed, and, unless closely watched, are difficult to detect in the new situation they may have chosen. I obtained the species at the Auckland Islands —in the seine in Musgrave Harbour, and by line in Carnley Harbour. On hauling up a speci- men from a crevice, amid sunken rocks, it was taken by a large Notothenia, and I obtained both fishes, as previously recorded. Bovichtus variegatus is known also from Campbell Island. The following is a description of one of the small rock-pool specimens :— BOVIS D5 VaAGIE aOR ASS Veal eo. cutkeenh Oto Oseults ote Length of head, 2-9; height of body, 5:0; and length of caudal, 4-5 in the total. The eye is large, lateral, its diameter 3-2 in the length of the head, and more than twice the interorbital space. The snout is a little shorter than the eye. The upper jaw is the longer, and the maxilla extends to below the front margin of the eye ; cleft of the mouth horizontal. The opercular spine is directed upwards and back- wards ; its length is equal to the diameter of the eye. Teeth.—Villiform, in the jaws on the vomer and palatines. Fins.—The base of the first dorsal fin is half the length of the second, and has a rounded margin ; the first five spies are subequal in length, the second being the longest, and 2-4 in the length of the head; the last three spines decrease almost vertically ; the soft dorsal anteriorly is of similar height, its rays decrease regularly in length. The anal is lower, its longest ray (the third) beimg 3-1 in that of the head. All the rays are simple; the last four are thickened, and similar in character to the lower pectoral rays. The ventral spine is flat and sabre-like, its length 1:4 in the head; the first ray is one-tenth longer, the rest rapidly decrease in length; the pectoral is pointed, the upper simple ray being the longest, and equal in length to the head. The caudal is subtruncate, its peduncle low, 2°6 in the height of the body. The head and body are naked ; the lateral line rises above the opercular spine, dips to the middle of the side, and runs between the bases of the central caudal rays. The tubules are directed alternately upwards and downwards, but run straight along the caudal peduncle. Colours.—The ground-colour is silvery, with red markings, the paired and caudal fins being orange; or the colour is olive, with brown markings, the fins named being bright yellow. All examples are silvery beneath. The head is brown, with darker brown and grey marks; some brown bars arranged below the eye pass forward over the jaws, forming five dots on the upper lip. Five distinct vertical bars on the body, below the lateral line, not coincident with similar, though less distinct, bars above. Some irregular white dots may be present along the anterior part of the lateral line, and interspaces below the line may be lighter than the ground in colour. The fins are marked with reddish bars, two wide oblique ones on the first dorsal and a black spot at the base of the vii—viii spines. Four oblique bars on the soft dorsal, * Waite, Rec. Cant. Mus., 1, 1907, p. 35. Pisces.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 597 and a bar across the four posterior anal rays. Two broad bars across the caudal. The ventral has two and the pectoral four bars; both these fins have, in addition, large spots at their bases. Length, 55 mm. Hab.—Australia, New Zealand, Auckland and Campbell Islands. Fam. BLENNIIDAK. TRIPTERYGION, Risso, 1826. Tripterygion varium, Forster. Blennius varius, Forster in Bloch and Schneider, Syst. Ichth., 1801, p. 178. Tripterigium robustum, Clarke, Trans. N.Z. Inst., x1, 1879, p. 292, pl. xv. TZ. genningsi, Hutton, Trans. N.Z. Inst., xi, 1879, p. 339. This species has been previously identified from the Auckland Islands, and I found it to be extremely common in the rock-pools. It was also encountered under stones between tide-marks, and was freely taken by means of the seine and dredge. A single example was taken at the Snares. The following is a description of a large specimen obtained in Musgrave Harbour, Auckland Islands :— Beavis VIEX xXolds A273 VIN; PTs C., 12-44. Length of head shghtly more than height of body, equal to length of caudal and 3:1 in the total. The diameter of the eye is equal to the length of the snout, and nearly twice the interorbital width, being 4:4 in the length of the head. A trifid tentacle over the eye, and a divided one at the anterior nostril. The jaws are equal, and the maxilla extends to within the anterior margin of the eye. Cleft of mouth slightly oblique. Teeth small, those in front scarcely enlarged. Fins.—The spines of the first dorsal increase in length backwards, the last being 1°8 in that of the head. The second fin is highest medially, its longest spine being 2-2 in the length of the head. The third fin is higher, and its last rays reach the base of the caudal. The anal rays are thickened at their tips, as are the lower ones of the pectoral. The latter fin is large, extends backwards to the third anal ray, and is one-third longer than the head. The ventral is slightly shorter than the caudal, which is rounded. The peduncle is narrow, its depth being one-third the length of the head. Scales.—The head and anterior part of the chest naked. The scales are moderate in size, and finely ctenoid. The lateral line runs high and parallel to the back ; it terminates beneath the fifteenth spime of the second dorsal. A depression, com- mencing on the fourth row of scales below, with occasional pores, runs along the middle of the side towards the caudal. Colour.—The colour varies greatly in different individuals ; an olive tint is com- mon, with traces of dark blotches on the dorsal surface. The first dorsal fin is sooty, with a black margin; the second and third are grey or red, the former with black margin; the anal is grey or pink, with lighter-coloured tips to the rays. Pectorals and ventrals sooty, the tips of the rays white or pink. Length of example described, 99 mm. 598 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Introduced Animals. The following table shows details of ten examples from Musgrave Harbour, Auckland Islands :— Total length (in millimetres) 809865 1 968 179m) hee70 63 68-2! 68 69-7| 77-8 Length without caudal aint 65 69 | 80 65 58 5] 52 54 56 63 Rays in anterior dorsal oe VI Wit |) Wal VI VI VI VI VI VI VI Rays in second dorsal .. 21+ 16214 1521+ 16 214 15/214 15/214 15214 15/224 15)21+4 16214 14 Rays in anal ae sy | Ws} | BS 28 28 | 28 28 28 28 28 Depth of body in length ae A-6)| 40) 425) 4-6) 5-45) 428 4-2) eee Length of head in length .. Sp a aes) oO) RAI Se) |) | DD). B2 Diameter of eye in length of 4.8) 4:3) 4:3) 47) 23:8)" 3:8) 40) 421) 40) aa head | | Interorbital width in length of Selly eee) TO Te} | SHO Gel SO) 83) YO) SS head | | Scales in longitudinal series | 82 86 above lateral line Seales in a transverse series.. | 9+ 1510+ 1510+ 17) 8+ 16 7+ 16) 7+ 14) 8+ 15) 8+ 16) 8+ 16) 8+ 16 Tubular scales in lateral line (214+ 1921+ 1921 + 20/19 + 17,184 16)18+ 15)19 + 1519+ 16)19+ 16194 17 | INTRODUCED ANIMALS. The only animal introduced to the southern islands from a purely commercial standpoint with success is the sheep, and the flocks thrive so well on Campbell Island that four shepherds are stationed there, while in the wool season several shearers take passage thither for the express purpose of gathering the fleeces. The introduction of other animals had a humanitarian rather than a utilitarian object. These isolated islands have too often been the scene of terrible shipwrecks, and in cases where life has been preserved the privations of the survivors make terrible reading. The Government has sought to minimise the hardships inseparable from the severe conditions by erecting huts, supplied with food and clothing, in several places on each group of islands, while many domestic animals have been introduced as additional sources of food. It was the unfortunate lot of the survivors of the ill-iated ““ Dundonald,” whom we had the good fortune to succour, to have been wrecked upon Disappointment Island, one on which neither shelter nor introduced food, living or dead, had been provided. I am indebted to Captain J. Bollons, of the ““ Hinemoa,” for much information on the subject lightly touched below, but I find I have little to add to the admirable notice published by Dr. L. Cockayne,* in which he discusses the “ Effect of Animals upon the Vegetation of the Southern Islands.” CATTLE. The cattle introduced on the Auckland Islands in 1850 were killed off by sealers. The animals later placed on Rose Island and eae Island epee to have thriven, * Cockayne, Trans. N.Z. Inst., xxxvi, 1904, p. 297 et seq. a Introduced Animals.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 599 for considerable numbers are now found there. The cattle landed on different oc- casions have, in company with the goats and sheep, died off from Antipodes Island. One consignment was said to have been killed—of course, legitimately—by the survivors of the French ship ‘“ President Felix Faure,” the animals providing wel- come food. SHEEP. Sheep are run on Campbell Island, and their presence must have a marked effect on the native vegetation. This aspect of the question has been investigated by Dr. Cockayne and by Mr. Laing, who make some interesting deductions therefrom. Sheep were also introduced at different times to islands of the Auckland Group, but the conditions proved unsuitable, or they were killed off by seal-poachers, for none are known to exist in the Aucklands to-day. From 1887 to 1904 sheep have been landed at Antipodes Island on the periodic visits of the Government steamers, but they do not appear to thrive, and die off in a short time. GOATS. Dr. Cockayne considers that goats, of all animals introduced by man, are the most destructive to vegetation. They were introduced on the Snares in 1900, but none are now known there. From 1850 onward, various consignments of goats have been placed on the Auckland Islands, at Enderby and Ewing Islands. In 1903 these animals were reported to be numerous on Ocean Island. T am not aware if the goats placed on Campbell Island in 1890, or their progeny, are still in existence. On their hali-yearly visits the Government steamers carry goats, among other animals, to the Antipodes, but Captain Bollons, of the “ Hine- moa,” says they soon die off. SWINE. The domestic pig runs wild on the Auckland Islands, having been introduced in 1807, more than a century ago; and when one has witnessed the fearlessness of so many different kinds of ground-haunting birds it is easy to comprehend how the introduction of an omnivorous animal may be an important factor in their exter- mination. There can be small doubt that the introduction of pigs to the Auckland Islands has already resulted in considerable havoc among the ground-nesting birds, by destroying both eggs and young. Traces of pigs were very plentiful, not only their spoor but their rootings also being abundantly apparent. Native plants are also suffering, for we found whole patches turned over, Bulbinella and Pleurophyllum evidently being favourites. On several occasions we came across the pigs themselves, but they were very wild, and were approached with difficulty. Of four seen on one occasion, one was black, two white, and one pied. One of them was shot, and proved to be a lean, long-legged, and long-snouted animal, apparently reverting to the characteristics of a wild type. Pigs were also introduced at different periods on some of the smaller islets of the Auckland Group, and I understand that their offspring are thriving ons Enderby Island. Captain Norman, who introduced pigs to the Auckland Islands in 1865, also placed a few on Campbell Island, but the latter appear to have died off. 600 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Introduced Animals. RABBITS. These rodents have been freely introduced into the Auckland Group, and are said to be common on some of the islets at the present day. They bred so rapidly on a small detached area that they were in danger of extermination owing to lack of food ; some of them were therefore removed to Rose Island, where they still exist. Rabbits also live on Enderby Island, and traces of them were seen by members of our expedition. RATS. Captain Bollons informs me that rats are very numerous around the homestead on Campbell Island, and he thinks, judging from their size, that they are brown rats (Mus decumanus). Mr. G. R. Marriner tells me that he met with “ grey rats ” at North-west Bay. It is quite possible that these are black rats (Mus rattus), which I have elsewhere* described as living in the Australian bush, while the brown rat is seldom found far from the habitations of man. HOUSE-MOUSE. It is not surprising that the mouse is acclimatised on Campbell Island, seemg that it is properly inhabited, four shepherds living thereon the year through. At the present day the Auckland Islands are visited normally but twice a year, and the chance of introducing mice is not great. Years ago, however, when whalers and sealers visited the islands, and doubtless lived thereon for considerable periods, the house-mouse was introduced, and has spread considerably. Traces were seen at the various depots established for the benefit of marooned sailors, while at the hut near our main camp kerosene-tins leit by former visitors, and containing a little fat at the bottom, yielded numbers of remains. Returning from the bush on one occasion, Mr. F. R. Field told me that he had seen a piebald mouse run under a log. and he piloted me to the scene. We dug under the log for a considerable space, and at length secured a young mouse; it was, however, of normal coloration. An adult example was secured on Masked Island. I am told by Captain Bollons that mice are very numerous at the Government depots on Campbell and Antipodes Islands, DOMESTIC POULTRY. Whether owing to unsuitable conditions or inability to exist when removed from the fostering care of man, the fact remains that the fowls and ducks placed on the Auckland Islands, and the game and guinea fowls similarly turned loose on Campbell Island, have failed to establish themselves; while the domestic poultry under the care of the shepherds at Campbell Island, where they are properly fed and housed, thrive equally well with the farmyard birds of less rigorous climes. * Waite, Proc. Zool. Soc., 1897, p. 857. ARTICLE XXVI.—THE CRUSTACKA OF THE SUBANTARCTIC ISLANDS OF NEW ZEALAND. By Cuarues Curtton, M.A., D.Sc., F.L.8., Professor of Biology, Canterbury College, University of New Zealand. INTRODUCTION. Tue collection of Crustacea actually made during the expedition was not very extensive, owing to the fact that very little dredging could be done, and that the Campbell Island station, where I spent most of my time, was not a favourable one for the littoral and marine forms. I have, however, been able to examine and incorporate in this report numerous species that were gathered at other times by Captain Bollons, Professor W. B. Benham, and Dr. L. Cockayne. To these gentlemen and to the various members of the expedition who so willingly collected Crustacea for me at places inaccessible to myself I wish to record here my best thanks. No separate list of the Crustacea from these islands has, so far as I am aware, hitherto been published, though several species have been recorded from them at different times by the various scientific expeditions that have visited antarctic seas, by the late Captain Hutton,* and by the late Monsieur Henri Filhol.t In this report I record the occurrence at these islands of seventy-five species—viz., Decapoda, 12; Stomatopoda, 1; Amphipoda, 34; Isopoda, 20; Tanaidacea, 1; Nebaliacea, 1; Entomostraca (Cirripedia, Copepoda, and Branchiopoda), 6. It will be seen that the greater number belong to the Amphipoda and Isopoda. Doubtless many other Decapoda would be added as the result of systematic dredgings round the islands. No proper attempt has yet been made to collect the Entomostraca. In the general classification I have followed that adopted by Dr. W. T. Calman in his recent work on the Crustacea for Ray Lankester’s “ Treatise on Zoology,” while in the Amphipoda I have mainly followed the classification in Stebbing’s “ Das Tierreich Amphipoda,” though with some slight modifications. The forms not hitherto described are few in number. I describe only one new genus and nine new species ; but, on the other hand, I have reduced a much greater number both of genera and of species to the rank of synonyms, for reasons that I hope will be considered satisfactory. This I have done chiefly in the Amphipoda, tor which group I have fortunately been able to consult most of the reports on the * Trans. N.Z. Inst., xi, pp. 340, 341. + “ Mission de Ile Campbell.” 602 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. recent expeditions to southern seas ; and, as many of the forms found at these islands are circumpolar in distribution, I have been in a position to correlate the various reports, and to decide in certain cases where the same form has been described under different names by different authors. In several groups of the Amphipoda the multiplication of genera has been carried to what appears to me an unnecessary degree, and characters have therefore been introduced into the generic diagnoses which, in some cases at any rate, are subject to individual variation. While the multiplication of species is bewildering enough, the unnecessary subdivision of genera creates still greater confusion. As regards the geographical distribution of the species concerned, the results seem to be somewhat important. Naturally, the chief interest attaches to the ter- restrial and fresh-water forms of the Amphipoda and Isopoda. Of these, there are in the islands three fresh-water and fourteen terrestrial species, including under the latter the sand-hoppers found on the shores. Of the fresh-water species,* two occur also in New Zealand; one of these (Idotea lacustris) 1s also found in Tierra del Fuego, the other (Chiltonia mihiwaka) is allied to species of the same genus found in Australia and Tasmania, and is represented in South America by the closely allied genus Hyalella. The third fresh-water species belongs to the genus Atyloides, of which two species have been described from the fresh waters of Victoria, but none as yet from the main islands of New Zealand. The genus as at present defined contains both marine and fresh-water species. Ot the fourteen terrestrial species, four belong to the genus Parorchestia, and three of these are peculiar to the islands, the fourth species being found on the Snares and in New Zealand (Stewart Island). Another closely allied species of this genus is very abundant in New Zealand, but has not yet been recorded with certainty from the subantarctic islands; but, as I have explained below, the identification of species in this genus is particularly difficult, and, in any case, we have here one terrestrial genus common to New Zealand and the islands, and represented by shghtly different forms in the different islands. Of the remaining terrestrial species, four are endemic, but are represented in the main islands of New Zealand by closely allied species, and five others are identical with New Zealand species; consequently the connection between New Zealand and the islands lying to the south of it is very close, and the existence of so many similar forms in the two localities pots, I think, undoubtedly to previous land connection. The remaining terrestrial species occurs also in South America, but has not been recorded from New Zealand. These terrestrial species, like the fresh-water ones, also show connection with those of South America, Falkland Islands, and other subantarctic localities. One species, Trichoniscus magellanicus, found in both Auckland and Campbell Islands, is, | think, identical with one found in Tierra del Fuego and the Falkland Islands, and is very closely related to T. verrucosus, which has recently been described by Budde- Lund from the Crozets. Both these species, together with 7. thomson (found in the Auckland Islands and also in New Zealand) and with some other New Zealand species, belong to a separate section of the genus Trichoniscus, confined in its distribution to * T have not included Parorchestia tenwis (Dana) among the fresh-water species, as it can also live in brackish or even in salt water. a Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 603 subantarctic regions. Another species, Deto aucklandiae, belongs to a genus of similar distribution, for species are known from New Zealand and the neighbouring islands, South America, Cape Colony, St. Paul (in the Indian Ocean), and Australia, and the genus is not known from any other locality. The Auckland Island species is peculiar to those islands, but is represented in New Zealand by D. novae-zealandiae (Filhol), which is very close to, if not identical with, a species described many years ago from Chili under the names Oniscus bucculentus (3) and O. tuberculatus (¢ ), Nicolet. The conclusions drawn from the above facts would be strengthened by a con- sideration of the distribution of the species of Hyale found on the subantarctic islands of New Zealand, for though, of course, marine, they are found only in shallow waters near the coast. Ot the three species, one, H. campbellica (Filhol), is known only from Campbell Islands, and is a doubtful species; of the other two, one cer- tainly extends to South America, South Georgia, and Kerguelen, and the other probably to South America. The marine Crustacea from these islands, omitting the Hntomostraca, number fitty. Ofthese, only one genus and four species are endemic, and two of these species are doubtful ; of the remainder, thirty are found in one or more of the other sub- antarctic or antarctic localities, and about eight of the others extend to Australia and other places beyond New Zealand, while only twelve are confined to the New Zealand region, and some of these are represented by closely allied species in other subantarctic islands. Some five antarctic or subantarctic species are found in the islands, which do not appear to extend to the main islands of New Zealand. It will thus be seen that the marine forms very considerably strengthen the evidence as to the large antarctic element in the crustacean fauna of these islands and to the close similarity of their Crustacea to those of the other subantarctic regions. In connection with the species of the terrestrial genus Parorchestia there is one point that seems worthy of note. It is a curious fact that although the male of P. sylvicola (Dana) on the main islands of New Zealand is very rare, nearly all the specimens captured being females, yet in the three species found on the Auckland and Campbell Islands the males appear to be almost as abundant as the females —at least, so far as the collections before me enable me to judge. I have noticed that with species of Orchestia and Talorchestia on the sea-shore of New Zealand the males are usually fairly abundant, and approximate in number to that of the females. Whatever may be the conditions producing this result in littoral situations, it 1s probable that the same conditions obtain to a considerable extent over the whole of these subantarctic islands, for, owing to the damp climate and the abundance of undergrowth, the soil is always more or less wet, and the strong westerly winds carry the sea-spray over the greater part of the islands. In islands like these it is easy to see that the transition from life on the sea-shore within reach of the sea-spray to terrestrial life at higher altitudes may be quite gradual, and be easily accomplished. I have not repeated all the references given in well-known works, such as Stebbing’s “ Das Tierreich Amphipoda,” but have given only those that appeared necessary in each case. 604 SUBANTARCTIC ISLANDS OF NEW ZHALAND. [Crustacea. iS OR Sib Cie s: Subclass MALACOSTRACA. Order DECAPODA. Suborder REPTANTIA. Section BRACHYURA. 1. Leptomithrax australis (Jacq. et Luc.). Auckland Islands and New Zealand.* is 11. 12. 13. 15. ie . Prionorhynchus edwardsu, Jacq. et Luc. Auckland and Campbell Islands, and New Zealand. . Cancer novae-zealandiae (Jacq. et Luc.). Auckland Islands and New Zealand. (A closely allied species in Chili.) . Nectocarconus antarcticus, Jacq. et Luc. Auckland and Campbell Islands, and New Zealand. . Hemaplax hirtipes, Heller. Campbell Island and New Zealand. . Halicarcinus planatus (Fabr.). All subantarctic seas. . Hymenosoma depressum, Jacq. et Luc. Auckland Islands and New Zealand. Section ANOMURA. . Porcellanopagurus edwardsi, Filhol. Campbell Island and Snares. (Allied species in Australia and Juan Fernandez.) . Hupagurus campbelli, Filhol. Campbell Island. . Munida subrugosa (White). All subantarctic seas. Suborder NATANTIA. Nauticaris marionis, Spence Bate. Auckland and Campbell Islands; New Zea- land; Marion, Prince Edward, and Falkland Islands. Palaemon affinis (H. Milne-Edwards). All subantarctic seas. Order STOMATOPODA. Lysiosquilla spinosa (Wood-Mason). Auckland Islands, New Zealand, and Indian Ocean. Order AMPHIPODA. Suborder GAMMARIDEA. . Nannonyx kiddert (S. 1. Smith). Auckland and Campbell Islands, New Zealand, Tasmania, and Kerguelen. Tryphosa kerqueleni (Miers). Snares, New Zealand, Victoria Land, and Kerguelen. . Tmetonyx stebbingr (Walker). Auckland Islands and Cape Adare. Phoxocephalus kerquelenr, Stebbing. Snares and Kerguelen. * By “ New Zealand ” in this list of localities the main islands only are included. Crustacea.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 605 18. Harpinia obtusifrons, Stebbing. Campbell Island, New Zealand, Victoria Land, Kerguelen. . Liljeborgia dubia (Haswell). Auckland Islands, New Zealand, Austraiia, &c. . Carolobatea novae-zealandiae (Dana). Auckland Islands, New Zealand, and Kereuelen. . Leptamphopus novae-zealandiae (G. M. Thomson). Auckland Islands, New Zea- land, Victoria Land, and Graham Land. . Bovallia monoculoides (Haswell). Auckland Islands, South Georgia, Graham Land, South Atlantic and Indian Oceans. . Pontogeneia antarctica, Chevreux. Auckland Islands, Campbell Island, and Graham Land. (Closely allied species in New Zealand.) . Paramoera austrina (Spence Bate). All subantarctic seas. . Atyloides serraticauda, Stebbing. Auckland Islands, Austratia, Victoria Land, and Graham Land (Flanders Bay). . Atyloides magellanica (Stebbing). Auckland Islands, Victoria Land, Tierra del Fuego, and Graham Land. . Atyloides aucklandicus, Walker. Auckland Islands. (Allied species in Victoria, Australia.) . Parapherusa crassipes (Haswell). Antipodes Island, New Zealand, and Australia. . Melita mnaequistylis (Dana). Auckland Islands, New Zealand, and Indian Ocean. . Paradexamine pacifica (G. M. Thomson). Auckland Islands and New Zealand. (A closely allied species in Graham Land.) . Orchestra serrulata, Dana. Auckland and Campbell Islands, and New Zealand. . Orchestra aucklandiae, Spence Bate. Auckland Islands (? New Zealand). . Orchestra bollonsi, sp. nov. Bounty Island, Snares, Auckland Islands, and New Zealand. . Parorchestia maynei, sp. nov. Auckland Islands. . Parorchestia nsularis, sp. nov. Campbell Island. . Parorchestia parva, sp. nov. Auckland Islands. . Parorchestia vmprovisa, sp. nov. Snares, New Zealand (Stewart Island). . Parorchestia tenuis (Dana). Campbell Island and New Zealand. . Hyale hirtipalma (Dana). Auckland Island, Macquarie Island, New Zealand, South Georgia, and Kerguelen. . Hyale novae-zealandiae (G. M. Thomson). Snares, Macquarie Island, and New Zealand. . Hyale campbellica (Filhol). Campbell Island. . Chiltonia mihiwwaka (Chilton). Auckland Islands, Campbell Island, and New Zealand. (Alhed species in Victoria and Tasmania.) . Allorchestes novae-zealandiae, Dana. Auckland Islands, New Zealand (? South America). . Aora typica, Kroyer. Atlantic, Pacific, and Southern Oceans. . Lembos kerqueleni (Stebbing). Campbell Island, Snares, Macquarie Island, New Zealand, Kerguelen, and Indian Ocean. . Jassa pulchella, Leach. Cosmopolitan. Suborder CAPRELLIDEA. . Caprellinopsis longicollis (Nicolet). Snares, New Zealand. and South America. 606 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. Order ISOPODA. Suborder ASELLOTA. — 48. Janira neglecta, sp. nov. Auckland Islands and New Zealand. 49. Iais pubescens (Dana). All subantarctic seas. 50. Haliacris neozelanica (Chilton). Auckland Islands and New Zealand. (An allied species in South Georgia and antarctic seas.) Suborder FLABELLIFERA. 51. Cirolana rossw, Miers. Auckland Islands, Campbell Islands, and New Zealand (? South Africa). 52. Livoneca novae-zealandiae, Miers. Antipodes Islands, New Zealand, Australia, Norfolk Island, and South America (? South Africa). 53. Serolis latufrons, Miers. Auckland Islands and Kerguelen. 54. Exosphaeroma gigas (Leach). All subantarctic seas. 55. Pseudosphaeroma campbellensis, nov. gen. et sp. Campbell and Auckland Islands. 56. Cymodocella tubicauda, Pieffer. Auckland Islands, New Zealand, South AeCOE EA Victoria Land, and Graham Land (? South Africa). 57. Dynamenella huttoni (G. M. Thomson). Antipodes Island, Snares, anal New Zealand. (A closely allied species in Tierra del Fuego and Kerguelen.) Suborder VALVIFERA. 58. Idotea elongata, Miers. Auckland Islands, New Zealand, and Falkland Islands. 59. Idotea lacustris, G. M. Thomson. Campbell Island, New Zealand, and Straits of Magellan. 60. Paridotea ungulata (Pallas). Subantarctic seas generally. Suborder ONISCOIDEA. 61. Trichoniscus thomsoni (Chilton). Auckland Islands and New Zealand. 62. Trichoniscus magellanicus (Dana). Campbell Island, Auckland Islands, South America, Falkland Islands (? Crozets). 63. Haplophthalmus australis, sp. nov. Campbell Island. (A closely allied species in New Zealand.) 64. Scyphoniscus magnus, sp. nov. Campbell Island and Auckland Islands. (A closely allied species in New Zealand.) 65. Deto aucklandiae (G. M. Thomson). Auckland Islands. (Closely allied species in New Zealand and South America.) 66. Oniscus punctatus, G. M. Thomson. Auckland Islands, New Zealand, Tasmania, and Australia. 67. Cubaris rugulosus, Miers. Auckland Islands, Campbell Island, and New Zea- land. Order TANAIDACEA. 68. Tanas novae-zealandiae, G. M. Thomson. Auckland Islands, Campbell Island, and New Zealand. we Ee Crustacea. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 607 Order NEBALIACEHA. 69. Nebalia longicornis, G. M. Thomson. Subantarctic and antarctic seas. Subclass CIRRIPEDIA. Order THORACICA. 70. Balanus porcatus, Da Costa. Auckland Islands, Australia, and North America. 71. Balanus campbelli, Filhol. Campbell Island. 72. Balanus decorus, Darwin. Auckland Islands, New Zealand, and Australia. Subclass COPEPOD A. Order EUCOPEPODA. 73. Deguernea antarctica (G. M. Thomson). Macquarie Island. 74. Zaus contractus, G. M. Thomson. Macquarie Island and New Zealand. Subclass BRANCHIOPODA. Order CLADOCERA. 75. Chydorus minutus, G. M. Thomson. Macquarie Island and New Zealand. Subclass MALACOSTRACA. Order DECAPODA. Suborder REPTANTIA. Section BRACHYURA. Genus Leptomithrax, Miers, 1876. Mstribution.—Auckland Islands, Tasmania, and Australia. Leptomithrax australis (Jacq. et Luc.). Maa australis, Jacquinot et Lucas, Voy. au Pole sud, Zool., i, Crust., p- ll, pl. u, fig. 1, 1853. Leptomithrax australis, Miers, Cat. N.Z. Crust., p. 7, 1876; Filhol, ‘‘ Mission de l’Ile Campbell,” p. 361, pl. XxXxXvll, 1885. Recorded from the Auckland Islands by Hombron and Jacquinot. It occurs also off the southern coasts of New Zealand, and there is a specimen labelled “Dunedin ” in the collections of the Canterbury Museum. An allied species, L. australiensis, Miers, has been described from Tasmania, and another, L. spinulosus, is known from Tasmania and Australia. 608 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Crustacea. Genus Prionorhynchus, Jacq. et Luc., 1853. Distribution.—New Zealand and adjacent islands. Prionorhynchus edwardsii, Jacq. et Luc. eee eer edwardsw, Jacquinot et Lucas, Voy. au Péle sud, Zool., , Crust., p. 8, fig. 1, 1853; Filhol, “‘ Mission de I’Ile Campbell, Zi "367, vl. xl, figs. lad 1885 ; Hodgson, “Southern Cross ” Crust., p. 230, 1902. This species has been recorded from the Auckland Islands by Hombron and Jacquinot and by Hodgson, and from Campbell Island by Filhol. It is abundant at both islands, and is also found at moderate depths off the coast of Otago and Stewart Island. Genus Cancer, Linn., 1766. Distribution.—Very widely distributed. Cancer novae-zelandiae (Jacq. et Luc.). Platycarconus novae-zealandiae, Jacq. et Luc., Voy. au Poéle Sud, i, Crust., p. 34, pl. im, fig. 6, 1853. Cancer novae-zealandiae, Miers, Cat. N.Z. Crust., p. 14, 1876 ; Filhol, “ Mission de l’Ile Campbell,” p. 371, 1885. One small immature specimen obtained from Musgrave Harbour, Auckland Island, by Professor Benham. Common on the New Zealand coasts. According to Miers, it differs from C. plebevus, Poeppig, found on the coast of Chili, only in the somewhat more acute tubercles of the hands and teeth of the antero-lateral margins. Genus Nectocarcinus, A. Milne-Edwards, 1861. Distribution.—Widely distributed in southern seas. Nectocarcinus antarcticus, Jacq. et Luc. Nectocarcinus antarcticus, Jacq. et Luc., Voy. au Péle sud, Zool., 1, Crust., p. 51, pl. v, fig. 1, 1853; Miers, Cat. N.Z. Crust., p. 30, 1876; Filhol, Mission de ’fle Campbell, “10, BB, WSS ¢ Hodgson, ‘ “ Southern Cross ” Crust., p. 229, 1902. Recorded from the Auckland Islands by Hombron and Jacquinot and by Hodgson. Also found off the coasts of New Zealand. Genus Hemiplax, Heller, 1865. Distribution.—New Zealand region. Hemiplax hirtipes, Heller Hemaplax lurtvpes, Heller, Voy. “ Novara,” Crust., p. 40, pl. xvi, fig. 3, 1865; Filhol, “‘ Mission de I’Ile Campbell, G10 385, 1885 ; Miers, Rep. Challenger, ” xvii, p. 251, 1886. Recorded from Campbell Island by Filhol. Common on the coasts of New Zealand. Crustacea. } SUBANTARCTIC ISLANDS OF NEW ZEALAND. 609 Genus Halicarcinus, White, 1846. Distribution.—Subantarctic seas generally. Halicarcinus planatus (Fabr.). Cancer planatus, Fabr., Ent. Syst., t. 11, p. 446, 1793. Halicarcinus planatus, White, Ann. & Mag. Nat. Hist., xvii, p. 178, pl. u, fig. 1, 1846; Miers, Cat. N.Z. Crust., p. 49, 1876 ; Haswell, Cat. Aust. Crust., p. 114, 1882 ; Filhol, “ Mission de ?Ile Campbell,” p. 396, 1885; Stebbing, P.Z.S., 1900, p. 524, pl. xxxvib, 1900; Hodgson, “‘ Southern Cross” Crust., p. 231, 1902. This species was found during the expedition to be very abundant between tide-marks in Perseverance Harbour, Campbell Island; and numerous other speci- mens were brought to me from various parts of both Auckland and Campbell Islands. It is evidently a species of wide circumpolar distribution, and has been many times referred to and described under different names, and even yet there are various points of uncertainty that have not been cleared up. A full discussion of most of these will be found in Mr. Stebbing’s paper quoted above. In his “* Mission de l’Ile Campbell” Filhol has distinguished H. planatus, White, and H. tridentatus (Jacq. et Luc.) from one another mainly by certain small characters of the tridentate front. This separation of the two species is upheld by Lenz, and by the writer of the account given of the Crustacea of the “ Mission du Cap Horn.” So far as the specimens at my disposal go, they certainly confirm Filhol’s statements as to the existence of slight differences between the two forms; and this is also con- firmed by the geographical distribution, for all the specimens in my possession from the Auckland and Campbell Islands belong to H. planatus as understood by Filhol, while all those from New Zealand itself and from Chatham Island, belong to H. tri- dentatus as described by him. ‘The differences are, however, only slight, and whether they are sufficient for specific separation is another question. Personally, I am, on the whole, inclined to consider the two forms as merely local varieties or subspecies of a widely distributed species. The differences are briefly as follows: In H. planatus the posterior lateral tooth on the margin of the carapace is always present and more or less acute, though situated a little below the level of the carapace. In the tridentate front the lateral teeth are somewhat widely separated from the median one, and they are produced a little downwards at the sides, so as to be slightly concave on the inner surtace, and in this way a slight recess is formed below the median tooth, in which the antennae rest, and there is practically no ridge on the epistome to divide the recess into two portions. In H. tridentatus, on the other hand, the three teeth of the front le closer together and more in the one horizontal plane, and the lateral teeth are flattened both above and below, so that there is no recess formed below the median tooth, and, on the contrary, there may even be a slight ridge on the epistome. In this form also both the teeth on the lateral margins of the carapace are obsolete, even the posterior one generally showing as little more than a slight projection. These differences, with the exception perhaps of the presence or absence of the posterior marginal tooth, readily escape observation, and naturally they are less marked in immature specimens. Consequently, it is practically impossible to tell from the descriptions of previous authors which of these two forms they had before 40—S, 610 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. them. So far as I can make out, Mr. Stebbing’s specimens from the Falkland Islands belong to H. planatus as described above, while the forms from Australia which he refers to H. ovatus, Stimpson, are probably the same as those from New Zealand which Filhol has described under the name AH. tridentatus ; but unfor- tunately Mr. Stebbing does not seem to have described the condit:on of the marginal teeth of the carapace in his specimens. The Cape Horn specimens undoubtedly belong to the true H. planatus, and in all probability so do those from Kerguelen Island and from the Cape of Good Hope. Ii this be so, 1t would appear that the true H. planatus is the subantarctic species of circumpolar distribution, while it is represented in the main islands of New Zealand and in Australia by the variety described by Filhol as H. tridentatus. Certainly, all the forms from the mainland of New Zealand that I have seen belong to the latter variety. It is true that Filhol speaks of H. planatus as being found on the whole extent of the coast of New Zealand ; but here I think he is rather quoting the results of previous authors, who had failed to distinguish between the two varieties, than giving the actual distribution of H. planatus as he understood it. Miers indirectly refers to the question in his report of the “Challenger” Brachyura, for he distinguishes between H. planatus (Fabricius) and H. ovatus (Stimpson), and refers specimens from Cape Campbell, New Zealand, to H. planatus, while he assigns to H. ovatus only some specimens from Australia. If my observations are trustworthy, it 1s more probable that the Cape Campbell specimens should be referred to H. ovatus, if that species is really the same as H. tridentatus ; and in cases like this, where minute characters are concerned, it is perhaps not desirable to attach too much importance to one single identification made when a large amount of material is being examined. While it is perhaps rash to hazard an opinion, I strongly suspect that Liriopea leachii, Nicolet (=Hymenosoma leachw, Guerin), and L. lucasw, Nicolet, described in Gay’s “ Historia de Chile,” both belong to H. planatus (sens. stz.). Genus Hymenosoma, Desmarest, 1823. Distribution.—Widely distributed in southern seas. Hymenosoma depressum, Jacq. et Luc. Hymenosoma depressum, Jacquinot et Lucas, Voy. au Pole sud, Zool., i, Crust., p. 62, pl. v, figs. 34-39, 1885; Chilton, Ann. & Mag. Nat. Hist., ser. 7, X1x, p. 148, pl. v, figs. 1-4, 1907. Recorded from Auckland Islands by Hombron and Jacquinot. Also found on the coasts of New Zealand. Section ANOMURBA. Genus Porcellanopagurus, Filhol, 1885. Distribution.—Snares and Campbell Islands, Australia, and Juan Fernandez. Porcellanopagurus edwardsi, Filhol. Porcellanopagurus edwardsi, Filhol, ‘‘ Mission de l’ile Campbell,” p. 410, pl. xlix, figs. 2-4, 1885; G. M. Thomson, Trans. N.Z. Inst., xxxi, p. 187, 1899. Crustacea. ]} SUBANTARCTIC ISLANDS OF NEW ZEALAND. 611 Originally described from Campbell Island, where it was taken by Filhol. I have one specimen dredged at the Snares in 60 fathoms by Captain Bollons. Filhol’s description and figure appear to have been taken from a female specimen. The one I have is a male, and agrees in general with Filhol’s description, but has the right chela (fig. 1a) very much larger than the left, its propod being about as long as the carapace is wide, the merus triangular, widening distally, and with both the inner and outer lower margins spinose. The carpus is about as broad as long, and is marked on the upper surface with transverse lines of setae arranged in short curves. The propod is some- what flattened and produced on the inner side so that it is considerably wider than the carpus; the inner margin bears a line of granules, and the upper surface is also finely granu- lar, the granules being more marked towards the margins, and the whole surface bears scattered tufts of short hairs; the movable finger has the upper portion of its surface granular, and bears on the inner edge about five strong blunt teeth, with tufts of hair Fic. 1.—Porcellanopagurus edwardsi, Filhol. 3 la. Right chela of male. between the bases of them 2 the fixed 1b. Left chela of same specimen. finger has a similar structure. The lc. Extremity of abdomen of same specimen. left chela (fig. 1b) has the propod quite small, narrower than the carpus; the fingers slender, their imner margins straight, fitting closely together, and without teeth. The abdomen is symmetrical, very short, and quite soft, with the exception of the terminal appendages (fig. 1c); these have the outer branch longer and broader than the inner, and the upper surface of both roughened with a fine scale-like ap- pearance ; the telson is soft, and has the posterior margin rounded. According to Filhol, this species lives among the marine algae and does not seek shelter in the empty shells of Gastropods. The structure of the terminal append- ages of the abdomen seems to indicate that they are for the purpose of enabling the animal to hold on to its surroundings, and probably the end of the abdomen is inserted into a crevice in the rocks, to which the animal holds on firmly by the roughened suriace of its terminal appendages. Mr. Whitelegge has described a species (P. tridentatus), dredged in 54-59 fathoms off the coast of New South Wales, which appears closely related to the above.” Another species of this genus (P. plater, Lenz) has been described from Juan Fernandez.+ * Results “ Thetis ” Exped., Australian Museum, memoir iv, p. 181. + See “ Zoological Record,” xxxix, p- 43 (1902), and Zool. Jahrb., Suppl. Vv, 612 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. Genus Eupagurus, Brandt, 1851. Distribution.—Widely distributed. Eupagurus campbelli, Filhol. Eupagurus campbell, Filhol, “‘ Mission de l’lle Campbell,” p. 421, pl. li, fig. 3, 1885; G. M. Thomson, Trans. N.Z. Inst., xxxi, p. 183, 1899. Taken by Filhol in Perseverance Harbour, Campbell Island, at a depth of 5 to 6 metres. I have not seen any specimens of this species. Genus Munida, Leach, 1820. Distribution.—In all seas. Munida subrugosa (White). Galathea subrugosa, White, List Crust. Brit. Mus., 1847. Munida subrugosa, Miers, Zool. “ Hrebus” and “ Terror,” Crust., p. 3, pl. mi, fig. 2, 1874 ; Hutton, Trans. N.Z. Inst., xi, p. 340, 1879; Henderson, Rep. ‘Challenger’ Anomura, p. 124, 1888; G. M. Thomson, Trans. N.Z. Inst., xxxi, p. 194, 1899 ; Hodgson, “ Southern Cross” Crust., p. 232, 1902; Chilton, Trans. N.Z. Inst., xxxvu, p. 320, 1905. ? Galathea gregaria, Fabricius, Ent. Syst., u, p. 473, 1793. ? Grimothea gregaria, Henderson, Rep. “Challenger”? Anomura, p. 124, 1888. G. novae- zealandiae, Filhol, “* Mission de l’Ile Campbell,” p. 426. This species is very abundant at the Auckland and Campbell Islands, and is widely distributed in subantarctic seas. The relationship of Munida subrugosa and Grimothea gregaria has been the subject of much dispute. Miers suggested that Grimothea gregaria is the immature form of Mumida subrugosa, and the question has been since discussed by Hutton, Henderson, Thomson, Hodgson, and others, without any definite conclusion being arrived at. Hutton was, I believe, the first to point out that the two forms are different in habit, Grimothea gregaria being pelagic, while Munda subrugosa lives at the bottom of the sea; and he argued from this, and from the fact that small forms are found with the maxillipedes shortened, as in Munida subrugosa, that the two species were distinct. Henderson and others have since likewise pointed out that small forms with the characters of Munida subrugosa are found, and have similarly upheld the distinction of the two species. The fullest discussion of the question has been given by Thomson, who gives measurements of various individuals, and shows that the difference in the length and development of the external maxillipedes is a comparative one, and that these appendages do not, after all, differ very greatly in the two forms—thus, in Mumnida subrugosa the relative length of the body to that of the external maxillipedes is about 5 to 2, while in Grimothea gregaria it 1s 5 to less than 3. He is therefore inclined to treat Grimothea gregaria as merely a stage in the development of Munda subrugosa. My own observations certainly lead me to confirm the measurements made by Mr. Thomson. The difference in the appearance of the external maxillipedes in the two forms is largely due, not so much to their actual size, as to the way in which oun Crustacea. ]} SUBANTARCTIC ISLANDS OF NEW ZEALAND. 613 they are folded back in Munida subrugosa, while they are kept extended in Grimothea gregaria ; and these positions are naturally associated with the difference in habit of the two forms. It is true that specimens with the external maxillipedes proportionally short and in-folded as in Munida subrugosa are sometimes found which are no larger than some of the specimens of Grimothea gregaria; and, on the other hand, large specimens which from their size should belong to Munda subrugosa are also met with which have the external maxillipedes elongated and showing the flattened and foliaceous form characteristic of Grimothea gregaria. This has already been recorded by Mr. Thomson when he says, “In several large males of Munda the joints all show the flattened and foliaceous form characteristic of Grimothea, as well as the densely fringing setae, while in one large female the joints are completely foliaceous.” Mr. Thomson speaks of these large forms as belonging to Munda, and in another place, speaking of the small Grimothea form, he says, “ Though I have examined hundreds of individuals, I have always found the sexual appendages: in a more or less undeveloped condition.” From the “ Nora Niven” collections I have a number of specimens varying from about 25 mm. in length to 54mm.; all of these have the external maxillipedes more or less flattened and foliaceous, though their length as compared with the length of the body seems to decrease a little in the larger forms; many of these large forms, however, varying from 35mm. to 40mm. in length, are mature females bearing eggs, and would un- doubtedly be considered as belonging to Munida subrugosa but for the character of their fohaceous maxillipedes, and it is doubtless large specimens of this kind which Mr. Thomson had before him when he made the statement in the first sentence quoted above. Considering these facts, it would no doubt be the simplest plan to say that there are two species, differing in the form and size of the external maxillipedes, and this is what has been done by Filhol and others. The general resemblance, however, between the forms is so great, and the length of the external maxillipedes varies so much in each form, that I cannot bring myself to agree with this view, but con- sider we are dealing, after all, only with two forms of one species; and this view seems to be confirmed by the fact that where one form js met with in any particular locality the other is also found somewhere in the neighbourhood. It is, of course, only natural that the immature form should be pelagic in habit, while the mature form inhabits the bottom of the sea; and it seems likely that in this case, just as in some other well-known animals, the immature stage may under certain circumstances be prolonged, and even become sexually mature without com- pletely losing its immature characters. I consider, then, that the foliaceous maxillipedes of Grimothea gregaria are associated with its pelagic habit, and that in the absence of favourable circumstances (eg., a suitable sea-bottom at moderate depth) it may continue pelagic, increase in size, and even become sexually mature without losing its foliaceous maxillipedes; but if it reaches a suitable locality it adopts a more sedentary life at the bottom of the sea, and in subsequent moults the external maxillipedes tend to become shorter and es foliaceous and are in-folded instead of being kept extended as in the pelagic orm. 614 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. Suborder NATANTIA, Genus Nauticaris, Spence Bate, 1888. Distribution.—Widely distributed in southern seas. Nauticaris marionis, Spence Bate. Nauticaris marvonis, Spence Bate, Rep. “Challenger” Macrura, p. 603, pl. evil, 1888. Hippolyte stewarti, G. M. Thomson, Trans. N.Z. Inst., xxl, p. 259, pl. xin, fig. 1, 1888. Merhippolyte australis, Hodgson, “Southern Cross’ Crust., p. 223, pl. xxix, 1902. Nauticaris stewarti, G. M. Thomson, Trans. Linn. Soc., Zool., vin, p. 445, pl. xxix, fig. 1, 1903. N. marionis, Calman, Ann. & Mag. Nat. Hist., ser. 7, xvu, p. 31, 1906. Two specimens agreeing closely with Mr. Thomson’s description of Nauticaris stewarti were dredged during the expedition in Perseverance Harbour, Campbell Island. These I have compared with Mr. Hodgson’s description of the “Southern Cross ” specimens (Merhippolyte australis, Hodgson), which were obtained at Auck- land Island in 10 fathoms, and can find no reason for continuing to keep the two species separate ; and as Dr. Calman, after a comparison of the type specimens, says that Mr. Hodgson’s species is the same as N. marionis, Spence Bate, it follows that Mr. Thomson’s species is also a synonym of this species. Mr. Thomson had himself pointed out that the species were very closely allied, if not identical. Besides the two specimens mentioned above, I have other specimens from Campbell Island, collected by Dr. Cockayne in July, 1903; and also specimens from the Bounty Islands, dredged by Captain Bollons, in 50 fathoms, in October, 1906. Mr. Thomson’s type specimen was dredged in Paterson Inlet, Stewart Island, in 10 fathoms. The species is found also at Marion, Prince Edward, and Falkland Islands, and is therefore circumpolar in distribution. According to Dr. Calman, who examined two of the type specimens, Hippolyte magellanicus, A. Milne-Edwards (“ Mission du Cap Horn,” Crust., p. F 46) should also be placed under Nawticaris, but differs from NV. marionis in possessing exopods on the third maxillipede. . Genus Palaemon, Fabricius, 1798. Distribution.—Cosmopolitan. Palaemon affinis (H. Milne-Kdwards). Leander affinis, H. M.-Edwards, Hist. Nat. des Crust., t. 11, p. 391, 1837;- Dana, U.S. Expl. Exped., xii, pl. xxxvill, fig. 5, 1852; Miers, Cat. N.Z. Crust., p. 85, 1876. Palaemon affinis, G. M. Thomson, Trans. Linn. Soc., Zool., vii, p. 450, 1903. Recorded from Campbell Island by Filhol. It is found also in New Zealand, and is a widespread species, being found at the Cape of Good Hope, Australia, Tas- mania, and the Falkland Islands. Crustacea. } SUBANTARCTIC ISLANDS OF NEW ZEALAND. 615 Order STOMATOPODA. Genus Lysiosquilla, Latreille, 1825. Distribution.—Widely distributed. Lysiosquilla spinosa (Wood-Mason). Coronis spinosa, Wood-Mason, Proc. Asiatic Soc. Bengal, p. 232, 1875. Lysvosquilla spinosa, Chilton, Trans. N.Z. Inst., xxi, p. 62, pl. x, 1891. Recorded from the Auckland Islands by Hutton, under the name Squilla laevis. Also known from various parts of New Zealand and from the Andamans (Wood- Mason). Order AMPHIPODA. Suborder GAMMARIDEA. Fam. LYSIANASSIDAE. Genus Nannonyx, G. O. Sars, 1890. Distribution.—Widely distributed in northern and southern seas. Nannonyx kidderi (S. I. Smith). Lysianassa kiddert, 8. I. Smith, Bull. U.S. Mus., in, p. 59, 1876; Stebbing, Rep. “ Challenger,” xxix, p. 694, 1888. JL. kréyerr, G. M. Thomson, Trans. N.Z. Inst., xi, p. 237, 1879. Nannonyx thomsom and N. kidderi, Stebbing, “ Das Tierreich Amphipoda,” p. 36, 1906. Lysianawx steb- bingi, G. M. Thomson, Proc. Roy. Soc. Tasmania, 1892, p. 19, pl. in, figs. 9-18, and pl. v, figs. 9, 10, 1893. Socarnoides kergqueleni and S. stebbingi, Stebbing, “Das Tierreich Amphipoda,” p. 47, 1906. “ Lysianassa?” A. O. Walker, Ann. & Mag. Nat. Hist., ser. 8, x1, p. 34, 1908. Numerous specimens taken between tide-marks in Perseverance Harbour, Camp- bell Island, in Carnley Harbour, Auckland Islands, and at the Snares, appear to belong to this species. From the brief description that Mr. Walker gives, they are certainly the same as a single specimen from Auckland Island described by him under the name “ Lysianassa?”’ They are also undoubtedly the same as the species common on the coasts of New Zealand which was referred to Lysianassa kroyerv (White) by Mr. G. M. Thomson, and which appears in “ Das Tierreich Amphipoda ” under the name Nannonyx thomson, Stebbing. Again, on comparing them with Socarnoides kerquelenr, Stebbing, I found that they agreed on the whole very closely and in some points quite exactly, except that in my specimens the telson was entire or only slightly emarginate at the extremity, while Mr. Stebbing describes and figures the Kerguelen species with the telson fairly deeply cleft. In other respects the telson in my specimens agrees closely with his description, even in the possession of a long and a short plumose seta about the middle of the lateral margin. This difference between the telsons appeared at first sight to be rather important, and puzzling, considering the close agreement of the two forms in nearly all other respects. The question was also complicated by the fact that the type of Socarnoides stebbingi (G. M. Thomson) which I was able to examine agreed closely with my speci- 616 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. mens, and this species is placed by Stebbing under the genus Socarnordes, as closely agreeing in most respects with S. kerguelen ; the differences which he points out are, I think, merely due to slight maccuracies, especially in the drawing of the first side plate, and to the fact that Mr. Thomson’s single type specimen is a male, while Mr. Stebbing’s description is evidently drawn up from the examination of female speci- mens. ‘The only points in which there is any real difference are that the outer plates of the maxillipedes are said to have the apex obtusely pointed in S. kergueleni, while in S. stebbingi they are apically rounded ; and that the third uropods in S. kergue- lent have the inner ramus not much shorter than the outer, while in S. stebbingi the inner ramus is described by Mr. Thomson as being “ short and quite rudimentary.” The first point is not deserving of very much importance, for the outer plate of the maxillipedes is rather delicate, and from an examination of the various specimens it appears that the shape of the apex is subject to some variation, while owing to its delicate nature it 1s not always easy to dissect 1t and draw it with accuracy. In the length of the rami of the third uropod, too, there is considerable variation, and after having compared Thomson’s type of S. stebbingi, in which the inner ramus is perhaps too large to be called “ quite rudimentary,” with Stebbing’s descriptions I have come to the conclusion that this species cannot be distinguished specifically from S. kerguelen. From this 1t would therefore follow that the Campbell and Auckland Islands specimens are the same as Socarnoides kerquelenr, Stebbing, as well as being identical with Nannonyx thomson, Stebbing. There remains, however, the difference in the telson to be considered. However, on turning up some drawings from New Zea- land specimens of NV. thomsoni that I had made many years ago I found that I had drawings showing in one case the telson only very slightly emarginate, while in another is was almost as deeply cleft as in Mr. Stebbing’s figure of S. kerguelenc. The telson is rather short, and curved upwards at the margins and extremity, so that it 1s concave above, with the lateral margin slightly raised and bearing the two plumose setae about the middle; at the ends the raised margins are slightly produced posteriorly, and usually bear a spine at the posterior angles, and the hind border between them may be straight or slightly emarginate or rather deeply in- dented—the differences probably corresponding to differences in age. My Camp- bell Island specimens are all small and probably immature, and in the specimens of them that I have examined the telson has the posterior border either straight or only shghtly emarginate. The type specimens of Socarnoides stebbingi (Thomson) is now mounted on a slide, and it is impossible to make out the telson clearly, but the postero-lateral angles are distinct, ending in setae, and the hind margin between them is probably somewhat indented. In his diagnosis of Socarnoides Mr. Stebbing says, “ telson small, narrowing distally, cleft not reaching the centre, dehiscent” ; while in that of Nannonyx he says, ~ telson entire,” although certainly some of the specimens of N. thomson: have the telson as much indented as his specimens of Socarnoides kerquelen. Again, in the description of N. goésw (Boeck), he says of the telson, “ apex truncate (notched, Boeck)’; so that in this species also it is probable that there is variation in the telson similar to that in N. thomsoni. In the genus Onisimus, Boeck, as described by Stebbing, it appears that the telson is entire in some species and incised in others, so that in these two genera it is not safe to include the amount of indentation of the telson as one of the generic characters. Crustacea. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 617 As the result of these considerations it seems clear that Nannonyx thomson, Stebbing, and Socarnordes kergueleni, Stebbing, must be united. The reasons for doing this seem to be confirmed by a consideration of the characters of Nannonyx kidderi (S. I. Smith) from Kerguelen Island. In 1888 Mr. Stebbing compared speci- mens of this species with his Socarnoides kergueleni, and apparently considered it as belonging to the same genus and as presenting some resemblance to his species, though he points out various differences, one of the most important of which is that the telson 1s slightly excavated, not cleft. In 1906, however, he placed Smith’s species under Nannonyzx next to N. thomsoni, to which he evidently then considered it to be closely related. The explanation of this apparent inconsistency is, I think, evident—viz., that Nannonyx thomsoni and Socarnoides kerquelent are both identical with N. kidderv. I think it is extremely likely that Lysianassa mtens and L. australiensis, Has- well, from Australia, both belong to this widely spread species, NV. kidderv. Miers has referred various specimens from Australia somewhat doubtiully to Socarnes kriyert (White), a species which was originally described under the genus Hphippiphora, and at the same time points out that his specimens are probably nearly related to Haswell’s species. It 1s probable that these specimens examined by Miers also belong to the species now under consideration, but I cannot refer this species to Ephippiphora kriyert, White, as described by Spence Bate in his “ Catalogue of the Amphipoda in the British Museum”; and as the type of that species in the British Museum when examined by Miers many years ago was dried, and lacked the appendages that are most desired for comparison, it is perhaps hopeless to endeavour to ascertain now what species was really described under that name by White. After I had come to the conclusions stated above I was pleased to find that they were to a large extent anticipated and confirmed by Mr. G. M. Thomson, for in a manuscript paper he had already combined Socarnoides stebbingi with Nannonyx thomsoni. As this species has already been fully described by various other authors, it is unnecessary to repeat the description here, but I may point out that it can generally be easily recognised by the structure of the third uropods ; these are small, but have the peduncle elevated above, either near the middle or towards the distal end, and the produced portion generally tipped with one or more small setae; the rami are both small, the outer with a terminal joint and the inner smaller than the outer. The male differs from the female in having the flagellum of the second antenna about two-thirds the length of the body; the terminal uropoda are the same as in the female, instead of differing, as in many allied species of the Lysanas- sidae. Genus Tryphosa, Boeck, 1871. Distribution.—Widely distributed in northern and southern seas. Tryphosa kergueleni (Miers). Lysianassa kerquelent, Miers, Ann. & Mag. Nat. Hist., ser. 4, xvi, p. 74, 1875. Tryphosa kerquelent, Stebbing, “ Das Tierreich Amphipoda,” pp. 69, 720, 1906; Walker, Nat. Antarct. Exped., Amphipoda, p. 16, 1907. Hoplonyx kerquelent, Walker, Journ. Linn. Soc., xxix, p. 51, 1903. A single specimen from the Snares, dredged in 50 fathoms by Captain Bollons. 618 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. The species is now known from Kerguelen, Cape Wadsworth, Cape Adare, and other localities in Victoria Land, and, although not hitherto recorded, it also occurs in New Zealand, for I have specimens from Lyttelton Harbour. I have been able to compare these and the Snares specimen with specimens in Mr. Thomson’s collection from Cape Adare collected during the “Southern Cross” Expedition, and can therefore feel confident about the identification. Genus Tmetonyx, Stebbing, 1906 (= Hoplonyx, G. O. Sars, preoccupied). Distribution.—Northern and southern seas. Tmetonyx stebbingi (Walker). Hoplonyx stebbingi, Walker, Journ. Linn, Soc., xxix, p. 52, pl. ix, figs. 52-57, 1903. Tmetonyx stebbingi, Stebbing, “ Das Tierreich Amphi- poda,” p. 720, 1906. I have two small specimens from Musgrave Harbour, Auckland Island, that I think must be referred to this species. They agree with Mr. Walker’s description in having the “ lateral angle of the head produced to the end of the first jomt of the upper antennae, the joint rounded,” and in having the fourth segment of the pleon neither dorsally depressed nor carimate and the posterior angle of the third segment not produced into a curved tooth. The eye is somewhat large and reniform; the upper antenna has the first joint very broad and swollen, but the second joint is not very short, being about two-thirds as long as the first but much more slender, the third joint is short; the flagellum is 11-jointed, the accessory appendage small, 4-jointed. The second antenna is very slender, the last two joints of peduncle sub- equal, flagellum 7-jomted. The first gnathopod agrees in general with Walker’s description, but has the palm somewhat more oblique and irregularly defined, with stout setae, and, though the finger has a projection on the inner side, it is not so deeply divided as shown in Walker’s figure. In the third uropods the peduncle is slightly produced at the upper distal angle into a subacute point. In other respects the specimen seems to agree closely with the description given by Walker, the few points of difference mentioned above being probably due to immaturity of my speci- mens. The whole of the body is darkly pigmented, the pigment being unaltered by the spirit. Walker's species was taken at Cape Adare; another species referred to this genus by Stebbing, 7. cicadoides (Stebbing), is found at Kerguelen Island, and is perhaps not so different from the present species as might be imagined from a com- parison of the descriptions given of the two species. Fam. PHOXOCEPHALIDAE. Genus Phoxocephalus, Stebbing, 1888. Distribution.—In northern and southern seas. Phoxocephalus kergueleni, Stebbing. Phoxocephalus kerguelen, Stebbing, Rep. “ Challenger,” xxix, p. 816, pl. ly, 1888; “ Das Tierreich Amphipoda,” p. 135, 1906. I have two imperfect specimens, dredged by Captain Bollons at the Snares, in 50 fathoms, which must, I think, be referred to this species. The gnathopods are Crustacea. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 619 in complete agreement with Mr. Stebbing’s description; and the same is true with regard to the other parts of the body and appendages, so far as they can be made out. The species appears to be very closely related to P. bassi, Stebbing, to which I have referred specimens taken with a surface-net in Otago Harbour, and I should not be surprised if the two species prove alterwards to be identical. According to Stebbing, they differ in many points, and the gnathopods alone suffice to distinguish them; his description of P. bassi is, however, taken from a single male specimen, and if I am right in referrmg my New Zealand specimens to this species its gnathopods are hardly so distinct from those of P. kerguelent as shown in Stebbing’s figure. However, the species in this family are so difficult to distinguish that I prefer to leave the decision of this point over for further investigation. P. kerguelent was taken by the “ Challenger” in Cumberland Bay, in Kerguelen Island, and Mr. Stebbing doubttully referred to this species a specimen from Marion Island. Genus Harpinia, Boeck, 1871. Distribution.—Widely distributed in northern and southern seas. Harpinia obtusifrons, Stebbing. Harpima obtusifrons, Stebbing, Rep. “ Challenger,” xxix, p. 820, pl. lvi, 1888, and “ Das Tierreich Amphipoda,” p. 143, 1906; Walker, Nat. Antarct. Exped., 11, p. 17, 1907. Numerous specimens were dredged in 8 fathoms in Perseverance Harbour, Campbell Island—one male with long second antennae and plumose third uropods, many females with short second antennae and third uropods nearly naked. The type specimens are from Kerguelen Island (55-220 m.), and Mr. Walker records _ the species from the “ Discovery ” winter quarters in McMurdo Strait. This species has not been hitherto recorded from New Zealand, having probably been confused with other species of the family; but I have numerous specimens taken in Otago Harbour with surface-net, others from Lyttelton Harbour (4 fathoms, H. Suter), and in Mr. G. M. Thomson’s collection there are specimens from the Bay of Islands. These specimens all agree so closely with one another, and also with Mr. Stebbing’s descriptions and figures of the “ Challenger ” specimens, that I have no doubt they are specifically identical; they differ, however, in the following points from the characters of the genus Harpina as given by Stebbing in “ Das Tierreich Amphipoda”’: The eyes are present and well marked, large, oval, black ; the second antenna in the male has a long flagellum, and bears calceoli on the terminal joints of the peduncle and also on the flagellum; the third uropods in the male have the rami plumose. Fam. LILJEBORGIIDAE. Genus Liljeborgia, Bate, 1862. Distribution.—Northern and southern seas. Liljeborgia dubia (Haswell). Eusirus dubius, Haswell, Proc. Linn. Soc. N.S.W., iv, p. 331, pl. xx, fig. 3, 1880. Liljeborgia haswelli, Stebbing, Rep. “ Challenger,” xxix, p. 985, 620 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. pl. xcu, 1888; Walker, Journ. Linn. Soc., xxix, p. 60, 1903; G. M. Thomson, Ann. & Mag. Nat. Hist., ser. 7, x, p. 463, 1902; Hutton, Index Faunae N.Z., p. 259, 1904. L. dubia, Stebbing, “ Das Tierreich Amphipoda,” p. 233, 1906; Walker, Nat. Antarct. Exped., 11, p. 35, 1907. A single small specimen was obtained in rock-pools at the Snares. The species is known from Tasmania and various parts of the east coast of Australia and from New Zealand. Mr. Walker has also recorded it from Cape Adare and from the winter quarters of the “ Discovery” in McMurdo Strait. It is probably pretty closely allied to L. consanguinea, Stebbing, from Kerguelen and Heard Islands. Fam. OEDICEROTIDAE. Genus Carolobatea, Stebbing, 1899. Distribution.—Contains only the following species, from New Zealand and Kerguelen. Carolobatea novae-zealandiae (Dana). Oediceros novi-zealandiae, Dana, U.S. Expl. Exped., xi, u, p. 934, pl. Lx, figs. 7 a-h, 1853-55. Halimedon schnerderr, Stebbing, Rep. “ Chal- lenger,” xxix, p. 839, pl. lix, 1888. Ocdzceros novae-zealandiae, Hutton, Index Faunae N.Z., p. 258, 1904. O. novi-zealandiae, Stebbing, “ Das Tierreich Amphipoda,” p. 270, 1906. Carolobatea schneider, Stebbing, “ Das Tierreich Amphipoda,” p. 252, 1906. A few specimens of this species, mostly of small size, were collected at Carnley Harbour, Auckland Island. They agree with a species occasionally taken with the surface-net on the coasts of New Zealand which I had previously identified with tolerable certainty as Oedzceros novi-zealandiae, Dana. It also agrees very closely indeed with Halimedon schneider, Stebbing, from Kerguelen, though Stebbing’s description was taken from a female, probably immature, and does not show the character of the second antenna as it appears in the fully matured male. In this species there is very great variation in the lengths of the an- tennae, particularly of the second an- tennae, and of the fifth peraeopods, according to the age of the animal. In younger specimens both antennae are quite short, and agree well with the description given by Stebbing. 4 Vy In older specimens the upper antenna .7?7/) is somewhat increased in length, and has the flagellum rather markedly Sins Se eee thickened, so that most of the joints Ob, Second (anathoneds of the flagellum become nearly as broad as long ; the second antenna is longer and more slender, and the flagellum is greatly elongated, until the whole antenna may about equal the length of the body ; Fie. 2.—Carolobatea novae-zealandiae (Dana). Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 621 the flagellum is very slender, with only a few fine setae, and is about the same width throughout nearly the whole of its length. In the same way the fifth peraeopod becomes greatly elongated in older specimens, so that it may reach far beyond the posterior end of the animal. It appears to be a fully developed specimen of this kind that is figured by Dana, and his figure represents pretty accurately the general appearance of such specimens, although it does not show the projection of the head characteristic of this family. The gnathopods (figs. 2a and 2b) and other appendages agree very closely with the description given by Stebbing, though in the older specimens the second gnathopod may have the propod more elongated than is shown in his figure. When examined alive the animal is seen to be almost transparent, with a few slight silvery marks on the back; the eyes are pale pinkish in colour, and appear to be quite coalesced and to occupy the whole front part of the head. Mr. Stebbing has suggested that Paracalliope fluviatilis (G. M. Thomson) is not improbably identical with Oediceros novi-zealandiae, Dana; at that time, however, Dana’s species had not been rediscovered in New Zealand, and it is evident from what has been already said that the two species are quite d stinct. Fam. CALLIOPIIDAE. Genus Leptamphopus, Sars, 1893. Distribution.—Only two species of the genus known—one from arctic and sub- arctic seas, the other irom antarctic and subantarctic seas. Leptamphopus novae-zealandiae (G. M. Thomson). Pherusa novae-zealandiae, G. M. Thomson, Trans. N.Z. Inst., xi, p. 239, pl. xe, figs. 2, 2 a-c, 1879. Panoploea debilis, G. M. Thomson, Ann. & Mag. Nat. Hist., ser. 5, vi, p. 3, pl. i, fig. 3, 1880. Leptamphopus novae-zealandiae, Stebbing, “ Das Tierreich Amphipoda,” pp. 294, 727, 1906. Oradarea longimana, A. O. Walker, Journ. Linn. Soc., xxix, pp. 40, 56, pl. x, figs. 77-89, 1903 ; Nat. Antarct. Exped., Amphipoda, p- 32, 1907; Stebbing, “ Das Tierreich Amphipoda,” p. 727, 1906; Chevreux, Expéd. antarct. francaise, Amphipoda, p. 54, 1906. A single specimen of this species was taken in Carnley Harbour along with Atyloides serraticauda and A. magellanica (Professor W. B. Benham). On the whole, this specimen agrees well with Walker’s description, but it has only the last segment of the peraeon and the first and second of the pleon dorsally produced. In the first antenna the second joint is produced distally, so as to form a subacute lobe or tooth on each side, as described by Mr. Walker. These lobes allow the third joint to move upon the second in a vertical direction but not laterally, and they thus act as a support to the third joint, guiding and strengthening its movements. There can be no doubt that Mr. Stebbing was right in pointing out the resemblance between Oradarea longimana, Walker, and Leptamphopus novae-zea- landiae (G. M. Thomson). My specimen from the Auckland Islands is certainly the same as the New Zealand species, and I cannot find any reason for considering it specifically different from Mr. Walker’s species. In his later work Mr. Walker has 622 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Crustacea. pointed out that in the New Zealand species as originally described the two posterior segments of the peraeon and the two anterior of the pleon are produced into two teeth, while in Oradarea longimana only the first two pleon segments are produced, each into one tooth. Mr. Walker had, however, himself pointed out in his original description that seme of his specimens had the first two segments of the pleon simple, and that these characters cannot be relied upon for specific dis- tinctions ; the later description by Stebbimg speaks of only one tooth on the last segment of peraeon and on each of the first two segments of the pleon, and only one tooth is present in my specimens. It is true that in the New Zealand specimen I cannot find an appendage to the first antenna, while this 1s present in Mr. Walker’s specimens; but, as I have poimted out elsewhere in this paper, the presence or absence of a vestigial accessory appendage on the antenna is subject to considerable variation, and Mr. Walker has himself drawn attention to the same fact.* In the uropoda and in all other characters the resemblance is so close that I feel bound to combine the two species. Leptamphopus novae-zealandiae evidently has a very wide range, being found in New Zealand, and at Auckland Islands, Cape Adare, McMurdo Strait, and other places visited by the “ Discovery”; while M. Chevreux has recorded it also from Flanders Bay, Port Charcot, and Wincke Island. Djerboa furcvpes, Chevreux, from Booth Wandel Island,+ shows a striking resemblance to this species in the gnathopoda and in several other respects, but differs in the details of the segments produced into dorsal spines and in having the telson deeply cleft. Kam. PONTOGENEIIDAE. Genus Bovallia, Pfeffer, 1888. Bovallia, Pieffer, Jahrb. der Wissensch. Anstalten zu Hamburg, 1888, p-. 96. Husiroides, Stebbing, Rep. “ Challenger,” xxix, p. 969, 1888. Distribution.—South Pacific, Indian, and South Atlantic Oceans. Bovallia monoculoides (Haswell). Atylus monoculoides, Haswell, Proc. Linn. Soc. N.S.W., iv, p. 327, pl. xvii, fig. 4, 1880. Husiroides caesaris, H. pompen, and H. crassi, Stebbing, Rep. “ Challenger,” xxix, pp. 970, 974, 977, pl. Ixxxvili-xe, 1888. H. monoculoides and H. crassz, Stebbing, “Das Tierreich Amphipoda,” pp. 345, 346, 1906. EH. caesaris, Steb., var. Walker, Rep. Ceylon Pearl Fish., ui, p. 264, pl. iv, fig. 22, 1904. Bovallia gigantea, Piefter, Jahrb. der Wissensch. Anstalten zu Hamburg, 1888, p. 96, pl. i, fig. 5, 1888; Chevreux, Expéd. antarct. francaise, Amphipoda, p. 54, figs. xxxi-xxxiil, 1906; Stebbing, “Das Tierreich Amphipoda,” p. 357, 1906. Three specimens from Carnley Harbour, Auckland Island—two oi large size (26 mm. in length), the third smaller (Professor W. B. Benham and Mr. H. R. Waite). * Ann, and Mag. Nat. Hist., ser. 7, vol. xvi, October, 1905, p. 464. 7 Expéd. antarct. francaise, Amphipoda, p. 74, figs. 42-44. Crustacea. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 623 The synonyms given above require some justification, seeing that Bovallia and Busiroides are placed by Stebbing in different families. I find that my specimens agree well with Stebbing’s description of #. crassi ; they also agree closely with his description of L. monoculoides (Haswell), except that the posterior border of the third segment of the pleon is not serrate, or shows only one or two minute indications of serration. The absence of this serration appears to be the only distinction between these two species, and probably examination of a fuller series of specimens would show that it does not always hold. Under the name “ #. caesaris, Steb., var.,”’ Mr. Walker has described from Ceylon a form that has only three teeth on the hind margin of the third pleon segment, and thus appears to be intermediate between E. crassi and EL. monoculoides, to the latter of which Mr. Stebbing has now united FE. caesaris. The supposition that EL. monoculoides and EH. crassi are forms of the one species appears to be borne out by the geographical distribution, for the single type specimen of H#. crassi was taken in the South Atlantic, off Monte Video, at a depth of 600 fathoms, while EZ. monoculoides is already known to be widely distributed in the South Pacific and Southern Indian Oceans, in Australia, and at Heard Island ; and one would naturally expect specimens from Auckland Islands to belong to this species rather than to one found only in the South Atlantic, if this were really a distinct species. From the above considerations I had come to the conclusion that it was inex- pedient to continue to look upon E. crassi as a species distinct from HL. monoculoides, from which Mr. Stebbing has himself said that it differs little; and I felt the more justified in combining the two because my specimens, although without distinct dorsal teeth, have the first and second segments of the pleon slightly produced and a little compressed dorsally, and in one specimen the seventh segment of the peraeon shows the same thing to a slight degree, in these respects agreeing with EH. mono- culoides rather than with EH. crassi. It appears evident that in this species, as in others to which attention has been directed by Walker and other writers, the pre- sence or absence of dorsal teeth on various segments of the body is subject to con- siderable variation. After arriving at this conclusion my attention was accidentally drawn again to the figures of Bovallia gigantea given by Chevreux, and to the great resemblance which they bore to my specimens; accordingly | compared them carefully with Cheyreux’s description and figures, and found that they agreed with these quite as well as they had done with Stebbing’s description of H. crassi. In Chevreux’s specimens the last segment of the peraeon and the first and second of the pleon are produced dorsally into teeth, though evidently to a greater degree than shown in my specimens. This difference in degree, however, would naturally be associated with the greater size of the specimens, as his figure was taken from a female measuring 32 mm. in length, while my largest specimen, also a female, measures only about 26mm. Pieffer’s specimens were even larger, being 45 mm. in length. In my speci- mens the setae and calceoli are arranged on the antennae in the same general way as described by Chevreux for Bovallia gigantea, though they are hardly so numerous ; but this, again, is naturally to be expected in smaller specimens. I find a very small one- jointed accessory appendage on the upper antenna, just as described by Stebbing in Eusiroides, though in mine it is of still smaller size than in his specimens; in Chevreux’s specimens it appears to be absent altogether, as might naturally be ex- 624 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. pected in such large specimens. My specimens have the hind margin of the third pleon segment slightly convex, with the lower angle produced into a very short tooth, just as shown in Chevreux’s figure of Bovallia gigantea. In all other respects I can find no difference between my specimens and those described by Chevreux, and I feel compelled to unite Bovallia gigantea and Husiroides monoculoides. It is true that Stebbing in his latest work has placed Husirowdes and Bovallia in different families, but a comparison of his diagnoses for these families will, I think, show that there is no distinct or constant point of difference between them. The two genera Husiroides and Bovallia were both published in the same year (1888), but as Stebbing in the second volume of his report on the “ Challenger ” Amphipoda, in which he describes Huszroides, refers to Bovallia (see p. 1653) it is evident that the latter genus has priority of publication, and I therefore adopt it, and place it, as Stebbing has done, in the Pontogenendae. Mr. A. O. Walker has stated that Atylus walkeri, Stebbing (= A. antarcticus, Walker),* from Cape Adare and other antarctic localities, is nearly related to Bovallia gigantea, Pieffer, though he gives various differences in the keels on the segments of the body and in other respects. In view of the facts stated above, some of these differences lose much of their importance, and it is quite possible that A. walkeri may prove to be near to, or perhaps even identical with, the species now under dis- cussion; the gnathopods, however, as figured by Walker, have the palm less oblique, and Walker makes no mention of the numerous calceoli which are so conspicuous a feature in the descriptions by Stebbing and Chevreux of the forms examined by them. Genus Pontogeneia, Boeck, 1871. Distribution.—Widely distributed both in northern and in southern seas. Pontogeneia antarctica, Chevreux. Pontogenera antarctica, Chevreux, Expéd. antarct. frangaise, Amphipoda, p. 69, figs. 40, 41, 1906. Numerous specimens of this species were gathered at Campbell Island and at Carnley Harbour, Auckland Islands, by Frofessor Benham in February, 1907, and Dr. L. Cockayne had previously given me specimens from the Antipodes Islands. These agree very closely with M. Chevreux’s figures and description. Tne species is very near to P. dana (G. M. Thomson), so common on the coasts of New Zealand, but differs in having every third joint of the flagellum of the upper antenna pro- duced below, and bearing a tuft of sensory setae, while in P. dana every fourth joint is produced, and the prominence thus caused is more evident. Chevreux’s specimens were from Flanders Bay and Booth Wandel Island. As at present known, the genus contains, in addition to the two species mentioned above, only P. wmermis (Kroyer), from the Arctic Ocean and North Sea. Another species, P. magellanica (Stebbing), is placed under the genus in “Das Tierreich Amphipoda,” but I think it fits better under Atyloides (see below, p. 627). * Journ. Linn. Soc., xxix, p. 58; and Nat. Antarct. Exped., Amphipoda, p. 34. On Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 62 Genus Paramoera, Miers, 1875. Distribution.—Subantarctic seas generally. Paramoera austrina (Spence Bate). Atylus austrinus, Spence Bate, Cat. Amphip. Brit. Mus., p. 137, pl. xxvi, fig. 4, 1862. Paramoera australis, Miers, Ann. & Mag. Nat. Hist., ser. 4, xvi, p- 75, 1875. Atylus australis, Miers, l.c., xvi, p. 117, 1875; Phil. Tr., elxviu, p. 208, pl. u, figs. 5, 5 a-g, 1879. Atylus (?) australis, Miers (?), S. I. Smith, Bull. U.S. Nat. Mus., No. 3, p. 61,1876. Atyloides australis and A. assimilis, Stebbing, Rep. “ Challenger,” xxix, p. 914, pls. Ixxv, Ixxvi, and p. 918, pl. Ixxvu, 1888. Megamoera fasciculata, G. M. Thomson, Ann. & Mag. Nat. Hist., ser. 5, vi, p. 5, pl. i, fig. 5, 1880, and Trans. N.Z. Inst., xxi, p. 261, 1889; Chilton, Trans. N.Z. Inst., XXxvill, p. 271, 1906. Maera fasciculata, G. M. Thomson, Proc. Roy. Soc. Tasmania (1892), p. 28, 1893. Atyloides australis, G. M. Thomson, Trans. N.Z. Inst., xxvu, p. 211, 1895. Maera fasciculata, Hutton, Index Faunae N.Z., p. 260, 1904; Stebbing, “ Das Tierreich Amphi- poda,” p. 741, 1906. Paramoera austrina, Stebbing, “ Das Tierreich Amphipoda,” p. 363, 1906. Paramoera austrina, var. Walker, Ann. & Mag. Nat. Hist., ser. 8, ul, p. 34, 1908. Stebbingia gregaria, Pieffer, Jahrb. Hamburg. Anst., v, p. 110, pl. u, fig. 7, 1888; Stebbing, Rep. “ Challenger,” xxix, p. 1654, 1888, and “ Das Tierreich Amphipoda,” p- 358, 1906; Walker, Nat. Antarct. Exped., Amphipoda, p. 33, 1907. ? Aucklandia enderbyi, Walker, Ann. & Mag. Nat. Hist., ser. 8, u, p- 35, pl. v, figs. 3, 4, 1908. Numerous specimens of this species were obtained at various localities both from the Auckland Islands and from Campbell Islands; it is recorded also from Macquarie Island by Mr. G. M. Thomson. It is known from Kerguelen Island and elsewhere, and is evidently a very common species in subantarctic regions. After careful comparison of specimens from various localities, I have no doubt this is the species that has been long known in New Zealand under the name Megamoera fasciculata, G. M. Thomson. Through the kindness of Mr. Thomson I have been able to examine a specimen from Tasmania referred by him to this species, and also to compare it with a specimen from Macquarie Island indentified by him with Atyloides australis, Miers, and I find that these both belong to the one species. Mr. Stebbing unites Atylus megalophthalmus, Haswell, with this species, though embodying in his description some points mentioned by myself in which it differs from the type. My description was taken from specimens collected in Sydney Harbour (the type locality), and I have no doubt that they belong to Atylus mega- lophthalmus, for they agree very closely with the description and the figures that Mr. Haswell gives of the gnathopods and of the uropods. These specimens I have re-examined, and, though they undoubtedly come very close to Paramoera austrina, they appear to differ in the following points: (1.) The head is produced into a rostrum about four-fifths as long as the first joint of the upper antenna. (2.) The appendage 41—S§, 626 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. on the first antenna is smaller and less distinct than in typical specimens, and is apparently fused with the last jomt of the peduncle, showing only as a slight pro- jection from it. (3.) The carpus of both gnathopods is shorter and more triangular than in typical specimens from the Auckland or Campbell Islands. (4.) The branches of the third uropods are slightly longer and broader, and are more abundantly sup- plied with setae and long fine hairs, and thus resemble more nearly the uropods of species cf Pontogeneva. In the specimens from New Zealand, Auckland Island, Campbell Island, and Macquarie Island there is no rostrum, nor is there one present in the Tasmanian specimen that I have examined; in all of these, too, the appendage on the first antenna 1s comparatively distinct and well marked. My specimens agree, on the whole, with the specific description given by Mr. Stebbing in “ Das Tierreich Amphipoda,” but there is at least one point in the generic description which does not always hold—vyiz., “ Uropods 1 and 2, outer ramus the shorter.” This is true for the second uropod ; but in the first uropod the outer ramus is hardly, if at all, shorter than the inner one. It appears that in this family of the amphipods particularly there has been an unnecessary multiplication of genera, and consequently some characters have been introduced into the generic description which are subject to individual variation. Mr. Walker has recorded a variety of this species from Port Ross, Auckland Island, but the points he mentions seem hardly sufficient to establish a new variety, especially as he was able to examine only a single female specimen. I think there can be no doubt that Stebbingia gregaria, Pieffer, from South Georgia, was based on specimens of Paramoera austrina in which the accessory ap- pendage was really absent or was not observed. In 1888 Stebbing was inclined to unite the two, but in 1906 he still keeps them distinct, and recognises Stebbingia as a separate genus. The specimens from Kerguelen Island described by 8. I. Smith in 1876 under the name “Atylus (?) australis, Miers (?),” are also shown by Smith’s description to belong to this widely distributed and variable species. On a single “‘female with ova, length 10 mm.,” from Hnderby Island, Mr. Walker* has founded a new genus, Aucklandia, for which he gives the followmg diagnosis: “Antenna 1 shorter than antenna 2, with short accessory flagellum. Gnathopods dissimilar. Otherwise as Paramoera.” In his diagnos's of Paramoera, however, Stebbing says, “Antenna 1 not longer than antenna 2,” and, as a matter of fact, in P. austrina it 1s occasionally somewhat shorter, so that this point is not of sufficient importance to rank as a generic character. As to the next point—* gnathopods dissimilar ”—this is also the case to a slight extent in Paramoera, though not expressly so stated in Stebbing’s diagnosis. Secing that these are the only two points on which Mr. Walker establishes his genus, and that he says “ Otherwise as Paramoera,” 1 cannot help thinking that his specimen is only one of Paramoera austrina in which the gnathopods are slightly more dissimilar than usual and the palm rather more transverse than is generally the case. The whole of his description applies almost exactly to Paramoera austrina, except perhaps with regard to the gnathopoda, and these are not altogether inconsistent with the supposition expressed above. * Ann. and Mag. Nat. Hist., ser. 8, 1, p. 35, pl. v, figs. 3, 4, 1908. ~I Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 62 Genus Atyloides, Stebbing, 1888. Distribution. Subantarctic and antarctic regions ; some species in fresh waters. Atyloides serraticauda, Stebbing. Atyloides serraticauda, Stebbing, Rep. “‘ Challenger,” xxix, p. 920, pl. Ixxviu, 1888; Walker, Journ. Linn. Soc., xxix, pl. i, fig. 90, 1903 ; Chevreux, Expéd. antarct. francaise, Amphipoda, p. 87, 1906; Walker, Nat. Antarct. Exped., Amphipoda, p. 33, 1907. A few specimens of this species were taken along with Atyloides magellanica in Carnley Harbour, Auckland Islands (Professor W. B. Benham). Stebbing’s original species was taken “off Melbourne.” Mr. Walker records this species from Cape Adare and Cape Wadsworth, and M. Chevreux has examined specimens from Flanders Bay, so that the species is evidently widely distributed in antarctic and subantarctic seas. My specimens agree well, on the whole, with Stebbing’s description, particularly as regards the telson; but in some of them the posterior margin of the third pleon segment is almost smooth, and it 1s evident, as Mr. Walker has pointed out, that the number of teeth on this margin is subject to variation. In this species, as in many others, the southern specimens appear to be larger than those found further north: mine do not measure more than about 6-8 mm., while the largest from Cape Adare was 15 mm. in length. This species bears a very close general resemblance to A. magellanica (Stebbing), and must be placed in the same genus. I doubt if there is any good reason for retaining the genus Atyloides as distinct from Paramoera, though I follow Stebbing, and retain it at present. A. serraticauda seems to differ from A. magellanica only in having both pairs of gnathopods much longer and more slender, and at one time I was inclined to consider the two as forms of one species —as, indeed, they may really be. My specimens of the two species were taken together, and it is worthy of note that during the “ Discovery ” Expedition the two species were taken from places not very far apart, and during the French Antarctic Expedition of 1903-5 both species were collected at Flanders Bay. Atyloides magellanica (Stebbing). Atylopsis magellanica, Stebbing, Rep. “ Challenger,” xxix, p. 925, pl. Ixxix, 1888. Pontogeneva magellanica, Stebbing, ‘‘ Das Tierreich Amphipoda,” p- 360, 1906; Walker, Nat. Antarct. Exped., Amphipoda, iu, p. 33, pl. xu, fig. 20, 1907; Chevreux, Exped. antarct. frangaise, Amphi- poda, p. 64, figs. 37-39, 1906. A few specimens of this species were taken in Carnley Harbour along with A. serraticauda, Stebbing. These agree well with the description given by Chevreux, and also with that given originally by Stebbing, though in his specimen the antennae were wanting. My specimens, however, have a small one-jointed accessory flagellum on the upper antenna. Chevreux says that in his specimens there is no appendage; but, as poimted out elsewhere, this difference is not sufficient to rank as a specific distinc- tion. In all other respects my specimens agree very closely with Chevreux’s description, and the telson and third uropod in particular appear identical with the 628 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacca. form figured by him. The gnathopods, which are considerably elongated, also agree well with the figure given by Walker for this species; but, on the other hand, he figures the telson with the sides somewhat convex and free from setae. This species, though originally placed by Stebbing under Atylopsis, was atter- wards transferred by him to Pontogeneia, and in this he is followed by both Walker and Chevreux. It appears evident, however, from Chevreux’s description that the joints of the flagellum of the upper antenna are not produced at every third or fourth jeint, as in the other species of Pontogeneva ; and, moreover, the telson hardly agrees with that found in these species, since in them the two terminal lobes of the telson are rounded and without setae. In my specimen there is a very slight widening of the alternate joints of the flagellum of the upper antenna, similar to that in Paramoera austrina, but to a much less degree. These points, together with the fact that the upper antenna may possess a one-jointed accessory appendage, seem to me to show that the species comes nearer either to Atyloides or Paramoera, ii, indeed, these two genera are to be kept distinct, and for the present I leave it under Atyloides. Chevreux has described under this genus a species, A. longicornis, which appears to be closely allied to A. magellanica, though the telson is rather more like that of A. serraticauda. In many respects both A. magellanica and A. longicornis come very close to Paramoera austrina (Spence Bate), and at one time I was inclined to place A. magellanica under Paramoera ; but it differs from typical specimens of P. austrina, the only species of that genus, in having the gnathopods longer and more slender, and in possessing fewer long setae on the antennae. Atyloides aucklandicus, Walker. Atyloides aucklandicus, Walker, Ann. & Mag. Nat. Hist., ser. 8, u, p. 33, Jol sv, wes. IL, BZ, ISOS, Several specimens were collected by Professor W. B. Benham from a small fresh-water pool on Enderby Island into which a small stream emptied. In the stream were found specimens of Chiltonia mihiwaka (described below), the present species having been found in the pool itself. I refer these specimens with some hesitation to Mr. Walker’s species. In most respects they agree with his description, but his account is hardly detailed enough to make one quite certain of identifying a species in this family, where there is such a close resemblance between all the members; moreover, as there is no statement to the contrary, Mr. Walker’s specimens are presumably marine. I am extremely doubtful whether the young imperfect specimen which he mentions really belongs to the same species as his other two specimens. My specimens agree well with the generic description of Atyloides given by Stebbing, except that the gnathopcds can hardly be called slender. From Mr. Walker’s description my specimens differ in the followimg points: (1.) The hind margin of the third pleon segment is almost straight (not convex); the posterior angle forms a right angle and is not produced into a small tooth, but the minute denticle or seta is present above it. (2.) The eyes are not visible. (3.) The first maxilla has the inner plate fringed with numerous setae, about eighteen or twenty— not seven, as described by Walker. (4.) The telson is deeply cleft, each portion narrowing distally and with the posterior end rounded; the outer margin bears _ Crustacea. } SUBANTARCTIC ISLANDS OF NEW ZEALAND. 629 about five or six small setae, one or two being situated at the rounded end ; in Mi. Walker’s specimens each half bears only two setae. The flagella of both antennae bear calceoli on the proximal joints. The gnatho- pods are described below and shown in the figures, and the rest of the animal calls tor no further detailed description. Figure 3a shows the first gnathopod. The side plate is slightly narrowed and rounded below, the margin bearing two or three fine setae; the basos is as long as the rest of the lhmb, its posterior margin bears five or six very long setae; the ischium has a small tuft of setae on its posterior mar- ea gin near the extremity ; the merus is rounded pos- teriorly, and bears near its SUNG angle a transverse row of bn eight or nine stout setae, AN ) ; ig aan with others on its posterior ‘atl margin towards the end of ‘ ae the joint. The carpus is y A rather longer than the pro- oe Oh pod, its anterior margin with am Ke if two or three setae and a Fie. 3.—Atyloides aucklandicus, Walker. small tuit at the distal 3a. First gnathopod of female (outer side). angle; the posterior margin 3b. Second gnathopod of female (outer side). bears three transverse rows of setae, the number in each row increasing towards the distal end of the joint ; the propod is oblong, the breadth more than half its length ; anterior margin rather convex, with a few setae on the distal half, and a tuft at the base of the finger ; posterior margin with three short transverse rows of setae similar to those on the carpus, and a small tuft on the outer surface some distance from the margin; the palm is transverse, straight, and bears four short stout setae near the point where the tip of the finger impinges; the finger is nearly straight, and fits closely on to the palm. The second gnathopod (fig. 36) is similar to the first, but slightly larger; the carpus is fully as long as the propod, and bears on its posterior margin four transverse rows of setae, and there are also four rows on the posterior margin of the propod. Mr. O. A. Sayce has described two species of Atyloides from the fresh waters of Victoria. Of one of these, A. gabrieli, I have been able to examine specimens kindly sent me by Mr. Sayce. It comes close to the species now under discussion, which, however, differs in the following points: The antennae bear fewer and smaller tufts of long hairs than those found in A. gabriela ; the inner lobe of the first maxilla bears numerous plumose setae, instead of only three; the second joint of the palp of the mandible is less broadened ; the gnathopods are rather less slender, and ap- parently have the posterior margin of the fifth joint less produced; and the palm is considerably more transverse than in A. gabrieli. Mr. Sayce has also described another species, A. fontana, which in some of these points approaches more nearly to the one from Auckland Islands, for the antennae do not bear the tufts of long setae, the second joint of the palp of the mandible is less expanded, and the inner lobe of the first maxillae bears numerous setae. 630 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. Fam. GAMMARIDAE. Genus Parapherusa, Stebbing, 1906 (=Harmonia, Haswell: see ‘‘ Das Tierreich Amphipoda,” p. 383). Distribution.—Contains only the following species, which is known from Aus- tralia, New Zealand, and Antipodes Island. Parapherusa crassipes (Haswell). Harmonia crassvpes, Haswell, Proc. Linn. Soc. N.S.W., iv, p. 330, 349, pl xix, ng. 3, 18795 Chilton, Wranse N:Z, Ist. xv.) ps S2eaolemae fic. 5 a, b, 1883; G. M. Thomson, Trans. N.Z. Inst., xxi, p. 261, 1888. Parapherusa crassvpes, Stebbing, “ Das Tierreich Amphipoda,” p. 383, 1906. A small female specimen was collected at Antipodes Island by Dr. L. Cockayne in July, 1903. I do not feel certain about the systematic position of this species, but pending turther investigation I leave it under the Gammaridae, where Stebbing has placed it. Genus Melita, Leach, 1813-14. Distribution.—In all seas. Melita inaequistylis (Dana). Amphithoe (Melita) naequstylis and A. tenwcorns, Dana, P. Amer. Ac., ll, pp. 214, 215, 1852. Melita tenuicorms, Dana, U.S. Expl. Exped., Xi, u, p. 963, pl. lxvi, figs. 5 a-m, 1853 and 1855; G. M. Thomson, Trans. N.Z. Inst., xii, p. 218, 1881; Chilton, Trans. N.Z. Inst., XXXvil, p. 271, 1906. Melita inaequistylis, Stebbing, “ Das Tierreich Amphipoda,” p. 429, 1906. Maera tenucorns, Sp. Bate, Cat. Amphip. Brit. Mus., p. 195, pl. xxxv, fig. 6, 1862; Walker, Rep. Ceylon Pearl Fish., pt. u, p. 273, pl. v, fig. 33, 1904. Melita zeylanica, Stebbing, “ Spolia Zeylanica,” 11, pt. v, p. 22, pl. v, 1904. Several specimens from Carnley Harbour, Auckland Islands (Professor W. B. Benham). This species is widely distributed in New Zealand, where it is met with both on the open coast and also in lagoons and estuaries where the water may at times be quite fresh. It is readily recognised both by the peculiar characters of the gnathopods of the male and also by the colour, which is either slaty or greenish-brown, and remains unchanged for a long time in spirit. Mr. Stebbing, m “ Das Tierreich Amphipoda,” points out that Dana’s and Thomson’s statements with regard to this species are in many respects contradictory. This, however, is merely due to the fact that Dana described the species as being without an accessory appendage to the first antenna, and to other errors in his description and figures. In 1904 Mr. A. O. Walker recorded this species from Ceylon, though poimting out one or two respects in which his speci- mens differed from Dana’s description, the chief beg that the first and second peraeopods were as long as the last two pairs, while in Dana’s description and figure the last two are much the longer; Mr. Walker’s specimens were dcubtless immature Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 631 (he gives the length as 5 mm.), for in such specimens the various peraeopods are of more uniform length, while in fuliy developed specimens the last twe pairs are, as described by Dana, much longer than the others. In the same year Mr. Stebbing described a new species, Melita zeylanica, also from Ceylon, which he could not reconcile with Dana’s description and figures. From a comparison of his description with that given by Walker and with the New Zealand specimens, there can, I think, be no doubt, however, that Stebbing’s species is the same as Dana’s, the colour and gnathopods especially being practically identical. Some of the contusion has no doubt arisen from the fact that the first and second enathopods of the male vary considerably, according to the state of their develop- ment: thus, m immature specimens the first gnathopod has the finger attached to the propod in the normal way, and the gnathopod does not then differ very much from the corresponding appendage in the female ; in fully matured specimens, how- ever, the end of the propod is somewhat excavate, the finger is attached about the middle of the distal extremity, it is much curved, and the postero-distal angle of the propod is produced into a rounded setose lobe against which the finger impinges. In the same way the peculiar characters of the second gnathopod, with the finger curved im on its inner concave suriace, are attained only in fully adult males; in younger forms this gnathopod is much more like that of the female, though it may have the propod much larger in proportion to the size of the body. There can be no doubt that Stebbing is right in saying that the figure which Dana labels as the female igs shown from the form of its second gnathopod to be a male ; the other figure, labelled “Male,” has the second gnathopod larger than it usually is in the-female, and was probably taken from a somewhat immature male, though Dana considered it to be the female, and accidentally transposed the sex signs in his plate. The teeth on the terminal segment of the pleon appear to be somewhat variable: in young specimens these segments are quite smooth; in others they have teeth on the fourth and fifth segments, as described by Walker; while in others again I have been able to find them only on the fifth.* This species appears to bear in colour and in other respects. a very close resemblance to MZ. palmata (Mont.), about which there seems to have been the same contusion as regards the shape oi the first gnathopod, but it differs from that species in the shape of the second gnathopod in the maie. M. inaequistylis is now known to occur in New Zealand, the Auckland Islands, and in Ceylon, and therefore must have a very wide range in southern seas. Dana had originally described two species, Amphitoe (Melita) inaequistylis and A. (Melita) tenucornis, though afterwards uniting them as the two sexes of one species under the name Melita tenwicornis, and this name has been used for the species until a few years ago. Stebbing has, however, revived the name M. inaequi- stylis, according to the rule of page precedence, and I have also used this name in order to avoid further confusion ; but, seeing that Dana himself and all subsequent writers had referred to the species as M. tenwicorns, I think it is a pity that the name M. wmaequistylis was not allowed to remain in the oblivion in which it had long been resting. *Canon A. M. Norman has called attention to a similar variability in the sculpturing of the pos- terior margins of the pleon in M. obtusata (Montagu): Ann. and Mag. Nat. Hist., ser. 6, iv, 1889, p- 132. 632 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. Fam. DEXAMINIDAE. Genus Paradexamine, Stebbing, 1899. Distribution.—New Zealand, Australia, and Graham Land. Paradexamine pacifica (G. M. Thomson). Dexamine pacifica, G. M. Thomson, Trans. N.Z. Inst., x1, p. 238, pl. xb, fig. 4. Paradexamine pacifica, Hutton, Index Faunae N.Z., p. 259, 1904; Stebbing, “ Das Tierreich Amphipoda,” p. 518, 1905; Calman, Ann. & Mag. Nat. Hist., ser. 8, 1, p. 233, 1908. A few specimens of this species were taken in Carnley Harbour, at a depth of 2 fathoms. They are identical with specimens from Port Chalmers, Lyttelton, and other parts of New Zealand, where the species is quite common. It 1s also known from the east coast of Australia. Monsieur Chevreux, in his report on the Amphipoda of the French Antarctic Expedition, has described a species of this genus from Booth Wandel Island, and also from Port Charcot, which appears to ap- proach very nearly to our New Zealand species. According to Monsieur Chevreux, however, his species differs in having a dorsal median carina on the last two segments of the peraeon and on the first segment of the pleon ; also, in the greater length of the lower antennae, and in having the propod of the last pair of peraeopoda only about half as long as the carpus. In his specimens, too, the last uropods are longer, and surpass the length of the telson. Fam. TALITRIDAE. Genus Orchestia, Leach, 1813-14. Distribution.—On all shores throughout the world. Orchestia serrulata, Dana. Orchestia serrulata, Dana, P. Amer. Ac., 1, p. 204, 1852, and U.S. Expl. Exped., xii, u, p. 870, pl. lv, figs. 7 ab (3), mo ( 2 ?), 1853 and 1855. O. serrulata (part), Stebbing, “Das Tierreich Amphipoda,” p. 535, 1906. I have numerous specimens of this species from several localities in the Auckland Islands and in Campbell Island. [t+ is also found in various parts of New Zealand, particularly in the south. Some of the specimens from the islands attain a large size, the males reaching as much as 40 mm. in length of body. These specimens undoubtedly belong to this species as described by Stebbing. It appears to be well characterized by the ridges or corrugations on the segments of the body (though these, of course, are best-marked only in the older males), by the shape of the second gnathopod of the male, by the character of the fifth peraeopod, and by the comparative absence of spines on the body and appendages. Stebbing united O. aucklandiae, Spence Bate, with this species. I have, however, as explained below, specimens which certainly agree well with Spence Bate’s species, but have no sign of ridges or corrugations even in large males, and, as there appear ee Crustacea. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 633 to be these two forms, it is perhaps best to keep them as separate species, though they are certainly very closely related. There is nothing mentioned in either Dana’s or Spence Bate’s descriptions about the corrugations. I am therefore following Stebbing in referring the corrugated form to O. serrulata, Dana. The distinctive characters of this species are to be seen only in fully adult males, the females and the immature males being very difficult to distinguish from those of other closely allied species of the genus. In the fully developed males the second antenna is about half the leneth of the body, with the peduncle smooth and somewhat angular in section, its last two joints very long ; the flagellum contains about twenty joints, and may be slightly longer than the peduncle, though often somewhat shorter; the whole appendage is smooth, being almost free from setae. The second gnathopod of the male (fig. 5b) has the propod very largely developed, nearly as broad as long, the palm nearly transverse, and defined by a stout sharp tooth, on the inner side cf which the finger impinges ; near the base of the finger the palm is deeply hollowed out, and near the middle it is produced to a subacute tooth, from which it extends almost straight to the defining-tooth ; the finger is strongly curved, so that its inner margin is very concave ; a few short setae are present on the basos and on the palm, but the rest of the joint is almost quite smooth. In the fifth peraeopod (fig. 5c) the basos has the posterior flat expansion somewhat narrow, its pos- terior margin being straight or only very slightly convex and only obscurely serrate, and the lower angle is pro- duced into an acute process reaching beyond the extremity of the ischium; the distal end of the merus is deeply concave, the anterior angle being pro- duced into a distinct tooth, curving shghtly imwards, the posterior angle being much less produced; the carpus is slightly longer and considerably broader than the propod, and both Sa bear a few short setae on the anterior Fic. 5.—Orchestia serrulata. Dana. margin ; the finger is short and stout. 5a. First gnathopod of large male. “In younger males the setae on the Becta eset hale, various appendages are much more prominent, the antennae are shorter in proportion to the length of the body, the second gnathopod has the propod smaller and the palm much straighter and the finger less curved; in the fifth peraeopod the basos has the posterior expansion shghtly wider in proportion to the length of the joint, its posterior margin some- what convex and more distinctly serrate, and the lower angle less produced, Fig. 4.—Orchestia serrulata, Dana. Side view of a large male. 634 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. forming often only a right angle; the merus is less broadened, and its distal extremity more of the usual shape. The females agree with the younger males in all these points, and apparently differ only in the gnathopods: in the first gnathopod the propod is about the same width throughout, with the palm very small and the finger projecting beyond it ; the second gnathopod presents no distinct features. Orchestia aucklandiae, Spence Bate. Orchestia aucklandiae, Spence Bate, Cat. Amphip. Brit. Mus., p. 17, pl. ia, fig. 3, 1862; G. M. Thomson, Trans. N.Z. Inst., xxxi, p. 201, 1898 ; A. O. Walker, Ann. & Mag. Nat. Hist., ser. 8, u, p. 36, 1908. Orchestia serrulata (part), Stebbing, “ Das Tierreich Amphipoda,” p. 535, 1906. Several specimens, both male and female, from Hnderby Island are certainly the same as those from the same locality referred to this species by Mr. Walkez, and I think he is right in referring them to Spence Bate’s species. They agree well with the description given by Mr. Walker, and, although they come very close to the preceding species, they differ from it in the absence of the transverse ridges or corru- gations on the segments of the peraeon even in large specimens, for some of the specimens are larger than specimens of the preceding species in which these trans- verse ridges are nevertheless present. The females have the distal end of the first gnathopod as wide or slightly wider than the base, and thus agree with the descrip- tion given by Mr. Walker, and differ from that given by Mr. Stebbing for O. serrulata. I am therefore following Walker in considering these specimens to belong to a species distinct from O. serrulata, though 1 quite realise that good reasons may be urged for combining the two. Mr. Walker states that his specimens might well be referred to O. gammarellus, Pallas, but for the absence of the expansion of the fourth and fifth joints of peraeo- pod 5,” and he adds that it is by no means certain that his specimens were sexually adult. Some of my specimens are adult females bearing eggs, and the largest males measure about 24 mm. in length, and are most probably also adult; but in none of them is there any sign of the expansions referred to. The second gnathopods of the male vary considerably in the appearance of the margin of the palm, according to age; but even in the largest specimens the excava- tion near the base of the finger 1s not great, while in younger specimens the palm is almost straight and the finger fits closely along it. Spence Bate describes the finger as being “ excavated near the base,” but this appearance is rather due to a curva- ture of the finger at that point, so that the inner margin becomes concave; in younger specimens the finger 1s straighter. The fifth peraeopod closely resembles that of the preceding species in having the posterior expansion of the basos angular and pro- duced downwards, the amount of this downward production varying greatly with the age and sex of the specimen, in young males and in females the lower inferior angle being a right angle; the posterior margin is straight or slightly convex and more or less serrulate, and the whole of the posterior expansion is clearly marked off on the outer side from the thicker anterior portion of the basos. The end of the merus is excavated, and the anterior and posterior margins produced downwards into two short spines, the anterior one being the longer, these spines being better Crustacea.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 635 marked in the older and larger specimens. The structure thus formed at the end of the merus is, however, much less marked than in the large specimens that I have referred to the two species O. serrulata (Dana) and O. bollonsi, sp. nov. In this species, as in O. serrulata and O. bollonsi, the carpus of the fifth peraeopod bends outwards on the merus so that when bent it les almost at right angles to the general longitudinal line of the body; it then rests on the ground, and the spinous processes on the end of the merus can be pressed into the sand, and the animal is thus enabled to stand firmly when walking or preparing to leap by suddenly extend- ing the pleon. Orchestia bollonsi, sp. nov. Male.—In general, like O. serrulata, bus with all segments of the body quite smooth ; the third pleon segment with the hind margin serrate and provided with short setae, the posterior angle a right angle or a little produced. The second an- tennae only about one-third the length of the body. The second gnathopod (fig. 65) with propod greatly enlarged, palm some- what oblique, defining-tooth long and acutely poited; a round prominence on the palm towards the base of the finger, from which it 1s separated by a slight depression; from PS this prominence the palm proceeds almost ) straight towards the defining-tooth, but is separated from it by another narrow depres- sion; a few short setae along the borders of the palm; the finger somewhat curved, with a very slight enlargement opposite the middle of the palm. Fiith peraeopod (fig. 6c) = : : 6 with the posterior expansion of the basos z somewhat narrow, the posterior margin ob- , scurely serrate and usually somewhat con- Fie. 6.—Orchestia bollonsi, sp. nov. cave about the middle, the lower angle See asin geelvopee! i wae lle, ; 6b. Second gnathopod of male. produced downwards into an obtuse lobe 6c. Fifth peraeopod of male. reaching considerably beyond the end of the ischium, the margins of this lobe being obscurely serrate like the posterior margin ; merus with the distal end deeply hollowed, the anterior angle produced into a long sharp tooth curved inwards, posterior angle broader and less produced; carpus about as long as the propod but considerably broader; a few short setae present on the merus, carpus, and propod. Female.—First gnathopod similar to that of O. serrulata, but with the propod slightly enlarged at its distal end; the fitth peraeopod not showing the special characters described in the male to the same degree. Length of largest specimens, about 28 mm. Hab.—Bounty Islands, under guano (Dr. L. Cockayne, July, 1903); Snares (Chilton, 11th November, 1907) ; Ewing Island (J. B. Mayne, November, 1907). Type in the Canterbury Museum, New Zealand. It is only with great hesitation that I propose this new species; but the speci- mens from Bounty Island are specially large and well marked, and, as they are the first Orchestra gathered from that island, it may cause less confusion to give them a 636 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. new name provisionally rather than to endeavour to identify them with any of the species already described. The fully developed males are easily recognised, but more immature males have the second gnathopoda smaller and with the palm nearly straight, and are then difficult to distinguish from the forms I have referred to O. aucklandiae, Bate. As stated below, I have specimens from the Snares and from Ewing Island that are evidently very close to the Bounty Island specimens; they differ, however, in being of smaller size and in having a more distinct enlargement on the middle of the inner margin of the finger—in the Bounty Island specimens there is only a faint indication of such an enlargement. I have similar specimens gathered in Preservation Inlet and other places in south-west Otago. Some of these seem to me to approach O. chiliensis, M.-Edwards. From the Snares and Ewing Island I have a considerable number of specimens, gathered at the roots of tussocks near the shore. In most respects these agree with the characters of Orchestia bollonsi as described above. None of them are of large size, the largest being about 16 mm. in length, yet most of them show tic charac- teristic shape of the basos of the fifth peraeopod. ‘They differ from the Bounty Island specimens referred to this species in the following points: The second gnathopod in the male has the palm rather more oblique; the rounded projection on the palm near the base of the finger is smaller and is followed by a rather deep but somewhat narrow depression; the palm thence proceeds almost straight to the acutely pointed defining-tooth. The finger is considerably curved, and bears on its Inner margin an enlargement which nearly corresponds in position to the depres- sion in the centre of the palm. The females have the propod of the first gnathopod only very shghtly expanded at the extremity, in this respect closely resembling the specimens of O. serrulata as described by Stebbing. None of the specimens show any sign of ridges on the seg- ments of the body. This species 1s named after Captain Bollons, to whom I am indebted for many species from the islands. Genus Parorchestia, Stebbing, 1899. Distribution._—A genus of terrestrial Amphipoda of wide distribution. Mr. Stebbing describes this genus as “ like Orchestia, but maxillipeds with fourth joint of palp distinct though very small, conical, and having a spine on the truncate apex.” ; In Orchestia, however, the maxillipedes may, as Stebbing himself states, have an obscure rudiment of the fourth joimt of the palp, and the presence or absence of this joint is therefore hardly sufficient to distinguish the two genera. At the same time, it is perhaps convenient to group the truly terrestrial species under a separate genus, and the species that I am acquainted with can, as a rule, be distinguished from species of Orchestia living on the sea-shore by the greater abundance of long slender spinelike setae on the antennae and the peraeopods, and by the more reduced con- dition of the pleopoda, especially of the third pair. Terrestrial amphipoda living far away irom the sea are already known to be very widely distributed, being found in New Zealand and the adjacent islands, Australia, America, the Hawaiian Islands, and other islands in the Pacific; while others, again, are known from various islands in the Indian Ocean. Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 637 Some of these species have been referred to Parorchestia, others to Talitrus, and others again to Orchestia ; and in some cases the same species appears to be very widely distributed: thus, Orchestia platensis Kroyer has been recorded from Rio de la Plata, in South America; the Atlantic coast of North America; Bermudas ; the Mediterranean ; the Lake of Tiberias, in Palestine; from the Hawaiian Islands, where it is found at a height of 3,000 ft.; and Mr. Walker has even recorded it from the Mahlosmadula Atoll, in the Indian Ocean, at a depth of 20 fathoms in the sea.* Some of these identifications, however, appear to have been founded only on female specimens, and, as is well known, it is in many cases almost impossible to distinguish species of this group by the females alone. In this paper I describe four new species of Parorchestia, three of which are distinguished by the characters of the adult male ; with two of these it is practically impossible to distinguish the females from one another or from those of P. sylvicola, the species so abundant on the main islands of New Zealand. It may be, of course, that in this instance we are dealing with one widely spread species with polymorphic males; but, as the forms so far as is at present known are confined to separate islands, I prefer to look upon them in the meantime as distinct species. T The fourth species, P. «mprovisa, of which only the female is known, is remark- able for having the propod of the first gnathopod enlarged and completely subchelate and similar to that of the second gnathopod of the males in other species, though developed to a less degree. Parochestia maynei, sp. nov. Male.—Body moderately compressed, side plates not very deep, fifth as deep as the fourth, pleon segments with the lower margins somewhat rounded, posterior angle of the third quadrate. Hyes round, as far apart as their width. First antenna reaching to the end of the fourth joint of peduncle of the second, its third jomt much longer than the first or second, which are subequal; flagellum six-jointed, as long as the third jomt of peduncle. The second antenna slender, more than half as long as the body ; fifth joint of peduncle one and a half times the length of the fourth, bearing a few fine setae; flagellum rather longer than peduncle, about thirty-jointed, joints rather longer than broad, slightly expanded distally, each with four stiff setae longer than the joint is wide. First gnathopod (fig. 7c) with basos slender; merus with a slight pellucid enlargement at the distal end; carpus considerably longer than the propod, its posterior margin produced into a rounded pellucid lobe, the *< The Fauna and Geography of the Maldive and Laccadive Archipelago,” vol. 11, Supplement 1, Amphipoda, pp. 923, 924. + I have spoken of these forms in the usual way as if they are simply males and females distinguished by constant differences. It is quite possible, however, that their relation is more complex, and that, e.g., the distinctive characters of the males are attained only during the breeding season, and that at other times the secondary sexual characters are poorly developed. In this case we would have some- thing approximating to the “‘ high and low dimorphism ” so common in the Lamellicorn beetles and to the case described by Geoffrey Smith in the crab Inachus mauritanicus ; he considers the condition of the “ middle-sized males” in this species as one of partial hermaphroditism, and draws attention to the partial hermaphroditism described by C. L. Boulenger in Orchestia. In these, however, the small ova appear to be developed only in young males in which the secondary sexual characters are not well developed, the gnathopods being scarcely differentiated. (See Geoffrey Smith, “ Cambridge Natural History,” Crustacea, pp. 102-4, and C. L. Boulenger, P.Z.S., 1908, pp. 42-47.) 638 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. junction of which with the propod is marked by an imperfect row of setae; propod also enlarged on the posterior margin; finger not reaching to the end of the rounded lobe. Second gnathopod (fig. 7d) very large and strong, basos not widened ;_ basos and ischium both slightly hollowed anteriorly to receive the propod when reflexed ; carpus very small, almost covered by propod, propod with greatest width nearly equal to the length, anterior margin very convex, posterior also convex and dis- tinctly marked off from the palm (though without defining spine), which forms a right angle with it; palm slightly oblique, produced near base of finger mto a strong prominent subacute tooth which overlaps the finger on its outer side, portion between the tooth and end of palm deeply concave ; this portion of palm is broad, the propod being here produced on the inner side into a prominent ledge ; finger strongly curved, its inner margin bearing towards the base a small subacute projection which impinges Fie. 7.—Parorchestia maynei, sp. nov. 7a. First gnathopod of female. 7f. Third uropod of male. 7b. Second gnathopod of female. 7g. Celson of male. 7c. First gnathopod of male. 7e. Hitth peraeopod of male. 7d. Second gnathopod of male. against the tooth of the palm; the inner margin of the finger and the margin of the palm bearing a few short setae. Third peraeopod shorter than the fourth, both with the basos rather narrow; fifth (fig. 7e) a little longer than the fourth, its basos well expanded, posterior margin slightly convex and minutely serrulate, propod longer than either the merus or carpus, which are subequal, all the joints supplied with long setae, none of the joints specially expanded, finger long and slender. Uropods 1 and 2 with spines on peduncle and at the end of branches and on upper margin of inner branch, third uropod (fig. 7/) very small, not reaching beyond the telson, two strong setae on upper margin of peduncle, the branch half as long as peduncle, and bearing three setae at its extremity ; telson (fig. 7g) with upper surface slightly concave, about two-thirds as broad as long, apex slightly emarginate, with three spines on each lobe, and two large spines near the middle of each lateral margin of telson. Crustacea. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 639 Female.—Differing from the male in the gnathopoda; the first gnathopod (fig. 7a) similar to that of the male, but with no dilatation on the merus, a small one only on the carpus, and slight indication of one on the propod ; second gnathopod (fig. 76) with basos not broadened, the merus having the posterior margin widely dilated and the carpus dilated in the same way, the whole of the propod expanded, the expansion being marked off from the rest of the limb by a row of small setae, the palm directed slightly outwards and the short finger not reaching the rounded end of the propod. Hab.—Several specimens, both male and female, from Norman Inlet, Auckland Island (J. B. Mayne), and one male and two females from Disappointment Island (Professor W. B. Benham). To this species I refer also some female specimens collected on Adams Island at a height of 2,000 ft. by Mr. R. Speight, and also others collected by myself on Auckland Island. Type in the Canterbury Museum, New Zealand. The female of this species is practically indistinguishable from that of P. sylvicola, so common on the main islands of New Zealand; but the male is easily recognised by the characteristic shape of the second gnathopod, though this is not easy to describe nor to represent clearly in a figure. I have named the species aiter Mr. J. B. Mayne, who gathered many Crustacea for me during the expedition. Parorchestia insularis, sp. nov. Male.—Similar to P. maynet in most points; the pleon segments, however, appear to have the lower margins rather more rounded, and the posterior angle of the third segment is a little rounded and produced. The second gnathopod (figs. 8b and 8c) large, with the anterior surface of the basos slightly concave and grooved, that of the ischium deeply concave, each of its two margins being produced into a rounded lobe and fitting up against the propod when this is bent back upon these joints; the propod large, shghtly dilated distally, its greatest width about four-fifths of its length, the palm shghtly transverse, well defined but with- out defining-spine, with a broad but prominent tooth about one-third of the length of the palm from the base of the finger overlapping the finger on its outer side, remainder of the palm from the aT Nhe tooth to the defining-angle concave, palm 8b. Bacon eoathopod ofaraalé (outer side). narrow and not produced inwards as in 8c. Second gnathopod of male (inner side). P. maynei; finger moderately curved, without projection on its inner margin, its tip fitting into a slight recess between the defining-corner of the palm and a slight prominence on the inner side of this. Length, 13 mm. Female.—Practically indistinguishable from that of P. mayne. Fie. 8.—Parorchestia insularis, sp. nov. 640 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. Hab.—Campbell Island (Mr. G. R. Marriner and Messrs. Des Barres and Chambers). Type in the Canterbury Museum, New Zealand. Specimens of a terrestrial hopper (belonging doubtless to this species) were reported by various members of the expedition to be extremely abundant on Camp- bell Island right up to the top of the highest hills; owing, however, to its very active habits, and to unfavourable opportunities for collecting, only a compara- tively small number of specimens were actually secured. This species evidently closely resembles P. maynei, but the second gnathopod of the male (though of the same type) differs, particularly in not having the palm broad and widely produced on the innner side of the propod. Parorchestia parva, sp. nov. Male.—Body small and compact, side plates not deep, pleon segments with lower margins rounded, posterior angle of third slightly rounded and produced. Fic. 9.—Parorchestia parva, sp. nov. 9a. Antennae of male. 9c. Second gnathopod of male. 9b. First gnathopod of male. 9d. Fifth peraeopod of male. First antenna reaching about half-way along the last jomt of peduncle of lower antenna, third jomt of its peduncle longer than the second, flagellum of four joints, shghtly longer than last jot of peduncle. Second antenna short, only one-third the length of the body, last joint of peduncle not much longer than the preceding one, flagellum longer than peduncle and contaiming about fifteen joints. First gnathopod of usual shape, second gnathopod with basos somewhat broadened, ischium and merus short and subequal ; propod large, broadly oval, widest about the middle, where it is almost as wide as long; palm transverse or slightly oblique, badly defined, curving gradually into the posterior margin of the propod; margin of palm straight or slightly convex, and fringed with small setae ; finger moderately curved, its tip fitting into a little recess on the inner side of the end of the palm, inner margin of finger without enlargement. Third peraeopod considerably shorter than the fourth, fifth somewhat longer than fourth, its basos only moderately dilated, the posterior margin being slightly convex and obscurely serrate, lower angle rounded Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 641 and only shghtly produced ; merus broad, especially towards distal end, so that the whole joint is triangular ; carpus wide and irregularly dilated, greatest width about two-thirds its length ; propod slightly longer than carpus, not dilated ; all the joints supplied with rather long setae, except on the posterior margin of the merus, which has only two at the distal end. In other respects closely resembling P. mayner. Female.—Resembling that of the two preceding species, except that the second antennae are shorter, being, as in the male, only about one-third the length of the body. Hab.—Several specimens from Norman’s Inlet taken in company with P. maynea (J. B. Mayne) ; others under logs on Auckland Island (Professor W. B. Benham). Type in the Canterbury Museum, New Zealand. This species is readily recognised by the short antennae, the large rounded propod of the second gnathopod with rts simple palm, and by the dilatation of the merus and the carpus of the fifth peraeopod ; -the peraeopods, too, are not so elon- gated as in the preceding species. Parorchestia improvisa, sp. nov. Male.—Unknown. Female.—In general shape resembling P. sylvicola. Body not much compressed ; first three segments of pleon with lower margin slightly convex, posterior angle of third segment right-angled, with corner a little rounded. Hyes rather large, as far apart as their width. First antenna reaching middle of fifth joint of peduncle of second antenna, third joint of peduncle longer than second, flagellum about seven - jointed. Second antenna long, slender, about three-fourths the length of the body, fifth jomt of peduncle nearly twice as long as the fourth, flagellum longer than peduncle, about 35-jomted. First gnatho- pod (figs. 10a, 10b) with the pro- pod moderately large, widened, width about half the length, with the palm oblique, occupying nearly Fic. 10.—Parorchestia improvisa, sp. nov. 10a. First gnathopod of female. one-halt the length of the whole 106. Palm of same, more highly magnified. posterior margin, the palm being 10c. Second gnathopod of female. defined by one or two stout setae situated on a little projection which is separated from the rest of the palm by a depression, another stout seta arising on the inner surface near the middle of this depression, the rest of the palm nearly straight or only slightly convex and bordered with a row of short setae, three other longer ones being situated on the surface near the palm and an oblique row of about six or,seven further,backon the outer surface of the propod ; the posterior margin of the propod also bears an irregular 42—S, 642 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. row of about six setae and is very slightly produced, forming a narrow pellucid area ; the carpus is triangular, of the usual shape, and is produced posteriorly into a some- what irregular pellucid lobe, on the inner side of which arise three long setae ; there is also a very slight indication of a pellucid lobe on the posterior margin of the merus. Second gnathopod (fig. 10c) of usual shape, the finger very short, the lobe of the ptopod being produced far beyond its extremity. The third peraeopod is shorter than the fourth, and in each the basos is rather narrow oval. The fifth peraeopod is slightly longer than the fourth, and both are considerably elongated, in large specimens being about three-fourths the length of the body. The uropoda and telson of the usual form. Length, 16 mm. Hab.—Snares, five female specimens (G. R. Marriner). Also found m Stewart Island. Type in the Canterbury Museum, New Zealand. This species seems very peculiar in possessing a fairly large and well-developed subchelate first gnathopod in the female very different from the usual form of this appendage, and more nearly resembling what is generally found in the second enathopod of the male. I have only about five or six specimens, but from these it appears that this form of gnathopod is only well developed when the female is mature, as the smaller specimens have the propod of the first gnathopod much more slender and more of the usual type, and in the smallest specimens the propod is of ap- proximately the same width throughout, not being dilated at all, and the palm is transverse or only very slightly oblique. T have long had from the north of Auckland a single imperfect female specimen of a Parorchestia which resembles the species now described in having the propod of the first gnathopod dilated and distinctly subchelate ; it differs, however, in having the propod suboblong, with a transverse palm. I had hitherto looked upon it as an abnormal specimen of P. sylvicola, but the existence of a somewhat similar form on the Snares shows that it may perhaps belong to a distinct species. Since the above description was drawn up Mr. G. M. Thomson has handed over to me all the notes and drawings of Amphipoda that he has accumulated during many years past, and among them I find drawings made years ago of the gnathopoda of an undetermined species of land-hopper from Port Pegasus, Stewart Island, which correspond closely with a somewhat immature female of Parorchestua improvisa, and evidently belong to that species, so that this species occurs on the mainland of New Zealand as well as at the Snares. Parorchestia tenuis (Dana). Orchestia tenuis, Dana, P. Amer. Ac., 11, p. 202, 1852, and U.S. Expl. Exped., xii, Hu, p. 872, pl. lx, fio. 1, 1853) and 1855. Parorchestia tenuis, Stebbing, “ Das Tierreich Amphipoda,” p. 557, 1906; Chilton, Trans. N.Z. Inst., xli, p. 58, 1909. I have a few small specimens, obtained on the sea-shore of Perseverance Har- bour, Campbell Island, at the mouth of a small fresh-water stream, that I think belong to this species, as defined by Stebbing ; they are the same as specimens from brackish and fresh-water streams in various parts of New Zealand. Although this species agrees with Stebbing’s description of the genus Parorchestia in having a small | Crustacea. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 643 —_—___———-- : - fourth joint on the palp of the maxillipedes, it does not appear to me to harmonize in all respects with the terrestrial species of Parorchestia known to me. The setae on the antennae and peraeopods are shorter and less prominent, the peraeopods them- selves are stouter and less elongated, and the finger is shorter and thicker and the pleopoda longer than in the truly terrestrial species. Genus Hyale, H. Rathke, 1837. Distribution.—On all shores. Hyale hirtipalma, Dana. Allorchestes hirtipalma, Dana, P. Amer. Ac., 1, p. 205, 1852, and U.S. Expl. Exp., xi, li, p. 888, pl. lx, fig. 4, 1853 and 1855. Nuzcea fimbriata, Mhomson, Trans. IN-Z5 Inst; x1, p. 2386; pl xb, tis. 2, 1879, yale fimbriata, Thomson, Trans. N.Z. Inst., xxvii, p. 211, 1895. H. hrti- palma, Stebbing, “ Das Tierreich Amphipoda,” p. 564, 1906. H. trigonochir, Walker, Ann. & Mag. Nat. Hist., ser. 8, i, p. 37, 1908. H. villosa, Smith, Bull. U.S. Nat. Mus., No. 3, p. 58, 1876; Stebbing, “Das Tierreich Amphipoda,” p. 574, 1906. Allorchestes georqianus, Pfeffer, Jahrb. Hamburg, Anst., v, p. 77, pl. 1, fig. 1 a—n, 1888. Hyale georgiana, Stebbing, “* Das Tierreich Amphipoda,” p. 572, 1906. I have specimens of this species from Enderby Island, and others from Persever- ance Harbour, Campbell Island, and also from the Antipodes Islands (collected by Dr. L. Cockayne in 1903), and from Macquarie Island (A. Hamilton). The specimens from Enderby Island agree precisely with Mr. Walker’s Hyale trigonochir, but this species is, I think, undoubtedly the same as H. hirtvpalma (Dana). Mr. Walker says that the form of gnathopod 1 in the male is the most salient cha- racter of his species. In large full-grown males the propod of this gnathopod is certainly more triangular than in the female or in immature males. I find, however, on comparison of different specimens that there is very considerable variation in the shape of this joint: for example, some males from the Antipodes Islands show the propod quite triangular, as in Mr. Walker’s figure, while in others it is more oblong, though these already possess the tufts of fine hairs characteristic of the male, and apparently only attained to the full extent in mature males. The species is found throughout New Zealand and the adjacent islands, in South America (Valparaiso, Island of San Lorenzo, Peru). i think there can be no doubt that Hyale villosa, 8. I. Smith, from Keguelen, and Allorchestes georguanus, Pfeffer, from South Georgia, also belong to the species, and that it is therefore widely distributed in subantarctic seas. Hyale novae-zealandiae (G. M. Thomson). Nicea novae-zealandiae, G. M. Thomson, Trans. N.Z. Inst., xi, p. 235, pl. x B, figs. 1 a-f, 1878. Hyale novae-zealandiae, G. M. Thomson, l.¢., xxvil, p- 211, 1895; Stebbing, “ Das Tierreich Amphipoda,” p. 567, 1906. Specimens obtained from Macquarie Island by Mr. A. Hamilton are referred by Mr. Thomson to this New Zealand species. I have several specimens from the Snares that appear also to belong to this species, which, according to Stebbing, is very near to H. grandicornis, Kroyer, from South America (Valparaiso). 644 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. Hyale campbellica (Filhol),. Allorchestes campbellica, Filhol, “ Mission de l’Ile Campbell,” p. 466, 1885. Hyale canvpbellica, Stebbing, “ Das Tierreich Amphipoda,” p. 562, 1906. Described by Filhol from specimens collected on the shores of Perseverance Harbour, but his description is insufficient for identification. Genus Chiltonia, Stebbing, 1899. Distribution.—New Zealand and adjacent islands, Australia, and Tasmania. Hyalella, a closely allied genus, is common in South America. Chiltonia mihiwaka (Chilton). Hyalella mehiwaka, Chilton, Ann. & Mag. Nat. Hist., ser. 7,1, p. 423, pl. xviii, re 1898. Chailtonia mihiwaka, Stebbing, “Das Tierreich Amphipoda,” p- 555, 1906; Chilton, Trans. N.Z. Inst., xli, p. 57, 1909. A few specimens of this species, both male and female, were collected by Pro- fessor Benham in a fresh-water pool not far from the sea on Enderby Island, and Fig. 11.—Chiltonia mihiwaka (Chilton). lla. First gnathopod of male of Auckland Island specimen. 116. Second gnathopod of male of Auckland Island specimen. others were taken at the exit of a fresh-water pool on Auckland Island; two small female specimens were found by Mr. Laing in a fresh-water stream on Campbell Island. These specimens all evidently belong to the one species, and are almost identical with specimens from the South Island of New Zealand, differing only in having the palm of both gnathopods (figs. 11a, 110) slightly more oblique than in the New Zea- land specimens. Two species of Chiltonia have been described from fresh-water streams in Victoria by Mr. O. A. Sayce; one of these, C. subtenwis, is apparently closely re- lated to C. mehiwaka, but differs in having shorter antennae and a more slender body. More recently Mr. Geoffrey Smith has found species of Chiltonsa abundant in the fresh-water streams and lakes of Tasmania.* * “ Naturalist in Tasmania,” Oxford, 1909, p. 136; and Proc, Roy. Soc. (B), Ixxx, p. 472, 1908, Crustacea.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 645 Genus Allorchestes, Dana, 1849. Distribution.—Widely distributed. Allorchestes novae-zealandiae, Dana. Allorchestes novi-zealandiae (¢), A. wtrepida (3), Dana, Pp. Jane Ac., 1853 and 1855 ; Thomson, Trans. N.Z. ‘Tnst., XX1, p. 260, pl xiii, fig. 3 1889 ; Stebbing, = Das iemeich Amphipoda,” p- 581, 1906 ; Walker, Ann. & Mag. Nat. Hist., ser. 8, 1, p. 39, 1908. Of this species I have several specimens, both male and female, from Ewing Island, collected by Dr. L. Cockayne in July, 1903. It was also taken at Enderby Island by Mr. Hodgson during the voyage of the “ Discovery.” The specimens agree well with the description as given by Stebbing in “ Das Tierreich Amphipoda.”’ The species is widely distributed about the coast of New Zealand, and probably extends also to South America. Fam. AORIDAE. Genus Aora, Kroyer, 1845. Distribution.—Contains only the following species, which is widely distributed in the Atlantic, Pacific, and Southern Oceans. Aora typica, Kroyer. Aora typica, Kroyer, Naturh. Tidsskr., ser. 2, 1, p. 328, pl. iii,' figs.!3 a-l, 1845; Chilton, Ann. & Mag. Nat. Hist., San ead, jo. SO. Tee Stebbing, “Das Tierreich Amphipoda,” p. 587, 1906. Further synonymy will be found under the reference last quoted. A single male specimen from Musgrave Harbour, Auckland Island, taken in seine net (E. R. Waite); also a female specimen collected by Professor Benham. The male specimen belongs to the form which I have described under the name “ Aora typica ¢, form 1” ‘(L.c., p- 370). Under the name Aora typica Mr. Stebbing has united various forms previously known. by distinct names, so that the species as now understood by him has a wide range, being known from the North Atlantic with adjoining seas (Europe), Pacific (South America, Australia, New Zealand), Southern Indian Ocean (Kerguelen Island). It includes at least three different forms of the male, differing in the character of the first gnathopod. Of this, Mr. Stebbing says, “The second joint in adult male (¢ypica) is said to have a triangular process on the front margin,” a state- ment somewhat unduly cautious, seeing that the triangular process in question was described and figured by Nicolet, mentioned by Spence Bate and by G. M. Thomson, and that as far back as 1885 I described and figured it, and also suggested a possible function for it. This particular form of the male is known from the coasts of Chili as well as from New Zealand, but I do not know of any record of its being found elsewhere. In New Zealand it is associated with a second male form, which appears to be the same as the form described as A. gracilis by Bate from British seas ; while in Australia there is a third form, differing in several small details. For these several male forms 646 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. only one form of the female is known, and it is, I presume, mainly for this reason Mr. Stebbing has united them all into one species. In this I am inclined to agree with him; but, if we are to be logical, the same method will have to be adopted in several other cases among the Amphipoda. Genus Lembos, Bate, 1856. Distribution.—In all seas. Lembos kergueleni (Stebbing). Antonoe kerguelem, Stebbing, Rep. “ Challenger,” xxix, p. 1087, pl. exi, 1888; G. M. Thomson, Ann. & Mag. Nat. Hist., ser. 7, x, p. 464, 1902. Lembos kerqueleni, Stebbing, “ Das Tierreich Amphipoda,” p: 598; 1906; Walker, Trans) Linn! Soc.) xm p: 337, ply xlun MieeG: IOS: A specimen that appears undoubtedly to belong to this species was taken at the Snares. Owing to an accident in mounting it, the specimen is in poor condition, but the shape of the first gnathopod of the male so closely resembles the figure and description given by Stebbing that I think there can be little doubt that it belongs here. The rest of the animal, so far as can be made out, is also in general agreement with his description, but the basos of the second gnathopod does not seem to be specially dilated. The species was first taken at Kerguelen Island, and has since been recorded by Mr. G. M. Thomson from the Bay of Islands, New Zealand. Quite recentlv Mr. Walker has recorded it from two localities in the Indian Ocean. I have also another specimen, dredged in about 8 fathoms in Perseverance Harbour, Campbell Island, that evidently belongs to this genus, and is, I think, only a more fully de- veloped male of this species, though the shape of the first gnathopod (fig. 12a) 1s considerably different. The carpus is much shortened and the propod very large, longer than the rest of the limb, and much swollen, forming a broad oval, with the palm oblique; the palm is defined by a stout tooth, and from the centre of the palm there arises a much longer tooth with broadened base, the finger fitting into a groove or depression on the inner side of these two teeth ; a few long setae are found scattered on the palm, and a tuft at the end of the anterior Fig. 12.—Lembos kerguelem (Stebbing). margin near the base of the finger and three 12a. First gnathopod of fully developed male. smaller setae are situated on the inner margin 126. Second gnathopod of fully developed male. of the finger, but the rest of the gnathopod is almost completely free of setae. The second gnathopod (fig. 126) also presents certain differences; the basos is slightly curved, broadly oblong, and has the Crustacea. } SUBANTARCTIC ISLANDS OF NEW ZEALAND. 647 distal angle of the anterior margin subacutely produced; the carpus is nearly as long as the propod and about the same width, and bears on its posterior margin four transverse rows of setae; the propod is suboblong, the width about three- fourths the length, the palm slightly oblique, defined by a prominent tooth, near the base of which is a slight rounded prominence on the palm, the finger fitting into the depression between the two, while the rest of the palm is slightly concave ; numerous transverse rows of setae are found along both margins and also on the side of the propod. The differences in appearance of the gnathopods in this specimen from Stebbing’s figure are so great that one would hardly think of suggesting that it belongs to the same species but for a knowledge of the great changes that may take place in appendages like this, which are specially modified in males; and I think it is very probable that Mr. Stebbing’s description was taken from an immature male, while the specimen described above represents a more mature male of the same species. : Since this was written I have received Mr. Walker’s paper quoted above, in which he also points out that the “‘ Challenger ” specimen was probably immature, and that an adult male from Kerguelen might appear very different from the specimens examined by him. The additional description that he gives agrees, on the whole, with my Snares specimen, and also with the Campbell Island specimen, except as regards the gnathopoda, and thus tends to confirm my opinion that the latter is an adult male of Lembos kergqueleni, Stebbing. This opinion has since been fully con- firmed by an examination of the New Zealand specimens which Mr. Thomson has kindly placed at my disposal, for among them I find forms quite intermediate between that described by Mr. Stebbing and the form with very large propod and oblique palm in the second gnathopod described above. After this was in type I found another specimen of this species in Mr. G. M. Thomson’s collection, in a tube containing specimens of Hyale novae-zealandiae trom Macquarie Island (collected by Mr. A. Hamilton), thus still further extending its range in antarctic seas. Fam. JASSIDAE. Genus Jassa, Leach, 1813-14. MNstribution.——In all seas. Jassa pulchella, Leach. Jassa pulchella, Leach, Edinb. Enc., vu, p. 433, 1813-14; Stebbing, ‘“‘ Das Tierreich Amphipoda,” pp. 654, 739, 1906. Bruzelvella falcata, Norman, Ann. & Mag. Nat. Hist., ser. 7, xvi, pp. 83, 92, 1905 ;; Walker, Trans. Linn. Soc., Zool., xii, p. 348, 1909. Full synonymy will be found in Stebbing’s “ Das Tierreich Amphipoda.” A specimen of this cosmopolitan species was taken on the carapace of Halicar- conus planatus, at Auckland Island. Mr. Thomson and I have frequently taken this species adhering to the body of Jasus edwardsw, and its wide distribution is probably to be partly accounted for by its habit of attaching itself temporarily to this and others of the larger Crustacea. 648 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. Suborder CAPRELLIDEA. Fam. CAPRELLIDAE. Genus Caprellinopsis, Stebbing, 1888. Distribution.— New Zealand and South America (Chili). Caprellinopsis longicollis (Nicolet). Caprella longicollis, Nicolet, Gay’s Historia di Chile, Crustacea, p. 251, pl. iv, fig. 5, 1849. Caprellina novae-zealandiae, Thomson, Trans. N.Z. Inst., xi, p. 247, pl. x v, fig. 6, 1879. Caprellina longicollis, Mayer, “ Caprelliden des Golfes von Neapel,” p. 27, 1882; Caprellidae Siboga Expedition, p. 30, 1903. Caprellinopsis longicollis, Stebbing, Rep. “Challenger,” xxix, p. 233, 1888. Two female specimens from the Snares. This species is known also from all the coasts of New Zealand, from South America (Chili), and from South Africa. Order ISOPODA. Suborder ASELLOTA. Fam. PARASELLIDAE. Genus Janira, Leach, 1814. Distribution.—Widely distributed in northern seas; known, as yet, only from New Zealand and Auckland Islands in the south. Janira neglecta, sp. nov. Specific Diagnosis. — Body oblong-oval, somewhat narrow, rather less than three times as long as broad; cephalon with sides nearly straight, posterior angles rounded, front very slightly convex between the bases of the antennules. Lateral margins of all the segments of the peraeon straight or slightly convex, entire, quite covering the coxal plates. Pleon oval, narrowing posteriorly, about as wide an- teriorly as it is long; margins smooth, entire, posterior margin very slightly produced between the bases of the uropoda. Hyes large, of ro. 7 / 3 df » Lif i 13 C j 2 13a. Leg of first pair of male. ee which contains about ten iss Ben nd onde joints. numerous ocelli. Antennules reaching to the end of the pen- ultimate jomt of peduncle of Antennae about as long as body; last two joints of peduncle subequal, with smooth margins with a few fine setae; the scale on the third joint narrow but dis- tinct ; flagellum much longer than peduncle. First pair of legs (fig. 13a) with the the antennae; first peduncular jomt broad, nearly circular ; basos slightly longer than the ischium and as long as the carpus, merus only half as second nearly as long, but much narrower ; third shorter, hardly distinguishable from the flagel- Fic. 13.—Jamra neglecta, sp. nov. Se ge re OO eee Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 649 long as the ischium, carpus slightly expanded in its distal half, the immer margin of this portion bearing about ten setae ; the dactylar claws of all the legs subequal. Uropoda (fig. 136) equal in length to the pleon ; base slightly expanded distally, as long as the outer ramus, which is about two-thirds the length of the inner; both rami rather slender, and provided with small tufts of long slender setae. Colour, yellow, dorsal surface densely mottled with dark markings. Length, 3 mm. Hab.—Carnley Harbour, Auckland Islands, 2 fathoms (Professor W. B. Benham) ; also known from Port Chalmers, and from Lyall Bay, Wellington, New Zealand. The Carnley Harbour specimen is a male, but probably immature; it has the first pair of legs less well developed than is shown in fig. 18a, which is taken from a Port Chalmers specimen. I have had specimens of this species from both Port Chalmers and Lyall Bay for many years, but it has hitherto remained undescribed ; in genera! appearance and in the appendages it appears to approach pretty closely to Jamra maculosa, Leach, the species common in the northern seas, but it differs in having the margins of the peraeon and pleon entire, and apparently the body is somewhat narrower than in that species. Genus Iais, Bovallius, 1886. Distribution.—Subantarctic seas. Iais pubescens (Dana). Jaera pubescens, Dana, U.S. Expl. Exped., xii, p. 744, pl. xlix, fig. 9, 1853; 8S. I. Smith, Bull. U.S. Nat. Mus., No. 3, p. 63, 1876. Jais pubescens, Stebbing, P.Z.S., 1900, p. 549, 1900; “‘ Spolia Zeylanica,” u, pt. v, p. 11, 1904. Full reference to the literature dealing with this much-described species up to the year 1900 will be found in Mr. Stebbing’s first paper quoted above. Numerous specimens were taken both at Auckland Islands and at Campbell Island on Hxosphaeroma gigas (Leach), and many were collected at Campbell Island creeping freely on the under-surface of stones in places where the Hxosphaeroma was abundant. These specimens agree closely with others taken under similar circum- stances at various parts of New Zealand. I have been able to compare them with specimens from Falkland Islands, and I find that they agree as closely with these also, as might naturally have been expected from the description given of these Falkland Island specimens by Mr. Stebbing. This little species, like its host, Hxosphaeroma gigas, 1s therefore widely distri- buted round the globe in subantarctic seas, while Mr. Stebbing also records it from Lake Negombo, in Ceylon, where it appears to have been associated with Sphae- roma terebrans (Bate). Its connection with the host has generally been spoken of as parasitic or semi-parasitic, though it perhaps would be more correctly described by the word “commensal.” Mr. Stebbing says of it, “ Apparently the small isopod makes use of the large one as a kind of floating island, affixing its eggs to it, and in adult life still clinging on but doing no harm to its animated lodging.” * In stating that the Zais attaches its eggs 's to the ESC, Mr. eel et pomp orarily © over- * P.Z.8., 1900, p. 550. 650 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. looked the fact that, like most other Isopods, Zais hatches out its eggs in an incubatory pouch under its own body. It is evident that he knew that this was actually the case in [ais pubescens from his reference to the marsupium upon the same page. The eggs are few in number (usually not more than five or six) and of comparatively large size, in both respects, therefore, differing from the conditions of things usually found in true parasites. The /avs can evidently leave the Hxosphaeroma if it wishes to, as is shown by the fact that I have frequently taken it away from the Hzosphaeroma. It would be interesting to know whether this separation is merely temporary for the purpose of seeking food, and if the Zazs returns to the Exosphaeroma for shelter when danger threatens. During a recent visit to the Sounds on the west coast of New Zealand I found Tas pubescens in great abundance in nearly all of them. I seldom failed to find it at the head of each Sound, creeping freely on the surface of stones at the mouths of the fresh-water streams; at low tide these stones would be washed with fresh water only, and some were above the reach of ordinary high tides. At high tide the animals would in most cases be covered with sea-water, but this at the head of the Sounds is often more or less brackish. In most of these localities I did not see any Sphaeromid from which the animal could have escaped. The Campbell Island speci- mens were also mainly collected near the mouth of a small stream, but in a place that would be completely covered with salt water at high tide. Mr. Stebbing’s Ceylon specimens were obtained from Lake Negombo, which Dr. Arthur Willey describes as “a salt-water lake having both fluviatile and marine connections.” It thus appears that [avs pubescens is a species that can live either free or in association with a Sphaeromid, and either in fresh water or in salt, and to this adapta- bility to varying conditions 1t doubtless owes its wide distribution and abundance. Genus Haliacris, Pfeffer, 1886. Distribution.—Antarctic and subantarctic seas. I provisionally adopt this genus for the following species merely to draw attention to its close relationship to H. antarctica, Pfeffer. The genus has been accepted by Miss Richardson, and provisionally by Mr. Hodgson, but in his report on the “Discovery ” Isopoda he says, “ It is much open to question if it is distinct from Munna. I think not.”* And with this opinion I quite agree. Haliacris neozelanica (Chilton). Munna neozelanica, Chilton, Ann. & Mag. Nat. Hist., ser. 6, 1x, p. 1, pl. 1, 1. Auckland and Campbell Islands ; also New Zealand. Several specimens—some males with the first pair of legs fully developed, others females with eggs in the brood-pouch—were taken by Professor Benham under stones on Masked Island, in Carnley Harbour, Auckland Islands. He had previously sent me one male specimen collected at Campbell Island in February, 1907. These specimens all agree closely with specimens from the type locality, Otago Harbour. This species is close to Haliacris antarctica, Piefier, from South Georgia, with which H. australis, Hodgson, from Cape Adare, is now united by Hodgson. Judging * Quoted from a proof of his report, which Mr. Hodgson has kindly sent to me. Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 651 from the figures and descriptions given by these two authors, the females of the two species are probably almost identical. Miss Harriet Richardson has recorded H. australis from Booth Wandel Island, and has described the male, which was previously unknown, as Hodgson’s three speci- mens were all females. In both species the male differs from the female in the great size and peculiar form of the first pair of legs, and from the figure given by Miss Richardson it is evident that these appendages in H. neozelanica (fig. 14a) differ from those of H. antarctica, being, as her figure (see fig. 14b) shows, of distinctly dit- Fic. 14.—Haliacris neozelanica. ferent shape. Pfeffer does, indeed, say that 14a. Haliacris neozelanica (Chilton), first leg of male. the males and females of his species resemble ae oe pestle. oy etal each other in the structure of the first pair of legs; but, as Miss Richardson says, it is probable that his specimens “ wer ganz schlechte Exemplare”’ were all females, and that the mature male with the fully developed first pair of legs was really unknown to him. Suborder FLABELLIFERA. Fam. CyMOTHOIDAE. Genus Cirolana, Leach, 1818. Distribution.—In all seas. Cirolana rossii, Miers. Crrolana rossw, Miers, Ann. & Mag. Nat. Hist., ser. 4, xvii, p. 228, 1876, and Cat. N.Z. Crust., p. 109, pl. ii, fig. 3, 1876; Hutton, Trans. N.Z. Inst., x1, p. 340, 1879. A few small specimens were obtained in Perseverance Harbour, Campbell Island, dredged in 8 fathoms; others, from Auckland Island, are in Mr. G. M. Thomson’s collection; and the species has also been recorded from Auckland Islands by Hutton. It is common on the New Zealand coasts, and is closely allied to, uf not identicai with, C. hirtupes M.-Edwards, from the Cape of Good Hope. Genus Livoneca, Leach, 1818. Distribution.—In all seas. Livoneca novae-zealandiae, Miers. Lironeca novae-zealandiae, Miers, Ann. & aa Nat. Hist., ser. 4, xvil, p. 227, and Cat. N.Z. Crust., p. 106, pl. ii, fig. 2, 1876; Filhol, “ Mission de l’Ile Campbell,” p. 450, pl. ly, he. 1, 1885. L. stewart, Filhol, “Mission de I’fle Campbell,” p: 450, pl. lv, fig. 6, 1885. Lwvoneca raynaudu, Whitelegge, “‘ Thetis” Scientific Results,” Australian Museum Memoir iv, pt. 11, p. 236, 1901. One specimen, found on Notothenia colbecki, from Antipodes Island (KE. R. Waite). This is a common species on the New Zealand coasts, where it is found parasitic on 652 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustaced. various species of fish. JL. stewarti, Filhol, is, I think, hardly sufficiently distinct to be maintained as a separate species. According to Mr. Whitelegge, L. novae-zealandiae was taken by the “ Thetis ” Expedition at several localities off the coast of Australia. I have also a small specimen from Norfolk Island that does not differ specifically from the New Zealand species. It has also been recorded by Miers from Portland Bay, Straits of Magellan.* Whitelegge unites this species with L. raynaudu, Milne- Edwards, from the Cape of Good Hope; this may very probably be correct, but Milne-Edwards’s original description is very short, and comparison of actual specimens is desirable before finally reducing L. novae-zealandiae to the rank of a synonym. Fam. SEROLIDAE. Genus Serolis, Leach, 1825. Distribution.—Widely distributed in subantarctic seas. Serolis latifrons, Miers. Serolis latifrons, Miers, Ann. & Mag. Nat. Hist., ser. 4, xvi, p. 74, 1875, and Cat. N.Z. Crust., p. 117, pl. i, fig. 7, 1876; Rep. “ Challenger ” Tsopoda, p. 44, pl. 1, figs. 1-4, 1884; 8. I. Smith, Bull. U.S. Nat. Mus., No. 3, p. 63, 1876. Recorded from “‘ Rendezvous Cove, Auckland Islands,” by E. J. Miers from specimens in the British Museum. ‘This species is known also from Kerguelen and the Crozets. Fam. SPHAEROMIDAE. Genus Exosphaeroma, Stebbing, 1900. Exosphaeroma, Stebbing, P.Z.8., 1900, p. 553. Distribution.—Widely distributed in southern seas. Exosphaeroma gigas (Leach). Sphaeroma gigas, Leach, Dict. Sc. Nat., xu, p. 346, 1818; Miers, Cat. N.Z. Crust:, p. 10; sve; Hutton, Wrans: INZo Insts, x1, ps 40; soe Haswell, Cat. Aus. Crust., p. 287, 1882; G. M. Thomson, Proc. Roy. Soc. Tasmania, p. 14, 1893; 8S. I. Smith, Bull. U.S. Nat. Mus., No. 3, p. 63, 1876. 8S. obtusa, Hutton, Trans. N.Z. Inst., xi, p. 341, 1879. Hxosphaeroma gigas, Stebbing, P.Z.S., 1900, p. 553, 1900, and “South African Crustacea,” u, p. 69, 1902; Hansen, Quart. Journ. Micr. Sci., xlix, p. 118, 1905. The species is widely distributed in subantarctic seas. Full references to the earlier papers dealing with this species will be found in the first of Mr. Stebbing’s papers quoted above; in it he considered Sphaeroma lanceolatum, White, as a synonym of Sphaeroma gigas, Leach, but in the paper on the “South African Crustacea ”’ he says, “On the whole, it now seems to me that the two forms ought to be kept specifically separate, and that the names allotted by Leach and White may conveniently stand, although 1t may not be absolutely certain which of the forms Leach had before him.” *P.Z.8., 1881, p. 77. A Bees . a Crustacea.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 653 Of this widely distributed and much-described species I have been able to examine numerous specimens from the Auckland Islands, Campbell Islands, and from the Macquaries, and to compare them with specimens from the main islands of New Zealand and from the Falkland Islands. My specimens vary very much in size, the largest being as much as 25 mm. in length ;* most of them are of a slaty colour, but some are of a light brown, with markings as described by Stebbing for the Falkland Island specimens. I have com- pared them with Mr. Stebbing’s description, and can find no real point of difference ; the straight ridge which he describes separating the vertex of the head from the occiput is well marked in the larger specimens, but indistinct or almost absent in the smaller ones, the front of the head being in them more produced and the angles of the first peraeon segment not produced quite so far anteriorly. There is very considerable variation in the shape of the terminal segment of the pleon : in some cases it is somewhat narrowly rounded and in others much broader. I noticed this when collecting specimens from the one locality on the shore of Perse- verance Harbour, Campbell Island, but I cannot associate the differences with that of sex. In the same way the amount of in-folding of the lateral margin of the terminal segment of the pleon on the ventral side also varies considerably. Captain Hutton identified some specimens collected from Auckland and Camp- bell Islands with Sphaeroma obtusa, Dana. Through the kindness of Professor Benham I have been able to examine these specimens, but find they are only Exosphaeroma gigas in which the terminal segment is somewhat narrow. Similar specimens labelled “S. obtusa”? were also in Mr. Thomson’s collection, which has been placed at my disposal. I have never satisfactorily identified any specimens with Dana’s S. obtusa. Genus Pseudosphaeroma (novum). This genus comes under the division Sphaerominae eubranchiatae, Hansen, though it differs from Hansen’s description of that group in the characters of the posterior end of the pleon, and to some extent also in the pleopoda, and may be defined as follows :— Exopod of third pleopod not articulated. Basal joint of first antennae not expanded, not much broader than the second. Uropoda with both branches equally developed in both sexes, and ovoid in shape. The male differmg from the female in having small tubercles on the first portion of the pleon, and two well- marked tubercles near the anterior margin of the second portion. End of pleon without emargination, ending in a flat or upturned projection. Female without tubercles on the pleon and with terminal portion less produced. Mouth parts not modified in the female. Hggs developed in internal pouches in the general body cavity. This genus, though by the character of the pleopods undoubtedly belonging to the Sphaerominae eubranchiatae, differs from it, and resembles certain species of Sphaeroma, Exosphaeroma, &c., in having the terminal segment quite without emargination, and the exopod of the fifth pleopoda mainly opercular and with only a n)} * Stebbing gives the leneth of his Falkland Island specimens as “ about 18 mm.,” while the speci- mens from Kerguelen examined by 8. I, Smith were “ of all sizes, from 5 mm, to 29 mm, in length,” 654 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Crustacea. few poorly developed transverse folds on the inner part near the base, so that at first I thought it was a species of Hxosphaeroma, and began to describe it as such. Except for the sexual differences and the lack of emargination in the terminal seg- ment, it seems to come pretty close to Dynamenella. Monsieur K. G. Racovitza* has recently described a genus Ischyromene which seems to be somewhat nearly re- lated, and also shows sexual differences, but these are of a different character to those found in the genus now under discussion. The type species is P. campbellensis, described below. Pseudosphaeroma campbellensis, sp. nov. Male.—Body fairly convex, widening a little posteriorly; head with the front depressed, vertex bounded in the front by a slight ridge and produced a little between the bases of the upper antennae, surface of head slightly uneven. First three segments of peraeon subequal, a little longer than the fourth, which is longer than any of the following three segments, the suriace of all nearly smooth, but with some slight granulations, specially towards the sides and on the last three segments. First portion of the pleon with two sutures at the sides, and bearing near the centre two small rounded tubercles. The second division of the pleon with two well- marked tubercles in the centre near the anterior margin, each tubercle being bifid at its extremity ; the segment is produced posteriorly, with the posterior border broadly truncate, with rounded angles and hinder portion usually sharply upturned, but im some cases nearly horizontal; whole surface of the pleon finely granular. The eyes rather small, subtriangular, occupying the postero-lateral angles of the head. First antenna with the first jomt large and broad, with mdentation on pos- terior margin, second joint nearly as broad but much shorter, the third much longer than second but much narrower, flagellum consisting of about eight jomts, which bear only a few setae. Second antenna longer than the first, last two joints of peduncle subequal and longer than the preceding one, flagellum containmg about eleven joints. First pair of legs shorter than the succeeding pairs, the carpus being shorter than the ischium and subtriangluar in outline, a few fine woolly setae on the inner surface of the merus, and larger setae on the ischium, carpus, and propod ; the suc- ceeding pairs of legs rather longer, increasing slightly m length posteriorly, with the carpus nearly as long as the merus, and the inner surface of the merus, carpus, and propod thickly frmged with short woolly setae; longer setae on the other joints, as shown in figure. The uropods with the inner branch reaching almost or quite to the end of the terminal segment, outer branch slightly shorter, both oval, with the extremities rounded. Female.—With the posterior segment of the pleon less produced and only slightly upturned, and without the tubercles on the dorsal surface. The immature males resemble the females in these points, but may have the tubercles on pleon partially developed. *“ Archives de Zoologie expérimentale et générale ” [4], 1x, notes et revue No. 3, p. Ix, figs. 1 to 3, 1908. Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 655 Colour.—Greenish-brown, with the margins of the segments reddish. Length of male, 7 mm.; breadth, 3 mm. Hab.—Perseverance Harbour, Campbell Island (November, 1907). Numerous specimens taken on the shore, at the mouth of a small fresh-water stream, in company with Hxosphaeroma gigas ; Auckland Island (Dr. L. Cockayne, 1903). Type in Canterbury Museum, New Zealand. The Auckland Island specimens are apparently nearly all immature; in the largest male the two tubercles on the terminal portion of the pleon are present, but are not so well marked, and are not bifid at the extremity, and in all the specimens the terminal segment 1s less broadly truncate. Fie. 15.—Pseudosphaeroma campbellensis, sp. nov. 15a. Antennae, with epistome and upper lip. 15d. Pleon and uropoda of male (from above). 15b. First leg of male. 15e. Pleon and uropoda of male (from below). 15c. Seventh leg of male. The mouth parts are, on the whole, very similar to those of Hxosphaeroma gigas. The epistome widens downwards, and the upper lip has the distal border straight. The mandible has the incisor process ending in one or two teeth, accessory plate stronger on the left than on the right mandible and ending in two or three teeth, the molar process strongly developed ; the palp not reaching beyond the extremity of the incisor process, its first segment broad, second as long as the first but narrower, with a row of pectinate spines, third rather strongly curved, the concave border with a row of short spines, those towards the end being longer and pectinate. First maxilla with the imner plate narrowing towards the end, which bears four plumose setae ; outer plate with about nine teeth at the extremity, the inner ones being smaller 656 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. and pectinate or dentate, the outer ones broad and smooth and brown in colour. Second maxilla with the two outer lobes subequal, each bearing at the end four strong pectinate setae; inner lobe as long as the two outer, its extremity oblique and fringed with numerous short plumose setae ; a few short setae also on its ner margin. The maxillipedes with the basos a little widened at the base and with its outer margin fringed with fine setae, the end of its lobe bearing six or seven stout short setae and numerous fine hairs; third joint short; fourth, fifth, and sixth produced on the inner side into a rounded lobe fringed with setae ; the seventh bearing on its apex five long plumose setae and some shorter ones. Fie. 16.—Pseudosphaeroma campbellensis, sp. nov. 16a. First pleopod of male. 166. Second pleopod of male. (Plumes on setae omitted.) l6c. Third pleopod of male. (Plumes on setae omitted.) 16d. Fourth pleopod of male. l6e. Fifth pleopod cf male. In a female in which the interior of the body was completely filled with eleven young the mouth parts were a little feebler than in the male, but were not specially modified. In the male the first pleopod has three coupling setae on the inner margin of the basal portion, the outer margin being fringed with fine setae; the exopod is long, oval, and is slightly longer than the triangular endopod ; both have the extremities and outer margins fringed with the usual long plumose setae, The second pleopod Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 657 is similar to the first, except as regards the endopod, which bears the male appendages ; these appendages are narrow, longer than the endopod, and widen slightly beyond its extremity, and then narrow again to the narrowly rounded apex. The third pleopod has the exopod oval, as in the preceding pleopods, but slightly shorter than the broadly triangular endopod, both consisting of one joint only. The fourth pleopod has the endopod provided with transverse folds, the exopod with outer margin fringed with numerous short setae and the transverse folds rather poorly developed on the inner half only. The fifth pleopod is similar in general shape to the fourth, but has the exopod 2-jomted and bearing the squamous protuberances as usual. The exopod is mainly opercular, and bears only a few poorly developed transverse folds on a rounded lobe of the inner margin near the base. Genus Cymodocella, Pfeffer, 1888. Mstribution.—Subantarctic seas. Cymodocella tubicauda, Pfeffer. Cymodocella tubicauda, Pieffer, Jahrbuch d. wissensch. Anstalten zu Televonlomen@s hy, jon IOS oll vin, inner, ee) oll Sealy tne NTIS IS Sphaeroma egregia, Chilton, Trans. N.Z. Inst., xxiv, p. 209, 1891. Cymodocea antarctica, Hodgson, “Southern Cross” Crust., p. 243, pl. xxxin, fig. 2, 1902. Cymodocella egregia, Hansen, Quart. Journ. Mier. Sci., xlix, p. 126, 1905; Richardson, Expéd. antarct. francaise, 1903-5, Isopodes, p. 7, 1907. C. tubscauda, Richardson, J.c., Iso- podes (2° mémoire), p. 4, 1907; Hodgson, Rep. “ Discovery” Iso- poda, p. 31. Recorded by Mr. Hodgson from the Auckland Islands from specimens collected by the “Southern Cross.” In his report on the Isopoda collected by the National Antarctic Expedition Mr. Hodgson records this species from Cape Adare, and gives a full description of it. It was originally described from specimens from South Georgia, and the French Antarctic Expedition of 1903-5 took it at Booth Wandel Island, and at Wincke Island, Flanders Bay. The species has therefore a wide range in subantarctic seas, and, according to Mr. Stebbing, is probably identical with Cymodocella (Sphaeroma) algoense, Stebbing, 1875, from Algoa Bay, South Africa. Genus Dynamenella, Hansen, 1905. Dynamenella, Hansen, Quart. Journ. Micr. Sci., xlix, p. 107. Distribution.—Widely distributed. Dynamenella huttoni (G. M. Thomson). Dynamene hutton, G. M. Thomson, Trans. N.Z. Inst., xi, p. 234, pl. xa, fig. 6, 1879. Cymodoce huttoni, Chilton, Trans. N.Z. Inst., xxxviii, P22.) 1906. Several specimens from Antipodes Islands, collected by Dr, Cockayne in July, 1903. Very common on the New Zealand coasts, 43—S, 658 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. This species appeared to me to be very closely allied to or identical with Dyna- mene eatonr, Miers, described from Kerguelen Island, and afterwards recorded by Dollfus from South America (“ Mission du Cap Horn,” Crustacea, p. F 66). Dr. Calman has, however, kindly compared specimens sent to him with co-types of D. eatoni, and finds that, although very closely allied, D. huttona is distinct in several respects: thus, D. eatoni is much less convex, and has the front margin of the head produced into a marked ridge hardly visible in D. huttoni ; the first segment of the peraeon is much broader as compared with the head ; the third segment of the peduncle of the antennule is longer and more slender, and on the terminal portion of the pleon the sinuous transverse groove found in D. huttona in addition to the seg- mental grooves which run in from the sides 1s absent. Suborder VALVIFERA. Fam. IDOTEIDAE. Genus Idotea, Fabr., 1798. Mstribution.—In all seas. Idotea elongata, Miers. Idotea elongata, Miers, Ann. & Mag. Nat. Hist., ser. 4, xvu, p. 225, 1876; Chilton, Trans. N.Z. Inst., xxi, p. 198, 1890. Two specimens, one male and one female, from Musgrave Harbour, Auckland Islands (EH. R. Waite). The female has the brood-pouch full of eggs, and the anterior segments of the trunk expanded, but not to so great a degree as it is sometimes met with. In the mature female with eggs in the brood-pouch the peraeon is “ much dilated in the middle, the second, third, and fourth segments being progressively broader and bluntly angled at the sides, fifth suddenly narrowing to less than half the width of the fourth.” The lateral suture on the pleon is often very indistinct, so that the pleon is almost or quite uniarticulate. The type specimens of this species are in the British Museum, and came from the Auckland Islands. It is common on the New Zealand coasts, and is always found on brown seaweeds, to which in colour and markings 1t presents a close resemblance. Miers has also recorded it from the Falkland Islands. Idotea lacustris, G. M. Thomson. Idotea lacustris, Thomson, Trans. N.Z. Inst., x1, p. 250, 1879; Miers, Journ. Linn. Soc., xvi, p. 39, pl. 1, figs. 11, 12, 1881; Chilton, Trans. N.Z. Inst., xxl, p. 194, 1890, and xxiv, p. 263, 1892; N.Z. Journ. Sci., TMS ily 06 UBIL, WSO A few specimens were gathered at Campbell Island: some, collected by Pro- ‘fessor H. B. Kirk, were taken in the lake or lagoon at the south end of the island, into which salt water may be driven at exceptionally high tides; others were gathered in fresh-water streams by Mr. R. M. Laing at a height of about 600 it. above sea-level. I have also specimens collected by Professor Benham eo a fresh- water creek on Mount Honey in February, 1907, Crustacea. ]} SUBANTARCTIC ISLANDS OF NEW ZEALAND. 659 The specimen from the lagoon is quite similar in colour and other external features to specimens taken at Tomahawk Lagoon and other places in the South Island of New Zealand where the water, though usually fresh, may at times be more or less brackish. The others, gathered in Campbell Island in fresh-water streams at a considerable elevation above sea-level, are rather lighter in colour, and differ slightly in the size of the eyes and in a few other points; though these differences are in the same direction, they are not so great as those exhibited by the specimens taken in mountain-streams around Dunedin, which I have distinguished as variety $8. This variety differs in the followmg points from the typical form, variety a, found at sea-level: (1) The eyes are much smaller, being only about half as large; (2) the front margin of the head has a slight depression in the centre ; (3) the inner antennae are rather more slender and longer, reaching to the end of the third joint of the peduncle of the outer antennae ; (4) the outer antenna is more slender, both in the peduncle and in the flagellum; (5) there is only one pair of sutures on the terminal segment of the pleon ; (6) the extremity of the pleon is more narrowed ; (7) the colour is usually lighter, being a light brown, with darker spots. and markings. I am well aware that the numerous small differences between these two varieties are quite sufficient to warrant one in making two species of them; but from the con- ditions under which the two forms are found it is perfectly clear that they are very closely related, one form having been derived from the other probably quite recently, and there is little doubt that the differences between them are associated with the different conditions under which they live—one in brackish water on the sea-shore, and the other in mountain-streams at a considerable elevation above the sea. If a separate species were made of the latter form its connection with the first one would be in danger of being overlooked; and, moreover, as I have already pointed out, the fresh-water forms from Campbell Island, though they are similar to those from the fresh-water streams of New Zealand, are not quite identical, and would pro- bably require to be also distinguished by a different specific name. In New Zealand the species is common in the southern part of the South Island and the adjacent islands. It was originally taken by Mr. Thomson in Tomahawk Lagoon, near Dunedin ; subsequently I took it at the mouth of a small stream running into Otago Harbour, and found the variety 6 in various mountain-streams around Port Chalmers and Dunedin. I have also specimens of variety a in my collection from Ruapuke Island (collected by Mr. T. Horan), and from Port Pegasus, Stewart Island (Dr. L. Cockayne), and more recently I found it in the West Coast Sounds of Otago at the mouth of nearly every fresh-water stream that I was able to examine. I have not succeeded in finding it at Lyttelton or on Banks Peninsula, though I have examined many likely places in these localities. A form which apparently belongs to this species is found at Port Henry, Straits of Magellan, and is represented by specimens in the British Museum, which were referred to I. lacustris by Miers in 1881. Miers gives one or two‘small points in which these specimens differ from Thomson’s original description, but these are not so great as the differences between varieties a and 8. Since the above remarks were written I have sent specimens of both varieties to Dr. W. T. Calman, of the British Museum, who has kindly compared them with the specimens in the Museum collections from the Straits of Magellan, and he informs 660 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. me that the brackish-water form (var. a) appears to be identical with the Straits of Magellan specimens, which have two sutures on the terminal segment of the pleon, Miers’s figure, which shows only one, being incorrect in this particular. Dr. Calman also says that the fresh-water variety seems to be about as good a species as most species are ; but for the reasons which I have given above I think it less misleading to look upon it merely as a variety which has risen from the brackish-water form, pro- bably comparatively recently. Untortunately, it is not known if the Patagonian specimens are from salt, brackish, or fresh water; but the facts now before us seem to show that Idotea lacustris is a species widely distributed on subantarctic shores, and is to be found chiefly in brackish water, but has in more than one place ascended fresh-water streams to a considerable height, and become slightly modified, in accord- ance with the different conditions to which 1t was exposed in these situations. Genus Paridotea, Stebbing, 1900. Paridotea, Stebbing, South African Crust., pt. 1, p. 58, 1900. Distribution.—Southern seas. Ihave followed Stebbing in adopting the genus Paridotea, the one of the main characters on which he founded it (the amount of segmentation indicated in the pleon) is subject to variation even within the limits of a single species, as I have elsewhere shown,* and the custom of establishing a new genus for a single species, unless done after full review of allied species, is apt to lead to needless complication, by making necessary the creation of other genera on equally small and unimportant points of structure. Paridotea ungulata (Pallas). Oniscus ungulatus, Pallas, Spicil. Zool. Fasc., ix, p. 62, pl. iv, fig. 11, 1772. Idotea ungulata, Miers, Journ. Linn. Soc. Zool., xvi, p. 52, 183i; Chilton, Trans. N.Z. Inst., xxu, p. 196, 1890. Paridotea ungulata, Stebbing, South African Crust., pt. 1, p. 53, 1900, and pt. i, p. 56, 1902; Chilton, Trans. N.Z. Inst., xxxvii, p. 272, 1905. One specimen was taken in Carnley Harbour by Professor Benham; it was taken on Ulva, and was, like the Ulva, of a bright-green colour. The species is widely distributed in the southern seas, being known from New Zealand, South Australia, Chili, Rio Janeiro, South Africa, and the Indian Ocean. In the paper last quoted I have given some account of the sexual differences in this species. Suborder ONISCOIDEA. Fam. TRICHONISCIDAE. Genus Trichoniscus, Brandt, 1833. Distribution.—World-wide. The section of the genus to which the following species belong is confined to lands in the Southern Bee * Trans, N.Z, Inst., xxii, pp. 197, 199, and 201. a i Crustacea. ]} SUBANTARCTIC ISLANDS OF NEW ZEALAND. 66]. Trichoniscus thomsoni (Chilton). Philygria thomsoni, Chilton, Trans. N.Z. Inst., xvui, p. 159, pl. v, figs. 1-6, 1886. Trichoniscus thomson, Chilton, Trans. Linn. Soc., vi, p. 118, pl. xin, fig. 1, 1901; Budde-Lund, Deutsche Siid-polar Exped., 1901-3, ix, p. 83, pl. iv, figs. 22-24, 1904. A few specimens obtained at Auckland Islands appear to belong to this species. They differ slightly from specimens from the main islands of New Zealand in having the surface of the body somewhat rugose, and in having the fiith joint of the peduncle of the antennae roughened with setose tubercles on its mner side. They are, however, too closely related to be distinguished as a separate species. Dr. Budde-Lund has pointed out that the species of Trichoniscus from New Zea- land, Tierra del Fuego, and other places in southern lands belong to a section of the genus which is distinguished by having the three ocelli of the compound eye sepa- rated from one another, and to this section he has added a new species, 7. verrucosus, from the Crozet Islands. Dr. Budde-Lund has also pointed out the near relation- ship between 7. thomsona and T. magellanicus, and the existence of the latter species on Auckland and Campbell Islands has enabled me to confirm what he says. T. thomsoni differs, however, in having the segments of the peraeon broader, the lateral parts especially being greatly dilated, and, as a result, the body is less convex than in T. magellanicus ; the uropoda, though closely similar, are hardly so long and slender as in 7. magellanicus. Trichoniscus magellanicus (Dana). Styloniscus magellanicus, Dana, U.S. Expl. Exped., xin, Crust., p. 736, pl. xlviu, figs. 7 a-g, 1853; Dolltus, “ Mission du Cap Horn,” Crust., p. F92, fig. 14 a-c. Trichoniscus magellanicus, Stebbing, P.Z.S., 1900, p. 566; Budde-Lund, Deutsche Stid-polar Exped., 1901-3, 1x, Zool., 1, p. 83, pl. iv, fig. 25, 1904. Several specimens from Auckland Island, and some also from Campbell Island, must, I think, be referred to this species, which is also known from Tierra del Fuego and from the Falkland Islands. My specimens agree very closely with the description of the species given by Stebbing, the only point of difference that I can find being that the flagellum of the outer antenna contains only six joints, distinctly marked, while his specimens were 7-8-joimted. The species also seems very close to 7. verrucosus, Budde-Lund, from the Crozets, which has four to six joints of the flagellum of the antenna; that species, however, appears to have the surface more rugose than in my specimens. Probably T. verrucosus will prove to be either identical with 7’. magellanicus or only a local variety of this widely distributed subantarctic species. Genus Haplophthalmus, Schobl, 1860. Distribution.—As yet recorded only from Europe and America and New Zealand, but I have also an undescribed species from Tasmania. The New Zealand species differ from the generic characters in one or two points (see below). 662 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. Haplophthalmus australis, sp. nov. Body oblong-oval, about half as broad as long, the epimeral plates of the peraeon being large and projecting horizontally, those of the first four segments somewhat widely separated. The head with the front triangularly rounded and tipped with small denticles, two rounded tubercles being situated about the centre of the head and an indistinct longitudinal ridge on each side ; lateral lobes small and subacute. Dorsal surface of the peraeon moderately convex, each segment with the surface somewhat roughened, and bearing a number of rounded tubercles ; these form a fairly well marked median ridge and a less well marked lateral ridge on each side, with indications of another poorly marked ridge one each side external to this. Surface of the pleon (fig. 17g) without tubercles, first three segments short, the first two without epimeral projections, the third with very small narrow epimera, fourth and fifth with epimera large and well developed, terminal segment broad, triangular, posterior border, straight. Antennae with the fourth joint of the peduncle as long as the two preceding and rather shorter than the fifth ; flagellum almost as long as the fifth, consisting of four or five joints, the last tipped with a pencil of long setae; the whole antenna covered with fine short setae. Uropoda with the branches equal in length, but the inner slightly more slender than the outer ; both covered with fine setae, and tipped with a pencil of long hairs. Length, 6mm.; greatest width, 2-5 mm. Colour.—Lightish-brown. Hab.—Campbell Island, on decaying wood and at roots of plants. Type in Canterbury Museum, New Zealand. The animal runs with great rapidity, and, although several were seen, I was able to secure only a very few specimens; further specimens were, however, afterwards obtained by Messrs. Chambers and Des Barres. Another species, Haplophthalmus helmsw, Chilton, has been described from Grey- mouth, in New Zealand. It differs from the generic characters given by Sars in having the first three segments of the pleon short and without epimeral expansions. In the present species the first three segments are also short, the first two quite with- out epimeral expansions, but the third has very small narrow expansions; in this species, too, the eyes consist of three ocelli, instead of only one, as given in Sars’s description ; there are also some slight differences in the maxillipedes, which approach still more closely than in other species to those of Trichoniscus. It would be easy, and perhaps justifiable, to place the New Zealand species under a new genus, and I had at one time thought of doing this, in the belief that all the southern species would differ from the northern species in the points mentioned. I have, however, an undescribed species from Tasmania which agrees with the northern species of Haplophthalmus in having only the first two segments without the epimeral expansions, while, as I have poimted out above, H. australis is to some extent termediate as regards the characters of the pleon, and, as there are only a few species hitherto described in this genus, I prefer to slightly widen the generic characters rather than to create a new genus. Sars has pointed out that Haplophthalmus comes very near to Trichonscus, and the species now described approaches still nearer, in having the eyes formed of three ocelli and in the character of the terminal portions of the maxillipedes. Crustacea. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 663 I append a more detailed description of the appendages. The antennules (fig. 17a) have the first joint large, slightly curved; the second joint not quite half the length of the first ; third joimt nearly twice as long as second, but very slender, tipped with about five or six sensory setae, but otherwise without setae ; the first and second joints bear a few fine setae. Fic. 17.—Haplophthalmus australis, sp. nov. 17a. Antennule. 17e. Second pleopod of male. 17b. Antenna. 177. Third pleopod of male. 17c. Seventh leg of male. 17g. Pleon and uropoda (from above). 17d. Extremity of same (more highly magnified). The antennae (fig. 17b) have the first three jomts increasing in length, the fourth as long as the two preceding and rather shorter than the filth, which is slightly narrowed and a little curved at its proximal end ; the flagellum is almost as long as the fifth jomt, and consists of four or five joints, the last one being tipped with a pencil of long setae; all the joints of the antennae are thickly covered with fine short setae, and one or two stouter setae are found at the end of the third and fourth joints of the peduncle. The upper (vp has the anterior border regularly rounded and fringed with short setae. The mandibles have the incisor process composed of three or four stout teeth, and the molar process prominent, covered with short stout setae, with a long plumose setae lying alongside it; in the left mandible the lacinia mobilis is circular at its extremity and bears a ring of short teeth or setae, at its base arises a single stout plumose hair; the right mandible has the lacinia mobilis ending in three stout teeth hke those of the incisor process, and two stout plumose hairs arise at its base. The lower lip is deeply cleft into two lobes, each fringed with setae along their outer border, and also round the margins of the clett. The first maxilla has the outer lobe narrow, ending in about a dozen stout curved setae, the outer ones being the stronger, and brownish in colour; the outer margin of the lobe bears numerous setae arranged in small tufts, and a few rather longer setae are found on the inner margin. The inner lobe bears the usual three setae at 664 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. its extremity, the two terminal ones being subequal, the proximal nearly twice as long. The second maxilla is of about the same width throughout; 1 is imdistinctly cleft at the end, the outer lobe being much narrower than the inner ; the whole of the extremity is covered with a thick fur of short setae; other setae are found along the inner border and on the distal portion of the outer border. The maaillipedes have the first joint short and the epipod short and rounded at the end; the second joint is large, expanded, ner margin straight, outer margin strongly curved, both fringed with long setae; the terminal part of the maxillipedes forms a single piece, with only faint indications of division into separate joints ; the epignath is shorter than the terminal lobe and narrower, and appears to have a separate small jot at the end, the whole of 1t being thickly covered with setae. The legs (fig. 17¢ and 17d) are all approximately the same size and shape, the seventh leg only slightly longer than the first; they bear long stout setae of the usual character, especially on the inner margin of the merus and the carpus, and all the joints are covered with fine short setae, which are particularly noticeable on the propod and on the dactyl; the dactylar seta has the same structure as in Trichoniscus—viz., it consists of a basal portion which divides into two branches, the inner one of which at once splits up mto further fine divisions, while the outer is continued as a long plumose seta ; the dactyl also bears on the inner side a secondary nail and a stout curved seta arising near it and extending a little beyond the extremity of the dactyl. The pleopoda appear to be similar to those of Trichoniscus. According to Sars, the first pair of pleopoda in the female of this genus are very small and rudimentary, while those in the male are well developed, with the inner ramus strongly produced, biarticulate, terminal jot spiniform ; I accidentally failed, however, to find the first pleopod in the single male specimen that I dissected. The second pleopod of the male (fig. 17e) has the inner branch long and narrow, the end fringed with setae, and apparently undivided and not marked off from the short but wide basal joint. The third pleopods (fig. 177) have the inner angle of the basal jomt somewhat produced and fringed with setae; the inner branch is much smaller than the outer, which is operculiform, and articulated to the outer portion of the basal joint, and has the outer margin fringed with fine setae. The fourth and fifth pleopods are similar to the third, though the shape of the outer branch and the proportional sizes of the two branches are not quite the same. The wropods have already been sufficiently described. Fam. SCYPHACIDAE. Genus Scyphoniscus, Chilton, 1901. Mistribution.—At present known only trom New Zealand and adjacent islands. In order to include the following species the generic characters, which were based on those of the single New Zealand species then known, will require slight alteration. The first character—‘body rather narrow, lateral parts not greatly developed ”—will have to be omitted, as the present species 1s fairly broad, especially in the female. The antennae should be defined as having the flagellum consisting of a few ill-developed joints, instead of “three ill-developed joints,” as the actual number of joints seems to vary to some extent Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 665 Scyphoniscus magnus, sp. nov. Specific Diagnosis.—Body in the female broadly oval, about half as broad as long; in the male much narrower, hardly more than a third as broad as long. Sur- face of cephalon and peraeon somewhat rugose, being covered with small roughened tubercles. Cephalon (fig. 18a) with the lateral lobes large, broadly rounded, front produced in the middle line into a shallow rounded lobe ; eyes well developed, lateral, containing about nine or ten ocelli. Margins of first segment of peraeon produced into rounded lobes reaching nearly to the eyes, posterior margin of first four seg- ments straight and with the posterior angles rectangular, those of the next three uy Fie. 18.—Scyphoniscus magnus, sp. nov. 18a. Head and antenna. 18e. Second pleopod of male. 186. First maxilla (highly magnified). 18/. Third pleopod of male. 18c. Seventh leg of male. 18g. Telson and uropoda. 18d. First pleopod of male. segments with posterior angles more and more acutely produced. Pleon with the first two segments short, the third to the fifth with moderate lateral expansions, last segment short, about twice as broad as long, subacutely produced posteriorly. Antennae with the last jomt of peduncle nearly as long as the two preceding; flagellum as long as the penultimate, and consisting of four fairly well marked sub- equal joints, the last ending in a small tuft of fine setae ; the peduncle with margins of all the joints fairly smooth and bearing only a number of short fine setae. Legs (fig. 18¢) short, the last pair only slightly longer than the first, and not specially modified in the male. The uropoda (fig. 18g) rather shorter than pleon ; peduncle stout, produced inwards into a small lobe bearing the inner ramus; inner ramus 666 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. rather more than half as long as the outer, which is more than twice as long as the peduncle, both ending in a few fine setae. Colour.—Pale-yellowish, with the dorsal surface more or less darkly pigmented. — Length, 10 mm.; greatest breadth (of female), 5 mm. Hab.—Campbell Island, abundant on the shore of Perseverance Harbour about high-water mark ; Ewing Island (Dr. L. Cockayne). Type in Canterbury Museum, New Zealand. This species differs from S. wactatensis, which is found on the coasts of the South Island of New Zealand, in being much larger and broader (in the female) and in having the surface of the body less tuberculated ; there are also slight differences in the antennae and some of the mouth parts. The mouth parts are, on the whole, similar to those of S. wactatensis, but in the first maxilla (fig. 18) the outer portion approaches somewhat more nearly to the normal type—viz., it bears at the end a number of stout slightly curved setae, suc- ceeded by more slender setae on the inner side; the outermost one, however, is thickly fringed on its outer margin with long fine setae curving slightly inwards, giving the same general appearance as in S. waitatensis, in which, however, the stout setae appear to be absent and the joint ends in a rounded lobe fringed with the long setae. The maxilliped has the palp with the various joints of which it is composed rather better indicated than in S. wartatensis. The other mouth parts call for no special attention. The male differs from the female in having the body much narrower, but there seem to be no secondary sexual characters in the legs or other appendages. The first pleopod of the male (fig. 18d) is specially modified, the exopod is small and of the usual shape, but the endopod is developed into a strongly chitinized organ which appears to be unjointed and to be fused with the basal portion of the pleopod ; it is more than twice as long as the exopod, and curves outwards towards the end, which narrows somewhat abruptly to the acute termination ; the single male organ appears more or less fused with the first pleopod, and is seen in the median line as a small narrow appendage less than half as long as the endopod. In the second pleopod (fig. 18e) the endopod is two-jointed, the first jomt short, broader than long ; the second joint reaching as far as the end of the exopod, narrowly triangular in shape, and ending acutely ; the exopod 1s articulated to the protopod towards its outer border, and forms a flat opercular lamella, with the inner margin strongly curved, the outer margin concave towards the end (which is fringed on both sides with fine setae) and convex near the base. In the third pleopod (fig. 18f) the endopod is branchial im.structure, nearly rectangular, with the angles rounded, and is not quite half as long as the opercular exopod, which has the same general shape as in the second pleopod, but is broader towards the distal end. The fourth and fifth pleopods are similar to the third. Genus Deto, Guérin, 1834. Distribution.—Widely distributed in subantarctic regions. Deto aucklandiae (G. M. Thomson). Actaecia aucklandiae, G. M. Thomson, Trans. N.Z. Inst., x1, p. 249, 1879. Scyphax (?) aucklandiae, Chilton, Trans. Linn. Soc., vin, p. 126, pl. xv, Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 667 fig. 2,1901. Deto magnifica, Budde-Lund, Deutsche Siid-polar Exped., ix, p. 86, 1906. D. robusta, Budde-Lund, l.c., p. 87, pl. iv, figs. 42-44, 1906. D. aucklandiae, Budde-Lund, l.c., p. 87; Chilton, Trans. N.Z. Inst., xxxviil, p. 273, 1906. Several specimens of this species were obtained on previous occasions at Ewing Island, in the Auckland Group, but unfortunately no additional specimens were collected during the expedition. It is a very large and handsome species, reaching a length of 23mm., and it shows very marked sexual dimorphism, the male (fig. 19a) differing from the female (fig. 195) in the greater development of tubercles on the dorsal surface and particularly in the size and thickness of the external antennae. These are very massive, and have the various joints fully twice as broad as those of the female; they have been figured by Budde-Lund under the name D. robusta. The joints of the flagellum, and to some extent also the terminal joimts of the pe- Fic. 19.—Deto aucklandiae (G. M. Thomson). duncle, are thickly covered with a fine OaauMfalel (Gousell cies: down of short setae. In the male, too, the 196. Female (dorsal view); x 2. tubercles on the dorsal surface are much better marked than in the female; each segment bears a row of about ten to twelve tubercles, the lateral ones being longer than those in the centre, and pro- jecting up from the dorsal surface like short blunt spimes*; in the pleon they are much shorter, and form slight tubercles only, similar to those found on the dorsal surface of the peraeon in the female. When preparing my paper on the terrestrial Jsopoda of New Zealand, quoted above, I had only a single female specimen of this species, and the male then remained unknown. There are three dried specimens of this species, from Auck- land Island (one of them imperiect), in the Dresden Museum ; these were examined by Budde-Lund, who has described them as forming two new species, D. magnifica and D. robusta, though he stated that the first one was perhaps not distinct from D. aucklandiae. J have no doubt that both}of his species must be referred to D. aucklandiae, his description of D. robusta being evidently based mainly on the examination of a male specimen, and agreeing closely with the male specimens in my collection. I had formerly considered Oniscus novae-zealandiae, Filhol, as a doubtful syno- nym of D. aucklandiae, but specimens since received from the Chatham Islands and from Stewart Island show that it is a distinct species, and must therefore be known as Deto novae-zealandiae. It differs from D. aucklandiae in the secondary sexual characters, for the male has the lateral portions of the first segment of the peraeon dilated into two bladder-like structures. A female specimen belonging * These are much longer than is shown in fig. 19a. 668 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Crustacea. to this species was obtained on Stewart Island during the expedition, and from a comparison of the figures given in Gay’s “ Historia di Chile” it is evident that Oniscus bucculentus, Nicolet, and Oniscus tuberculatus, Nicolet, from Chili, are male and female either of this same species or of one very closely allied. Other species of the genus are found at the Cape of Good Hope; at St. Paul, in the New Amsterdam Group; and in Australia. Fam. ONISCIDAE. Genus Oniscus, Linnaeus, 1767. Distribution.—The genus Oniscus is cosmopolitan, but the section to which the following species belongs is, so far as at present known, confined to Australia and New Zealand. Oniscus punctatus, G. M. Thomson. Onscus punctatus, G. M. Thomson, Trans. N.Z. Inst., xi, p. 232, pl. xa, fig. 3, 1879; Chilton, Trans. Linn. Soc., vin, p. 133,. pl. xvi, fig. 2, 1901; Trans. N.Z. Inst., xxxvin, p. 273, 1906. Numerous specimens were obtained at Auckland Island, and are quite the same as those found in New Zealand, where the species is very widely spread. The same species, or one very closely allied, is also found in Tasmania and Australia. This species does not come strictly under the genus Oniscus as now restricted. Dr. Budde-Lund, by letter, some years ago told me that he was establishing a new genus, Phalloniscus, for its reception, but I am not aware whether this genus has yet been published. Genus Cubaris, Brandt, 1833. Distribution.—Cosmopolitan. Cubaris rugulosus, Miers. Cubaris rugulosus, Miers, Ann. & Mag. Nat. Hist., ser. 4, xvii, p. 225, 1876 ; Cat. N.Z. Crust., p. 96, pl. 1, fig. 5, 1876. Armadillo rugulosus, Chilton, Trans. Linn. Soe., vin, p. 147, pl. xvi, fig. 7, 1901. Spherillo rugulosus, Budde-Lund, Revision of Crust. Isopoda Terrestria, 1, p. 65, pl. vil, figs. 37-39, 1904. This species appears to be common in the Auckland Islands. Professor Benham gathered it near Port Ross, in the north of Auckland Island ; under logs at Carnley Harbour, in the south ; and one specimen on Disappointment Island. It also occurs in Campbell Island, where it was collected by myself and by Messrs. Chambers and Des Barres. These specimens appear to be quite the same as those from the main islands ot New Zealand, where the species is widely distributed, and is particularly abundant in the more southern parts. Dr. Budde-Lund has revived Dana’s name Spherillo for a section of the genus hitherto known as Armadillo, but as I do not quite under- stand the characters given by him in the definition of Spheridlo, and as the name Cubaris is older and has been adopted by Stebbing and Calman, I also have used it. cs vas aaeel Ee gs ES PPO Crustacea. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 669 The species grouped by Dr. Budde-Lund under Spherillo belong mainly to Australia and the various islands in the Pacific, south-east Asia, and the Malayan Archipelago ; while those which he retains under Armadillo are nearly all inhabitants of Europe, Asia Minor, Africa, and America. Order TANAIDACEA. Fam. TANAIDAE. Genus Tanais, Audouin and Milne-Edwards, 1829. Distribution.—Cosmopolitan. Tanais novae-zealandiae, G. M. Thomson. Tanas novae-zealandiae, G. M. Thomson, Ann. & Mag. Nat. Hist., ser. 5, iv, p. 417, pl. xix, figs. 5, 6, 1879; and Trans. N.Z. Inst., xi, p. 207, fig. 3, 1881. One specimen, from Perseverance Harbour, Campbell Island, taken on the shore with Hxosphaeroma gigas, &c. It is darkly pigmented. Another was taken at the Snares. The species 1s common on the coasts of New Zealand, and at present is not known elsewhere, though it may prove to be more widely distributed. Order CUMACEA. IT am not aware that any Cumacea have been described from these islands. In his report on the Cumacea collected by the National Antarctic Expedition, 1901-4, Dr. W. T. Calman mentions that a species of Cumacean was obtained from the Auck- land Islands ; it was not included in his report, as it is not antarctic, but I do not know whether it has yet been described elsewhere. Order NEBALIACEA. Fam. NEBALIIDAE. Genus Nebalia. Distribution.—Probably cosmopolitan, but best represented in northern and in southern seas. Nebalia longicornis, G. M. Thomson. Nebalia longicorms, G. M. Thomson, Ann. & Mag., Nat. Hist., ser. 5, iv, p. 418, pl. xix, figs. 7-9; Sayce, “ Victorian Naturalist,” xvii, p. 151, 1902; Thiele, Rep. “ Valdivia’ Crust., vii, p. 9, pl. iv, figs. 66-69, and Nat. Antarct. Exped., 1901-4, in, Leptostraca, p. 1. One small specimen, from Musgrave Harbour, collected by Professor Benham. This species was originally described from a male specimen taken in Dunedin Harbour, New Zealand, and appears widely distributed in the southern seas ; Sayce 670 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Orustacea. has recorded it from Port Phillip, Victoria, and, according to Thiele, the specimens collected by Dr. Willey in the Friendly Islands and at New Britain also belong to this species. In his report on the Leptostraca collected by the “ Discovery,” Thiele states that N. longicornis, var. magellamca, has been found at McMurdo Strait, and that it was also taken by the German Antarctic Expedition at the winter quarters at Gaussberg. The species is very close to N. bipes, the northern form, and Dr. Willey’s specimens were referred to N. bepes by Mr. Stebbing. Ohlin also placed in this species specimens gathered in Magellan Straits ; these, however, Thiele con- siders to be a variety of N. longicornis, distinguishing it under the name magellanica. In the Berlin Museum there are specimens from Cuba which Thiele also refers to © this species, considering them, however, to represent another variety, which he has named soror. Subclass CIRRIPEDIA. No special attention was devoted during the expedition to the Cuirripedia and the other groups of the Hntomostraea. A few species were previously known from the islands, and one more 1s now added. Order THORACICA. Fam. BALANIDAE. Genus Balanus, Lister. Distribution.—In all the warmer parts of the globe. Balanus porcatus, Da Costa. Balanus porcatus, Da Costa, Hist. Nat. Test. Brit., p. 249 (1778) ; Darwin, ‘* Monograph Cirripedia,” Balanidae, p. 256, ‘pl. vi, figs. 4a—4e, 1854. Hutton, Trans. N.Z. Inst., x1, p. 328, 1879 ; Filhol, ‘Mission de I’Hle Campbell,” p. 487, 1885. Recorded from Campbell Island by Hutton and by Filhol. ‘The species also occurs in New Zealand, and is widely distributed in Kurope, America, &c. Balanus campbelli, Filhol. Balanus campbelli, Filhol, “ Mission de I’lle Cann. ” p. 487, 1885. Described by Filhol from specimens gathered at Campbell Island. Balanus decorus, Darwin. Balanus decorus, Darwin, “ Monograph Cirripedia,” Balanidae, p. 212, pl. u1, figs. 6a, 6b, 1884; Hutton, Trans. N.Z. Inst., x1, p. 328, 1879. Specimens were collected at Auckland Island during the expedition. ‘The species is common in New Zealand, and is found also in Australia. Crustacea. } SUBANTARCTIC ISLANDS OF NEW ZEALAND. 671 Subclass COPEPODA. Order KUCOPEPODA. Fam. CALANIDAE. Genus Deguernea, G. M. Thomson, 1902 (= Guernea G. M. Thomson, 1895). Deguernea antarctica (G. M. Thomson). Guernea antarctica, G. M. Thomson, Trans. N.Z. Inst., xxvu, p, 213, pl. xiv, 1895. Deguernea antarctica, Hutton, Index Faunae N.Z., p. 270, 1904. Guernella antarctica, Mrazek, Hamburgher Magalhaensischen Sammel- reise, Siiswasser-Copepoden, p. 25. Hab.—Macquarie Island, in a fresh-water pool (A. Hamilton). Fam. PELTIDIIDAE. Zaus contractus, G. M. Thomson. Zaus contractus, G. M. Thomson, Trans. N.Z. Inst., xv, p. 106, and xxvu, p- 213. Hab.—Macquarie Island (A. Hamilton). This species occurs also at New Zealand. The Macquarie Island specimens differ in some respects from the type, and Mr. Thomson has suggested the name Z. hamultom, should they prove to form a distinct species. Subclass BRANCHIOPODA. Order CLADOCERA. Fam. DAPHNIIDAE. Genus Chydorus. Distribution.—Cosmopolitan. Chydorus minutus, G. M. Thomson. Chydorus minutus, G. M. Thomson, Trans. N.Z. Inst., x1, p. 262, pl. ii, fig. E3, 1878, and xxvii, p. 211, 1894. Specimens from a fresh-water pool in Macquarie Island (A. Hamilton) are con- sidered by Mr. Thomson to belong to this New Zealand species. ARTICLE XXVII.—ON LAND PLANARIANS FROM AUCKLAND AND ENDERBY ISLANDS. By Artuur Denpy, D.Sc., F.R.S., F.Z.S., Sec. Linn. Soc., Professor of Zoology in King’s College (University of London). I RECEIVED from Professor Benham four specimens of land planarians—two from Auckland Island and two from Enderby Island—unfortunately without any notes of the colours in life. I understand that Mr. George Marriner also collected a few land planarians on Campbell Island in November, 1907. I have received his notes on the colours of these, but have not seen the specimens themselves, which I am therefore unable to report upon. The only land planarian hitherto described from the subantarctic islands ap- pears to be Geoplana moebiusi, Grafi,* of which four specimens were collected by Krone in the Auckland Islands. Geoplana subquadrangulata, Dendy, var., enderbyensis, n. var. A single specimen of this variety of a widely distributed and very variable New Zealand species comes from Enderby Island. The specimen (in spirit) measures 14 mm. in length by 34mm. in maximum breadth (about the middle). The body is much flattened above, and only slightly convex below, not distinctly quadrangular, but with rather sharply angulated margins ; tapering fairly gradually in front and behind, but with the posterior tip broken off. The peripharyngeal aperture is at about the middle of the ventral surface and the genital aperture about hali-way between it and the posterior extremity. Hyes numerous, distinct, and arranged as usual. The dorsal surface is pale yellow, with three longitudinal stripes of brown—a median and a pair.t The median stripe is narrow, rather pale, slightly discontinuous, and evidently composed of numerous small round specks of varying size and in- tensity. Hach paired stripe lies midway between the median stripe and the margin of the body ; it is about twice as wide as the median stripe and much darker, and also made up of numerous confluent round specks of brown pigment. The much wider bands of ground-colour between the median and the paired stripes, and again between the paired stripes and the margins of the body, are sparsely flecked with small round specks of dark-brown pigment similar to those of which the paired stripes are composed. The ventral surface is almost colourless, pale yellow, with a faintly marked submarginal band of a very light uniform brown colour on either side, which pro- bably represents the ill-defined lateral surface. On the inner side of each of these submarginal bands is a faint trace of mottling on the ventral suriace, absent in the median region. * “ Monographie du Turbellerien,” 1: Tricladida terricola (Landplanarien), Lepzig, 1899, p. 386. { In the living animal the paired stripes are reddish-brown.—W. B. B, Land Planarians.]} SUBANTARCTIC ISLANDS OF NEW ZEALAND. 673 Geoplana antarctica, n. sp. Body (in spirit) about 12 mm. in length by 2°5 mm. in maximum breadth (in the middle). Strongly convex, but evenly rounded above; rather more flattened below; tapering fairly gradually behind, but rather more bluntly than usual in front. Lateral margins well rounded. Eyes numerous, and arranged as usual. Peripharyngeal aperture about in the middle of the ventral surface, and genital aperture at about one-third of the distance from it to the posterior extremity. Dorsal surface almost colourless; very pale greyish-yellow, and without any markings. There is, however, a distinct rather broad marginal band of olive-grey. Ground-colour of ventral surface like that of dorsal, but with a narrow lght- grey stripe on either side of the mid-ventral line, and traces of pale-grey mottling submarginally. The specimen (in spirit) is strongly flexed ventrally, almost forming a circle. It contains a large egg-capsule, which causes a great bulge on the dorsal surface above the genital aperture. The single specimen is not mentioned in the list sent to me by Dr. Benham, but was found in a tube numbered 10, which also contained a land nemertine from Enderby Island. The presumption is that the planarian came from Enderby Island also. Geoplana aucklandica, n. sp. Body (in spirit) considerably elongated, about 41 mm. in length by 4:5 mm. in ereatest breadth; broader behind than in front; tapering very gradually to the horseshoe-shaped anterior extremity, rather more suddenly behind. Dorsal surface strongly convex. Ventral surface flat, or even concave, separated from the dorsal surface by very pronounced narrow lateral margins, which may be slightly crenate. Eyes numerous, and arranged as usual. Peripharyngeal aperture well behind the middle of the ventral surface. Genital aperture rather nearer to the peripharyn- geal aperture than to the posterior extremity. Ground-colour of dorsal surface pale yellow or orange, with numerous short dashes or specks of brown arranged in discontinuous and ill-defined longitudinal stripes. Of these stripes, at least four may be recognised, arranged in two pairs, and dividing the ground-colour into five approximately equal bands. The outer stripe on each side contains more numerous and more darkly coloured specks than the inner one. There are faint traces of another pair of stripes (or, better, bands of specks), one on each side of the mid-dorsal line, in the median band of ground-colour. The ventral surface is pale-yellowish, without markings. The body in spirit becomes strongly curved ventralwards, so that the extremities approach one another, and the integument of the dorsal surface has a strong tendency to wrinkle transversely. There are two specimens of this species from Auckland Island,* closely re- sembling one another. Their shape and general appearance in spirit remind me strongly of Geoplana fletcher, Dendy,t a Victorian and South Australian species, to which it is probably closely related. * Other specimens were obtained under logs on Auckland Island and on Enderby Island. In life the stripes are bright reddish-brown.—W. B. B. t Figured in Proc. Roy. Soc. Victoria, 1893, pl. x. 44S, ARTICLE XXVIII.—A DESCRIPTION OF TWO NEW SPECIES OF LAND NEMERTEANS FROM THE AUCKLAND ISLANDS; TOGETHER WITH SOME OBSERVATIONS ON THE ANATOMY OF THE PROBOSCIS IN GHONEMERTES AUSTRALIENSIS AND G. NOVAH-ZHALANDIAE. By A. D. Darsisuire, M.A., B.Sc., Demonstrator of Zoology in the Royal College of Science, London. THE two worms which compose the collection of land nemerteans were collected by Dr. Benham during this expedition to the Auckland Islands, and prove to be new species of the genus Geonemertes (I have named them G. spirospermia and G. caeca) ; but in view of the fact that there is only one specimen of each species, and that the worms were not preserved with a view to minute histological examination, I shall confine myself in this paper to a description of such characters as warrant the creation of the new species, and shall postpone a full consideration of the systematic position of these two forms until a larger number of specimens in a better state of preservation is available. DESCRIPTION OF THE TWO NEW SPECIES. Geonemertes spirospermia, n. sp. The single specimen was obtained at Auckland Island. External Characters. The single specimen of this species is 15°5 mm. long and a little over 2mm. across at the anterior end, gradually de- creasing in breadth backward to the posterior end, which is a little over 1mm. Just in front of the posterior end is a con- striction, which at first led me to take what is really the tail for the head. It was by finding the eyes on what I had thought was the tail that I discovered the mistake. The eyes are four in number ; they consist of a larger anterior pair, and a smaller posterior pair closer together, just behind the front pair. The ground-colour of the spirit specimen is a rich cream, on which are two longitudinal stripes of a dark-brown colour, the form and extent of which is shown in fig. 1. The curious interruption and anastomosis of the two stripes just in front : of their middle is probably due to some injury received at _ Bo Te an earlier stage in development, SE Ee SO cas > Land Nemerteans. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 675 Internal Characters. The Reproductive Elements.—Perhaps the most striking peculiarity of this species is the possession of spiral spermatozoa. The conclusion, based on their existence in the testes, that these really were spermatozoa was confirmed by the result of a staining reaction with Flemming’s triple stain. If this stain is used with tissues which contain spermatozoa in various stages of development it will be found that the ripe spermatozoa have a very strong affinity for the gentian violet, with the result that they assume a purple colour of a much greater intensity than that taken on by any of the other tissues. I have tested this effect of this stain with G. australt- ensis, in which the disposition and contents of the testes have been described by Dendy, and found that it picked out the ripe spermatozoa with extraordinary sharp- ness. The effect in G. spirospermia was no less striking. The only specimen of this species is a male. But, in view of the possibility of the occurrence of protandrous or protogynous hermaphroditism in these worms, | do not consider that this solitary observation justifies the statement that G. spiro- spermia 1s dioecious. The Proboscis. The number of proboscideal nerves is 18. The number of reserve stylet- pouches is 4; and these are seen in a transverse section to be arranged, as it were, at the four points of the compass—that is to say, there are four single stylet-pouches equidistant from one another. The accessory lateral nerve exists as a ganglionic strand running along the dorsal surface of the nerve-cords. Geonemertes caeca, Nn. sp. The single specimen was obtained at Enderby Island. External Characters. This worm is 10 mm. long; it is 2 mm. broad at the anterior end and 1:25 mm. at the posterior. The proboscis had been everted, but probably only partly. The colour of the spirit specimen was a very pale cream, and there were no traces of markings on it whatsoever. The specimen was examined, both in spirit and in cedar-wood oil, for eyes, but Professor Dendy and I were both unable to detect any trace of these organs. There is no note that this animal was subjected to treatment different from that to which G. spirospermia was subjected, so that there is no ground for sup- posing that the absence of markings and of eyes are other than the natural characters of the worm. Internal Characters. The single specimen contained ova only. ‘The number of proboscideal nerves is 17; and the number of reserve stylet-pouches is 5. The exact arrangement of these is difficult to make out, because the proboscis was cut longitudinally. It is not possible to state that the 5 pouches are equidistant from one another; but it can easily be seen that there is no association of neighbouring pouches of such a kind that the two come to be close together, 676 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Land Nemerteans. NOTE ON THE NUMBER OF PROBOSCIDEAL NERVES AND OF RESERVE STYLET- POUCHES IN G. AUSTRALIENSIS AND G. NOVAE-ZEALANDIAE. The observations which I am about to place on record were made on a collection of land nemerteans made in Australia and New Zealand by Professor Dendy, and kindly placed at my disposal by him. Many of the descriptions of new species of animals and plants are made, as in the case of the two new species described in the present paper, on solitary specimens of the new forms. If this description is not soon superseded by one based on a larger number of specimens, the impression that the characters of the species are not subject to variation takes a hold which becomes firmer with each repetition of the original description. One of the characters used in the classification of the species of the genus Greone- mertes is the number of the proboscideal nerves. Now, a number of this kind when used for classificatory purposes 1s valueless unless the range of variability, if any, to which this number is subject is known. Let us fix out attention on a particular instance, G. australiensis : this was first described in detail by Dendy,* who does not actually state the number of proboscideal nerves, but publishes a drawing of a transverse section of the proboscis from which this number—18—can easily be made out. Coe, in his monograph on G. agricola,t has drawn up a synoptic table of the species of Geonemertes, and states that the number of proboscideal nerves in G. australiensis is 18. Punnett, in his account of G. arboricola from the Sey- chellest, publishes an amplification of Coe’s synoptic table, in which G. australiensis again appears with 18 proboscideal nerves. All that I am concerned with in pointing out here is that the impression that the number was invariable, which the reader who did not read the original de- scription would be likely to gather, would be simply a result of the fact that the first description of this character was based on observations made on a single specimen. My observations on Professor Dendy’s specimens prove, however, that this number is by no means constant. There proved to be two specimens with 16 nerves, two with 17, seven with 18, and one with 19. So that the specimen whose proboscis Dendy figured happens to have represented the commonest type. The state of affairs in G. novae-zelandiae, the number of whose proboscideal nerves has not yet been recorded, 1s very curious. The two actual specimens on which I have been able to count the number of proboscideal nerves are already familiar to those who are conversant with the litera- ture relating to land nemerteans. The species was first named by Dendy,§ from two spirit specimens—one found in a jar of “land planarians”’ which had been collected in various unspecified localities in New Zealand ; the other (a much more periect specimen) found in a bottle labelled “ Land Planarians, Toitoi, Southland ; Miss T. G. Rich.” Dendy soon afterwards found living specimens of this species, and gave a brief account of it.'| The two specimens I have examined are the spirit * Proc. Roy. Soc. Victoria, vol. iv, p. 85 (1892). + Proc. Boston Soc. Nat. Hist., vol. xxxi (1904). { Trans. Linn. Soc. London, 2nd ser., Zoology, vol. xii, p. 1 (1997). § Trans. N.Z. Inst., vol. xxvii, p. 191. || Trans, N.Z. Inst,, vol, xxvii, p. 214, Land Nemerteans.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 677 specimen from Toitoi and the first living specimen of the species found by Dendy. As the coloration of this species has not before been figured, I have thought it desirable to publish a drawing of it (fig. 2). The number of proboscideal nerves in this specimen is 17; the number in the specimen from Toitoi is 13. Punnett suggests that the known land nemerteans of the world fall into two groups, of which one, including Geonemertes agricola and G. chalicophora, is characterized by the absence of an accessory lateral nerve, by the pos- session of an excretory system, by the fact that they are not striped, and that they possess a small number of proboscideal nerves: G. agricola has 13-15, G. chalicophora has 12. The other group includes G. palaensis (from the Pelew Islands), and G. rodericana (from Rodriquez), G. arboricola, and G. graffi: all these forms possess an accessory lateral nerve, but have no excretory system ; they are also all striped; and the number oi proboscideal nerves in them is large, varying, according to the infor- mation available at the time at which Punnett’s paper was written, from 18 (australiensis) to 23 (graffi). With regard to the two remaining species, Punnett says (loc. cit., p. 61), “Tt is not impossible that G. novae-zealandiae may turn out to belong to the same group (v.e., the latter group) when more is known about it. G. australiensis resembles the forms mentioned in the presence of an accessory lateral nerve, but differs in having an excretory system.” The three specimens of G. novae-zealandiae which I have examined exhibited the accessory lateral nerve; I have not, however, been able to detect any traces of an ean excretory system. But I do not attach any value, as — Geonemertes novae-zealandiae. evidence, to this purely negative result. Deere w Nts) The presence of an accessory lateral nerve and the well-marked striping point to the conclusion that G. novae-zealandiae belongs to the second of Punnett’s groups. The only fact which makes against this conclusion is the small number (13) of proboscideal nerves in one of the specimens examined, which is regarded as characteristic of the other group. I am of opinion that until a very much larger number of specimens of all the species have been examined it is unwise to attach any systematic significance to the number of proboscideal nerves in isolated specimens. The conclusion that the number of proboscideal nerves characteristic of Punnett’s two groups formed a discontinuous series, according to the evidence available at the time at which Punnett’s paper was written, was reached by reason of the imperfection of that evidence. According to it, this number varied in the one group between 12 (chalicophora) and 15 (agricola), and in the other between 18 (australiensis) and 23 (graffi). Now we know that this number in australiensis varies between 16 and 19; and it is a noteworthy fact that the gap between the two groups should have been bridged over by the species which presented the lowest number in the group 678 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Land Nemerteans. characterized by the larger number of nerves. The barrier between the two groups is further broken down by the possession by G. novae-zealandiae of numbers (13 and 17) characteristic of both groups. Let us now turn to the question of the number of reserve stylet-pouches. In the table in Punnett’s paper G. australiensis 1s stated to possess 2 and novae- zealandiae 4. Now, Dendy, in his monograph on G. australiensis, says (p. 98), “they are not definitely two in number, as appears to be the case in G. chalicophora, but their number varies; I have counted as many as five in a single specimen ” ; and, further, with regard to G. novae-zealandiae, “The calcareous stylets have the usual form, and I detected four reserve sacs in the usual position. Probably, however, this number is not constant.” The explanation of the apparent contradiction between Coe’s and Punnett’s table and Dendy’s original account is probably of the following kind. I have been able to examine the reserve stylet-pouches in six specimens of G. australiensis, and find them to vary in number between 2 and 5; but that even when there are more than 2 the pouches are always arranged in two groups which are vis-a-vis to one another, and in which the component individual sacs are closely apposed to one another. The details of the association of these sacs in the two groups, which we may call « and y, in the six specimens (a to /) are as follows :— Group x. Group y. Total. a 1 yy 3 b 2 3 5) Cc 2 3 5 d Ys, 3 5 e 1 1 D, ji ai Y, 3 a It is therefore only true that G. australiensis has 2 reserve stylet-pouches in the sense that it sometimes actually has only 2, asin e; and that when it has more they are always, so far as the present evidence goes, associated in two groups. In the only specimen, however, of G. novae-zealandiae which I have which shows the dis- position of these sacs the four sacs are not associated in groups, but are, as in G. spirospermia, at the four points of the compass, as seen in transverse section. It is possible that the disposition of the reserve stylet-pouches may afford a further distinguishing character between the two groups. But it is a noteworthy fact that the only two species of the other group which agree with the agricola- chalicophora group in having their reserve sacs only 2 in number or arranged in two groups are—(i) G. australiensis, which bridges the gap between the two groups by the number of its proboscideal nerves; and (i) G. graffi, which possesses the greatest number of proboscideal nerves in the whole genus—namely, 23. COMPARISON BETWEEN THE NEW SPECIES OF THE GENUS GEONEMERTES AND THOSE ALREADY KNOWN. G. sprrosperma differs from any other known species of land nemertean in the pattern of its pigmentation. There is no other species with two stripes. The pos- session of spiral spermatozoa is also a character peculiar to it. It agrees with the Land Nemerteans. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 679 average character of the “ striped” group in being striped, in the large number of proboscideal nerves (18), in the possession of an accessory lateral nerve, and (possibly) in the possession of 4 disparate reserve stylet-pouches. G. caeca differs from any other known land nemertean in being totally blind. It is also the smallest species of Geonemertes hitherto recorded. It seems to approxi- mate most closely to G. australiensis, with which it agrees (possibly) in the number of its proboscideal nerves, in the absence of striping (possibly), in the possession of so many as 5 reserve stylet-pouches, and perhaps in the absence of eyes, if the diffuse arrangement of these organs which is peculiar to G. australiensis may be regarded as the first stage in their disappearance. It seems not unreasonable to regard any departure from centralisation in an organ like the eye as an indication of ap- proaching dissolution. ARTICLE XXIX.—THK GEOLOGY OF CAMPBELL ISLAND AND THE SNARES. By Professor P. Marsan, M.A., D.Sc., F.G.S., School of Mines, Otago University; assisted by Mr. R. Browne, Wanganui. CAMPBELL ISLAND. THE information previously possessed in regard to Campbell Island was mainly derived from the work of Filhol.* Buchananft made some observations on the rocks of the island, and still later Hectort stated the general results obtained from his visits. Summaries of these have been made use of by Suess§ and Hutton.|| The results here set forth were obtained irom a five days’ visit to the island, and are necessarily somewhat incomplete and disconnected. Owing to the tangled nature of the wind-resisting scrub, expeditions were in some measure circumscribed. For almost the whole time the weather was boisterous, and often wet, and snow fell more than once. The following are the days’ journeys, in the course of which observations were made and material collected: (1.) From the camp near the head of Garden Cove to the western side of Mount Dumas, Monument Harbour, South-east Harbour, the summit of Mount Honey, and Filhol Peaks. (2.) From camp to North-west Bay and the south-west coast. (3.) From camp to Mount Menhir, Mount Paris, the south coast, and Mount Dumas. (4.) From camp to St. Col Peak, North-east Harbour, Lyall Pyramid, and Mount Beeman. (5.) From camp to Terror Point, alterwards to Mount Azimuth. This list shows that the east and north coasts were not visited. Inspection from the deck of the “ Hinemoa” showed that this portion of the coast-line is formed of nearly horizontal lava-flows separated by scoria-beds. GEOLOGY. In general outline Campbell Island approaches a circular form more nearly than any other simple figure, though the regularity is much interfered with by the great indentation of North-west Bay. A portion of the coast-line nearly three miles in length is fairly straight, and faces north-north-east. The cliffs in this portion of the island appear almost 200 ft. high, and are perfectly abrupt. These features are maintained in the adjacent * Filhol, “ Mission de l’ile Campbell.” + Buchanan, J., ““ Campbell Island and its Flora,” Trans. N.Z. Inst., vol. xvi, p. 394. t Hector, J., Trans. N.Z. Inst., vol. xxvii, p. 736, 1896. § Suess, E., “ The Face of the Harth,” vol. u, p. 146. || Hutton, Captain F. W., “ Index Faunae Novae Zealandiae,” p. 16. oe Geology. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 681 portion oi the coast-line between the north-east and south-east points, though two small islands lie off this part, and two deep inlets—North-east Harbour and Per- severance Harbour—strike far into the recesses of the island from the east. One hy Gabbzey Courrejolles Point = — Miocene. Volcanic Breccia. Volcanic Lavas. z North-east Harbour == = a Yvon Villarceau PeakoB—= SES — = Beemanra~ = — p\ GEE ue = SAc0_ NY “a \ =~ SS ——————— aS = slit H Honey= “Sey Pk: 942 He Geological Map of CAMPBELL ISLAND. 3 : Ss r ; Scale of Nautical Miles. of the islets mentioned is situated at the north-east corner, and the other at the entrance of the North-east Harbour. The south coast is far less regular. Here in several places the land falls pre- cipitously from a level of 1,500 ft. high to the sea-margin, though the highest cliffs 682 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Geology. are sometimes separated by quite low saddles. At the east of this part of the coast an inlet-—Monument Harbour—reaches two miles into the land. The entrance to a lagoon at the head of the bay is obstructed by a boulder-bank, across which the waves break in rough weather and at high tide only. On either side of Monument Harbour the land rises abruptly, and off its entrance are two needle-shaped islands and a larger block, Jacquemart Island. Further west the coast-line, for the time directed north, strikes straight across Mount Eboulé in a vertical cliff, and further westward Mount Dumas presents a vertical front for nearly its whole height of 1,650 ft. Thence the cliffs become low, and there are several outlying islands ; but in the south- west corner Mount Paris is intersected in the same way as Mount Eboulé and Mount Dumas. Thence as the coast-line stretches north and east it is relatively low until a point 1s reached west of Tucker Cove. In the extreme south-west corner of North- west Bay, where a mass of gabbro rock forms a protecting headland, there is a small sandy beach, protected from the violence of the heavy westerly swell by Dent Island, as well as by the westerly point of the bay. The remainder of the coast-line is ex- tremely high and steep, with several outlying rocks. In the north it stretches some- what to the west, and thus forms Courrejolles Point, a jagged promontory separated from the rest of the island by a low neck. From this general account it will be noticed that the cliffs are far higher and the outlying islands more numerous on the west and south than elsewhere, and that the only accumulation of shore drift is at the entrance to Monument Harbour lagoon and in the corner of North-west Bay. As will be seen afterwards, the differ- ence in the hardness of the rocks cannot be held to be sufficient to account for the change in height of the cliffs in different parts of the island. The only adequate explanation is to be found in the prevailing south-westerly swell and current, which have effected the removal of large rock-masses in the south and west of the island, and the coast on these sides now strikes across what was the heart of the island. The entire absence of terraces in the harbours and on the exposed headlands offers a strong presumption that the island has not been submitted to a movement of elevation since its present features were acquired. The nearly complete absence of bars of drift suggests that the present elevation has not been long maintaimed, for with such a strong swell and prevailing current from the south-west the drift from even such a short coast-line would be expected to build up a gravel-bank ex- tending north-eastwards from Courrejolles Point. Of this no sign was seen, nor 1s there any gravel-bank in the island, except at Monument Harbour, where it is but 50 yards long, and composed of material derived from the adjacent slopes of Mount Kboulé. If the absence of raised beaches and rock shelves is accepted as proof that there has been no recent elevation, and the absence of bars of sand and boulders is proof that the present level has not been long maintained, it is probable that a movement of depression must be in progress. Of such a movement there appears to be sufficient evidence in the existence of such inlets as North-east Harbour and Perseverance Harbour. That both of these are due to an agent of surface-erosion there is no doubt, and, as will be shown later, the present level of their floors may be due to depression since their formation. The large number of outlying islands also appears to be due to the action of depression, which has left the higher summits only above the sea-level, or possibly to the greater resistance offered by the harder rocks to wave-action. It was impossible to visit the islets, so this point could not be settled. Geology. SUBANTARCTIC ISLANDS OF NEW ZEALAND. 683 SURFACE. Undulating and hilly country forms the surface of Campbell Island. In the south Mounts Puiseux (1,322 it.), Eboulé, Dumas (1,650 it.), Paris (1,526 ft.), and Yvon Villarceau rise almost vertically from the ocean-shore. In the south-east Mount Honey (1,867 ft.), with its plateau extending eastward, occupies all the area between Perseverance Harbour and South-east Bay. Lyall Pyramid extends as a ridge east and west, and separates North-east Harbour from Perseverance Harbour. Mounts Azimuth (1,600 ft.), Faye (1,155 ft.), and Fizeau (1,555 ft.) form a nearly continuous plateau north of the North-east Harbour, and the first of them, rises vertically from the western shore. Lesser heights—St. Col (984 ft.) and low ground elsewhere—tringe the west ; while the curious conical Menhir stands between Dumas and Yvon Villarceau. Another small conical hill—Mount Beeman—rises from the low ground east of Tucker Cove. North-east Bay is separated from the western coast by ground 1,000 ft. high; but only undulating country, nowhere more than 500 ft. in elevation, separates Perseverance Harbour from North-west Bay. Fic. 1.—Panoramic SketcH or CAMPBELL ISLAND. FROM THE TOP oF Mounr Parts. As noted by Filhol, the surface of this island presents some striking features. The main elevations approach generally the same level, but are quite separated by low saddles, though Mount Azimuth and Mount Faye are exceptions, as they are situated on the same plateau. Usually the elevations have peak-hke summits, though in some instances, such as Mount Lyall and Mount Honey, they have wide plateau-like shoulders. All the higher hills are fringed in their upper portions by a series of terraces, which rise like serried ramparts from the sloping ground below, and are often so abrupt as to present formidable obstacles to all who attempt to scale them. It will be noticed that this description agrees perfectly with that given of Kerguelen by Renard.* Some of the lower hills have not this fortified appearance. The Menhir has gently sloping sides until the crowning rock of druidical appearance, to which its name is due, is reached at the summit. Mount Beeman, too, has no succession of ramparts, but for much of its height it has precipitous walls of rock. To the most casual observer it is evident that the terraced form of the hills is due to the outcrop of nearly horizontal flows of lava, separated by the less-resistant scoria-beds, the material of which was emitted between the periods of lava-flow. The drainage of the island presents some peculiar features. The greater part of the rainfall flows to the east, and this is particularly marked at the head of North- * Part vu, “ Physics and Chemistry of Voyage of ‘ Challenger,” p. 3, 1. 684 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Geology. east Harbour. “ eigen 1-57 100°39 100:98 Beeman Hill is composed of basalt of rather a different kind; it is por- phyritic, but augite as well as feldspar occurs among the larger crystals. In the field occurrence this rock has a distinct columnar structure. The summit of Mount Paris is composed of a basalt that has a few porphyritic crystals of labradorite, but the base of the rock is very fine, and the augite is pale grey. At the top of St. Col Peak the basalt is rather similar to that at the top of Mount Paris. Several of the dykes on the western shore of Perseverance Harbour are formed of a basalt that is closely related to diabases in structure. These rocks do not contain any olivine. They are very dense. The augite is in plates of small size, but sometimes encloses a few feldspar microlites ophitically; it has a violet tint, which probably indicates that it is titaniferous. The feldspar is in the form of microlites only. There is a good deal of glass and some serpentinous matter. Several of the smaller dykes have a purely glassy selvage. The dykes on the shore of Perseverance Harbour numbered 2, 14, 23, 24, 25, 4, 12, 9, and 22 are formed of this type of rock. Some of the basic rocks are coarse enough to call dolerites. Most of them, such as that at Terror Point, are different from the basalts in texture only; but one, a light-grey rock, found near the summit of Lyall Pyramid, has porphyritic crystals of olivine and augite. Dyke 7 is similar to it, as well as another dyke on the west side of Mount Dumas. Geology. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 697 Meuitire BASALT. A dark-grey dense rock found at Lyall Pyramid. Section : A few feldspar crystals, little twinned, but one individual is distinctly andesine. A few colourless augites and some olivine crystals with crusts of limonite. With the olivine a few flakes of reddish-yellow mica, probably anomite. Ground- mass very fine-grained. Colourless augite magnetite and olivine are present in about equal quantities ; they are imbedded in plates of a teebly bi-refringent colourless mineral. The characters of this mineral suggest melilite, and its presence is proved by the separa- tion of gypsum needles from a solution in dilute hydrochloric acid on the addition of sulphuric acid. Very similar rocks occur at the summits of Mount Dumas and Mount Honey, but in them melilite appears to be wholly replaced by ESNCISTORIE, 0" go I aS A e/a ~—— Basalt dykes. wecee Trachyte dykes. coeacacnn Nature of dyke not determined. Analysis. SiO, 54-44 Al,O. 12°83 Fe,0, 8°32 FeO 5:41 CaO 6:90 MgO 2-66 HOES 2°68 Na,.O 4:34 K,0 - Po) 5s H,O ae 50 1°58 100-69 Limestone. (Fig. 13.) As a rule, the limestone is somewhat lamellar, but appears to the naked eye to be_ periectly homogeneous. In section a few grains of glauconite are occasionally seen ; except for these, the whole PLAN OF F orESHORE OF Heap OF PERSEVERANCE rock consists of extremely fine cal- Harpovr, Camupesenn Isnanp, SHOWING THE careous matter, in which tests of Noumper anp Direcrion or Dykus. Foraminifera can be seen. 698 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Geology. MUDSTONE. On the eastern side of the head of Perseverance Harbour, in the locality where Filhol records schist, there is now exposed a lamellar mudstone. In section 1t is seen to contain abundant angular grains of quartz, imbedded in extremely fine- erained dusty matter, in which occasional flakes of muscovite are distinct. QUARTZ. Pebbles of this mineral obtained from the conglomerate beneath the limestone consist entirely of interlocked grains of the same appearance that is usually pos- sessed by the quartz of schists. It frequently shows undulose extinction. MARBLE. At North-west Bay the limestone is represented by a crystalline rock which proved to be a fine-grained marble. The recrystallisation was not quite complete, for traces of the tests of Foraminifera still remain. FLInt. Abundant in some of the limestone. Many flints are exposed on the shore at the head of Perseverance Harbour and along the middle part of the coast of North- west Bay. From the foregoing statements the sequence of volcanic rocks will be found to be as follows :— The first effect of volcanic action was to form a great thickness of tufts. The earliest lava-flows consisted of trachytic rock. Two distinct flows of this rock can be seen on the flanks of Mounts Dumas, Lyall, Eboulé, and Azimuth, and it is possible that further work would reveal their presence on the other hills. On Mount Dumas the lowest flow is nearly 100 ft. thick. The trachyte lavas lie at rather a higher level on Mount Dumas and Mount Azimuth than on Mount Honey and Mount Lyall. Rapid fluctuations of the barometer prevented any accurate observations being made as to exact elevations. The trachytes are succeeded on the greater heights by coarsely porphyritic basalts or dolerites. This lava decomposes rather easily, and does not present a steep scarp, so its thickness could not be estimated, but it is relatively thin. On Mount Lyall the porphyritic basalts are followed by melilite basalt, and on Mount Dumas and Mount Honey by extremely fine-grained basalts which closely resemble the melilite rock, though the feldspar represents the melilite. Ordinary basalts do not appear in this succession, but the separate mass of Mount Beeman is formed of this rock, and it is also found in numerous dykes and in rather a coarse form at Terror Point. | The remarkable agreement in the succession of lavas in the different hills is in itself almost sufficient to prove that they have all issued from the same point of eruption. The slight slope in the level of the trachyte-flow to the east may be sufh- cient to prove that the main point of eruption was to the west of the present land. Geology. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 699 It will be seen that this rock-succession is very different from that mentioned by Filhol. The following is an attempt to summarise his statements as to the nature and succession of the lavas :— (1.) Andesites with labradorite, but without olivine. The base of Mount Dumas, Mount Honey, and the mass of Mount Lyall. (2.) Andesitic basalts, with labradorite, olivine and pyroxene. The terraces of Mounts Dumas, Puiseux, Fizeau, and the upper mass of Mount Honey. (3.) Andesitic basalts, with anorthite, olivine, and pyroxene. The small peaks between Mount Dumas and Mount Honey. . (4.) Labradorite basalt. Dykes. (5.) Labradorite basalts. Dykes like those at Terror Point. Filhol’s idea of the source of the volcanic rocks and the points of emission is also altogether different from that outlined above. These are his words: “ Hn jetant les yeux sur la Carte jointe a ce travail on peut voir que les eruptions volcaniques, dont Vile Campbell a été le siége, se sont produites en des points excessivement nombreux, et, d’autre part, on peut constater a quel faible degré les formes exté- rieures des divers centres eruptifs ont été modifiées par le temps. Si je tragais Vhistoire de chacun de ces massifs volcaniques resserrés les uns contre les autres je serais conduit a répéter constamment et dans le méme ordre des observations iden- tiques.”” : It will be remembered that the form of the hills, which was considered of such importance by Filhol in considering their origin that he suggested that they were simple volcanic cones, is here ascribed to the action of denudation. The slight slope of the lavas from the west has just been mentioned as a reason for thinking that the point of eruption was westward of the present land. This con- clusion is supported—(1.) By the fact that the converging point of the dykes on the foreshore of Perseverance Harbour is also westward of the inlet, though apparently not far from its margin; however, the dykes that were examined show no exact point of convergence. (2.) Dykes are more numerous in the limestone cliffs of North-west Bay than elsewhere, and particularly so in its middle portion. (3.) The limestone becomes more and more metamorphic as it is followed westward: this might be ascribed to the action of the gabbro, if it is regarded as younger than the Cainozoic rocks ; but, as previously stated, the horizontal position of the limestone (which does not appear to have been disturbed even where it rests on the gabbro), the decomposed surface of the gabbro, and the absence of any dykes of crystalline rock in its neighbourhood are all facts that tell against the probability of the gabbro being intruded into the rocks which now rest on it. Though it cannot be said that the evidence in favour of important structural movements was particularly clear, some information in regard to this was obtained. At Camp Cove the limestone dips west at an angle of 15°, though on the west coast it 1s almost horizontal. In the western part of the bay it becomes somewhat argillaceous, more distinctly laminated, and highly inclined, with a north-and-south strike. It shows exactly the same features near the head of Garden Cove. It is believed by the writer that these mark the position of small faults. In the high western cliffs between St. Col Peak and Mount Azimuth there is, as far as could be seen in the boisterous weather experienced, 2 sudden change from limestone to breccia, and between the two a vertical intrusion of glassy trachyte, which thus 700 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Geology. appears to follow a plane of fracture, which in this case has a strike nearly east- and-west. From the relative elevations of the limestone and the breccia, there was roughly estimated to be a displacement of 500 ft. or more. PALA H ONT OmiOG NC But few fossils have hitherto been recorded from the rocks of Campbell Island. The only identifications that have been recorded are those given by Filhol. Wald- heimia campbellica and a Pentacrinoid stem are the only forms described, but mention is made of the occurrence of several kinds of Foraminifera in the limestones. Hector more lately states that fossil wood has been found, but no locality is mentioned. The locality from which Filhol obtaimed his fossils are—(1) the limestone at Point Duris, near the head of Perseverance Harbour; (2) the breccia near Terror Point. Both of these places were searched carefully by the members of the expedi- tion, and, in addition, a large number of fossils were found by Messrs. Chambers and Des Barres from volcanic breccia at two places on the south coast-line. The following forms were represented in the collection :-— CRINOIDEA. Prentacrinus, Miller. Pentacrinus stellatus, Hutton, Cat. Marine Tert. Moll. & Echin. of N.Z. Distrib.—Curiosity Shop, Chatham Islands. ECHINOIDEA. MicrastEer, Ag., Brissopsis, Ag., sp. ind. Loc.—Limestone, Perseverance Harbour. BRACHIOPODA. MaGELLANIA, Bayle. Magellania lenticularis, Hutton, Trans. N.Z. Inst., vol. xxxvu, p. 476. Terebratula lenticularis, Deshayes, Mag. Zool., 1841, pl. xu. Waldhewmia len- ticularis, Adams, Gen. of Recent Moll., 1, 575. Magellana lenticulars, var. ovalis, Hutton, Trans. N.Z. Inst., vol. xxxvu, pl. xlv, fig. 2. This appears to be identical with Waldheimia campbellica of Filhol. GASTROPODA. CoNCHOTHYRA, Hutton. Conchothyra parasitica, Hutton, Trans. N.Z. Inst., vol. xxvi, p. 358, pl. xliu. The specimen is not well preserved, but it appears to be correctly placed here. Loc.—Camp Cove. Distrib.Canterbury, N.Z. Geology. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 701 PELECYPODA. Prcten, Muller. Pecten zealandiae, Gray, Cat. Marine Moll., p. 81. Represented by a small fragment only, and the identification is somewhat doubtful. Loc.—Common in later Cainozoic of New Zealand. Pecten delicatula, Hutton, Cat. Marine Tert. Moll. & Hchin. of N.Z., p. 30. Several specimens of this species were found as impressions in the flints on the shore of Perseverance Harbour. The best specimens were given to me by Mr. Chambers. Loc.—Castle Point and Chatham Islands. Pecten triphooki, Zittel, Voy. ““ Novara,” Palaen., p. 52, pl. xi, fig. 4. My specimens were obtained near Terror Poimt. They differ slightly in measurement from those given by Zittel, but agree with those quoted by Hutton in his Catalogue of the Marine Tertiary Mollusca. Loc.—East coast of Wellington and Waitotara. Pecten polymorphoides, Zittel, Voy. ‘‘ Novara,” Palaen., p. 49, pl. x, fig. 1. A fragment only was found in a flint near Camp Cove. Loc.—Widely distributed in New Zealand Miocene rocks. Lima, Brug. Lima colorata, Hutton, Cat. Marine Tert. Moll. of N.Z., p. 33. This species was represented in the breccia at Terror Point and on the southern coast. Loc.—Awamoa. GLYCIMERIS, Da Costa. Glycimeris chambersi, n. sp. Orbicular, thick, equilateral, with about 25 broad radiating ribs nearly equal in breadth to the intervening grooves; both crossed with concentric striae. Near the margin the ribs and striae both become indistinct. Ligamental area with 13 fine striae on each side, which entirely occupy all the space between the two series of teeth, of which there are only 3 on each side. Height 2-5in.; length, 2-5 in. This species appears to resemble G. traversi, Hutton, from the Chatham Islands, but differs from it in the number of ribs. The characters of the hinge of G. traversi are not described. All my specimens were found in breccia on the south coast by Mr. Chambers. VENERICARDIA, Lam. Venericardia australis, Quoy. A single cast of this species was found in the breccia near Terror Point. It occurs in many localities in the later Tertiary rocks of New Zealand. 702 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Geology. Mereteix, Lam. Meretrix multistriata, Sow. A cast somewhat doubtfully belonging to this species was obtained from the breccia near Terror Point. The species does not appear to have been found in the fossil state in New Zealand. ALCYONARIA. FLABELLUM, Lesson. Flabellum sphenoideum, Ten.- Woods. Two portions of casts were found in the breccia of Terror Pot. The specific identification is doubttul. FORAMINIFERA. Many forms are represented in the limestone. It was found impossible to identify the species in the rock-sections, and all efforts to isolate the tests failed. The genera. Nodosaria, Globigerina, Textularia, and Cristellaria are undoubtedly represented. AGE OF FOSSILIFEROUS ROCKS. Of these fossils, none, with the doubtful exception of Conchothyra parasitica, are found in rocks older than the Miocene formation m New Zealand. It is well known that different authors have expressed very different views as to the age of the Cainozoic rocks of New Zealand. The following statement embodies the main points of the classification adopted by the authors named :— HeEcror*— Cretaceo-tertiary, Hocene, Upper Miocene, Lower Miocene, and Pliocene. Hurront— Oligocene: Includes Ototara and Mawhera series of Cretaceo-tertiary of Hector ; _ also his Eocene formation, except the Mount Brown series. Miocene: Includes Waitemata series of Cretaceo-tertiary of Hector, the Mount Brown series of his Hocene, and his Upper and Lower Miocene. Pliocene. ParK{— Miocene: Includes most of Hector’s Cretaceo-tertiary, Hocene, and Upper and Lower Miocene (Hutton’s Oligocene and Miocene correspond to this formation). Plocene: Includes the Motanau series, placed by the other authorities in the Miocene. If an attempt is made to correlate the Campbell Island limestones with divisions of these classifications, it is probable that this would be their proper position : Hector’s classification, Cretaceo-tertiary ; Hutton’s classification, Oligocene ; Park’s classification, Miocene. * Hector, “ Outline of New Zealand Geology.” Government Printer. 1886. + Hutton, Trans. N.Z. Inst., 1899, p. 169. { Park, Trans. N.Z. Inst., 1905, p. 492. Geology. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 703 The present writer is inclined to think that the Miocene period is more likely to be the European equivalent than any other period, though, as frequently insisted by Hutton and Tate, exact correlation is impossible in the present state of our know- ledge. The other fossiliferous formation — the volcanic breccia —is undoubtedly of younger age. Although this is clearly shown by the geological occurrence of the rocks, no section was found in which an actual junction occurred. The fossils found in the breccia indicate an early Pliocene or late Miocene age, probably the latter. THE SNARES. These islands are situated in lat. 48° S. and long. 166° 35’ E., distant sixty-five miles from Stewart Island. The only geological knowledge previously possessed in regard to the group was contained in a paper by Armstrong* and in a short abstract written by Hector.t Armstrong records the occurrence of basalt and decomposed quartz por- phyry and jasper. Hector records only granite, which is mentioned as having a red colour ; it is compared to that at Chalky Inlet. From the appearance of columnar structure on the west coast, doubt is ex- pressed as to the granitic structure of the island as a whole. The coast-line has the normal features of exposed islands composed of un- stratified rocks. On the west coast the cliffs rise everywhere nearly are Fic. 14.—Sovurn-rast Port Osa ee OF THE SNARES Group, cally to a height of some The level surface of the island marks a plain of marine erosion. 300 ft. On the east side the declivity is much less, and in one place the cliffs are so low that a landing can be made easily in good weather. Since a small stream enters the head of the landing- cove, it is probable that the cove is the submerged portion of 2 stream-valley eroded when the level of the land was somewhat higher than now. The surface is entirely covered with peat to an average depth of about 8 ft. In the major part of the island the surface slopes gently to the eastward, but terminates * Armstrong, H., “ Cruise of the Brig ‘ Amherst.’ ” t Hector, J., Trans. N.Z. Jnst., xxviii, p. 736. 704 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Geology. in abrupt cliffs almost everywhere. The smaller south-west portion of the island is united to the main portion by a razorback ridge. The surface of this portion is almost flat, and is elevated about 500 ft. above sea-level. It appears to be an old plain of marine erosion, and indicates a former lower level of the land. The only rock found on the island is a moderately coarse muscovite granite of an even-grained character throughout. In colour it is almost white, but where weathered it acquires a pink or even red tint. In section the structure is gneissic, and many crystals show strain-shadows. The feldspar is mostly orthoclase, but there is much oligoclase and a very little perthite. In a few spots a little micro- eraphic structure is to be seen. The muscovite is in large plates, sometimes bent, and there are a few plates of biotite and a few crystals of garnet. Though this particular type of granite has not been recorded from Stewart Island or elsewhere in New Zealand, it is not fundamentally different from many of them, especially from some of those in Stewart Island, and it is reasonable to regard the Snares as part of the same plutonic complex of which Stewart Island and the south-west of Otago are composed. The greater portion of the island was traversed in search of specimens, but only the one type of rock was found, and specimens from different localities were remark- ably similar. The west coast in particular was visited, and it was found that the columnar structure remarked by Hector was a2 result of the jointed structure of the granite. PHYSIOGRAPHY AND GEOLOGY OF THE AUCKLAND, BOUNTY, AND ANTIPODES ISLANDS. ARTICLE XXX. By Roserr Speient, M.A., B.Sc., F.G.S., Lecturer on Geology, Canterbury College, and Assistant Curator, Canterbury Museum; assisted by A. M. Frytayson, M.Sc., Exhibition 1851 Science Scholar, New Zealand University. INTRODUCTORY. THE importance of the geology of the southern islands of New Zealand is enhanced by the fact that biologists have suggested that the existence of a land connection between New Zealand and a subantarctic continent would explain certain anomalies in the distribution of plants and animals. When viewed in that light their geology becomes most interesting. Separated now by wide stretches of sea from the nearest large land-masses, some of them, such as the Snares and Bounty Islands, have apparently all the characteristics of oceanic islands. They are mere dots on the ocean. The Auckland Islands are larger, but they, too, are small when compared with other islands which are truly oceanic. However, evidence will be put forward which will show that they have formed part of a much larger land-mass, which in all pro- bability was of such extent that the land connection asked by zoologists existed either in a continuous form or in a series of slightly separated islands. The evidence as to the date of this extension is not so satisfactory. The following facts proving the existence of a large continental area in the region will be referred to in the report, and after due consideration they will probably be found convincing that such actually existed; but there is no decided evidence from the Auckland Islands to show that the extension was a recent one. There are facts which suggest it, and nothing has been found to contradict it. The following conclusions have been arrived at as a result of the work of the Auckland Island members of the geological party of the expedition :— 1. There is evidence from physiographical considerations that the land in recent times was higher, as only on those conditions could the valleys of the eastern water- shed have been eroded. 2. The islands were undoubtedly subjected to a moderately severe glaciation. This may be explained by an elevation of the land; but it is an unsatisfactory ex- planation, and one that must be accepted with great reserve, seeing that glaciation was fairly general in south temperate regions in Pleistocene times, and that it was in all probability due toysome general cause about which we cannot speak with certainty at the present. 3. Large masses of plutonic rocks occur at the Auckland Islands, the Snares, the Bounty Islands, and at Campbell Island. 46—S. 706 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. 4, Granites, gneisses, contorted schists, and gabbros occur in a conglomerate at Carnley Harbour. This is a fact of the highest importance, as showing the pre- sence of a continental area in the neighbourhood. 5. It seems fairly certain that the great outpourings of basalt which form the main mass of Auckland Island were subaerial, and not submarine, even in the earlier stages. No marine tufis were found underlying the basalts, and a coked peat-bed between two flows of lava not only proves undoubtedly that there was land in the region at that time, but also suggests that the climatic conditions were approximately the same then as now. . With this introduction, indicating the special points which will be referred to more thoroughly later, the geological history of the Auckland Group may be con- sidered in detail—z.e., as far as our limited investigations enable us to express definite opinions. AUCKLAND ISLANDS. ELOY SiO Ga AGE sr ve INTRODUCTORY. The Auckland Islands are situated almost due south of New Zealand, between 51° and 51° 30’ S. lat., and between 165° 50’ and 166° 15’ E. long. They are thus 5 degrees south of the southern part of New Zealand. The depth of the sea between them and New Zealand is nowhere greater than 200 fathoms. An elevation of 1,200 it. would therefore connect the islands with New Zealand. ‘The sea is swept by constant heavy gales, as it lies well within the belt of prevailing westerly winds that encircles the globe between latitudes 40° and 60° 8. The winds have exerted a marked effect, directly and indirectly, not only on the vegetation, but also on the coast-line of these small islands. The group consists of two principal islands, of which Auckland Island proper is by far the largest, being twenty-five miles long and seventeen broad at its widest part. This island will be referred to in this physiographical account as “the mainland.” Adams Island les immediately to the south of 1t, and is separated from it by Adams Strait, or Carnley Harbour, as it is more commonly called. Adams Island is some- what long and narrow, lying parallel with the general direction of the shore of the mainland. ‘This island is nearly fifteen miles long, and about five wide in its widest part, but its average width is little over two miles. At the north end of the group, lying round the entrance to Port Ross, there is a small archipelago, of which Enderby is the chief island. The other islands are Ewing Island, Ocean Island, and Rose Island. To the west, separated by about five miles of sea, is Disappointment Island, of comparatively small size, beg not more than two miles across. The general shape of the group is like a blunt-ended pear, with the thick part pointing south. Coast FEATURES. The sea from which the islands rise is comparatively shallow. Along the north- eastern coast there extends a fringe of submerged or half-submerged reefs of con- siderable area, and stretching out in some cases for several miles from the present coast-lme. These no doubt form part of the island, now much depressed from a former elevation. They are extremely dangerous to navigation. All round the coast de- tached rocks are visible, occasionally far out to sea. Some sunken reefs are known Physiography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 707 to exist—e.g., Bristow Rock, lying five miles north of the entrance to Port Ross ; and no doubt others occur whose positions are at present unknown. Their fre- quent occurrence points distinctly to shallow water in the neighbourhood, and to the island being a remnant of a much larger land. A sounding taken by the “ Hinemoa ” twenty miles north of Enderby Island gave a depth of 95 fathoms. The theory of the former extension of the land is supported by the form of the present coast-lme. The eastern side is a paradise of safe and commodious harbours. Dumont D’Urville called Port Ross, at the north end of the group, the finest harbour in the world. Carnley Harbour, in the south, disputes this claim, for it is a magnificent sheet of water. Between these two there are a number of inlets, of which Norman’s Inlet, Cascade Inlet, and Smith’s Harbour are the chief. While Carnley Fie. 1.—West Coasr or AucKLAND IsLAND, NEAR CAPE Bristow. The precipitous cliffs are 600 ft. high, and are composed of regular flows of basalt and beds of scoria dipping west at very low angles. Harbour occupies "the site of an old volcano, modified subsequently by subaerial erosive agents, the others are drowned valleys, eroded at a time when the land was higher, first by stream-action and then were modified by glacier-action. They are now depressed, so that the sea occupies their lower reaches. The western coast presents a marked contrast. Instead of a series of bays and inlets, it presents an almost unbroken wall of rock for a length of over twenty miles. The perpendicular cliffs, composed of numerous and regular flows of basalt, rise to a height of over 1 200 tt. (fig. 1). It is extremely probable that this coast was 708 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [| Phystography, Geology. somewhat indented in former times, but the prevailing westerly gales and heavy seas have cut back the promontories and left a fairly simple coast-line. The eastern coast, being the sheltered side, has not suffered so much from marine erosion. The south coast of Adams Island is also com- posed of precipitous cliffs, with only one small indentation—Fly Harbour—near its eastern end. This bay, according to Chapman, affords perfect shelter imside a remarkable bar of kelp which crosses the harbour at its entrance. The most remarkable physical feature of the coast-line is Carnley Harbour (fig. 2). This runs almost east-and-west, with numerous arms running to the north, but with a fairly uniform coast-line on the south. The two largest indentations on the north are Musgrave Harbour and North Arm. The latter runs up for nearly six miles, and is about three miles wide at its widest part. To the west of it, and separated from it by Circular Head, hes Musgrave Harbour, and just inside the western entrance is a sheltered sheet of water called Western Harbour. The main harbour has two entrances—the eastern, a mile and a half wide; and the western, extremely narrow and partially blocked by a small island, close by which strong tides ebb and flow in a constricted and somewhat shallow channel. The western entrance is not safe for large vessels. There is a slightly indented bay outside the narrows, which has no doubt formed part of a valley running west. ‘The force of the waves must be very great at this pomt; they run up into this funnel-shaped bay with the weight of the westerly gales behind them, and water pours in enormous masses over parts of the low ridge 50 ft. high (fig. 3.) The height to which it reaches in storms is shown by the fact that starfishes have been found cast up on the cliffs 150 ft. above sea-level. The rdge is in process of rapid demolition from the action of the sea (fig. 4). A similar case occurs at the head of North Arm, but the destruction has not reached such an advanced stage. Barely a mile of land separates the northerly Fic. 2.—PanoraMa oF CArnupy HaArsour. Coleridge Bay in the middle, Musgrave Peninsula in the distance, and the Eastern Entrance in the far distance. ee te ee ae) haere 709 Physwography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. Fie. 3.—SourHERN SpE or Bay ourstpE WeEstrERN Enrranck To CarNnury Harpour. Over those in the foreground the seas vour The cliffs are composed of regular flows of basalt and beds of scoria. during strong westerly gales. Fic. 4.—WeEsteERN ENTRANCE TO CARNLEY HaArpour. 710 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. extension of the harbour from the western coast, and the draimage is exclusively easterly. The western coast-line is thus being eroded, so that large areas of the upper basins of the valleys are being continually removed. The depth of Carnley Harbour, as appears from the Admiralty charts, is nowhere ereater than 50 fathoms, and that is just inside its eastern entrance ; and, from a general examination of the soundings, it does not show the form of a typical fiord. The large bays usually terminate in pebbly beaches, though sandy ones occur occa- sionally. A considerable area of shallow water occurs at the head of North Arm and Musgrave Harbour. The general form of the harbour has resulted from the opera- tion of stream and glacier erosion on the enlarged hollow of a great volcanic crater. Ete. 5.—Portr Ross, wrrn PRovision-DEPor. The harbour of next importance is Port Ross. This is evidently a drowned valley running north-east and south-west, with its unsubmerged portion prolonged to the western coast, so that its upper part has been cut off by the eastward move- ment of the western coast-line. Numerous small sheltered coves fringe its shores, and round its entrance is a circle of small islands, which give unrivalled protection in all weathers (fig. 5.) Norman’s Inlet, Smith’s Harbour, and Cascade Inlet are small harbours with steep sides and deep water, reproducing in miniature all the features of the West Coast Sounds of New Zealand. Their form will be more fully dealt with when treat- ing of the glaciation to which the land-surface has been subjected (fig. 6.) 711 SUBANTARCTIC ISLANDS OF NEW ZEALAND Physiography, Geology. | “LHINT S\NYNUON—9 “SIT 72 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. The coast-line of the islands, as would be expected in such a storm-exposed land, is bold and precipitous, the western side especially so. For miles it reaches a height of not less than 600 ft., and at times it rises to more than 1,200 ft. in one steep perpendicular face. At one place only, near the head of Port Ross, where there is a narrow Cleft, is ascent possible. Magnificent geological sections are thus given of the structure of the island on a line running north-and-south. The cliffs are seen to be composed exclusively of regular flows of basalt and beds of scoria. At their base there sometimes}lies a talus slope of débris ; sometimes the cliffs run down vertically into deep water, but in many places submerged ledges run out from the shore (fig. 1.) This almost inaccessible coast-line was in former times the haunt of the fur- seal. Sealers were lowered by ropes from the tops of the clifis, and, after despatching their victims, were hauled up with their load of pelts, which they carried down by well-known tracks to the sheltered harbours on the east coast. The eastern coast-line is not so high, although it is precipitous in many places. Round the entrance to Port Ross it is very flat, the islands friging its entrance being but a few feet above sea-level. They form part of a plain of marine denuda- tion which has recently been elevated—part of it, indeed, being just above the sea. These areas grade into the dangerous platforms and reefs which fringe the north- easterly coast. A few stony beaches occur, and mside Enderby Island there is a fine sandy beach fully half a mile in length, extending back from the shore to a line of low dunes. These dunes are composed very largely of shell-remains, great piles of which—mostly Mytilus in a more or less perfect condition—are to be found occasionally on the exposed coast-line of the island. SURFACE FEATURES. As a general rule, the highest part of the land is near the western coast-line. It rises at times to just over 2,000 ft. as measured with the aneroid, the highest point of the mainland being in all probability a hill between the head of the North Arm and Western Harbour, just east of Cape Bristow; but elevations of nearly equal altitude occur near the head of Norman’s Inlet, and between it and Carnley Harbour. Adams Island forms a long high ridge, with an elevation of just over 2,000 it. Many of the hill-tops are remarkably flat, and are formed of level flows of basalt. Large areas are covered with shallow pools of water, owing to the want of drainage facilities. Here, too, there are signs that the disintegration due to frost-action plays an important part in subaerial denudation. In some places a considerable portion of the loose surface-covering consists of fragments of augite weathered from the basalt. This is specially marked where the last flows are more doleritic in cha- racter and contain numerous large crystals. The sides of these flat-topped hills are frequently precipitous where the edges of the flows have been exposed. Round the head of Musgrave Harbour they form perpendicular faces between 400 ft. and 500 ft. high, with splendid examples of vertical basaltic columns. In the northern part of the mainland the elevation is not so great, and the Hooker Hills, between Port Ross and the northern coast, rise to a height of 950 ft. The valleys are all occupied by small streams, fed by the prevailing rains and mists, and flow in their upper portion over the solid rock of amphitheatre-shaped Physiography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 713 basins, but in their lower part they pass through irregular masses of angular boulders of all sizes, which have all the appearance of old glacier-moraines. In some cases, as at the head of Norman’s Inlet and Cascade Inlet, they enter the sea in falls coming from valleys high above the floor of the main valley. These are doubtless true hanging valleys, due to glacier-action. As mentioned previously, nearly all drainage is easterly, from the very edge of the western cliffs in many places. At times small tributaries have been captured by the encroachments of the sea, and drain west, falling over the cliffs in tiny waterfalls, which are blown upward and eastward by the strong westerly gales. Fic. 7.—Smani Lake, with OLp Lareran Morarns, CoLertpcre Bay. Owing to the small size of the island, lakes of any extent could not be expected, but small ponds occasionally occur. The largest seen was on the side of the valley leading to Coleridge Bay, a small bay on the north side of Carnley Harbour, between Musgrave Harbour and Camp Cove. It is several acres in extent, the water being retained behind an old lateral moraine. (See fig. 7.) At the back of the lakelet the basaltic cliffs rise almost perpendicularly for several hundred feet, with their lower slopes covered by stunted forest. A small stream flows over a low part of the margin, and empties itself into Coleridge Bay. The party was told by one of the Maori boatmen of another small lake, on the north side of Adams Island, and near sea-level. From the description given by him it seems likely that this lake was ponded back by the terminal moraine of a small glacier coming from a valley on that island. 714 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. SURFACE-COVERING. Nearly the whole surface of the island is covered with vegetation ; only near the tops of the hills and on the sea-shore is solid rock exposed. The upper portion is covered with peat derived from Danthonia and other plants; but where contacts of the peaty covering with the solid rock were visible there seemed to be little true soil produced from the decay of the underlying rocks. On the very tops of the hills, where peat 1s frequently absent, there is a residual clay a few inches in depth, no doubt derived from the basalt, and such occurs in basins on the hill-sides, to which it has slipped or been swept by water-action. The covering of peat seems to exert a protective influence on the rocks, so that beneath it denudation does not proceed at such a rapid rate as it would on surface exposed to general atmospheric weather- ing in a moist climate. The same statement is no doubt true, in a modified form, of the lower parts of the slopes when they are protected by forest and a peaty cover- ing derived from liverworts and other plants of similar character. This protection has important bearing on the rate of weathering, and thus, directly or indirectly, on various biological problems in connection with the group. CLIMATE. The climate of the islands is dealt with in detail in a separate report. (See p. 789.) GLACIATION OF THE AUCKLAND ISLANDS. There 1s evidence that in recent times the Auckland Islands were subjected to a moderately severe glaciation. The direct positive proof of this rests on the fre- quent occurrence of lateral moraines on the sides of both main and tributary valleys. Within the Carnley Harbour drainage-basin good examples are to be seen on the northern slopes of Adams Island, and perhaps best of all on the southern side of the valley running into Coleridge Bay (fig. 7). Here there is a very distinct lateral moraine belonging to an old glacier of the valley type, behind which is a small lake several acres in extent. Close alongside it are two moraines belonging to a more recent tributary glacier. These latter moraines are very similar to the small ones on Ruapehu, and evidently belong to the last phases of the glaciation. On the eastern side of Circular Head Peninsula there are well-defined terraces, several hundred feet above the sea, and running for half a mile in a direction parallel to the axis of the valley. They occupy a position sheltered by a projecting spur at their upper end, and, judging from their appearance as seen from below, they are in all probability old lateral moraines; but this cannot be stated for certain, as op- portunity did not permit of a closer examination. It is not at all likely that they are old river-terraces. On the eastern side of the group, in the larger valleys at the head of Norman’s Inlet and Smith’s Inlet, just north of it, there are lateral moraines of much larger size. These are well seen on the north side of the Norman’s Inlet valley, and again between this inlet and the one to the north, for it is almost certain the Norman’s Inlet glacier invaded the basin of its neighbour through a marked depression in the valley-wall and the moraine formed under the sheltering influence of the bluff which ends the dividing ridge. In Norman’s Inlet, too, there are several morainic Physiography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 715 ridges one within the other, showing that glaciation was becoming less intense ; such ridges are frequently seen in those parts of the Southern Alps where glaciers are retreating. In addition to this undoubted evidence from the lateral moraines, there are certain collections of rough angular blocks of all sizes which he across the lower end of several of the tributary valleys, and at times pass into the lateral moraines. These valleys are distinguished by their cross-section. They are usually wide and U-shaped ; they are never narrow; and the streams which occupy them have no well-defined channel till they cut through the angular débris lying across their mouths. During this portion of their course the streams never expose solid rock, and the boulders are of such a size that they could not owe their present position to the transporting agency of these streams. They are undoubtedly the terminal moraines of small glaciers, and are now so completely covered with peat and scrub that they can only be examined satisfactorily by wading down the beds of the torrents which cut through them. The heads of the valleys show the typical amphitheatre form which charac- terizes the upper part of glacier-troughs, so that they were once occupied by ice, and the irregular accumulations lower down the valleys are the rubbish-heaps of small valley or corrie glaciers. At times the amphitheatre form is double, one being above the other, indicating in all probability a halting-stage during the recession of the ice. Similar landscape forms frequently occur in the valleys heading the sounds of western Otago. There appears, therefore, to be satisfactory evidence that the islands under- went a glaciation of moderate intensity ; but there is evidence that at an earlier date it was much more intense. No large terminal moraines are visible, but they are probably submerged at the present time. Although the floor of Carnley Harbour does not show from its soundings the form of a fiord, yet the sides in places—and particularly the two large peninsulas, Circular Head and Musgrave Peninsula— show the rounded shape which would be produced by the rasping action of a glacier. No grooves or striae were observed; but these could not be expected to be visible, owing to the complete covering of peat. The harbours on the eastern coast north of Carnley give important additional evidence of this glaciation. Norman’s Inlet, one of the most beautiful of these, reproduces on a small scale almost every detail of the great fiords on the south- west coast of New Zealand (fig. 6). There are no overlapping spurs; the valley- sides run down steeply into deep water; the upper end of the valley termi- nates in a great wall-sided amphitheatre; a typical hanging valley occurs on the southern side of the inlet, pouring its water into the sea in a series of cascades. The dividing-wall between the main valley and its tributary hanging valley ends in a sheer bluff like that at Milford Sound, but on a smaller scale. Soundings show that the sound is deeper near the head and shallower towards the entrance. It therefore possesses all the characters of a fiord. Even more striking is Cascade Inlet, just south of it. This is a double valley—one a well-marked hanging valley, with a waterfall from which ships can fill their tanks directly from the falling water. Its sides are precipitous, and the soundings show some remarkable features. The upper portion of the harbour shows a submerged basin, with the soundings reaching 32 fathoms ; but it is crossed by a bar half-way down, where the soundings are only 716 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. 6 and 7 fathoms. Thence to the mouth of the harbour the water gradually deepens. This bar suggests a submerged terminal moraine, and the deeper basin above it an ice-eroded valley. Similar occurrences can be observed in the sounds of Otago, the most noteworthy case being in Dages Sound, where there is a well-marked bar stretching right across the sound about half-way from the entrance. 8) 65 RIE pg i & g Z5 /8 7 19 [§ 20 iz 7 /0 /A i Baie 25 VAL J g 6 Fie. &.—Cascapr Inner, sHowine Sounpines rrom Apwiratty Carr. Although Cascade Inlet shows this character to the greatest degree, others do so to a less extent. The Admiralty chart shows that the sound to the north of Smith — Harbour has the same characters: it is deeper than the sea just outside. Fie. 9.—Norman’s Inner. Fic. 10.—Inuer Norru or Smire Sounp. {Figs. 9 and 10 are on a smaller scale than fig. 8.} It seems, therefore, almost certain that these valleys were filled with ice during the time of maximum glaciation. It is possible that the break in the valley-wall between Norman’s Inlet and Smith’s Inlet may have been due to ice-action, and that it was formed in the same way as the connecting passages between adjacent sounds in western Otago. As the northern part of the group was only imperfectly examined, nothing for or against the occurrence there of traces of glaciers can be definitely said. Port Ross is evidently a drowned valley, but it is uncertain whether it has been sub- jected to glaciation or not. In all probability it has, if glaciers were so extensive in the neighbourhood of Carnley. It must be strongly urged, however, that there is no proof that an ice-sheet existed on these islands in fairly recent geological times, and there is no foundation for the statement made by Professor Park on page 43 of Bulletin No. 7 of the New Zealand Geological Survey, that the expedition found “that the Auckland and Campbell Islands had suffered intense erosion by an overriding sheet of ice that must have had a polar origin.” No boulder-clay or till, no erratics, no ice striae crossing divides, none of the other phenomena usually associated with glaciation by an ice- sheet were discovered, and it does not seem probable that subsequent examination Physiography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. Tt will find them. Glaciers certainly filled the valleys, but it is extremely unlikely that they ever approached even in a modified from to the character of a continuous ice covering, and, further, the biological evidence is totally opposed to its occurrence. The establishment of the fact that these islands were glaciated within recent times is of the utmost importance in its bearing on the origin and present character of their fauna and flora, and especially on their flora. Biological considera- tions are as weighty as geological ones when discussing the cause of the glaciation. This glaciation can be explained most easily by supposing that the land was much higher—certainly several thousand feet higher—than it is now. If we postulate a general refrigeration, from whatever cause it might be—e.g., variation in the heat of the sun, shifting of the earth’s axis, &c.—we are at once faced with the fact that the present subtropical element in the flora would have been wiped out unless it re-established itself subsequently to the glaciation by migration from warmer lands. This would imply a post-glacial elevation in order to make a land connection. In order to explain the appearance of glaciers at sea-level in the locality without eleva- tion of the land, the climate must have been arctic in severity, and this would have been fatal to the rata and tree-ferns which flourish now in sheltered spots on the island. The case of the Franz Josef Glacier, on the west of the Southern Alps, which is so frequently instanced as an illustration of the association of a subtropical vege- tation with glaciers, hardly applies in this connection. The Franz Josef takes its origin in vast snowfields, which are concentrated into a somewhat narrow valley of very steep grade, so that the glacier moves rapidly, and reaches a low level before it melts. The circumstances are quite different in the Auckland Islands. At the present level of the land it would seem impossible for glaciation to occur without destroying the subtropical element in the vegetation, seeing that very few indeed of our New Zealand alpine plants can stand the rigour of the climate of even Kew or Berlin. We are driven, therefore, to consider elevation of the land as an important factor, but not necessarily a predominating factor, in the extension of the glaciation. If it were the sole cause of the glaciation, the land must have reached at least 7,000 ft. in height. Ii we consider there are no present-day glaciers on Mount Anglem, in Stewart Island* (height 3,200 ft.), though they existed there formerly, and large glaciers are rare in the sounds country of Otago, and that it is only when the mountains reach a height of more than 8,000 ft. that glaciers of considerable size occur in the Southern Alps, it would seem reasonable that the Auckland Islands were at least 7,000 ft. high: they may have been much higher. This would allow the eastern valleys to be filled with snow, and would probably account for all their special features; and, further, as the land sank it would fully account for those features due to a waning glaciation. This theory would apparently account for the facts in the present case; but when it is considered that glaciation was fairly severe over a large part of the Southern Hemisphere in Pleistocene times, and elevation cannot be cited in every case as an explanation of it, this apparently simple solution of the difficulty in the case of the Auckland Islands must be accepted with considerable reserve. But there is proof of a slight elevation, whatever the cause of the glaciation. *“ Botanical Survey of Stewart Island,” by Dr. L. Cockayne, 1909, p. 6. 718 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. (55 Captain Hutton accepted the “ elevation hypothesis ” when accounting for the glacier period in New Zealand, and when explaining the biological relationships of the New Zealand fauna and flora,* but he was unaware of the former glaciation of the subantarctic islands. If this hypothesis is correct, his case is greatly strengthened, since, while he used the biological conditions of the off islands to explain the glaciation of New Zealand, we may now use the fact of the extension of the glacial conditions to explain their biological relations. If an elevation of the Auckland Island group by at least 5,000 ft. is thus demanded, and this elevation extended over the whole area south of New Zealand, it would be sufficient to connect the Snares, Auckland Islands, Antipodes, the Bounty Islands, and perhaps Campbell Island, with New Zealand. If it extended further south it would join on Macquarie Island, and thus go far to explain the connection of New Zealand with a former antarctic continent. STRATIGRAPHY AND PETROLOGY. The oldest rocks found in the Auckland Islands consist of masses of granite and gabbro, which are overlaid by flows of trachyte and penetrated by dykes of similar character, as well as by dykes of diabase and porphyrite. Then follows a conglomerate containing fragments of these granites, gabbros, and trachytes, and also of gneisses and granites of different type. A series of porphyrites and diabases form the next group in order of time, although they may occur before the conglomerate; and the whole sequence is closed by vast outpourings of basalt, of which the main mass of the group has been constructed. These basalts are penetrated by a series of dykes of diabase and basic porphyrites which sometimes approach augite-camptonites in composition and general character. Disappointment Island is apparently composed of melaphyres and diabases, which no doubt form part of the fourth series of the main group. The general sequence may be expressed thus in tabular form :— (5.) Younger basic series, with dykes of diabase and basic porphyrites. It is impossible in the present state of our knowledge to assign a definite age to any of these series, but (2), (3), and (4) are almost certainly pre-Tertiary, and (5) is almost certainly of middle or late Tertiary age. (1.) PLutontc Rocks. Granite. The exposure of granite on the Auckland Islands is of very limited extent, but its existence in the group has been previously reported in Musgrave’s account of his adventures in this island, and also by Armstrong and Sir James Hector. The out- crop occurs near the isthmus of Musgrave Peninsula, on the east side of Carnley *“ Origin of the Flora and Fauna of New Zealand,” Ann. & Mag. Nat. Hist., series v, vol. xv. Physiography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 719 GEOLOGICAL SKETCH MAP OF SOUTHERN HALF OF AUCKLAND ISLANDS. = Probable northerly limit of Carnley basalts, ~ — — — - { Cape Bristow Cascade Inlet ~~, Circular ¢ Hd. “ \ HA eer \ Fly Haréour == VI ANU ALA Gabbro. — -Trachyte. VIL im Gi Granite. 39, 5+) o. ;+ Moraines. Iv v a Big ==] Younger basic series. Big WV Older basic series. NOTE.—The boundaries indicated on the map are in many cases very doubtful. This specially applies to the older basic series, and above all to the area to the west of North Arm. This may subsequently be classed with the younger basic series. Moraines, no doubt, also occur in localities not indicated on the map. 720 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. Harbour, and the rock is visible on the shore for nearly a quarter of a mile on either side of that peninsula. It is found also in large boulders in a conglomerate on the opposite shore of Carnley Harbour, near Camp Cove. The rock is very white in colour, but when exposed to the weather it appears brownish at times, owing to a staining of iron-oxides. Large white and pink crystals of feldspar are porphyritically distributed through it, which occasionally reach a length of 5 cm., and sometimes exhibit idiomorphic forms of the Baveno habit. There is very little muscovite present, but a considerable amount of biotite-mica, so 1t must be classed as a biotite-granite. The amount of quartz varies considerably, and the rock is sometimes quite syenitic in character; a considerable amount of plagioclase also occurs. It resembles somewhat the granite from the Bounty Islands, but differs altogether from that from the Snares. The following analysis, by Mr. A. M. Wright, F.C.S., chemist to the Christchurch Meat Company, Islington, shows its chemical composition :— Per Cent. S10, at: tw. Ls ce ne So (ado Al,O, oe i oe Es A: ue OMIENG Fe,0, | 1-44 FeO 3°36 MnO 0-19 CaO 1:97 MgO 2°02 K,O 2-90 Na,O 3-26 Oe Trace Oy a a ae my ate Ree OA CO, : i i ie Oeil Water lost below 100°C. (H,O—) .... cf .. 0-14 a above 100° C. (H,0 +) e x he 0:30 100:65 Specific gravity on 3 Oe ie A partial analysis of the rock by Professor Jarman gave a silica percentage of 73°94. This rock, like the Bounty granite, shows a rather low percentage of alumina, which is also a marked feature of the trachytes occurring later in the Auck- land Islands. It seems highly likely that the granite forms part of a wide great mass of which the Snares and the Bounty Islands are remnants; and perhaps the granite in Stewart Island and in south-western Otago near Preservation Inlet and Chalky Inlet are but local manifestations of the same widely extended plutonic mass. The occurrence of granite in the Auckland Islands is evidence that these islands once formed part of a continental area. Gabbro. The gabbro is found in situ at the end of Circular Head, and also at McClure Head, but its actual extent was not determined owing to lack of opportunity to thoroughly examine the localities. Large boulders of gabbro are found plentifully Physiography, Geology.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. el along the shore of Camp Cove and of the bay between it and Musgrave Harbour, and are of such a size that they indicate the immediate neighbourhood of large masses of the rock in position. Boulders of gabbro were found at Fairchild’s Garden, near the western entrance of Carnley Harbour, and on the beach on the north side of Enderby Island, in such positions that they could not have been carried far by any natural transporting agency. These boulders indicate either the immediate neighbourhood of gabbro in position underneath these widely separated parts of the island, or that they have been brought up from below as inclusions in the more recent basalt. The former alternative appears to be the more probable. These facts taken im conjunction with the occurrence of a sumilar gabbro at Campbell Island, and of allied rocks (norites) at the Bluff, in the extreme south of New Zealand, seem to indicate the existence of great mtru- sive masses of gabbro over wide areas. This intrusion is perhaps contemporaneous with the intrusions of dunite and other basic rocks occurring on the western side of the South Island of New Zealand, and passing through the centre of the Nelson Province towards D’Urville Island, in Cook Strait. Such rocks also occur in New Caledonia, with a connecting-link of norites found recently in the Auckland Penin- sula, west of the Bay of Islands. They thus form part of a great convex arc facing the Pacific Ocean, and enclosing the Tasman Sea area. As the intrusion occurs in Lower Mesozoic rocks near Nelson, and it nowhere penetrates the Upper Cretaceous beds in other parts of the country, it seems reasonable to assign its occurrence to either the Jurassic or Lower Cretaceous period. It is probable that the intrusive masses of the Auckland Island are of that date as well. It seems likely, too, that they are connected with those movements which formed the Southern Alps and brought about the severance of New Zealand from Australia, which no doubt occurred in Jurassic or early Cretaceous times—a movement which in all probability was intimately related with the formation of the Tasman Sea. The petrological character of the gabbro varies a good deal. In some cases it is a dark rock with large crystals of pyroxene standing out on the weathered surface. These frequently show the bronzy lustre of diallage on fracture. There is also a finer-grained type of the rock, which is extremely tough and hard to break. Under the microscope the rocks appears to be an olivine-gabbro with large crystals of augite, which frequently show diallage structure. At times, especially in the coarser variety, schiller structure is well developed ; the augite is shghtly pleochroic in purple tints. The olivine is in rounded grains, common in some sections and almost absent in others, sometimes fresh, sometimes serpentinized, and, again, par- tially replaced by magnetite aggregates. It is frequently completely included in the augite plates as poecilitic grains. The feldspar is labradorite with the usual twinning ; the amount present varies greatly. In the finer varieties it far exceeds the other minerals in importance, but in the coarser type it is subordinate to them. It is usually fresh, in broad plates, but occasionally it is broken up into a mosaic of small grains; small fragments are also included in the augite, in some cases with apparently definite orientation of graphic intergrowths. Primary hornblende is a rare constituent, but a boulder from Enderby Island contains a fair amount of both primary and secondary horn- blende derived from the augite, and thus resembles somewhat the Bluff norite. It is possible that this specimen has been brought from the Bluff as ballast for a sailing- 47—S. 722 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. ship. Buiotite occurs freely, and in the coarse varieties it 1s an important constituent. It is usually in ragged and irregular forms, moulded on the feldspar, and forming zones round the olivine and iron-ores, but frequently numerous flakes are included in the pyroxene. LHnstatite appears occasionally, especially in the coarser types. It was not seen in sections from the boulders at Camp Cove, but it appears in speci- mens taken from the solid rock. Magnetite occurs freely in irregular masses and square-sided grains ; apatite is also present. The composition of one of the coarser specimens is shown in the following com- plete analysis by Mr. A. M. Wright :— Per Cent. Si0. Ne Re © ie 7 Ree .. 49°44 A Ony sume oe. a a - me co. ESS Fe,0, os ne os i irs el ea FeO a S Me - a peal 2r eet MnO Pe Pe ts a ve eo Osa, CaO #0 Ma an oe i mi 7°30 MgO e - Be Es e . 19-26 K,0 ao Bt Rs ae! e eae Le Na,O ue 2 re = a Be 2:91 P.O; < te * i as (O12 TiO, . & - , i oe CO, Ree te ms is Fi Wace Water lost below 100° C. (H,0 —) cf Ce Go ae OUND & above 100°C. (H,O+) .. at ao) Llinace 100-70 A silica determination of another sample of rock gave a result as low as 42°44, which no doubt belongs to an exceptionally basic variety. The specific gravity of a number of specimens was found to range from 3-02 up to 3°23. One interesting feature of the gabbro is its association with a marked magnetic disturbance. The magnetic records show that the general declination of the needle in the islands is about 18° H., while in the Carnley area the following readings occur: Adams Islands, 18°; Camp Cove, 17° 46’; Coleridge Bay, 16° 21’; McClure Head, 7° 49’; North Arm, 20°; Norman’s Inlet, 18° 32’. The marked disturbance will be seen to coincide exactly with the known occur- rence of gabbro in position, or with the close proximity of it. Similar phenomena are associated with the gabbro in the west of Scotland.* The special features of this case seem to be the normal behaviour of the com- pass on a highly ferriferous basalt, but its abnormal behaviour when near a gabbro which is overlaid by the same basalt. In the present state of our knowledge nothing further can be said about this disturbance, but some light may be thrown on it when the final results of the mag- netic survey of New Zealand are fully discussed in connection with the geological structure of the country. * Harker, “ Tertiary Igneous Rocks of Skye.” Physiography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 723 (2.) TRACHYTES. Overlying the granite at Musgrave Peninsula is a trachyte tuff, and above it a trachyte in solid flows and also in dykes and sills penetrating the granite. Tra- chyte dykes occur at Circular Head, running in a north-west and south-east direction —1.e., directly toward Musgrave Peninsula, as the probable centre of disturbance (fig. 11). As they penetrate both the granite and the gabbro, they are un- doubtedly younger than both. The tuff which'rests “on the granite of Musgrave Peninsula, on the eastern side of the isthmus, is white in colour, varying in coarse- ness, with numerous fragments of partially rounded grains of quartz and plates of Fic. 11.—Granite Rocks, wirh Dykes anp Sinus or Tracuyrr, Muserave Preninsuna. feldspar (albite-oligoclase). The rest of the rock is a cementing-medium of kaolin, calcite, and other products. Where the rock is finer in grain, the quartz and feldspar fragments occur more sparingly, and the bulk of the rock is a fine-grained mass of irresoluble matter. It thus shows all the characters of a trachyte tuff largely derived from the erosion of the granite similar to that found beneath it. The trachyte flows and dykes which overlie and penetrate it are much decomposed, and stained with limonite. The microscopic examination is somewhat unsatisfactory on that account, but it appears to be a trachyte of distinctly alkaline type. The phenocrysts are of sanidine, anorthoclase, and sparingly plagioclase. The ferro-magnesian minerals are quite weathered, and are represented by aggregates of iron-oxides whose origin 724 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. could not be determined; but a boulder in a conglomerate at Camp Cove is in a much better state of preservation, and contains an apparently alkaline hornblende of greenish colour with strongly marked pleochroism. The groundmass consists of quadratic-shaped feldspars, and also a fair amount of free silica (fig. 14,1). The dykes occurring at Circular Head are of similar type, with slightly more quartz and a considerable quantity of chlorite, and one specimen shows the characters of a coarse-grained bostonite. The feldspars are arranged in sheaf-hke masses, or exhibit flow structure, and anorthoclase is a prominent pheno- cryst. The analyses of these two rocks show them to be of a distinctly alkaline type, with a very low percentage of Al,O,. This percentage I had confirmed by the analyst, who had no knowledge when they were submitted to him that one rock was from a flow and the other from a dyke from the same volcanic series. The two analyses, A and B, will be found also to be mutually confirmatory :— A. B. C. SiO, he a .. 66:02 62:81 59°52 NO by na Peu-83 10:36 11-48 Fe,0, ‘6 e .. 2:08 592 1-62 FeO Se ae so Arehs) 2°59 9-93 MnO Ly a 5 pple 0-11 0:22 Ti0, a i i061 0-71 0:27 CaO By été se LOS: ibe 7/8) 4-11 MgO % ‘ . 0-95 0:47 1-03 K,0 yy: i be SE) 3-64 2-03 Na,O m i 2 BRS 7:73 5-31 P.O; - a on lieve Trace None CO, oe ee ,o Irae 0-17 0:4] Water lost below 100° (H,O —) 1-21 0°81 0°82 Be) Mabove 1002 (HERO) = a l-07 9-44 2:88 100-46 99°55 99-63 Specific gravity wie .. 2:48 2°44 els] A: Dyke, Circular Head, Carnley ; analysis by A. M. Wright. B: Flow, Musgrave Peninsula, Carnley ; analysis by A. M. Wright. C: Pitchstone dyke, east side of Musgrave Peninsula, Carnley; analysis by A. M. Wright. An examination of these analyses show the rocks to be of very pronounced alkaline type, approaching a pantellarite or comendite in general character, with the exception of the low percentage of iron-oxides. It seems, therefore, most likely that the iron-oxides’and chlorite have been derived from an alkaline hornblende or augite. A pitchstone dyke belonging to this series was found on the east side of Mus- grave Peninsula, and a boulder evidently from another similar dyke was found be- low Musegrave’s old camp on the western side of the peninsula. The rock is of dull glassy’ appearance, and under the: microscope it appears as a brown glassy rock crowded with slender microlites showing flow structure, especially im the neighbour- Physiography, Geology.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 725 hood of the phenocrysts. The microlites are of a pale-green, alkaline augite, with high extinction-angle. They are elongated and straight, frequently with sheaf-like termination, or with a ragged border along one side. The phenocrysts are of an oligoclase-andesine, a green slightly pleochroic augite in fairly large crystals, and with included grains of magnetite. An analysis of this rock is given above (C). It shows distinct relationship to the trachytes, but is less alkaline in character and more basic, with a very high percentage of ferrous iron and the very low percentage of alumina which apparently characterizes rocks of this series. This rock is ex- tremely like a trachytic pitchstone which occurs as a small dyke at Governor’s Bay, near Lyttleton, New Zealand; in fact, the trachytes of Carnley Harbour show a close resemblance to certain alkaline trachytes from this locality. There occur in the trachyte at times dark-coloured bands, apparently com- posed of coarse black augites and small decomposed olivines, in a black crystalline base. Under the microscope the groundmass of the rock is even-grained and fairly coarse, with an ophitic structure in places. It 1s composed of long narrow twinned laths of labradorite, grains of violet augite, and grains of olivine with borders and seams of limonite, and a good deal of magnetite in squares and irregular grains. The phenocrysts are of augite, in coarse, fresh, pleochroic crystals, with occasional inclusions of small plates of feldspar, of olivine altered to serpentine and iron- oxides, and feldspar as occasional small rectangles of fresh labradorite. This rock is mineralogically a dolerite, and may be a dyke, but its field occurrence did not suggest this as at all likely. In Hartmann’s account of the basalt of Auckland Island he mentions the occurrence of a series of trachyte older than the basalts, but without giving the localty.* As the German expedition had its headquarters at Port Ross, and as he does not mention the presence of gabbro or granite, it seems to be a safe inference that there is an exposure of a trachyte near Port Ross in the same stratigraphical position as that at Carnley. Want of time and adverse weather-conditions pre- vented a thorough examination of the northern end of the main island. Specimens of trachytes evidently derived from dykes were picked up on the shore of Coleridge Bay, but they could not be traced to their origin. The anortho- clase in some of the specimens is very apparent. Another small specimen which evidently belongs here, though it is more decidedly basic, contains numerous crystals of a rhombic pyroxene ; this is the only case noted of the occurrence of this mineral in the volcanic rocks of the group. The fragment was very vesicular, and deeply weathered. Otherspecimens were found in the same locality with much chalcedonic quartz filling vesicles, and the feldspars completely altered and partly replaced by small crystals of epidote. Some of these specimens undoubtedly belong to the trachytes, but others belong to the lower basic series, the underlying gabbro being in close proximity to the place where the specimens were found. (3.) Camp CovE CONGLOMERATE. The next formation in point of age is a conglomerate which is best developed at Camp Cove, but extends in a northerly direction through the peninsula behind the depot. The conglomerate is apparently an old river-deposit, or perhaps a shore- * “ Ueber Basalte der Aucklands Inseln,” by Max Hartmann: L.J., 1878. 726 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. deposit, though the former seems more likely. It is full of rounded pebbles of granite and trachyte like that in position at Musgrave Peninsula, of gabbro, and of various gneisses and schists. The latter are at times contorted, and with well - marked schistose structure. There are granites of a different type from that at Musgrave Pen‘nsula, but they are nearly all biotite-granites, with small flakes of muscovite occasionally visible in some slides; no specimen like the Snares granite was dis- covered. The gneisses contain little feldspar, and under the microscope do not show marked schistose structure, being of a granular type. The trachyte occurs occasionally in large boulders, one being over 2 ft. in diameter. The conglomerate is penetrated by the same dykes that penetrate the overlying basalts, one being exactly similar to a dyke of camptonitic type occurring on the opposite shore of Adams Island. These facts show that the conglomerate was formed subsequently to the gabbros, granites, or trachytes, but earlier than the basaltic flows. No pebbles of diabases or porphyrites were found; but this does not say that they do not occur, as only one morning was spent in examining the locality. The most noteworthy fact about the conglomerate is that its contents prove absolutely the proximity of a continental area at a time immediately anterior to the outbreak of volcanic action. In what direction this land lay is at present undetermined, but it is possible that the granite rocks at the Snares and Bounties are remnants of the area which supplied the pebbles to this conglomerate. The presence of con- torted schists is especially important, as these show the probable presence of an area of metamorphosed sedimentary rocks. No traces of any fossil remains were - discovered to help to fix its age. At the extreme end of Musgrave Peninsula a large mass which has been included in the basaltic rocks and brought up from below shows a mixture of fragments evi- dently derived from a granite area. This is apparently a tuff contaiming fragments of quartz, orthoclase, and plagioclase feldspars, and occasional fragments showing distinct granophyric structure. (4.) Tue OLpEeR Basic SERIES. The basic rocks in this area are divided in all probability mto an older and a younger series. The older is best developed at Musgrave Peninsula, on the east side of Carnley Harbour. ‘This peninsula is about a mile in length, and half a mile wide at its broadest part. A rapid examination was made at various points of the shore-line, and the peninsula was traversed from one end to the other, but no rock was seen outcropping on the top, owing to the complete covering of peat and scrub. The whole peninsula is apparently composed of basic rocks, with the occasional exception of rocks of more acid type, and is built up on the foundation of granite and trachyte which is exposed near the isthmus. The general dip of the beds 1s towards the south, but there may be alterations which were not noticed. Although it is built up principally of flows, dykes of a similar type are extremely frequent, and it is at times very difficult to tell the dykes from the flows, especially as some of the latter are very coarse in texture. Rocks which evidently belong to this series occur at Disappointment Island, and it is extremely likely that they occur on the west side of North Arm, on Circular Head Peninsula, and on the spur terminated by McClure Head, as fragments of porphyrites were picked up in the latter locality, although the rock was never seen in position. It is possible, therefore, that an Physiography, Geology.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 727 extensive area has been covered by these rocks, and they may be of the same age as the porphyrites and diabases from Macquarie Island described by Marshall,* and also by G. T. Prior.t Some of the specimens got from the end of Musgrave Peninsula, so strongly resemble certain of the recent basic series that they may well belong to it. One interesting discovery was the occurrence of a thin bed of impure anthracite between two flows. This had a distinct metallic lustre, and the rough columnar structure frequently observed in cases where sedimentary rocks have been subjected to contact metamorphism. This, no doubt, has been formed from a coating of peat which has been coked by the heat of a superjacent lava-stream ; when burnt it left a large quantity of reddish ash. Its occurrence is noteworthy, as showing that even in the period of volcanic activity peat occurred over parts of the country, proving that the climatic conditions were in all probability the same then as now ; and, further, that the volcanic flows were not submarine, and that a land surface existed in the locality at the period of volcanic activity. There are also persistent reports that similar deposits of such coal occur in the north-western part of the eroup. In connection with this occurrence of peat turned into coal, 1t may be mentioned that Captain Bollons, of the Government steamer ‘‘ Hinemoa,” presented a member of the expedition with a piece of coal from Antipodes Island. It shows the effect of heat, in that itis hard and has columnar jointing, but the peaty structure is still quite apparent. Moseley, in his “ Notes by a Naturalist” (p. 173), mentions the occurrence of coal between basaltic flows at Kerguelen Land, and his description would exactly fit the occurrence at Carnley. The oldest rock of the series is a typical diabase, but the most common rock is a diabase-porphyrite, which has the following composition, the analysis having been made by Mr. A. M. Wright :-— Per Cent. Si0, oe = i fe! se .. 474) Al, O03 Ne o = si oe na 6208 Fe,0, oh me ne ce ee e292 FeO 8°37 MnO 0:27 Ti0, 0-31 CaO 4:03 MgO 1:16 K,0 215 Na,O 3:49 les 0) ee me 2 & Me -. race CO, * rs +f 2 Oe) 202K Water lost below 100° C. (HO —) = r i. 1:42 above 100° C. (H,0 +) ~ re - 1:38 99°30 Specific gravity ee a fs vs 2200) * “ Some Rocks from Macquarie Island,” Rep. A.A.A.S8., vol. x, 1904. 7G. T. Prior, Nat. Ant. Exped., Geology, p. 109. 728 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. The analysis shows the rock to be of a basic type, with a remarkably high per- centage of iron-oxides. This is confirmed by the microscope, as the groundmass contains an abnormal amount of titaniferous magnetite. There are phenocrysts of andesine-labradorite, and occasional small phenocrysts of augite, grains of which mineral also occur in the groundmass. This rock is probably an old lava-flow. The special petrological characters of the diabases appear to be the presence of purple pleochroic augite, often in large ophitic plates, and a remarkable amount of titaniferous magnetite. Hpidote is a common constituent of some of the rocks, and secondary hornblende and mica are sparingly found. In many cases there is much chlorite present, probably derived from the augite ; apatite needles and occa- sional grains of quartz also occur, and in some of the later flows there is much ~ secondary calcite. Besides the true diabases, diabase-porphyrites also occur, and these grade into feldspar-porphyrites of more acid type. The age of these rocks cannot be exactly determined, as they are associated with no sedimentaries, but from their petrological character it is almost certain that they are older than early Tertiary—how much it is impossible to say. The dykes penetrating these rocks are of varied character; there are dolerites with much olivine in large crystals (no doubt belonging to the later basic series), porphyrites, and diabases (fig. 14, ii), and a large development of a coarse-grained porphyritic olivine-diabase. This rock contains large feldspars (labradorite), up to lem. in length, and occasional augite phenocrysts im a coarse holocrystalline groundmass composed of feldspar laths, altered augite, and olivine. The ground- mass also contains a considerable quantity of chlorite derived from the F.M. minerals, and much titaniferous magnetite in grains and comb-shaped forms ; a good deal of the latter is no doubt of secondary origin. Secondary carbonates occur, and apatite needles are occasionally seen. The following analysis, by Mr. A. M. Wright, shows its composition, and accords with the microscopical characters :— Per Cent. SiO, mo ie a _ = .. 49-16 Al,O; oa eS ne nee f aah ease Fe,O, “ te fe ee i 1) e864 FeO Se es fe nee eis ne 7°22 MnO me ee a “4 Me a. 0:29 CaO ue 5-26 MgO 8-34 K,O 0-61 Na,O 2°72 Ol Trace Ti05 a x i a - 0-48 CO, i . - a a OT Water lost below 100°C. (H,O—) .. a .. 0°44 - above 100° C. (H,0 +) i A ve OOS 100°12 Specific gravity a ie Ai os ae 0 Physiography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 729 (5.) THe YouncER Basic SERIES. The basalts which have built up the greater part of the group were poured out principally from a centre at Carnley Harbour. Other foci no doubt existed to the north, but of that little can be said definitely. On the west coast, nearly opposite Disappointment Island, there is a structure exposed in the cliffs which may be the pipe of a volcano. In any case, the northern limit of the Carnley lavas appears to be just north of an east-and-west line through Norman’s Inlet, as there appears to be a change in the dip of the flows to the northward of this line. The northern part of the island was not examined sufficiently to give any definite opinion of the place of origin of the flows. The following descriptions therefore apply to the southern end of the group, although the conditions obtaining in the north appear to have been somewhat similar. Carnley Harbour is no doubt an explosion crater, or caldera, which has been subsequently modified by marine and subaerial erosion. The eastern entrance forms the barranco, while the western entrance has been formed by the destruction of the crater-ring, chiefly by marine action. The breadth of the base of the cone at sea- level is now seventeen miles, and no doubt it stretched far out east and west and south in former times, even supposing the land as a whole was not more elevated than at present. There is distinct evidence, however, that it was much higher. The estimate given by Sir James Hector that the volcano was fully 12,000 ft. high is based on a misconception.* He believed that the centre of the volcanic system of the group was at Disappointment Island. This is now known to be incorrect. His estimate of elevation based on the slope of the flows is also in error, for the reason that the dip of the flows on the west coast, especially in the south of the group, is distinctly westerly, at very low angles. The present distance across the Carnley crater-ring 1s now about seven miles from north to south, but stream-action has considerably modified any regularity of outline if 1t formerly existed. Even allow- ing for a good deal of denudation, the height of the volcano could never have been much more than at present, unless the islands were raised as a whole. The appear- ance of the harbour is strongly reminiscent of Akaroa. Instead of one pear-shaped peninsula, with its core of syenite, as at Akaroa, there are two similarly shaped peninsulas, with a base, one of gabbro and the other of granite; nevertheless, the height of the crater-ring, the character of the lavas, and the marine erosion of the outer coasts cause a striking resemblance between the Auckland Island volcano and that of Banks Peninsula. The whole system at Carnley has been built up of successive flows of basic lava, with intervening tuff-beds of remarkable regularity. The sequence is best seen on the western cliffs, near the head of the North Arm. Here one cliff, 1,200 ft. in alti- tude, gave exposures of forty-two successive flows ; another, of 800 ft., gave twenty- nine flows. They are distinguished, too, by their remarkable flatness, so that the lava must have been ejected in a very fluid condition. This suggests that the island must have been of very great extent in former times. The flat points stretching far out into the sea on the eastern and north-eastern coast, which suggest by their form a plane of marine denudation, may, perhaps, be the upper almost-level sur- *“ Note on the Geology of the Outlying Islands of New Zealand,” Trans. N.Z. Inst., vol. xxvii; 1896. 730 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. Fig. 13 —Curmrs or Conumnar Basairt, Enprrpy Isnanp. Physiography, Geology.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 731 face of one of these flat extensive flows. The hill-tops, too, show the effects of this horizontal character. They are very frequently almost flat over wide areas, so that distinct peaks are exceptional (fig. 12). When their edges are exposed on the sea-shore, or round the heads of valleys, columnar structure is exhibited in its most perfect form. The south shore of Enderby Island (fig. 13) and the head of Cole- ridge Bay and Musgrave Harbour afford splendid examples of this structure. The lava-flows are cut by a series of dykes which are of basic character— viz., diabases, and basic porphyrites approaching augite-camptonites. None of these were observed cutting the higher flows, although the hill-tops were easily and extensively examined, but dykes were frequently seen on the shore. They may extend upward for some distance, as the thick covering of scrub and tussock and peat completely masks the surface for neazly 1,500 ft. PETROLOGY OF THE CARNLEY BASALTS. In order to determine, if possible, the sequence of the flows, specimens were taken from all exposures up a ridge from sea-level to the summit, both at Adams Island and at the head of North Arm. Unfortunately, these are necessarily very im- perfect, owing to the surface being hidden by a close plant covering. In order to give a clear idea of the sequence, a description is given of each specimen taken ; but it must be remembered that these are from flows exposed on the shore, and then with a ereat hiatus—from those exposed for the upper 300 it. or 400 ft. or more. Thus the middle and lower portions are unrepresented. Again, difference in texture of the rocks may be due to the position from which specimens were gathered from a flow: those taken from the middle would, im general, be coarser than those from the edge. ADAMS ISLAND SERIES. No. 1. Hornblende Basalt. Macroscopic.—A dense rock, very fine grained; only an occasional feldspar phenocryst present. Microscopic.—The groundmass is fine-grained, with occasional portions of slightly coarser grain. It is partly glassy, and contains small laths and microlites of oligoclase-andesine, simple or once twinned, scattered grains of magnetite, and a good deal of secondary chlorite derived from hornblende. There are occasional fresh twinned plates of plagioclase between oligoclase and andesine, and small grains of brown hornblende. Norr.—This is here the lowest member of the basic series which is exposed, and is very like a rock lying immediately over the conglomerate at Camp Cove. No. 2. Basalt. Macroscopic.—A black fine-grained rock, with small phenocrysts of feldspar and, in lesser proportion, augite. Microscopic.—The groundmass shows a slight flow structure, and is especially fine-grained in the neighbourhood of phenocrysts. It contains a large proportion of feldspar laths, a good deal of violet augite and magnetite, and a little brown hornblende. An occasional large grain of augite occurs, but the dominant pheno- cryst is andesine-feldspar in fresh squares and elongated plates. 732 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. No. 3. Dolerite. Macroscopic.—A black rock, with numerous coarse pale-green crystals of olivine. Microscopic.—The groundmass is fine, and composed of feldspar laths, granular augite, and magnetite grains. The phenocrysts are augite in pleochroic crystals, often with a violet border, a little highly serpentinized olivine, and andesine-labradorite feldspars. The term “dolerite” is applied in this report to a coarse-grained basalt (fig. 14, ii). No. 4. Dolerite. Practically the same as No. 3, except that the augite crystals are very coarse and the olivine in smaller grains with limonite border, and the feldspar is consider- ably weathered with the formation of secondary calcite. No. 5. Augite-olwine Andesite. Macroscopic.—A fine-gramed dense rock; the phenocrysts are small, chiefly of feldspar. Microscoprc.—The groundmass is of medium grain, composed of feldspar laths showing flow structure, fine magnetite, and granular augite. The phenocrysts are oligoclase-andesine feldspars, highly weathered olivine, and a few pale augites. No. 6. Dolerite. Macroscopic.—A black rock, with numerous phenocrysts of augite and olivine of varying size. Smaller feldspars also occur. Microscopic.—The groundmass is fine, and made up of feldspar laths, augite, olivine, and magnetite, the last in both coarse and fine crystals; the granules of olivine are somewhat altered. The phenocrysts are coarse, and consist of olivine in various stages of alteration, slightly pleochroic augite with characteristic dark border frequently developed, and tolerably fresh labradorite. No. 7. Dolerite. Macroscopic.—The rock is black, with a good few phenocrysts of medium- sized olivine, augite, and feldspar. Microscopic.—The groundmass is fine, composed of feldspar laths, grains of magnetite and violet augite, plates and irregular patches of chlorite. There are a few small pieces of pleochroic brown hornblende with included grains of magnetite. The phenocrysts are of fairly coarse pleochroic augite with violet border, labradorite often finely zoned with strings of secondary chloritic products, and olivine generally fresh but sometimes serpentinized. No. 8. Dolerite. Macroscopic.—This rock is decomposed, and full of vesicles. Microscopic.—Similar to No. 7, except that chlorite has completely replaced some original ferro-magnesian constituent. No. 9. Dolerite. Macroscopic.—A coarse-grained rock, full of large augite and olivine crystals. It is much weathered. — Physiography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 733 Microscopic.—The groundmass is composed chiefly of granular augite, olivine, and magnetite, with a lesser proportion of feldspar laths. Phenocrysts are of very coarse slightly pleochroic augite, occasionally twinned, and coarse round olivine erains, with a border and seams of serpentine and limonite. This rock is similar to that found capping the hills in other parts of the group, and from it the large augites have weathered which are found occasionally on the surface in considerable quantity. The following are analyses made by Mr. A. M. Wright—No. 1, a basalt from shore at Adams Island; No. 4, a dolerite from a height of 1,700 it. at Adams Island :— Noma: No. 4. Per Cent. Per Cent. SiO, ue i: * .. 46:18 47-14 BOs ee y ie .. 16°82 19°35 Fe,0, i - : pe 08 561 FeO ie =) a .. 13°46 10°82 MnO 5a ane ba ae 0:27 0-18 TiO, a - .. O51 0-68 CaO Se aN ae ay 4°81] 5°42 MeO i x! es oy ce 3:98 K,0 1:83 1-60 Na, O 5-02 238 P.O; 1-05 0-14 CO, ae 0-31 0-26 Water lost below 100° ©, (H,0—) 0-89 0-84 above 100° C. (H,0 +) 0°82 0:96 99-73 99°36 Specific gravity sh 2°88 Te 96 These two analyses were made ote. eles belonging to the fine type and the coarse type. They show no marked difference in basicity, but a high per- centage of ferrous iron ; the percentage of magnesia is somewhat low. In No. 1 we have a very high percentage of soda, suggesting the presence of a felspathoid, but no sign of this could be observed. There seems to be no well-defined order in the character of the lavas extruded, but merely an alternation of coarse- and fine-grained types right from the earliest rocks erupted. A similar series of specimens, taken at the head of the North Arm, was prac- tically basaltic throughout, with the exception of a basic andesite occurring near the top of the series. The series shows the same alternation of coarse- and fine-grained rocks. The following features of the minerals of the basaltic series may be noted: The augite is notably titaniferous, as shown by the frequent pleochroism and dark-violet border. Ilmenite or titaniferous magnetite is of very common occurrence not only in the basalts, but in the diabases and diabase-prophyrites which occur as dykes penetrating the basalts. The feldspars are not of very basic character, and vary 734 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geoloqy. for the most part between andesine and labradorite: this gives certain of the rocks an andesitic character which the analysis does not support. Analysis No. 1 (given above) is from a rock of this character. The marked absence of any F.M. minerals, except augite and olivine, is very noteworthy. Hornblende is very scarce, even in the earlier flows. These series may be taken as typical of the basaltic rocks of the southern por- tion of the group. As a rule, the coarser-grained types seem to be more common than the fine ones, though the finer types seem to be relatively more abundant in the eather flows. This statement may, however, have to be modified after further lavestigation. Samples of the basalts of the northern end of the group were col- lected by the German Transit of Venus Expedition in 1874, and were thoroughly described by Hartmann ;* and there is also a reference to the basalts of Port Ross in the geological report of the “ Discovery ” Expedition. Nothing is known about the rocks of the middle portion of the main island—+.e., the part north of Norman’s Inlet and south of Port Ross—but there is reason to expect they are the same as the rocks at the north and the south ends of the group. It is almost impossible to assign an age to these basalts, but from their fresh- ness and from the well-preserved form of the caldera-walls it may be inferred that the eruptions took place in middle or even late Tertiary times. VOLCANIC DYKES. The voleanic dykes at Carnley may be classified as follows: (1) Trachyte dykes of an alkaline character found penetrating gabbro and granite and not found elsewhere ; (2) porphyrites of a relatively more acid type found penetrating gabbro and the porphyrites and diabases of the lower basic series; (3) diabases and basic porphyrites found in gabbro and basalt; (4) dolerites found in the lower basic series. This classification mdicates the relative age, with the exception that (3) and (4) may be more or less contemporaneous, and date from the last phase of the volcanic action. The trachytes and more acid porphyrites have been referred to previously, but the third series has not been mentioned. The dykes of this series were observed on the shore of Carnley Harbour, near Camp Cove, on the end of McClure Head and Circular Head, and on the north shore of Adams Island. Those observed had a northerly strike, varying from north- east to north-west, but the harbour-foreshore was not covered sufficiently to determine their general orientation. The diabases are very fine grained, and without the typical diabase structure; they are rarely ophitic, and are occasionally much more decomposed than the basic rocks they penetrate, and for no apparent reason. They are almost certainly of Tertiary age, so that the term “ diabase ”’ 1s not used here in its strict signification, but is applied in this case to a decomposed rock of dolerite affinities. They are sometimes somewhat acid in character, but at other times they contain much granular augite. The porphyritic dykes occasionally approach augite- camptonites in character, with violet-coloured augite and very rarely with hornblende *““ Uber Basalte der Aucklands Inseln,” by Max Hartmann: L.J., 1878. +G. T. Prior, Nat. Antarct. Exped., Geology, p. 109. Nore.—I have not seen the paper on the subject in the Mineralogical Mag. (1899) by the same author. Physiography. Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 735 IV. Fic. 14. I. TrAcHyrE rrom Dyker, Crrcunar Heap. II. Basaur, Carntey Harpour. III. Drapasz, Muserave Prenrinsuna. 1Y. Bastc Porpuyrire, Dyke, Apams Istanp. 736 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. as well in the groundmass, but they have very few idiomorphic augite phenocrysts, - the phenocrysts being an occasional andesine-labradorite feldspar. The following analyses, by Mr. A. M. Wright, are typical of these rocks (A is a basic dyke from Adams Island, and B a basic porphyrite dyke in gabbro from Carnley Harbour) :— A. B. Per Cent. Per Cent. Sin ee a te a .. 43°62 47-61 AOR eee e i a 6 IGS 17-47 ieOngn a - ve i eac88 3-44 FeO ne ne ne be on 9°36 8-78 Win) oh ne big Bes 0:22 0-13 MO. ow - e mt Oe 0-87 C20t a as ak so Os 7:04 MgO ... “ e eo Je 706 2-49 KOO of “A # .. 1:60 1-60 NER One x Ri , ee 4-49 ic Oa ¥ a ad dle @ Cis 1:10 GOs 2m te = tp .. 0:28 0-28 Water lost below 100° C. (H,O—) .. Seager al 0-84 » above 100° C. (H,O +) .. ane 0°84 3°71 99-56 99°85 Specific gravity .. Be si ek 28s 2°68 The first analysis shows a close approximation to augite-camptonite ; the rock appears under the microscope as even-grained and holocrystalline. It is largely composed of feldspar laths; much granular augite, frequently with a violet tinge, and distinctly pleochroic ; decomposed olivine, with selvage and seams of limonite ; and a good deal of magnetite. Occasional apatite needles are present. These phenocrysts are of labradorite, with an occasional augite, which does not exhibit idiomorphic outline. There are also numerous cavities in the rock filled with secondary carbonates. The second analysis is from a rock of more acid character, and is typical of the basic porphyrites. The groundmass is composed of feldspar laths, granular violet augite, much fine magnetite, and needles of ilmenite and apatite. The augite is largely decomposed, with formation of chlorite and ilmenite. A good few pheno- crysts of andesine occur (fig. 14, iv). The dolerite dykes were only observed penetrating the lower basic series at Mus- erave Peninsula, but no doubt they occur in other parts of the area. The rocks are generally coarse-grained, with conspicuous phenocrysts of augite, olivine, and sometimes feldspar. Under the microscope the augite phenocrysts are usually shghtly pleochroic, the olivines altered to serpentine and limonite. The feldspar is labradorite. The groundmass is usually fairly coarse, and is composed of twinned laths of labradorite and grains of titaniferous augite and magnetite. Comb-like forms of ilmenite (titaniferous magnetite) occur frequently in some specimens. Physvography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. DISAPPOINTMENT ISLAND. Disappointment Island (fig. 15) is a small island, from one and a half to two miles across, lying five miles to the west of Auckland Island. It is not so high as Auckland Island, but the coast-line is, in general, bold and precipitous. The sea be- tween it and the mainland is comparatively shallow, with one prominent reef rising well above sea-level, about half- way across. The general physiographical conditions are the same as those of the mainland. The island was visited by Dr. Marshall during the course of the expedition, and the following statements are based upon his notes. Sir James Hector assumed that the island represented the centre, or core, of a great volcano, and its preservation was due to the greater power of resistance of the plutonic mass forming that core. This is apparently not the case. The main mass of the island appears to consist of diabases and melaphyres somewhat similar to those at Musgrave Peninsula. The structural planes visible at the landing- place have in general an east- and-west strike, and a dip to the south of 60°. In other places these beds appear flat, and then again bent into folds. Cutting these rocks there are nearly vertical dykes of white rock striking north-west and south-east. The petrological character of the rocks makes it almost certain that they belong to 48—S. Fic. 1—Disaprointmunr Isuanp. 738 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. the Musgrave Peninsula older basic series. They are all dark-coloured in the hand- specimen, some appearing coarse and others very fine grained, the latter contaiming at times a marked amount of pyrites. The microscope shows that one of the coarser specimens is principally composed of even-sized deeply kaolinized feldspar (labra- dorite) laths, with biotite and chlorite probably derived from interstitial augite ; there is no apparent sign of ophitic structure. Much apatite occurs in the form of needles, and much ilmenite in quadratic grains and skeleton crystals. Another specimen of rock is a diabase-porphyrite, with large phenocrysts of labradorite and purplish augite. The other rocks appear to be melaphyres, and are much decomposed. The dykes are white, very hard, and somewhat flinty in appearance. Under the microscope they prove to be much more acid than any rock found on the main- land, and must be classified as rhyolites. The phenocrysts are of highly weathered untwinned feldspar and much quartz in irregular fragments. These are contained in a felsitic groundmass. A small amount of glass is visible under higher powers, but the sections are obscured by the alteration of the constituents. Minute zircons are occasionally seen, and, in addition, very small fiakes of a mineral pleochroic in ereenish-yellow tints and small extinction-angle measured from the cleavage, in all probability an alkaline hornblende. The iron-ores assume mossy forms, and are no doubt wholly or partly derived from a hornblende or related mineral. These dykes may be connected genetically with the mailand trachytes, which contain a small quantity of tree quartz, and are of a distinctly acid variety ; but the textures of the rocks are quite different. BOUNTY ISLANDS. The Bounty Islands are a group of small islets lying in long. 179° EH. and lat. 47° 43’ 8. The group takes the shape of a rough semicircle facing north, the length from east to west being about three miles and the width from north to south about two miles. There are nine islets of appreciable size, the largest being about three- quarters of a mile long by half a mile wide. Besides these there are numbers of rocks, and also outlying reeis belonging to the group. The highest point is only 290 it. above the sea, but several of the islets rise to over 200 it. They show the marks of severe marine erosion, and are cut into in-a remarkable manner; but there is no bay or indentation which affords even moderate shelter in stormy weather. During calm ‘weather the sea breaks heavily on the weather shore, but in severe storms no part is beyond the reach of the waves and flying spray. The rocks near sea-level are worn smooth not only by the action of the breakers, but also by the polishing-action of the feet of the seals and millions of penguins and other sea-birds which make the islands their breeding-place. The general rock- surface is as slippery as glass, and exceedingly difficult for man to travel over. Immense quantities of guano are deposited on the islands during the breeding season, but during winter storms it 1s swept off, with the exception of that which accumulates between the boulders. A little brackish water impregnated with guano collects in hollows on the higher parts of the islands, but there is no regular supply of fresh water, except that afforded by the almost constant rams. Owing to the unfavourable conditions, only one species of fresh-water alga has managed to maintain a foot- hold on the exposed surface of the rocks, but great masses of seaweed are torn away ™ Physiography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 739 by the storms and piled up above the reach of ordinary waves, and this teems with the lower forms of animal-life. No member of the expedition visited the group during the 1907 trip of the “ Hinemoa,” but Captain Bollons brought back speci- mens of the rocks on this and on other occasions. These prove to be granites of a fairly constant type. They are whitish or light brownish on fracture, and show plainly quartz, feldspar, and biotite-mica. The amount of this last mineral varies somewhat in different specimens, and in some it is by no means a prominent con- stituent. Under the microscope it shows the characters of an ordinary biotite-granite. A small amount of plagioclase feldspar (albite) and a very occasional minute flake of muscovite also occurs ; needles of apatite are also visible. The amount of feldspar present is sometimes much in excess of the quartz, though occasionally the latter mineral approaches the former in importance. The feldspar shows at times the development of sericitic material along the cleavage-planes. This. granite shows a much closer resemblance to that found am situ at Carnley Harbour than it does to the Snares granite. An analysis by R. Morgan, of the Auckland University College, shows that the rock has the followimg chemical composition :— Per Cent. SiO, * ly a ats a a5 | eros! AlO, A ie Yi be: He Se ed lilegle He50,, me ae ne = a ag . FeO Ss Ks br re $ a 212 CaO ie an sd be 43 aA 1-93 MgO a a ay ¥s Bs ee 0223 K,0 ns : e fs ie 6:57 NaO me ee io Ne ae Be 3°58 Eo Oe ee a A se ae os 0-01 Water lost below 100° C. (H,0 —) ae me Oe 55 above 100° C. (H,O +) a: oe a O28) Cl ae ss ét ne a Paeeirace 100-02 This analysis presents no remarkable peculiarity, except the low percentage of alumina and the high percentage of alkalies, two properties which are also exhibited by the trachytes of Carnley Harbour. Judging from the petrological character of the rocks of the Bounty Islands, it seems probable that they are part of a great granite mass of which the granites at the Auckland Islands, the Snares, and pos- sibly of Stewart Island and south-west Otago are unsubmerged remnants. Con- clusions as to this being certainly the remains of a wide continental area based on the mere fact that they are plutonic rocks do not seem to be altogether well founded unless supported by collateral evidence; but this evidence is supplied from other islands in this area. The fact that the sea just east of them rapidly reaches a depth of 2,000 fathoms shows that they are near the edge of the continental shelf, and probably mark approximately the furthest eastward extension of the former continental area. * The iron present is all estimated as ferrous iron. 740 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. ANTIPODES ISLAND. [Norn.—No member of the expedition visited Antipodes Island on the present occasion, and the following facts about it have been collected from various authors; but additional matter has been supplied by Captain Bollons, whose intimate acquaintance with the locality makes his contribution specially valuable. The outline of the accompanying map has been taken from the chart recently made by H.M.S. “ Ringdove.”’| Antipodes Island is situated in long. 178° 43’ H. and lat. 49° 41’ S. Recent surveys by H.M.S. “ Ringdove ” show that its shape is not as shown originally on maps and charts. Its length from north-east to south-west is about five miles, and its greatest breadth is about three miles. These dimensions are those on the chart, but there is some doubt as to the scale on which it is drawn. ‘There are no openings in the coast-line, with the exception of Ringdove Bay, on the south-east corner, and a shallow bight on the north-east between North Cape and Reef Point, where there is a somewhat insecure anchorage. Landing is effected on a beach composed of rough blocks of rock, behind which the cliffs rise We a height of 80 it., but with easy ascent in several places (fig. 16). The north-west coast-line is bold and precipitous, and from 400 ft. to 500 ft. in height, rising to its greatest elevation in North Cape (500 ft.). This fine coastal feature, sometimes known as Perpendicular Head, has a close resemblance both in structure and appearance to Godley Head, at the entrance to Lyttelton Harbour. The south coast is from 80 ft. to 150 ft. high, and the east coast is also precipitous, with cliffs 300 ft. to 400 ft. in height. A number of reefs and small islets fringe the shore, the most important being the Bollons Islands, which he about a mile to the north-east of North Cape. There are two of them, and they are disposed in the shape of a horse-shoe, with the convexity facing north-east. They are ex- tremely rugged and precipitous, and rise to a height of nearly 500 ft. On one the erosive action of the sea has produced a remarkable arch. These islands are inaccessible except in the finest weather, the only landing within recent years being effected by Captain Bollons, after whom they are named. The surface of the main island forms a rough undulating plateau. It slopes up from the north-east and southern coastal cliffs towards the west. The highest point is Mount Galloway (1,320 it.), on the top of which a lake 13 or 14 acres in extent is reported to le. Other cones of lesser height also occur. The surface of the plateau is cut up by a number of deep narrow gullies occupied by small streams. The largest of these runs north-east, passes through a large swamp about a mile in length and breadth, and falls over the cliffs near the landing-place as a considerable body ot water. The hut for castaways 1s placed near the edge of the swamp, not far from the stream. Other parts of the island are bogey, notably those near the south coast, and its surface is covered with great depth of peat made of herbaceous plants, and not of Sphagnum. On this grow long tussock-grasses and very occasional patches of low scrub, but through it solid rock outcrops in many places. Kirk says that all the exposures he observed were basaltic in character. The islands appear to be entirely volcanic. No plutonic rock has been recorded from them, and the pro- bable occurrence of sedimentary rock mentioned by T. Kirk, and supported by Sir James Hector, seems very doubtful indeed, and is not based on close examination of the outcrops. Fragments of coal were found at one spot by Captain Bollons, Physiography, Geology.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 741 Bollons Is. SKETCH MAP _areatl pt of the ee ANTIPODES ISLANDS. ILaahes Atop lnt S). Ronen 17343) 3: North Cape (Perpendicular Hd. ) 500 ft. ae Cs Land gradually ci Ae sloping up ‘ towards S.W Wiese all ETE eel Windward Is. Mt. Galloway 1,320 ft. Leeward I. 'S) 400 ft. High undulating plateau (800 ft.) covered with tussock grass, occa- sional low scrub and bog. Ringdove Bay ee x ve) oe “€crpitous, 80 to 150 ft. Albatross Pt. Lava flows ex posed South Islet () 2 Statute Miles. 742 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. but they have all the appearance of a peat that has been partially coked by the heat of an overlying lava-stream. They exhibit rough prismatic jointing, but the vegetable structure has not been destroyed, and the streak is decidedly brown. Sir James Hector says that there were at least four centres of volcanic activity, but Kirk apparently was of the opinion that there was only one. He mentions, however, a volcanic crater near the south coast in a good state of preservation. It is apparent that fragmental matter has played a prominent part in the structure of the island. The clear sections given by the cliffs show that 1t has been Fic. 16.—NorrH-rast Point or AntTIpopES IsLAND, SHOWING JLiANDING-PLacE AND SMALL ISLET COMPOSED oF IncLINED Brps or ScorraA AND Fnows or Daya. constructed of regular flows of lava and beds of scoria (see fig. 16). Captain Hutton was of the opinion that it started as a submarine volcano. Specimens collected by Captain Bollons on the coast south of Reef Point are scorlaceous In character. One piece of rock from a lava-flow 25 ft. thick, and running north-east, shows a well- preserved ropy surface, which suggests that the volcanic activity was of very recent date, a conclusion also arrived at by Sir James Hector from a consideration of the impossibility that the fragmental matter of which the islands are largely built could resist for long the attacks of the heavy westerly seas. Physiography, Geology.| | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 743 The blocks on the northern beach are of hard compact basalt. The presence of dolerite and phonolite is also recorded, but it is very probable that the fine-grained basalt has been mistaken for the latter, as there is no record of the microscopic examination of these rocks having ever been made. PETROLOGY. It is impossible at present to give a satisiactory account of the petrology of the islands, but the character of the specimens collected by Captain Bollons is as follows :— No. 1. Basalt. (From the beach near the landing on the north-east of the main island.) Macroscopic.—The rock is dark in colour, slightly vesicular, with well-marked flow structure. It breaks under the hammer into irregular quadratic pieces. Crystals of augite are visible in a dark groundmass. Specific gravity, 2°85. Microscopic. — The only phenocrysts are of brownish augite and olivine in roundish crystals, usually clear, but occasionally seamed with limonite. No feldspar phenocrysts were seen in the sections made. The groundmass consists of colour- less glass, in which are microlites of feldspar (labradorite), magnetite in grains and square-shaped sections, and much granular and short lath-shaped augites similar to the phenocrysts. The rock is a basalt of a somewhat basic type. It contains occasional inclusions of a basalt, containing much feldspar in the groundmass ; this shows that probably the underlying rocks are of that type, the later flows being more basic. Nos. 2 and 3. Basalt Glasses. (From the eastern side of Reef Point.) Macroscopic.—Both these specimens are dark in colour, and very scoriaceous. No. 3 shows the ropy surface of a lava-flow. The rocks are glassy in appearance, with no phenocrysts showing. Microscopic.—The main part of both specimens is a very dark glass, with small feldspar microlites showing through it. There are numerous small phenocrysts of clear olivine and brownish augite. These rocks are evidently basalt glasses, the fragmentary and vitreous equivalents of No. 1, and show no difference in compo- sition from it, being probably the products of the same eruption. Dr. Marshall informs me that he has in his possession a specimen of basalt from these islands which shows large crystals of augite and olivine in a very fine ground- mass, with much feldspar. 744 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Physiography, Geology. BIBLIOGRAPHY. Armstrong, H. “Cruise of the Brig ‘ Amherst.’”” N.Z. Govt. Gazette, Prov. of Southland. 1868. Chapman, F. Rk. “The Outlying Islands South of New Zealand.” Trans. N.Z. Inst., vol. xxi, p. 491. 1891. Cockayne, L. “ Botanical Survey of Stewart Island.” 1909. Hartmann, Max. ‘“ Ueber Basalte der Aucklands Inseln.” L.J., 1878. Hector, Sir James. “ Notes on the Geology of the Outlying Islands of New Zea- lenge’ Wrans., INZ. lies, soll, m, S70, Hector, Sir James. “ Note on the Geology of the Outlying Islands of New Zea- land.” Trans. N.Z. Inst., vol. xxvii. 1896. Hutton, 7. W. “ Origin of the Flora and Fauna of New Zealand.” Ann. & Mac. Nat. Hist., series v, vol. xv. 1885. Kirk, T. “On the Botany of Antipodes Island.” Trans. N.Z. Inst., vol. xxi, p. 436. 1891. Marshall, P. “Some Rocks from Macquarie Islands.” Rep. A.A.A.S., p. 206. 1904. Moseley, H. N. “ Notes by a Naturalist on the * Challenger.’”’ London, 1879. Musgrave, T. “Cast away on the Auckland Isles.” London, 1866. Prior, G. T. “‘ Basalts and Dolerites from the Auckland Islands and Macquarie Islands.” Nat. Ant. Exped., Geology, p. 109. 1907. Prior, G. T. “ Mineralogical Magazine,” vol. xii, pp. 70-73. 1899. Suess, Professor Edouard. ‘‘ Face of the Earth ” (Knglish translation), vol. ui, p. 149. . = ARTICLE XXXI1.—THE SOILS AND SOIL-FORMERS OF THE SUBANTARCTIC [ISLANDS.* By B. C. Aston, F.1.C., Chief Chemist, Department of Agriculture. INTRODUCTION. GENERALLY speaking, the soils of New Zealand, excluding those of swamps, littoral situations, or limestone districts, are deficient in those humus or peaty substances so essential to intensive agriculture. The manner in which the humus content of British soils is supplemented is largely by frequent applications of stable manure. In a country where the stock are not stabled to any extent, and where the absence of cities precludes the possibility of obtaming large quantities of stable manure or other organic refuse, the problem of maintaining a proper proportion of humus in the soil is one which sooner or later must be faced. One of the most feasible ways of supplementing the store of organic matter in the soil, and one practised with marked success in America, is the ploughing-in of quick-growing green crops. The principle is a difficult one for the New Zealand farmer to act upon. The temptation to turn his stock into a paddock of clover or mustard, instead of ploughing it underground, is too hard to resist, except it be in very small areas, as an experiment, which, how- ever successiul, is too soon forgotten. This report deals with the composition of types of soil which, so far from suffering from the above defect, are at present unsuited to the uses of higher agriculture owing to a plethora of that very constituent which is lacking in the mainland. Should the facts outlined in these pages throw any light on the reason why these southern islands are so bountifully supphed with humus, it may suggest methods of treatment by which the stores of humus in the mainland soils may be profitably reinforced. PEAT AND HUMUS. The terms “ humus” or “ peat” are synonymously used by most authorities.+ Little is known of the precise chemical composition of the different forms of decom- * The investigation is incomplete. A number of specimens remain to be analysed, particularly those soils and rocks containing rare earths, which require more detailed investigation. In a subse- quent paper it is intended to treat of the mainland humus soils, and to summarise the results of the whole research. + It is, of course, often impossible to conjecture how large areas of vegetable matter have originated, out it is advisable to as far as possible restrict the term “ peat,” in the substantive, to plant-remains presumably formed in stagnant water, and containing a very small proportion of pure ash. 746 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Sozls and Soil-formers. posed plant-remains which are variously designated “humus,” “peat,” “black earth,” “mould,” or, in America, “tule,” or by the ineuphonious “ muck.” This final stage of decomposition of ligneous tissue results in the formation of brown matters which are known as ulmic compounds; a further stage results in black matter—humic substances: both of which classes are insoluble in water. Further decomposition results in a soluble brown acid (apocrenic acid) and a colourless acid (crenic acid). Although the chemistry of these bodies is so little understood, their value in agriculture is universally recognised, and the application of decaying vegetable substances to soils has from time immemorial been recognised as a fitting treatment. By the generation of carbonic anhydride and ammonia (the first rendering soluble and the latter supplying soluble plant-food); by absorbing moisture from the air, and thus keeping the soil damp; by increasing the porosity of the soil, so that in cases of exceptionally heavy rainfall the water may drain off, thus pre- venting the crop from rotting; by enabling more of the sun’s heat to be absorbed, owing to the black colour, and thus making the soil warmer—are some of the ways in which humus benefits the land. : Peat is generally understood to require stagnant water for its formation. The aquatic plants growing on the surface of a pool in summer die in autumn and fall to the bottom, and the pool, yearly becoming smaller, is eventually filled with plant- remains, which then allow of the growth of Sphagnum and other mosses, and thus, with a deficiency of oxygen, the moss finally becomes consolidated into a bed of peat. One-seventh of the entire surface of Ireland, a basin-shaped country, low in the interior and high at the coast, is covered with peat bogs. The growth of peat is often comparatively rapid (5 in. or 6 in. a year is a common thing in Ireland), and it is composed mainly of forms of plant-life such as Sphagnum mosses (a low order of plants, having the faculty of continuing in growth upwards as they die below, thus possibly enjoying an eternity of existence which may rank them among the oldest of hving plants) and Hypnuwm mosses, the former a calcifuge, the latter a calciphile. Other plants, such as Hriophorum, Carex, Phragmites, Scurpus, Hquisetum, and heath-plants, such as Erica, are undoubtedly peat-formers, though they perform their work more slowly. Hallmeén, quoted by Nystrom (6), classifies peats as follows: Moss peat— (a) Sphagnum peat, from Sphagnum moss; (6) Hypnum peat, from Hypnum moss, 8 to 30 per cent. of ash; (c) forest moss peat, from mosses, heath-plants, and the residue of forests, 5 to 8 per cent. of ash. Grass peat—(a) Sea-peat, from Phragmites, Scirpus, Equisetum, Menyanthes, Nymphaea, &c., 8 to 10 per cent. of ash ; (6) Carez, from Carex family, mosses, and other plants; (c) Hrophorum, from Hriophorum (cotton sedge), 0-75 to 4 per cent. of ash. Bjorlng and Gissing (8) pomt out that the formation of peat is dependent on a special combination of climatic and topographical conditions, and state that the principal factors are—(a) growth of aquatic and moisture-loving plants; (6) a soil or subsoil which will retain moisture at the surface; (c) sufficiently humid atmo- sphere to prevent too-rapid evaporation; and (d) a temperature high enough to allow a profuse growth of vegetation, yet low enough to check too rapid a decay of vegetable matter. » Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 747 Most of the analyses of peat have been made with regard to its value as fuel, and not as a soil. The nature of the organic constituents present, beyond a bare statement of the amount of ash, has seldom been given. As it is the soil aspect which this report principally deals with, it has been rather difficult to learn any- thing from the literature on peat which would assist to compare soils of the sub- antarctic islands with other vegetable deposits. COMPOSITION OF HUMUS AND PEAT-ASH. In the accompanying tabular statement the analyses of a few typical island soils have been recalculated to show the percentage of the inorganic constituents, calculated on the ash, in order that they may be compared with peat-ash analyses given by Groves and Thorpe (38), quoted below. The authors say, “ It is said that carbonates of the alkalis never occur in peat.” The amount of ash in European peats varies from | to 33 per cent. ANALYSIS or Asx (calculated from Total Analysis). Rye | IR aA ic mtecalas BE AA SIO eal Silica (by difference) 35-80) 46-20) 31-89) 53-74) 43-77) 24-11) 39-91] 51-25) 43-58) 36-19 Tron-oxides and alumina, anil. oxides of 28-87| 14:98] 27-41) 40-76} 29-16) 19-10} 23-31) 36-88) 47-64) 43-85 rare earths Calcic oxide Le md .. | 3:18] 4-41) 9-92) 0-89} 3-93)17-19} 9-51) 14:34) 1-16) 0-92 Magnesic oxide ae a .. | 3:60) 3-24) 14:35) 1-35} 4-96] 13-13) 13-83} 1-34] 0-89) 1-31 Potassic oxide : a, .. | 3:81) 3:89] 4:17) 0-85) 1:14] 2-86) 1-84} 1-12) 0-94) 14-00 Phosphoric anydride . . ae .. | 0:84) 1-16} 3-91) 0-93) 4:20) 6-20) 4-91) 2-52) 2:56) 1-36 Sodic chloride Ne sv .. | 12-38] 17-05} 5:22) 0-64) 5-21) 6-44! 3-07) 0-69] 0:43) 0-66 Sodic sulphate me! a .. | 2°89) 0-77] 1-83) 0-42) 1-90) 8-59; 2-09) 0:31) 0-24) 0-62 Sodic oxide ie iy .. | 8:63) 8-30} 1:30) 0-42) 5:73 2-38 1-53) 1-65} 2-56) 1-09 The following is an analysis of tiaee samples of peat-ash from Tuam, west Treland, with ash varying from 3°69 to 4°81 per cent.; average, 4:54 per cent. (No. 1 is a light peat from Ballinderry Bog; No. 2 is a somewhat older peat from Claretuam Bog; and No. 3 is a light peat from Weir’s Bog) :— No. 1. No. 2. No. 3. K5O an at a ee Oa olltes 3°56 5:19 Na,O ta an Aes oO 0°37 0:76 (CEH eae - ys 5 eB eD 34°79 36°60 MgO we a aoe als re 3°60 iiiikes Al, O53 Mis Be ae 2. 4°42 2°80 3°44 Fe,O, ee a a Se) OG, 6°34 6°50 Cl en sa a Pee Woy) 0:43 0:596 SO. eu sale ae Pe 12°86 19°70 16°46 lene a as He ae 4S) eS) 3°56 2°82 CO, we ty, ae ip ore 6°76 Telly Sol. S10, up ee en A 3°34 4°90 Imsgol. .. ne oe Moe Ree 12°92 6°81 748 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Sozls and Soil-formers. Fream (39) gives the following analysis of peat-moss to be used for litter :— Per Cent. Moisture .. ae Bt he bs s5 | LAG Organic matter an ue = me e. SONG EB OF hy? ae a: uM iy. a 0-02 K,0O ne ae xr et $i, qe 0-01 CaO ae ss ae ae an ae 0:15 MgO, &e. .. i ts . H L048 810, ta is at me a ae 0-72 N bee Ea as ae Pa a 0:81 An analysis of three layers of peat from a bog in Ireland is given below :— No. 1. No. 2. No. 3. ke Ole ne eh 2.) 0-49 0-21 0:25 Na,O .. Me vs eel Gr 0°65 0°50 ClO ae A .. 33:04 20-71) @4-04! We .. aa ig es aS, “1-20 1:28 NIGOR. ie 2 69 0:30 0:36 Bess. - 2 oe: 918-28 20°37 19-40 AOR 3: te a lA 1:07 0:24 SOe ns a ie .. 20:08 22-66 10°74 FIC g. ate ae ey leis 0-44 0:34 Sol. S10, ifs Be a 25 0:64 1:08 oinsolyes i ey, ee IS 11-18 26-79 COR: a) a A Soon 10:78 13°89 From these figures it will be seen that the Irish peats differ greatly from the island humus soils in the high amount of lime and sulphates and the low amount of alumina and iron in the former, contrasted with the low amount of lime and sulphates and the high amount of iron, alumina, and rare earths in the latter. Peat-ashes are usually very poor in potash, and usually so in phosphoric acid. In a number of Saxon peats Stockhardt (39) found 0-25 to 0:5 per cent. of alkaline salts, 0°5 to 2 per cent. of phosphate of lime, 3 to 10 per cent. of calcic carbonate, and 3 to 10 per cent. of calcic sulphate. An analysis of peat soils given in the Annual Report of the U.S.A. Department of Agriculture for 1881 gives,— A (Top Layer). B (2 ft. below). Ash a ye ee sa fal 6°5 IKGOs Wei es a ». Traces Traces P.O; ts on . pa OS 0-78 NGS a we a ie 0°87 1-40 Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 749 Mendeléeff (7) gives the following two analyses on air-dried soil :— Black Earth Peaty Soil near from Seimbirsk. St. Petersburg. Per Cent. Per Cent. Na,O ea x =: peo Oelal 0-04 K,O oo a aa v5. ORBLE) 0:05 MgO a ee Ms Sa OL, 0:07 CaO of oH wa ae Welles 0-11 EsOz a he ae See ORO 0-03 ING 23 i) aE oF .. O44 0:16 See ae a a ee Ons 0:06 Fe,0, ae P 3°41 0:46 The black earth has been exhausted by cultivation, a cause which accounts for the extremely low amount of phosphoric acid present. Ganong (15), remarking that the flora of American raised bogs is, on the whole, very similar to the hochmoor of Kurope, quotes Drude (41), who gives the analyses of hochmoor and flatmoor peats :— Chiemsee Chiemsee Hochmoor. Flatmoor. : Per Cent. Per Cent. Nea ae ae 4% a5 OPIS} 0-269 K,O ies ate ae .. 0:°024 0°052 CaO ae ae as oa Orla 3°267 MgO Be ae a .. 0°034 0-092 lee ae so OPANG 0°320 The humus or “ muck” soils of the Florida Peninsula, U.S.A., are described by Persons (22). They are formed in partially submerged swamps either from the remains of laurel or bay-tree, or from Cladiwm effusum, Nymphaea flava, Panicum Curtis, Pontederva cordata, Cyperus sp., Osmunda sp., Malva sp., Andropogon sp., Saggitarva sp.—all plants considerably higher in the vegetable kingdom than mosses. A number of complete analyses by Dr. H. W. Wiley are given, which show that these soils contain from 3 to 60 per cent. of inorganic matter. Those low in in- organic matter (3 to 6 per cent.) contain* from 0-01 to 0:1 per cent. potash, 0-08 to 0-3 per cent. soda, trace to 1-7 per cent. lime, trace to 0:28 per cent. magnesia, and trace to 0-115 per cent. phosphoric anhydride. The nitrogen varied from 2-74 per cent. at the surface foot to 3 per cent. at the second foot, 2-76 per cent. at the third foot, and 1-01 per cent. at the fourth foot. It was found that the nitrogen varied regularly with the volatile matter. Organic matter was ereater at depths of 2 ft. and 3 ft. than at the surface. These soils are, therefore, considerably lower than the Auckland Island soils in mineral plant-food, except lime in the case of saw-grass (Cladiwm effusum) humus. In other Florida humus soils the lime is present only as traces. Taylor (33), in five samples of Indiana peat, found the nitrogen range from 2-22 to 3:9 per cent., the P.O; from 0-96 to 1-9 per cent., and the K,O from 0-82 to 1:56 per cent. * Presumably soluble in hydrochloric acid. 750 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Soils and Soil-formers. Vogeler (34) found that the average analyses of cultivated Bavarian moor soils showed a variation in potash content from 0-0544 to 0-089 per cent., the average amount being 0-033. In spite of the fact that these soils are well supplied with this ingredient, it is apparently im a very unavailable form, as they are much benefited by liberal applications of potash fertilisers. All authorities agree on the advisability of applying lime to peaty soils to obtain the best results. Woolney has shown that calcium-humate will oxidize much more rapidly than uncombined humic acid placed under similar conditions. To the absence of calcitum-carbonate and mineral salts generally Hall (13) thinks may be ascribed the tendency of humus to accumulate and persist in very light sandy heaths. The simplest treatment* of peat and humus soils for agricultural purposes depends on the following principles: (a) Drainage, to take away the surplus water ; (b) removal of the peat or humus layer, and exposing the underlying clay or sand; (c) mixing the sandy or clayey base of the bog with a certain amount of humus. This procedure may be considerably modified, and, instead of exposing the base of the deposit, large applications of gravel (350 tons of gravel per acre) or marl (carbonate of lime with clay) were recommended in the report ot the Royal Commissioners, 1809-14). Thomas phosphate (basic slag), and on Irish bogs potash-salts, are the most efficacious fertilisers to apply. Farmyard manure, itself an organic manure, and green manure are also similarly success- ful on humus soils—King (36) and Whitson (37). Burning peat in heaps and spreading the ashes on the soil is also a well-known treatment to render such soils fertile. Other uses to which peat is put are the manufacture of—(a) alcohol; (6) fuel; (c) peat wool for clothing; (d) products of destructive distillation (oil, tar, paraffin, and pitch) ; (e) peat mull or powder for litter, absorbent for food mixtures, molasses ; (f) paper; (g) building-materials, roofing-material; (h) brown dye; (2) tanning- material; (7) manure; (&%) packing-materials ; and (/) ammonia. Different erades of peat have different uses, and thus the upper layers of a peat bog, and layers containing much cellulose and unaltered plant-remains, would be more suited for use as absorbents, manufacture of alcohol, paper, peat clothing feeding-stuffs, and packing-material, while the well-humified layers rich in carbon and nitrogen are better adapted for use as fuel, as fertiliser, or as a source of ammonia. Paturel (31) finds that peat yields most of its nitrogen as ammonia when heated by superheated steam, and may be used to promote intense nitrification in nitre- beds. * Humus acid is said to decompose silicates (Rodzyanko, J.C.S., 1892, vol. Ixii, p. 1373), and by a reaction between it and ammonia from ram-water, nitrogen from the air, and silica from the soil a series of silic-azohumic acids is formed (P. Thenard, Compt. Rendus, vol. Ixx, 1412, 1870) whose alkaline salts are easily soluble. Hyen quartz may be corroded by this class of organic solvents (EH. V. Hayes, Bull. Geol. Soc. Amer., 8, p. 213, 1896). See also Clarke (20). Norr.—Recent publications are leaflet No. 203, “ On the Utilisation of Peat Lands,” published gratuitously by the British Board of Agriculture ; and Bulletin No. 2, vol. m, of the Imperial Institute, “ On Utilisation of Peat.” Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 751 Muntz and Girard (35) have patented a method of obtaiming ammonia and nitrates from peat by means of superheated steam. Were this Dominion cut off from supplies of nitre, as France once was, humus soils so rich in nitrogen might prove the only immediately available source of nitre for making gunpowder. It is somewhat remarkable that, as far as can be ascertained, all southerly stations ravaged by salt-laden gales produce humus soils without the aid of mosses. Darwin (43) speaks of the Falkland Islands* as an undulating land with a desolate and wretched aspect, everywhere covered with peaty soil and wiry grass of one monotonous brown colour. The Falkland Islands country rock is clay, slate, and sandstone. Speaking of the Chonos Archipelago, Darwin (43) says, “ Hvery patch of level eround is covered with two species of plants, Astelia pumila and Donatia magellamiea, which by their joint decay compose a thick bed of elastic peat. In Tierra del Fuego, above the region of woodland, the former of these eminently sociable plants is the chief agent in the production of peat. Fresh leaves are always succeeding one to the other round the centre tap-root, the lower ones soon decay, and in tracing a root downwards in the peat ‘the leaves yet holding their places can be observed passing through every stage, until the whole becomes blended in one confused mass.” Astelia 1s assisted by Myrtus nummularia, Empetrum rubrum, Juncus grandi- florus—plants possessing a close general resemblance to the Hnglish species of the same genera. The climate of the southern part of America appears particularly favourable to the production of peat. In the Falkland Islands even the coarse grass which covers the whole surface of the land becomes converted into this sub- stance; scarcely any situation checks its growth. Some of the beds are as much as 12 ft. thick, and the lower part becomes so solid when dry that it will hardly burn. Although every plant lends its aid, yet in most places the Astelia is the most efficient. It is rather a singular circumstance, as being so different from what occurs in Kurope, that Darwin nowhere saw moss forming, by its decay, any portion of the peat in South America. This is so significant, and illustrates so exactly what is taking place in many parts of the subantarctic islands that I reproduce his views here. The “Challenger” Reports (vol. 1, p. 888) describe the Falkland Islands as a treeless expanse of moorland bog and bare and barren rock. Bolax glebaria forms “balsam bogs,” the plant simulating in appearance the elastic cushions of Azorella Selago at Kergulen Island, another peat-ridden country. The Falkland Islands are in the same latitude as London and Campbell Island, and are subject to thick fogs. Dangerous overflows of peat bogs, or “ bog-slides,” have occurred at the Falkland Islands (44). A curious effect of the peaty soil on the feet of animals is noted in the “Challenger” Reports: the hoofs of horses * The common grass there is a species of Dactylis, a sister species to the cocksfoot (Dactylis glo- merata). This forms tussocks in the same way as the Danthonia and Poas do in the Auckland Islands. The fact that the Falklands are noted for their cattle-raising industry suggests that the grasses of those islands might with advantage be tried on the soils of the Auckland and Campbell Islands, where the indigenous grasses are rapidly being eaten out by stock. Mutual exchange of the grass-seed of these islands might benefit both. 752 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Soils and Soil-formers. quite commonly grow out and become turned up like a ram’s horn, owing to the softness of the soil. The same phenomenon was seen at the Campbell Islands with the feet of sheep. It may be noticed that the same peaty covering clothes the Shetland and Orkney Islands (crystalline, metamorphic, and Old Red Sandstone formation), in the Northern Hemisphere, as 1t does in the subantarctic islands, THE AUCKLAND ISLANDS.* The soils of the islands treated of in this report are for the greater part not formed in the manner in which peat in its special sense is formed, inasmuch as— (a.) They are derived chiefly from the decay of the higher plants (Filices, Juncaceae, Grammeae, Araliaceae, Umbelliferae, Compositae, Liliaceae, Rubiaceae, Myrsineae, and Epacrideae) on the uplands and unforested areas, and from Filices, Myrtaceae, Epacridaceae, Araliaceae, and Compositae in the forests. (b.) They are not formed in or under stagnant water. Many of the soils must have been formed on a steep hillside, with every advantage of drainage which such a position could afford. (See fig. 7, p. 713, and figs. 12 and 13, p. 730.) Any water retained with soil must be solely through the absorbent quality of the organic matter composing it. In no case, in digging samples on Auckland, Antipodes, or Camp- bell Islands, did water flow into the hole.t This bears out Cockayne’s (10) state- ment with regard to Campbell Island, and is an experience contrary to that of Buchanan (11), whose assertion that on Campbell Island wherever a plant is dug out with a knife the hole immediately fills with water is a considerable exaggeration. Buchanan’s statement as to the quality of the pasture of Campbell Island 1s likewise inaccurate. (c.) They support a vigorous growth of higher plants, generally understood to require a soil rich in mineral nutrients. Cockayne (10) calls attention to the fact that the Auckland Island forests are ““ wanting some of its usual constituents, but otherwise, except for its peculiar physiognomy . . . is a true New Zealand rata forest.”t Treating of the soil on which an herbaceous section of the Auck- land Island flora grows, the same observer (p. 537) remarks, “ All that can be said is that, notwithstanding the soil is altogether peat, 1t must be much more favourable for plant-life than the peat of an ordinary bog, or even a dry heath.” Thus, the fact that these soils differ so largely in their vegetation from those which are usually accepted as peat soils as to be manifest to a non-chemist observer confirms me in classifying them as humus soils. The large areas of peat lands in Europe, known as “ hochmoor,” contain re- latively little nitrogen near the surface, but much more a few feet below. Detmer (4) * These include Auckland Island, Adams Island, Enderby Island, Disappointment Island, and Masked Island. ft Of course, it is quite possible that on all the islands, in limited areas, instances where this is untrue would occur; but in collecting samples representing types of soils my experience is as stated, and therefore it is Just to assume that for a humus soil with a daily or almost hourly rainfall the islands are naturally well drained. t See fig. 10, p. 214, and fig. 15, p. 223. > i Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. foo calculated on the ash-free peat from Jessbeck, in Schleswig-Holstein, the layer at the surface contained 0-08 per cent.; at 7 ft., 2-1 per cent.; and at 14 ft., 4:05 per cent. nitrogen. The original peat contained 2-72 per cent., 7°42 per cent., and 9°16 per cent. of mineral matter respectively at the different depths. A more detailed exami- nation of the non-nitrogenous matter of peat was made by von Feilitzen (18), and his results confirm those of Detmer as regards the gradual increase in the proportion of carbon concurrently with the increase in depth. Notwithstanding that the peat rapidly loses two prominent non-nitrogenous constituents—cellulose and furfuroids— the peats investigated by von Feilitzen were of the hochmoor variety, derived from vegetation which thrives when the surrounding water is deficient in lime and other plant-food. Absence of lime has been shown by Hilgard to be favourable to the accumulation of carbon. Miller (5), in discussing Detmer’s results, states that we possess no evidence that the vegetation from which the lower layers of peat were derived was identical in composition with that of later growth, but the high percentage of nitrogen at a depth of 14 ft. must be mainly due to losses of oxygen in combination with carbon, as well as losses of hydrogen, probably as marsh-gas. Hall (13) remarks that the nitrogenous portions of vegetable matter resist decay longer than the carbohydrates, and hence tend to accumulate, so that an old sample of deep-seated peat is richer in nitrogen than a more recent sample taken from nearer the surface.* The total nitrogen in these island soils is exceptionally high, even for a humus soil, at the respective depths given. Analyses of European peats quoted by Nystrom (6) give the lowest nitrogen content as from 0:95 per cent. to 1°7 per cent., and one instance is given of a sample as high as 2°7 per cent. Eliminating the inorganic portion of those series of soils which were taken in successive layers at Camp Cove, Snares, and Campbell Island, it will be seen that the top layer is the richest in nitrogen, but some of the lower layers are richer than those nearer the top. The record is, of course, fragmentary. No face of humus was sampled lower than 8 ft., and if it were possible to sample a section to 14 ft. much more conclusive evidence might be obtained. A remarkable difference between the soils of Auckland Island (which is also true of the Campbell Islands, so far as they have been examined) and those of Treland is that in the case of the last-named country the high bogs, according to Ryan (9), contain little ash, but the low or flat bogs contain more ash; whereas in the Auckland Islands it is the upland soils which contain the most ash. This is probably due to bog-slides taking place more frequently on the uplands of these islands, and carrying down much mineral débris, which finally becomes incorporated with the soil. If there is a check in the declivity, or a dense forest, it 1s easy to see how the humus soil in the vicinity may be considerably enriched by the clay and débris being spread over it. On the other hand, if the gradient is unchecked, and the slide has sufficient impetus to carry it through the forest, it may sweep all before it to the sea, in’ which case it is possible that on the old scar we may get a Plewrophyllum meadow ‘established at sea-level, as at Fair- child’s Garden. Evidence of both kinds of bog-slides was seen at Auckland and * The italics are mine. It will be observed that he does not say. “ at the surface.” 49—§, 754 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Soils and Soil-formers. Adams Islands. On the ridge above Sealers’ Camp, Adams Island, a considerable extent of country had slipped away, leaving a perfectly bare scar in a Danthonia meadow. At Circular Head, in Carnley Harbour, the effects of huge slips were observed coming down through the forest to sea-level. Several sheep were observed browsing on this bare place.* The unusual matting-together of the roots of plants growing at littoral habitats (on rock, J 805-7) and elsewhere (on clay, or clay and humus, J 819-20, 816-8, 821-2) deserves notice. It would seem necessary for plants to send their roots through several feet of top soil in order to obtain some of the mineral nutrients lacking in the overlying humus. Usually in soils there is a gradual diminution of the amount of roots towards the centre of the earth—that is, the proportion of roots varies inversely with the depth.t The antiseptic action of humus deserves brief mention in this chemical report. The wrecked “ Dundonald” survivors clambered ashore over sharp rocks, in the dark, and received many minor wounds, but these rapidly healed owing to the hands and feet being constantly in contact with the humus soil. The body of the first mate of the ill-fated ship was buried on Disappointment Island for eight months and a half, at the end of which time it was exhumed and reinterred at the Port Ross Cemetery. The remains were in an excellent state of preservation. Members of the expedition suffered from the usual cuts and scratches incidental to “ roughing it,” but these all healed up remarkably quickly. * An excellent example of bog-slides is shown on the slopes of the rounded hill on the extreme right of the panoramic view of Auckland Island (fig. 2, p. 708). The hill is deeply scarred on every side visible. t For instance, Thiel found that at the depth of 10 in. the number of roots was 100; at 14 in. the number was 53; at 18in., 33; at 22im., 25; at 26in., 20; at 30im., 11; at 341in., 6; and at 37 im., 2. In a clover-field examined by John six times as many fine roots were found in the uppermost 7 in. of soil as were found in the soil below that depth. Hellreigel gives two instances of exceptions to the general rule (Storer, i, p. 56) that the number of roots diminishes regularly with the depth. A layer of soil contained a greater proportion of roots than one above it :— - (a.) In winter wheat on low land (Storer, i, p. 61): The soil was rich in humus at the surface, and consisted of 1 ft. of sandy loam, below which there was from | ft. to 2 ft. of clayey river-loam overlying clear sand. At a depth of 81in., in the soil proper, there were found 558 root-fibres; at 15 in., in the river-loam, 218 root-fibres ; at 26 in., in sand, 83 root-fibres; at 33 in., in sand, 106 root- fibres ; at 41 in., water-table, no root-fibres. (b.) Red-clover on low-lying land (Storer, iti, p. 62) : The soil consisted of sandy loam 3 ft. deep, of which the uppermost 18 in. contained humus. Below the sandy loam there was from 3 ft. to 3-5 ft. of clayey river-loam, and below that clean sand. At a depth of 9 in., in the loam, there were found 729 root-fibres ; at 20 in., in the sandy loam, 87 root-fibres ; at 24 in., in the sandy loam, 56 root-fibres ; at 32 1n., in river-loam, 34 root-fibres ; at 40in., in river-loam, 32 root-fibres; at 43 1n., in the sand, 74 root-fibres ; at 51 in., water-table, no root-fibres. In both of these instances a considerable number of roots were developed in the lower moist sand after the plant had been at the trouble of pushing through the clayey loam of the subsoil. Korticheff (quoted by Hilgard), in the black earth of Russia, “ tchernozem,” found a gradual decrease of roots with increasing depth, in close correspondence with the humus content of the dif- ferent layers. It is pointed out that the surface soil may contain less humus than the subsoil, and, we may infer, less roots, where the surface soil is porous and favouring slow combustion, the conditions of normal humification being fulfilled at a lower depth. Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 755 Instances of this antiseptic action are well known in Hurope. Two bodies were found in a Thuringian bog; from their clothing it was deduced that they were German merchants of the time of Julius Caesar or Augustus. A mummy-like body in the Copenhagen Museum, found fastened to a pole in a bog in Jutland, appears to have been that of Queen Gunhilde of Norway, who in A.D. 965 was enticed to Denmark by King Harold on a promise of marriage, and who was assassinated and sunk in a bog in the above manner.* The ground outside the castaways’ camp at Disappointment Island (see fig. 13, p. XXxlil) was strewn with the half-devoured bodies of fat young mollymawks— the six months’ accumulation of camp-refuse. These had not putrefied, but were covered by a growth of red mould or fungus. According to Kostyschoff (12), moulds alone produce the dark-coloured matters which give soils rich in vegetable matter their colour. The soil of this island is jet-black in colour. Although the breaking- down of vegetable matter in ordinary soils into humus may usually be ascribed to the action of bacteria, it is easy to imagine that in those soils where we have— (a) an acid reaction, deleterious to bacteria; (6) a high percentage of salt (sodium- chloride), which is, according to Hall (13), particularly harmful to bacterial action in soils; (c) fairly low temperature ; (d) soil and atmosphere saturated with water ; (e) absence of carbonate of lime—then the presence of bacteria may be reduced to a minimum, or altogether wanting, and their function usurped by moulds or fungi. A bacteriological analysis of these soils would, I am sure, prove highly interesting, even if the results were entirely negative. As soon as the medium becomes decidedly acid bacteria do not thrive. Clarke (20) states that the formation of humus acids appears to take place by a fermenta- tive process, which eliminates some carbon, hydrogen, and oxygen in the form of carbon-dioxide, methane, and water. Micro- -organisms play some part in pro- ducing the changes observed. On this point, however, there is some doubt, Frih and Schroter (21), for example, regarding the microbian influence as very small. It is somewhat singular that no species of the family Conzferae occurs on these islands, although coniferous trees form a great part of the forest of Stewart Island. Endemic leguminous plants are wanting alike at both these habitats. Clovers were observed only on Enderby Island, where there is an abundance of calcareous sand and phosphate of lime in the soil. A study of the soil’s action on bacteria may pos- sibly enable one to say if members of those families which depend to some extent on nodular colonies of bacteria on the roots could exist in such a soil. * The action of peat and humus on bone appears to vary considerably. In Ireland numerous skeletons of the great Irish elk have been obtained from the bogs, although the animal itself was extinct before the beginning of the authentic history of the country.—Geikie (28). In New Zealand, Hutton (29) and Booth (30) explored the Hamilton (Otago) peat-moss, “a small dry basin 50 ft. in diameter, and from 5 ft. to 6 ft. deep in the lowest part, excavated out of a bed of clay. This small basin was filled with peat and bones imextricably mixed, and forming a compact layer 2 ft. to 4ft. thick, and before being disturbed its surface was rather higher than the surround- ing country, which was quite flat for a distance of 200 yards. Out of this small hole was taken about 7 tons weight of moa-bones, more than half of them quite rotten, the remains of at least four hundred birds. Mr. A. Hamilton also informs me that bones taken from a swamp at Swampy Mill, Dunedin, were quite rotten. 756 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Sozls and Soil-formers. METHODS OF ANALYSIS. ANALYSIS OF PEatT SOILS. Loss on Air Drying.—The soil was exposed to the air in thin layers on shallow trays, and as the soil dried 1t was broken by hand. When fine enough it was put through a 2mm. sieve, and then kept on the tray until it ceased to lose weight— an operation extending over several days. Moisture.—The soil was heated to 100°C. in a platinum dish until it was of constant weight. Hydrocarbons.—A weighed quantity of soil was placed in a platinum crucible, well covered, and heated to low redness until the production of volatile inflammable vapours ceased. Fixed Carbon.—The residue from estimation of hydrocarbons was heated to dull redness, with the cover of the crucible removed, until the carbon was entirely consumed. Water-soluble Salts—To each 10 grams of soil 100 c.c. of water were added, placed in mechanical shaker and shaken for five hours, and filtered. In an aliquot part of the filtrate the chlorine was estimated as silver-chloride. Total Soluble Salts.—An aliquot part was exaporated to dryness in a platinum dish heated at 100° C. till constant weight, and weighed. Soluble Organic Matter.—The dried residue from total soluble salts was ignited at low temperature, the loss being calculated as organic matter. The residue from the ignited salts was analysed for sulphuric acid, silica, iron and alumina, lime, magnesia, phosphoric acid, and alkalis. Citric-acid-soluble Salts.—These were extracted by Dyer’s method, Hall’s modification (J. Chem. Soc. Trans., vol. Ixxxix, pt. 2, 1906).* Hydrochloric-acid-soluble Salts.—Twenty grams of the dried soil was ignited, taken up with hydrochloric acid, evaporated to dryness, repeated, filtered, and washed with dilute HCl, made up to 250 c.c., and aliquot parts taken for estimation of iron and alumina, lime, magnesia, phosphoric acid, and potash. Total by Hydrofluoric Acid.—The ash, after estimation of hydrocarbons and fixed carbons, was digested with hydrofluoric acid and sulphuric acid, and residue redissolved in hydrochloric acid. Silica, iron and alumina, lime, magnesia, phos- phoric acid, and alkalis estimated in aliquot parts of the solution. The “iron and alumina” also contains manganese-oxide, titanium-oxide, zirconia, and possibly oxides of other rare earths. Total Nitrogen.—¥ive grams of the dried soil were digested with sulphuric acid and zine dust, with the addition of mercury, until colourless. The ammonia in the ammonium-sulphate formed was estimated by distillation into normal sulphuric acid. * Wood and Berry (42) remark on the probable necessity for raising the limit below which the economical use of phosphatic fertilisers 1s indicated, for citric-acid- soluble phosphoric acid in peaty soils. This is probably due to the citric acid dissolving organic matter, including organic phosphorus compounds, if the soil is rich in humus. Peat from the Isle of Ely yielding as much as 0-05 per cent. P.O; to 1 per cent. citric acid responded very profitably to phosphatic manuring. Citric acid of 1 per cent. strength appears not to dissolve as much organic matter from these island soils as water does, Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. (D7 NOTES ON THE ABOVE SAMPLES AND ANALYSES. Micacrous Humus Sorts (15 per cent. inorganic). J 791-3.—Collected on the 15th November, 1907, on the main Snares Island, at an altitude of 150 ft. on the west side, on a slope facing north-east. J 791-3 are a series taken from the face of a bed of humus which is exposed. The samples were obtained by cutting several inches into the face and rejecting the portion exposed to the weather. J 791, representing the top 18 in. of soil, contains much undecomposed remains of the two grasses, Poa litorosa and P. foliosa (trace), which constitute the phanerogamic flora of this soil. This soil was honeycombed by bird holes and runs. J 792, collected below J 791, and representing the next 2 ft. 6in., containing peaty soil, with some undecomposed grass-leaves, together with an amount of mica present in fairly large particles. This sample was collected in a canvas bag; the amount lost on air drying does not, therefore, represent the field samples in this particular. J 793, collected below J 792, represents the next 4 ft. of peat, containing mica. This was also collected in a canvas bag. The most interesting fact brought out by these analyses is the gradual increase in the salts from sea-water from the surface downwards. Thus, below 4 ft. under the surface there is twice as much soluble salts as on the surface layer. In the most soluble of the sea-salts present—magnesic chloride—the sample from the lowest 4 ft. contains five times the quantity found in the sample from the layer immediately above it. K 2660 was collected from a Poa foliosa formation, on the 6th January, 1909, on the Snares main island, a few yards from where J 792 was taken over twelve months previously. The flora was entirely Poa foliosa, and the sample was taken to 1 ft. in depth. This formation is much more compact than that of Poa litorosa, and easy to walk on. K 3049 is the subsoil, taken to the depth of 1 ft. below K 2660. Hitt Humus Soits (25-30 per cent. organic). J 795-6.—Hill soils collected on Adams Island on the 18th and 19th November, 1907. J 795 was got at an altitude of 700 ft., in a Plewrophyllum Hookeri meadow. The flora was P. Hookert, Danthonia bromoides, Coprosma cuneata, Gentiana, Celmisia vermcosa, Veronica Bentham, Bulbinella Rossi, Dracophyllum longifolium (stunted), Carpha alpina, Hierochloe Brunonis, Lycopodium fastigiatum, Oreobolus pectinatus, Hymenophyllum multifidum. This was a black well-decomposed soil, having a northerly aspect, and was taken to a depth of 9 in. J 796 was collected at 1,400 ft. elevation, taken on a different spur, and to the west of situations of previous samples, facing north-west. The flora was a nearly pure Danthonia bromoides (snowgrass) formation, but containing also Plewrophyllum Hookeri, P. speciosum, Ligusticum latifolium, and Hymenophyllum multifidum. A good many bare patches were seen on these spurs, the soil and vegetation having 758 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Soils and Soil-formers. apparently slipped down the gullies, which may be considered evidence of the occur- rence of bog-slides on the islands. These two soils, taken from localities about a mile apart as the crow flies, are very similar in composition. J 810 is a soil taken to a depth of 9 in. in two places on the 24th November, 1907, which is closely related in composition to J 795-6, differime chiefly in the amount of precipitate classed as iron and aluminium oxides, but which in this case are titanium, zirconium, and other oxides of rare earths, precipitated by ammonia. It was collected from the hill above Camp Cove, at 1,000 it. elevation, and from a locality facing the previous samples, being on the opposite side of Carnley Strait. The flora was chiefly Danthonia bromoides, mixed with Gnaphalium, Acaena, Veronica Bentham, Hrerochloe Brunonis, Coprosma parviflora, Pleurophyllum Hookeri, Coprosma cuneaia, Celnusia vernicosa, Hymenophyllum multifidum. See fig. 11, p. 217, for a view of the locality near where this was collected. Lowianp Humus Sorts (less than 10 per cent. organic). J 794, 797, 809, 816, 819, 825.—These are samples from similar localities as regards altitude, flora, and distance from sea. The figures obtained are all very similar, with the exception of those for iron and aluminium oxides, a point which will be discussed elsewhere. J 797 was collected on the 22nd November, 1907, to a depth of 3ft., on the North Arm track to the west coast, about half a mile from the inland side of the sea. The flora was chiefly Panax simplex, Cassima Vauvilliersw, Danthoma bromordes, and moss. J 809 was collected on the 24th November, 1907, to a depth of 1 ft., from two places on Skua Gull Flat, situated on an open space of several acres, 150 ft. above the boat-depot at Camp Cove. The flora was Danthonia bromoides, Cyathodes empe- trifolia, Coprosma repens, Dracophyllum longifoloum (stunted), Oreobolus pectinatus, Coprosma parviflora, C. foetidissema, Cassinia Vauvilliersu, Suttonia divaricata, Hymenophyllum multifidum, moss, and lichens. J 816 was the first 18 in. of a shaft which was sunk on the 26th November, 1907, to ascertain the depth and character of the subsoil on the highest part of this flat (see fig. 3, p. xvi). J 819 was the first 18 in. of another shaft, sunk on the lowest portion of the flat. These four soils are characterized by the excessive amount of organic matter (96-97 per cent.), and the consequent minimum amount of inorganic or mineral matter. They are, moreover, all collected from what will probably be known phy- tologically as the “ open-lane formation ” of the Auckland Islands. The forest on these islands is in many places cut into lanes, probably by excessive wind-action. This is well seen in photos fig. 11, p. 217, and figs. 3 and 4, p. xvi. The lanes, or gaps, Support a sparse growth of scrub, and the ground is covered with herbaceous plants, filmy ferns, lichens, and mosses. J 825 was sampled on the 29th November, 1907, to a depth of 1 ft., near sea- level, on the Port Ross track, through dense rata (Metrosideros lucida) torest, the predominant plants being rata, Dracophyllum longifoliwm, Cassinia Vauvilliersi, Soils and Soil-formers.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 759 Suttonia divaricata, ferns, and mosses. The sample was collected in a canvas bag, hence the water lost on air drying cannot be considered accurate. The sample bears a close similarity in composition to J 797. J 794 was sampled on the 17th November, 1907, to a depth of 9in., on a hill about 500 ft. above Camp Cove boat-depot, Auckland Islands, above the rata-bush line. The flora was Danthonia bromoides, Dracophyllum longifolium, moss, Hymeno- phyllum multifidum, Myrsine dwaricata, Ranunculus pinguis—mostly the first named. Aspect southerly. K 2641 was taken on the Port Ross Track, on bush-line, at 300 ft. elevation, on the 7th January, 1909. Sampled to a depth of 1 ft. The flora was Danthonia bromoides, Dracophyllum longifolium, Panax simplex, Metrosideros lucida, Oreobolus pectinatus, and lichens. This should be classed at a “ lowland humus ”’ soil. K 3044 is the subsoil taken below K 2641. Ricw Humus Soixs (37-17 per cent. imorganic). (Pleurophyllum Meadow.) J 811-4, 824.—J 811 is the first 18in. layer of soil, collected near sea-level, at Fairchild’s Garden, west end of Adams Island, on the 25th November, 1907, in a Pleurophyllwm meadow,* the flora being entirely herbaceous, and consisting of Lagusticum latifolium, Stilbocarpa polaris, Pleurophyllum criniferum, P. speciosum, Carex trifida, Aspidium aculeatum var. vestitum, and a very few Danthonia bromoides lants. : J 812 is the subsoil or 18 in. layer of soil below J 811. J 813 is the 18 in. top layer of soil taken about 100 ft. altitude above J 811, on the same hill, and on same date. The flora was as follows: Plewrophyllum crini- ferum, P. speciosum, Ligusticum latifolium, Poa foliosa, Carex appressa, C. trifida, and scattered tussocks of Danthonia bromoides. This soil contained a few rounded stones. J 814 is the 18 in. layer of subsoil below J 813. The aspect of these soils is north-east—that is, facing in opposite direction to the prevailing gales, and protected from them by a high ridge, on the windward side of which, and about half a mile away, J 815 was taken. J 824 was collected to a depth of 9in., on the 28th November, 1907, at Disap- pointment Island, a few feet above sea-level, on a hillside facing north-east. Being collected in a canvas bag, it may have lost a little water. The flora was very similar to that of Fairchild’s Garden—entirely herbaceous plants—as follows: Ligusticum latifolium, Stilbocarpa polaris, Pleurophyllum crimferum, P. speciosum, and Poa foliosa. Fig. 18, p. 574, shows this type of soil, and fig. 13, p. xxxili, shows the upland soil of the same island. K 2645.—Collected at Adams Island to a depth of 2 ft. 6 in., at a situation a few feet above sea-level, and above the boat-depot, having a north-west aspect, on the 10th January, 1909. This soil is adjacent to what is known as “ Fairchild’s Garden,” but supports a shrubby growth of Veronica elliptica and Dracophyllum longrfolium, * Fig. 15, p. 223, shows a Plewrophyllum meadow growing on a raised gravel beach near sea-level, Adams Island, opposite Camp Cove; but the soil of this cannot be compared to a Fairchild’s Garden or a Disappointment Island soil. 760 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Soils and Soil-formers. with Acaena beneath. This soil differs from that of Fairchild’s Garden in the much larger amount of mineral matter, especially silica, which is present as an angular quartz sand. Owing to the greater depth to which this was taken, it must be com- pared with J 811, together with its subsoil J 812. It would be interesting to as- certain if the content of angular silica is constantly correlated with the occurrence of shrubby flora in this locality. Lirrorat Humus Sorzs (less than 10 per cent. inorganic, and more than 0°25 per cent. salt). J 803, 805, 806, 815, 834.—J 803 was collected at the littoral, Camp Cove depot, Auckland Islands, on the 23rd November, 1907, to a depth of 18in. The soil was resting directly on the rock (basalt), which was only weathered to a slight depth, a mere skin of about 4in. or so (J 801), where the rootlets were resting upon it. Below this was the unweathered rock (J 802). J 805 was a brown soil with many roots, taken on the 24th November at the littoral of Masked Island, a small island off Camp Cove depot, to a depth of 18 in. Aspect east. Flora: Poa foliosa and Stiubocarpa polaris. (See fig. 7, p. 202.) J 806 was the lower 18 in., a black soil with few roots, taken below J 805. This soil rested directly on basalt, which was only slightly weathered at the point of contact—about Sin. thick (J 807).* Below this was the unweathered rock (J 808).* An interesting fact observed in sampling these soils was that the top soil con- tained many roots and the subsoil few roots, but on the weathered rock below this there rested a mat of small roots. It will be seen that the greatest amount of plant- food soluble in hydrochloric acid is found in the top layer, while the greatest amount of salt (sodium-chloride) is found in the bottom layer. J 815 was taken to a depth of 1 ft. on the west end of Adams Island, at the entrance to Carnley Strait, from a pure formation of Poa litorosa, about a chain from the sea. This grass, usually found in tussocks, was here—where it met the full fury of the westerly salt-laden gales—densely matted together, forming a dense and thick sward. The amount of salt present is high enough to place these soils with the Snares soils. On barer places, near the situation of J 815, are plants of a decided halophytic nature, such as Samolus repens, Plantago Brown, Tillaea moschata, Cotula plumosa, while the nearest bush is mostly Veronica elliptica (a littoral species) and Draco- phyllum longifolium ; and the rata (Metrosideros) is not so plentiful, and, where present, is stunted. The tolerance which Poa litorosat and its near ally Poa Astonia exhibit towards salt, and the avidity with which the Enderby Island cattle have almost exterminated the former and the Bluff stock the latter, suggest their economical use as a pasture in salt, wind-swept areas. Hxperiments are being made to test the value of these grasses, live specimens of which I collected and sent to the experimental farms of the een ee of rs * J 807 and 808 were not andiyeeass but J 799 and 800 are very similar in every respect to them. + A fine specimen of this grass is shown in fig. 1, p. 187, with Dr. Cockayne standing. Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 761 EXAMINATION OF RELATION BETWEEN SURFACE SOIL AND SUBSOILS AND ROCKS. . J 816-8, 821, 822.—In order to study the soil-conditions at different depths below the surface, on the 26th November, 1907 (the last day of the camp at Camp Cove), a trench 12 ft. long was cut, under my supervision, by the Maori boat’s crew, on Skua Gull Flat, by which name is designated a shallow basin-shaped cavity on the shoulder of the hill 150 ft. above the boat-depot at Camp Cove. The situation was chosen owing to its proximity to camp, its fairly level character, the absence of timber and roots, and, lastly, its elevation. An account of the top soils and their flora is given under the heading ‘‘ Lowland Humus Soils.” Figs. 3 and 4, p. xvi, show well the scene of operations. The top soil (J 816), the first 18in., was of a brownish colour, and composed for the greater part of decomposed vegetable matter. The subsoil (J 817), the next 14in., had a brownish-yellow colour, and contained many thick undecomposed roots, while towards the junction with the clay, at the lowest part of this 14 in. layer, was a thick mat of roots. J 818 was the 12in. layer below J 817, and consisted of clay and partially decomposed fragments of basalt. J 821 and J 822 were the clay layers, of 2 ft. each, resting below J 818. Thus the trench was sampled to a depth of 7 ft. 10in. Roots of Danthonia bromoides, the characteristic grass tussock (allied to the New Zealand snowegrass) of the islands, were traced down below the surface in the clay to a depth of 4 ft. 9 in. Portions of the weathered basalt having a white or pinkish earthy appearance (J 833A), and unweathered basalt (J 8338), were taken from the trench at 3 ft. 7 in. for analysis. J 819-20.—About a chain lower than the above trench another hole was sunk to a depth of 3ft. Here the clay and rocks were found at that level, and the massing of the roots occurred just as in the previous trench, but only 1 ft. 8 in. below the surface, being thickest at 2 {t. below the surface. A plastic clay was met with at a depth of 3it. The two samples, soil and subsoil, were 18 in. in thickness. The bottom layer is much richer in total mineral constituents soluble in hydrochloric acid. A remarkable exception to any other island soil is the high content of sul- phates of the subsoil. Humus Coarse Sanpy Sort (more than 50 per cent. siliceous inorganic matter). J 798 is a coarse sandy soil from “salt meadow” flats at the head of North Arm, Carnley Harbour. Collected on the 22nd November, 1907, to a depth of a few inches only. The vegetation was Montia fontana, Cardamine hirsuta, Ranun- culus acaulis, moss, sedge, Poa annua, Poa pratensis. In its content of inorganic non-calcareous matter this soil, which represents only a limited area, differs from any other met with. K 2644 was collected on the windward side of Enderby Island, above basaltic cliffs, on the 8th January, 1909, to a depth of lft. It contaims much clay and mineral matter, probably weathered in situ. Flora: English grasses, Tllaea moschata, Ranunculus acaulis, Cerastiwm glomeratum, Rumex neglectus, Scirpus. K 3047 was the subsoil taken 1 {t. below K 2644. 762 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Soils and Soil-formers. Humus CALCAREOUS SOIL. J 823 was sampled on the 27th November, 1907, in three places, to a depth of 9 in.—in front of depot, on south side (sheltered side) of Enderby Island, on slope of basaltic cliffs (see fig. 13, p. 730); on a sward composed of introduced grasses, clovers, and Bellis perenns (daisy). Aspect easterly. The sample was collected in a canvas bag; the water-determination may therefore be low. This is a type of soil which is sharply separated from all the others obtained at these islands, as it contains 26-4 per cent. calcic oxide, equal to 47 per cent. calcium-carbonate. It is there- fore a highly calcareous soil, due to admixture of the beach-sand with humus. Analysis of the beach-sand, which is given under J 826, shows it to consist of 91 per cent. calcium-carbonate, and to be, therefore, an almost pure calcareous sand. The indigenous flora of Enderby Island in the vicinity of J 823 is for the most part composed of Dracophyllum longifolsum, Veronica elliptica, some Veronica buxifolra, Metrosideros lucida, Suttona dwaricata, Scirpus cernuus (?). K 2642.—Collected on the lee side of Enderby Island, 20 yards from sandy beach, on the 8th January, 1909, to a depth of 9im. Flora: Bellis perennis, Rumex neglectus, Lagenophera petiolata, Agrostis sp., Scorpus sp., Ranunculus acaulis, Saguna procumbens, Deschampsia, Cerastuum glomeratum, Holcus sp. K 3045 was the subsoil of K 2642. K 2643 was taken a chain higher up the hill, supporting the same flora. K 3046 was the subsoil of K 2643. The high content of total and available phosphoric acid in these calcareous sandy soils is an interesting feature. CAMPBELL ISLAND SOILS. At my visit to Campbell Island in January, 1909, time only permitted of the collection of two top soils—K 2650, a soil similar to the “lowland humus soils” of Auckland Islands; and K 2647, intermediate in composition between Auckland Island “hill humus” and “rich humus” types. So few of the Campbell Island soils were collected that it would be inadvisable to discuss the results at any length. They have a higher lime content than the Auckland Island soils. This island is not so precipitous as Auckland Island. The formation of deep beds of humus is therefore favoured. Sphagnum bogs also occur on the island, and may have been of much greater extent in the past. The only one of any size observed was 250 ft. altitude, at North-west Bay, had an area of about 50 it. by 30 ft., and contained scattered plants of Poa ltorosa, Dracophyllum, Bulbinella, and Carex appressa growing through it at long intervals. No Sphagnum bogs were observed at Auckland Island, neither can I find any mention of such having been observed by Cockayne. K 2650-3.—These are a series of samples taken on the 13th January, 1909, from an 8 ft. face of humus and plant-remains which had been excavated to allow the buildings of the homestead at Perseverance Harbour, Campbell Island, to be erected (fig. 1, p. 485, 1s taken near the locality). K 2650 is the ‘surface 21] in. ; K 2651 is 8in. of black humus of greasy texture below K 2650; K 2652 is 2 ft. of plant-remains below K 2651; K 2653 is a further 2 ft. of plant-remains below K 2652. The flora was Bulbinella Rossuw, Poa litorosa, Aspidivum aculeatum var. vestitum, and Dracophyllum scrub. Soils and Soil- ee SUBANTARCTIC ISLANDS OF NEW ZEALAND. 763 K 2647 was the first 18 in. of soil taken on the 13th January, 1909, at 300 ft. elevation above Perseverance Harbour, Campbell Island, at Zircon Creek, and above the subsoils and samples K 2618-20, 2648. The soil was resting on dead logs, and was supporting a flora of Poa litorosa, Aspidium aculeatum var. vestitum, Bulbinella Rossu, Dracophyllum sp., Helichrysum prostratum, Carex appressa, Lomaria. J 834 was the first 121m. of soil taken on the 25th November, 1907, by Mr. Marriner at 40 ft. elevation above Perseverance Harbour, Campbell Island, on an open hillside facing south-west, and about 100 yards from the sea. Flora: Tussock (Poa foliosa), Bulbinella Ross, Pleurophyllum speciosum, Aspidium aculeatum var. vestitum, Coprosma ciliata, Veronica Benthama, mosses, and liverworts. (For pano- ramic view showing class of country see fig. 6, p. xix, and fig. 3, p. 491.) For notes of other samples see tabular statement. ANTIPODES ISLAND SOILs. K 2655.—Taken about 200 ft. above sea-level to a depth of 16 in., below 6 in. to 8in. of undecomposed fibrous plant-remains, in a swampy inland situation, on the 15th January, 1909, on Antipodes Island. Flora: Aspidiwm aculeatum var. vestitum, Carex ternaria, Lagusticum antipodum, Poa litorosa, moss. K 2656.—Subsoil, taken to a depth of 1 ft. below K 2655. These soils are both singularly rich in total and available phosphoric acid. They are derived from recent volcanic lavas and scoriae, very rich in plant-food. K 2657 is a peculiar red earth, containing about 30 per cent. of phosphates, occurring on the edge of the clifis near the boat-depot. It is undoubtedly derived from the red scoria K 3001. Both of these contain titanium-phosphate, hitherto unrecorded in nature. This phosphate must rank as one of the most insoluble phos- phates known. It resisted the action of six successive fusions with bisulphate of potash without appreciably diminishing in solubility. K 3001 (red scoria) and K 3002 (basalt) both contain an abnormal proportion of phosphoric acid, which in the scoria is in the form of titantum-phosphate. K 3003 contains relatively little phosphoric acid. K 2657. | K3001. | 3003. ted Earth| K 3002. ee | eee e sok from Basalt. corla. Ncorla. Scoria. | Silica (Si0, ) aks Ne es av | 41°95 40°50 2768 | 43°15 Alumina : ae ree) S40 14°85 | 10-48 | 15°45 Ferrous and ferric oxides be oe |) ayes 14:30 | 27-40 14°85 Phosphoric acid (P,O;) .. 3 a Zea O52 lO 70. a k08 Lime (CaO)... ss . S910 Oma tes On 22880 Magnesia (MgO) Bic a <- 5°58 A32) |, (0-79 7:10 Manganese-oxide ae 203) ae Ag 0:25 0-28 Ilage | Org Titanium-oxide (TiO,) .. of Ly. Dr As) 0:46 3°44 Sodium-oxide (Na,0 Vee rite oe ed Te 4:13 2:25 4:10 Potassic oxide (K,O) A x 156 | 0-96 0-60 1 55 Moisture at 110° ae Pe a O36) 1010 4°35 113 Loss en ignition a By Sh age 5:00 Titanium. phosphate (insoluble) Me AER esto} 5 Rromes ns S400 Zirconium-oxide Bt Re Fa ee ey 0-41 | 764 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Soils and Soil-formers. The matter insoluble after treating with HCl and HNO, was evaporated to dry- ness, treated with HF and H,SO,, dissolved in HCl, and fused with KHSO,, equalled 8-4 per cent., and was found to consist of titantum-phosphate.* ROCKS AND SANDS. Of considerable use in applied agriculture are various rocks and sands. One need only mention phosphate rock, the source of mineral superphosphate ; lime- stone or chalk, from which quicklime is produced; gypsum (“land plaster”’), yet to become generally used by farmers in New Zealand, but well known to fertiliser- vendors as a suitable diluent for fertilisers ; and siliceous sands and gravels, exten- sively used in peaty Huropean soils. The chief desideratum in the work amongst rocks was, of course, the discovery of a phosphate-deposit. The Campbell Island limestone bears a considerable re- semblance to Milburn (Otago) rock phosphate, but analysis proved that the nuclei which gave it so much the resemblance of phosphate were merely siliceous. The content of silica in the samples of limestone examined was, indeed, enough to con- demn the stone as a source of limestone for “burning.” However, no doubt deposits of suitable lime could be found on further search. Hector (23) mentions the occurrence of true chalk, with large flints, at West Bay, Campbell Island, where the cliffs are composed of chalk and flints on limestone. Small samples of chalk (K 2649) were found by me, interspersed with clay, at Perseverance Harbour. If the West Bay chalk is similar to this specimen in texture and composition, it would make an admirable dressing for Campbell Island soils without previous burning, if it were merely crushed up and spread on the soil, or harrowed in.f At the same spot in the clay were found beautiful blue granules of vivianite (hydrated phosphate of iron). The occurrence of this is probably due to the bones of animals, iron having replaced the calcium of the bone, a molecular change which has been observed in moa-bones exhumed from swamps in New Zealand. t Of less interest to the agriculturist 1s the occurrence of zircons (K 2620) in this clay, from which they can easily be washed with water. Zircon (zirconium-silicate) is a precious stone containing the rare element zirconium. Zirconia (the oxide) is used to some extent in the incandescent method of gas lighting. An attempt has been made to collect evidence as to the method of weathering of the rocks from these islands. Analyses of samples which showed weathering and unaltered rocks side by side are given. * This has the composition 37:1 per cent. P,O;, 61°82 per cent. Ti0,, 1:08 per cent. undeter- mined; is hygroscopic; and has a specific gravity of 275. It may be the phosphotitanic acid of W. Knop (Watts Dict. of Chem., Second Supp., p. 976), 3T10,.P,0;, which contained 37°16 per cent. P,O,. + Chalk also occurs at Oxford, in Canterbury. For analysis, see Cox (24). t Vivianite has been reported from several localities on the mainland of New Zealand by Skey (Colonial Laboratory Reports). An extensive deposit is said to have been found at Mercer, Auckland in 1892, recorded by Park (25). Sorls and Soil-formers.]| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 765 CAMPBELL Istanp ZIRCONS AND ASSOCIATED Rocks. | K 2649. cect a, | K 2610 Blue Clay con- We eathered Te seri | White Rock —— _ taining Zircons, Siliceous Rock Campbell associated Vivianite, associated Tedeael with and Chalk. with Zircons. ‘ Zircons. Lose on een ne ict Nh oe ALTO) 3°30 as 3°05 Silica (Si0,) nf a) 2a%68 57°60 33°20 94-93 Alumina (A1,0.) 9: ¥ 14-25 20°64 0-70 0:98 Ferric oxide (Fe,0.) en ls cO0. 4-32 Trace W752 Calcic oxide (CaO) ie 1-20 2-60 au 0:52 Calcic carbonate (CaCO,) .. 5-40 a | ay Sodic oxide (Na, 0) at 1-84 7-16 | Dipotassic oxide (KO)... 1-20 2°35 Titanium-oxide (Ti0,) e 3°10 1-20 Manganese-oxide .. an Trace Trace ne | Zirconium-oxide (ZrO), &c. | 2:75 0:50 66:10 | a Phosphoric anhydride (P,O;) 1-66 0-10 Trace Magnesic oxide (MgO) a 172 1-59 | dikes Specific gravity .. f a a; 4: 63 | Samples found associated with fe clay at Titeon Grove aL ft. ng lens tthe Dice collected 13th January, 1909. Zircons have been previously recorded by Cox (24) from the western slopes of the Southern Alps, and possibly from Doubtful Sound, Otago. They are also pre- sent in Snares granite (q.v.). K 2618 contains a few microscopic zircons. K 2619 is probably a weathered flint. Locality, same as K 2649 (chalk). TiraNIFEROUS TRONSANDS. one a “K 2654. | K2766. | 9393. Loss on ignition .. a me es 0-60 1-02 0:25 Magnetic iron-oxide (Fe, 6 $j 46:00 | 36°80 | 76°73 Titanium-oxide (TiO,), owe probably small quantity 42°18 31-58 9-20 zirconium-oxide (ZrO. ) Calcic oxide (CaQ) Le a ee al 0:68) 0 4:70 1-80 Magnesic oxide (MgO) ee i: eh OB Se OL 2°77 Alumina (Al,0;) A ie sa) | DAD 2°85 4-00 Phosphoric noid (Ps 0. :) ee Ke eo races diace 0:09 Dipotassic oxide (K,0)_.. oe ae 0-14 0:36 0-90* Manganoso-manganic oxide (Mn,O ) Bie = 2:92, 9) 2:21 0:34 Cupric oxide (CuO) a cf a Palate nie s-: Ae Stannic (tin) oxide (SnO,) o is Oh 0:32 | 0-21 Wi Silica (Si0,) Se ba jal lees) 16°23 3:90 Sulphuric anhydride (SO, Ys Mes is eg here = 0-01 * Includes alkalis and undetermined. 766 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Soils and Soil-formers. K 2654 is a coarse sand freed from gangue ; it was collected on the sea-shore, near the woolshed on Perseverance Harbour, Campbell Island, on the 13th January, 1909. The sample, which was very coarse, was freed from gangue before analysis. The high proportion of titanium (and possibly zirconium), and the presence of tin in small quantities, is the most interesting feature of these ironsands. Taranaki ironsands contain only a low percentage of titanium. Maclaurin (26) found 9-2 per cent. TiO, in a sample from New Plymouth beach (sample No. 9323). K 2766 is a fine sand, analysed without removing gangue. It was collected on the beach at North-west Bay, Campbell Island, on the 12th January, 1909. It is a sample of much finer grain than K 2654, and no attempt was made to separate any gangue or siliceous matter. CARBONACEOUS Rocks (LIMESTONE AND CHALK), CAMPBELL ISLAND. as |. 8a | 32 ee od PE | ub BOS s ae Se |Sute Sie | 2828 VE sie me = eo a BS 6a | ed aa 1S) pa Ay Water te Ma: - Se 2°51 1-95 3°80 1-44 Silica (SiO,) . . 255), ORD 1-91 64:06 | 57-46 Tron-oxide (Fe,O,), alumina, and titanium- 3°64 3°03 1-44 1:48 oxide (Ti0,) Calcic carbonate (CaCO, ) a ae 73°80 90°33 27°90 39-26 Calcic oxide (CaO) .. z ~ op aA ss Magnesic carbonate (MgCO,) .. 0:75 0:95 0:30 0°36 Magnesic oxide (MgO) 2 oy af a 1 a Phosphoric anhydride (P,O;) >> | divaces 1:21 Trace Soda (Na,O) and potash (K,0) 4 oe am 0:25 Ferrous carbonate (FeCO,) .. ie . on 2:08 Sulphides and carbides en ee fe ne Trace Specific gravity ue a * o me 2°28 2°15 K 2711, trom Perseverance Harbour, Campbell Island, collected on the 12th January, 1909, is a very hard, light-grey rock, resembling flint, weathered to white for about 4in., covered with iron-stains. A smell of carburetted hydrogen is given off when the rock is treated with hydrochloric acid. It is hard enough to scratch glass. Contains traces of sulphides. It is very similar in composition to a speci- men of hard, compact, siliceous limestone from a vein in basalt at South Harbour (Perseverance Harbour), Campbell Island, collected by McCormick, and described by Prior (14). A thin section showed that it consisted largely of the remains of Foramimfera (Globigerina) in fairly good preservation. A note is given of the palaeontology of the specimen, Soils and Soil-formers.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 767 K 2649, chalk, was found associated with the clay, 24 ft. below the surface, at Zircon Creek, 300 ft. above sea-level, at head of Perseverance Harbour, Camp- bell Island, at same spot, and below where samples of soils K 2647-8 were taken on the 13th January, 1909. K 2662 is a white siliceous limestone, collected on the 13th January, 1909, from a face eroded by the sea, in a bay at the head of Perseverance Harbour, Campbell Island. Samples of it are soft and white on the outside, but contain a hard and close-grained nucleus resembling lithographic stone. Basattic Rocks, ARGILLITE, AND IRON Pan, CAMPBELL ISLAND. fis ee , ie ee a) Oe aes a eee a Se eee ce gases - : ; ( ae a aL = 63 Silica .. x S, .. | 49°75 | 51:65 | 58:40 | 44:81 | 16-80 Alumina s Semele We 1912s 14:00) 13-95. 1.51 53:00 Ferrous and amie oxides 58 PBR B a 6-60 7:20 23°10 52-00 Calcic oxide ie és. 5:65 5:50 8°35 3°00 0:86 Magnesic oxide .. a x 133 0:76 2-29 2°49 0°51 Potassic oxide .. ss we 1:6] 1:92 0:50 OE AS Stolen Sodic oxide : ie on Neon T:47 4:42 a a Phosphoric anhydride Be 0:80 0:98 | Trace 1:02 1:18 Titanic and probably zirconic oxides | 3°75 3°50 1:90 4°60 Se Manganese-oxide a ies 0:60 0:55 0:23 | Trace 0:80 Loss on ignition ue Bee la 2-30 O25 | G5 25-40 Zirconium-oxide a Ms ae By he | 1:23 as K 3005 and K 3005 were collected from a very hard basaltic boulder with erystals of olivine, from the sea-beach at the homestead, Perseverance Harbour, Campbell Island, on the 13th January, 1909. The analysis of the weathered and unweathered portions show that the weathering has chiefly resulted in a diminution of iron and magnesia. K 3004 is an altered sedimentary rock (phyllitic argillite) containing quartz, mica, and pyrites, collected on the 12th January, 1909, from Perseverance Harbour, on beach under Mount Honey. K 2671 is an iron pan from the same locality. It differs from other iron pans chiefly in the excess of alumina and the presence of the rare earths. It is interesting to compare it with K 2616, a bog-iron, which was collected at the Ocean Beach, Bluff, below a layer of humus at the littoral, in January, 1909. The iron in this sample exists as hydrated ferric oxide, with a small quantity of organic matter. It is similar in appearance and analysis to the bog-iron ore mined in Sweden and Ireland. (See Watt’s Dict. of Chem., vol. ui, p. 338.) 768 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Soils and Soil-formers. BasaLttT FROM Camp Cove, AUCKLAND ISLAND. K 752. | | Weathered dab d | Possession pes __ Red Devens | X799.| K 800. | K801.| K802.| Island | Exterior ene | Basalt. Portion. Belsale Water oe ec 8°32 O-1l 6°20 | 2°55.| 4:26 | 2°81 | 2°34* Silica a ae 42°36 55°75 52°35 | 46°80 | 52:95 | 49-60 45°61 Calcic oxide .. 35 2°60 4°30 aioe | 6:00) 3:05) 5:05 6°34 Ferric and ferrous oxides 911 | 14°76 4:95 |16:20 | 5:40 | 14-99 13°46 Titanium-oxide oe 2°11 0:75 5°10 | 2°45 | 6°20 | 2°81] | 3°48 Maenesic oxide - 1:86 0:54 | 0:36 | 1:47 | 0-20 | 1°87 4:84 Manganese-oxide Be 0:20 0:20 | 0:06; 0-11 | 0:04 0:09 | Trace Sodic oxide .. ae 4°53 5°90 S01 o3955) Ion eco | 5°06 Potassic oxide es 2°43 | 223) a) 2: OU 2276 | e229 om 2°67 Alumina is a 26°48 15:93 17°80 | 15°90 | 16°94 | 14°70 15:70 Phosphoric anhydride... | Trace Oni 0:52 | 0-79 | 0-67 | 0°86 * On ignition. K 7524 and K 3005 are very high in alkalis and low in lime for basalts. K 752 and K 752a are from a basalt boulder weathered by sea, from the beach at Camp Cove boat-depot. The exterior scale was hard, red, and polished, and extended for 4 in. into K 752a. The Possession Island (Victoria Land) analysis 1s that of a basalt collected from that locality by the Ross expedition, and analysed by Prior (32). J 799 and J 800 are weathered and unweathered basalt collected on the 23rd November, 1907, from the littoral at Camp Cove, Auckland Island ; taken under a humus soil. J 801 and J 802 were taken near the same locality on the same date. These analyses show the effect of the humus soils in weathering basalt, plus the action of a certain amount of sea-salts which would drain through the soil in such a situation. BasaLt FROM AUCKLAND ISLAND. | J833a. | | Weathered : en | to a Pink ee 656. | K 2999. | Powder. Silica ce es 3 Py se220 45°85 4214 | 44:55 Alumina as ee te ae 13°65 TSO | OS 15°45 Ferrous and ferric oxides HM be 1:50 14:94 | :12°61 15°60 Titanic oxide. . a ie 2°70 4-10 4-90 3°70 Calcic oxide .. ca Ke 0°55 9-02 ellOs32 8°81 Magnesic oxide a ae Ae, fy eile? 6°69 9°47 5°40 Manganese-oxide ips es Gaal a 0-09 0-12 0-11 Sodic oxide he or feos 6°66 3°45 3 27 3°18 Potassic oxide * = ieee eT 2-02 1:80 1:80 Phosphoric anhydride . . | Trace 0-11 0°40 0-64 Loss at 100° C. te | 0°33 0°20 0°28 1:26 Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 769 656 is an olivine basalt from cliff between the Gap and Horseshoe Bay, Ross Archipelago ; analysis by Prior. Merrill (27) gives a number of analyses of basalts and their residuals on weather- ing, and calculates the percentage of each ingredient lost in the altered product. Potash is one of the ingredients invariably lost in large amounts. In the samples analysed from the Auckland and Campbell Islands it will be seen that the reverse is the case; the humus acids evidently hinder the decomposition of the potassic compounds. J 833a and J 833B were collected at a depth of 3 ft. 7in. from shaft sunk at Skua Gull Flat, Camp Cove, on the 26th November, 1907. K 3000. K 2645a. a Basalt, Enderby | Rotten Rock, Island. Auckland Island. Silica (Si0,)_ .. se ah i: me 45°65 42°30 Alumina (Al,O3) “ es aE - 14°10 18-05 Ferrous oxide (FeO) .... 2 af Salil 13-95 | a Ferric oxide (Fe,0;3) .. a e mel 10°55 Calcic oxide (CaO) Me n Hi: 9-10 1-00 Magnesic oxide (MgO) .. * ee i 7:02 1:38 Manganous oxide (MnO) oe aes oF 0-08 M Phosphoric anhydride (P,0;) oe aa 0-52 0-25 Titanic oxide (Ti0,)_ _—.... es ae ie 3°70 4:10 Potassic oxide (KO) .. ae ae oe 1-01 | 1:05 Sodic oxide (Na,0) at: ie a Re Boy een | 3°38 Loss on ignition a es os be 1-30 18°40 K 3000 is unweathered basalt from Enderby Island beach. K 26454 is much-decomposed rock from below high-water mark on beach, Camp Cove boat-depot, Auckland Island. GRANITE FROM SNARES ISLAND. | | K 1297. on K 751. Weathered by Atmo- Unweathered. spheric Influences and Sea-water. Silica a3 4) ae ay eal 72:00 ieasOe Alumina... - a er = 12°96 13°56 Ferric oxide bi 2 Eat 3 as 1:98 1:80 Titanic and zirconic oxides... e as 1:56 | 1-20 Manganese-oxide e ie ES aa 0:30 | 0-30 Calcic oxide ee fe is 1 1:12 0°82 Magnesic oxide ee oe he BN: O41 0:28 Sodic oxide .. _ By a8 be 5eB> | 5°35 Potassic oxide a wg Bs reo 4:06 | 3°95 Loss at 110° C. ee oe i ) *SJOOI MO} poulezUoD (;) "16 £ Se syoor Aueur sv soul, eg paulejzuoD (v) *1Z.0 ‘OpIxo olsouseuUL £0zZ-0 “eprxo O10[eo *e.( ‘aplxO olsouseU { 9ZT-0 “eplxo oI0;eo *S4001 JO 9¢-0 |88:0 |29-0 |9F1-0 | €1-0 90:0 G0:0 | S1-0 | 01-0 91-91 |00-8L FE8 £ os 2-4 06-0 |0&-0 | 61-0 | 80-0 | ST-0 | T1-0 | 10-0 | 10-0 i: II-0 | $0-0 | Gu-0 08-21 (00-84 99S ‘poy | €¢:0 [FS-0 |€¢-0 | 9€:0 | GI-0 |O1Z-0 |SE1-0 |4E0-0 |LE0-0 | 60:0 GZ0-0 |S80-0 [920-0 F671 |00-€8 LEIS ‘poe Ajsu0yg | LF-0 |0F-0 |LF-0 | SL-0 | 20-0 [022-0 \SZI-0 |LZ0:0 |L10-0 | 40-0 880-0 |S€0-0 |1¢0-0 98-11 |00-08 0996 ‘prey | OFT 1&0 |0F-0 |860:0 | G0-0 |O81-0 |O€T-0 |I110-0 |F10-0 |920-0 Gg0:0 | #0:0 |800-0 ZISI {00-8 169% ‘poe Apysys | 10-1 [68:0 |SF-0 |S90-0 | ¥0-0 |I9T-0 |OS1-0 [620-0 |€10-0 | 11-0 G00 | ¢0-0 | €0-0 9-1 [00-38 GS9G poy | 09:1 (16c-0 |Zc-0 | IL-0 | 20-0 |F61-0 |ZS1-0 \Sz0-0 |¢c0-0 | 10-0 80:0 |9L0-:0 |L20-0 99-21 |00-18 E96 780 (€P-O |1E:0 |L80-0 | 01-0 G00 | 20-0 | 80:0 92:0 | 04:0 | 20-0 OL-E1 |00-8L | (a)6F0E 09:0 68:0 |ZG-0 |91-0 |¢L0-0 Z0-0 | 10:0 | ¥0:0 6&0 | 82-0 | #00 84-61 |00-69 | (u)099% #60 [PS:0 |1L-0 | L0-0 |Fg0-0 10:0 |S$10-0 | ¢0:0 GLO:0 |#80-0 |90-0 96-:LZ [00:88 | ()IF9S ZL:0 |0G:0 |€1:0 | L0-0 |c€0-0 ¥0-0 |Z80-:0 | 80-0 80:0 |#80-0 [600-0 FG-11 |00-cL | (x)PrOEM ‘poy | GeO |0F-0 |1S-0 | €1-0 |890-0 1z0:0 |610-0 oe ¢0:0 | 10-0 ae 9L-18 |00-SL GZ8 ‘poe Ajurey | 08: |PFO |68:0 | ZG-0 | €L-0 90:0 G00 | 91-0 ee 08-E1 00-94 $28 ‘oulTeyTY | OLS I1-@ |0F-9Z| 0:0 | 80-0 GG0-0 |910-0 3 ZL-0 | SLO : 0Z-E1 |00-6S £28 a 08: 08:0 'FZ-0 | 11-0 |¢F0-0 é +0:0 “| 10-0 06:01 |00-62 G68 ‘prow Aguieg | 09: [79:0 |1F-0 | 62-0 | Z1-0 ae ¥0-0 “| 0-0 91-01 [00-8 128 ‘poe ABYSS | 0¢-6 j\8€0 |PEO | ZI-0 | S10 OF:0 60:0 | Lz-0 | SL-0 OFFI “| ()068 Se 0ST 1:0 |6z:0 |§60:0 |Z€0-0 Be a 61:0 ae GFE ae (618 = 00:7 €8-0 |22:0 | 01-0 |SF0-0 €1-0 re Be GPP 00-SL | (W818 yy 06:0 0:0 01-0 |190-0 |€z0-0 60:0 “| g0-0 | 90Ra], 88:61 [00-88 | (a)LI8 te 0Z-0 (0¢:0 |#Z-0 |980-:0 |6¢0:0 80:0 “| €0-0 3 69-S1 00-28 | G)9TS ‘poy | ogg [G80 |§€-0 | 81-0 | IL-0 0:0 G00 | 6z-C : F1L-ZE (00-08 GIs ‘poe Apurey | O18 [86:0 80 | 86-0 | 600 G0:0 |600:0 | ¢0-0 90:0 | 80:0 | 20-0 8E-FI |00-:0L | (r)FI8 ‘prey | 00:8 |0€:0 |81-0 | 29:0 | 01-0 90:0 | 20:0 | 90:0 eo | OO | °° FIST |00-LL | G)sts 4 0€-SI 09:0 09:0 | 89:0 | TI-0 G0-0 | 10:0 | $00 : G0-0 | 70:0 6G-Z1 |00-€9 | (,)Z18 a 06 |PS-0 |S8-0 | 09-0 | O1-0 80:0 |F10:0 | 01-0 0:0 a oS 29ST |00-8L | ( )II8 ‘pov Ayuiey | 00-4 |8¢-0 02-0 | 81:0 | 40:0 10:0 $0:0 ee St 90:91 |00-6L O18 ee 81-0 0F-0 |1€:0 | 80-0 | G0-0 80:0 £0:0 ; 0:0 89:F1 |00-E8 608 Bs 0G: €6-0 (0S-0 |60:0 | 40-0 ¥0:0 ae 61-0 a OF-LL |00-€8 | (v)908 a 79:0 |L&-0 |€9-0 | 71-0 | 7T0 610 ‘ 6Z:0 | PLOT (00-18 | (0)S08 poy | 08:1 ({0F-0 |0F-0 | G20 |z90-0 x ‘ cL-0 : Z6-F1 \00-FL | 08 06:6 |PS-0 |90:1 | €2-0 | F1-0 #2 90:0 | P10 | 90e4, 80-11 |00-FL 86L OL-1 ZP-0 |9Z-0 | G00 |8F0-0 cL-0 oc ae 23 91-S1 00-8 L6L i 08: [72-0 |LL-0 |191-0 |S¢0-0 0L-0 ge 10:0 09-81 |00-98 96L ‘poy | €9:0 (|SI-0 |1¢-0 | 42-0 | €1:0 IL-0 €0:0 | #0:0 | 9%ez], ¥9-S1 00-08 G6L 0-0 (0F-0 |1€-0 | IL-0 |SI1-0 10:0 be “* | €0-0 ZL-91 |00-98 F6L ‘poy | 81-0 (02:0 |€9-0 | 80-0 | 81-0 a LPI | 19:0 | €1-0 ZS 100-66 | (a)E6L Fe 06:0 (|Sh-0 |0S-0 |160-:0 | 0S-0 60:0 | 91:0 | €2-0 08-0 | GL-0 Pe FFS1 (00-19 | (v)e6L ‘prow Ajsu04g | ZL:0 (6&0 |1¢-0 | 1-0 | 02-0 68:0 91-0 | O10 | 21-0 ¥6-F1 00-14 16i£ BOPSe lies | es | Ss | ae ie = Bg Ba | 36 2 os jose | eee | & e5 2 |Sseel| she SI ae Ex | oa | © Se | OS | oe ) as | OS | 8 fo) Ba | 5H Z | as 2 “SHUT aS gail el x Ee @ 3 > gs = z BS | & BOS S Aso| * 3 04 BE EG B | BS a | BS Bl ae ee aes ae ee fe WOOO, > : s ‘ z : oo) 6G Zz (‘0 .00T 98 pelup [log WO pozye[Mo[vo) Ploy x (‘OD .00L 3% pelap [fog Wo) (9 ,OOL 48 Palp [Log Wo peyefnoqeo) 10}}eAY aq NJOS-10y" AA “1o4U\\ dO;OoIpAH, Bu0wyG Aq popovszxe 10}7vTT PLOY OGIO juep Jed T ur efqntos “GNVIVEZT MON FO SANVIST OILOUVINVENG AHL FO STIOg 4oO NOILISOdINO/) 773 Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. ‘prow Apsu0g | EZ-g FL-0| 12-0 | 9-0} €1-0 | £0-0 | GI-0 | €1-0 |L10-0 “s ** | G00 | OT-:0 | €0-0 | €0-0 | IT-0 | 60-0] OF-0 | 00-91 ee SP9G ‘TRINON | F6-Z &6:0| GF-0 | FF-0| OL-:0 | 61-0 | 91-0 | 0-0 | 10:0 | ¢0-0 “* | GO-0 | L0-0 | Z0-0 “+ | $L:0 | IL-0] GF-0 | GL-F%| 00-08} 9993 ‘plow Ayquieg | ¢L.Z 96:0 | 0S-0 | GF:0| LO-0 | Gé-0 | LI-0 | 40-0 | 10-0 | €0-0 ** | 0:0 | 80-0 | 20:0 ** | $L:0 | LI-0| 6F-0 | 0€-0Z} 00-E8} ¢Eg9z % 06°8 96-0| 97-0 | GS-0| 80-0 | 1-0 | 81-0 | ST-0 | €0-0 | L0-0 “+ | OL:0 | 70-0 | 40-0 | 70-0 | 80:0 | LE-0| FL:0 | OL-L | 00-1¢) LOE 0¢-F1 6€:0| 89:0 | €G-0} L0-0 | LI-0 | 02-0 | 8T-0 | 80-0 | OT-0 ti 90:0 | G0-0 | L0-0 | 90-0 | F1-0 } 9€-0| €8:0 | 00:8 | 00:LF| FF9Z ‘TRINON | OG-F 67:0} OL-G | 88-0| F0-0 | c-0 | €8-0 | TZ-0 |L00-0 | Z0-0 “* | 70:0 | 60:0 | 20-0 | 20-0 | 90:0 | #2-0| GG-0 | OT-FZ) 00-LL| 9F0E ‘pre AQuieg | gL-F | 6-0) OL-G | 16-0} €0-0 | 81-0 | 92-0 | OF-0 |F10-0 | 10-0 “* | 70:0 | 60:0 | €0:0 | 0-0 | 90:0 | ¥2-0) €F-0 | 8Z-ST| 00-EL| &F9Z ourjeyye AUNT — OZ-E | OF-0| 0G-F | GS-0} 90:0 | FZ-0 | E81 | 12-0 | 10-0 “* | 70:0 | G0-0 | €1-:0 | €0-0 | Z0-0 | 60-0 | 02-0} LZ-0 | F0-LT| 00-69} SHOE “OulTexTV | 08-§ | 69:0} 00-FL} 8&1} 90-0 | €1-0 | Sé-€ | IT-0 | 10-0 ae 40:0 | 0-0 | O1:0 | €0-0 | 20-0 | L0-0 | 8-0} 8F:0 | 98:6 | 00:09) ZFOTH | Sell 4 Q g y 2 a ~ i a a a = wR ma ) 8 aH Bee oe eee | eter lees |e lec] Per ee eee eee ee ea) ae |} 2082 | 8 S Ze] &@ 5 S au a S 8 Pi se. We be = B. = |e | Bie 5 \Bobe | & Sf ese. jos 3 ge | Eg S g iS et |p eRe = | 3 Bees eres | ae. *snU}IT Ses pS a aS x ° a aS ° — a g a} ° 2 el ey 2a ne S 0} WOIoweT "REE | & s : & Ss : : & = ; S as 5 g sis a 2 "(CO .00T 98 PalIp [10g U0) ploy ___°CO 001 48 parp [og wo) 463) hie See RAED HOPED METERS) Sa or ROUEN cena ° dMOTYIOIpAH SuoAyG Aq poqowszXe Jaqgey_ | Ploy o1yIp “quay sed T ur efqnjog | | PanuyuooI—ANVIVEZ MAN AO SANVIS]T OILOUVINVENG GHL AO STIOQG AO NOILISOdNO;) [Soils and Soil-formers. SUBANTARCTIC ISLANDS OF NEW ZEALAND. 774 yonur sureqzu0D (p) “17, pUe Ty, ‘Anois |y pue oq 94 JO S[eyour Ul Yor Ayjetoods@ (0) “OMT 87941] “euUTUIN[Y YOnuT sureqUO) (7) “BOUL YONUL SUTYAUOD : MOI WOT, dary AT[ROIQOVIG (“)) “COLL WIM pues xorTq sureyzUOD (a) *‘sugive o1vi pue OlL “WOM 9199] ALOA *Vorur YonuT surezU0) (x) | | -ogeuogiey | -ogemoqie) oIsousey | o1oTeD ¢0-0 IL-1 €o-1 0S-0 Ge-0 61-0 g¢-0 0OL-0 0¢-0 60:0 10:0 90-0 ¥60 02:0 9L-0 SGV 00-0 G0 90:0 £00 10-0 c0-0 GT 6-0 99-0 €S-0 0r-0 ¢0:0 09:0 OL-0 cP0 OFT 60:0 OL-0 b0) 0:0 GOT 8e1 (Caat 91:0 90:0 €1:0 ; 80:0 90:0 91:0 €1-0 96:0 GZ:0 90:0 (OO) 10:0 | 02:0 | SL-0 | 69:0 610 61:0 €10 C0) 0) Go-0 GLO 60:0 60:0 (0) 0) 10-0 80:0 66:0 9¢-0 S10 OL-0 C10 (0) 0) FL-0 10-0 09:0 4) GAO) OIPOS F6-0 cr0 61-0 g¢-0 IL-0 8c-0 ¢1-0 F1:0 L1-0 L1-0 81-0 | S01:0 €€-0 91:0 (acm 01-0 68:0 Ge-0 91-0 81-0 STO 1:0 €1-0 91-0 8L-0 CFT 1F-0 6:1 ¢0:0 Gel 0-0 Sia 11:0 81-0 O10 1-0 40:0 GG-0 O10 60:0 F110 C10 86-1 0€:0 81-0 64:0 91-0 6:0 61:0 0€-T GG:0 18:0 61-0 9¢:0 OL-0 91:0 FE0 Ge-0 Lé-0 96:0 I¥'0 €€:0 60:0 OF:0 06:0 80-0 ST-0 GG:0 S10 16:0 0z:0 CTO gG-9 0c:0 69°G 81:0 GL-T G10 Ss BS ‘ageyding|eploypo | 2S orpog ES ie) 81-0 ¥S-0 69-0 €1:0 FL-0 60-0 81-0 L¥:0 6E:0 66:0 G10 80:0 0) LO:7 ¥9-0 €0-T L1:0 L0-0 80:0 01-0 L0-0 (0) OF-0 cé-0 GS:0 9€-0 0€-0 90-0 G10 61:0 60-0 L9-0 80 0 06:0 6&0 91-0 Gé:0 09:0 $S:0 “apIxO dISSeIOg -(ygeq-UURAIG TO FYSTIOM JULISUOD 09 PelIp [og UO spr) “q' A Aq SISATeUY [FO | 0F-0 | OL-0 0&-6 GS-0 GLE | L9-1 8L:6¢ | 68-4 IL-S6 : Te Pegnt IF-0 | O8-€ CG.1Ll | 9-6 (TLE | OF-1 LL-Sb | 89-66 | GEL | ~~ “* OF9Z 8L:0 | 8-0 66:6 GF-8 F866 | 98T 9L-9F | PIGS | 98-LL L9G 19-0 | ¥9.0 0S-1 81:0 L8-0F | 66:1 IL-@G | &hP LG:S6 0996 64:0 | ¢S-0 88:1 1F-0 1E-68 | SET 9¢-¢¢ | C64 80:96 oS GFO | 6F-0 rere 16:0 Corr | Loi $9-0E | 98-¢ F1-96 ECS IG 8Z:0 ) 99-0 99-2 8¢-1 L¥-6E 1¢-1 6L-3S | &F-9 LG.86 20 BERK SF-0 19:0 66:2 6¢-P 06:68 | 2& | S6F | 9F-II FG-88 “* 6F0E tF-0 =| 6F-0 Gaal LUG 6G-9E L&E | GO-ES | 3G.9 SF-€6 “* 099% 19:0 | 12-0 6L-1 9L-0 FL86 | Sal 6.-L9 | 6&-€ 19:96 1F9G 19:0 1z-0 L&T SLT 89:S§ 8aT 0€-6¢ | FL-€ 96:96 FFOEM GF-0 | 8¢-0 $L-0 6F-0 G9-GE itor 6Z:09 | SL GEL6 | °° 20 EB €6-.0 | ¢8-0 04:9 GL-9 OLS | 26S | €89r | ST-LT | S8es | - 20 6S GLO | 29:% 1¢:L 0G-L 10:91 16:0 | &@ST 6L-L9 166 * (3)868 0G:0 | 2-0 OF-0F | LEZ | OSI | 0-0 a3 91:98 | 8ST eS 96-1 19: CELE | FEE | GOLT 11-0 = F838 | OLLI |" LCS 1g0 | 9€-0 FOP 6L-1 GG-eg | FET 9G-LG | SSL GL:26 | SCS 8g-0 | 96-0 8¢-1 €8:0 G0:8 IL 6G:9E | S6-€ G0:96 | °° GIS ZE0 | 88-0 G0:91 69:ST OL-e6 | 80-1 LE-OF | G8 | ST-L9 ()818 1g0 | S10 96:0 OFT 0¢:88 | 9F-T 6L:9E | GZ-€ GL:96 | °° cea Alc) 09:0 | 8&0 61:0 66-1 0g-98 | TLT FLs¢ | GB-¢e GL-96 | “918 1g0 | ¢¢-0 6G-€ 08-€ 08-8z | 89:2 | LESS | SL-OL | ¢8-68 oo GIS OF-0 | 66:0 98:36 | F816 | OG-ST LG.1 €6-66 | 00:89 | 0O-LF | ** (p)FI8 €¢-0 | &F-0 ¢9-L1 FL-91 gg-ee | LoS | SO-LG | GOLE | G69 | -- Beare (i 8g-0 | 19-0 G8-LZ | 19-66 | S&F CLT GZ-e% | GZ-19 | GLs8E |: "+ (p)618 &F0 | 9F0 F811 68-41 co-Fe | 8¢-6 | 69-0F | 80:68 | G6L9 | -° ems GZ-0 | &%-0 C891 ORE || Geers || Siete | Hieiap || GIO | GEN) | >> (5)0L8 GF-0 | 16-0 91:0 08-1 00-FE | O1-% | F9-09 | 96: PL:96 | 7° “* 608 cg-0 | 8-0 Le 9L-1 00-2 | La% | 91-6¢ | LE-9 Ch86 | “908 gg-0 | GL-0 08:0 éL-1 OF-ce | 96:2 | GSFLE | 61-7 18:6 “G08 68-0 1¢:0 0&-% eats CE-PE | 00:7 | 89-SE | LBL e166 | °° encos 19:0 | 96-0 0€-6% | S9-61 GZ-01 G61 88rs | S669 | GOLF |. - oo GR, GS-0 | Lz-0 raat 0¢-0 0¢-66 | 8&1 ZL-0L | 00-€ 00-16 8 Bi, GE:0 | 12:0 09-6 GL:ZI | Sl-9% | 62-6 TL-LP | GS:86 | SP-OL | iee 9G), 0:0 | 16:0 20:01 GI, | 68:86 | 68% | L8-Sh | G8oo | GILL | ~~ “ G6L G¢.0 | 8&0 G0-1 SLI ies | GLI ZEZI | EB L196 | °° -- -$6L 29:0 | OF-0 06:2 OF-9 1-98 | O6L 9¢-7r | LO-6I €6:08 (q)86L 0¢-0 | 89:0 18-2 CLL SETS | 86S | GOLF | GP-S 8o-F8 ¢ (q)@6L 1¢0 | SFO 80:4 FE-9 PLE | Shs | OLSP | SLT 18-98 | ° (w)16LL | eee hens Pome ne Pag ee sae ae = ; Boce | | “Opix “OprIx [sn “MOq1®: a .| “STOg1ed ae ies Pee g “BONIS are | REN ae | teres 10} a2 =o) | | Att Vv IRIOA ‘ON AIOPVIOGL'T Zope eon Te4O.L ‘GNVIVEZ MON fO SANVIS] OILOUVINVANG THL FO STIOG 40 SINMALISNO) TVLOT, 775 Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. | | | PIL 80-0 IL-0 IL-0 | 99-0 GS-0 L8-0 GE-8 GL-6P 8:11 99-T | 86:76 CS-EL SP-L8 | -° SRIS | 9:0 80-0 06-0 CL-O LT-0 9L-0 €9-0 OF-9 Or-S 66°GE FSG 88° 1¢ 66ST TL98 | °° 9992 Lg-0 10-0 0G-0 L9-0 LT-0 86-0 16-0 VES 9L-P LG-9E 00-€ 09-LF €L-S1 L898 | ** GG9Z 8-0 G8°E FL-0 LT-0 | 86:6 ILT LOS 98°§ ¥6°CE 1S-0€ €6-G §L-0 6S-F1 GS-64 8P:0G “LOE T@1 86°E OT-0 0¢-0 IL-1 6ST £9°G PEE 69-&€ OL-1E ELF €L-0 6F-F1 F0:08 96-61 | °° #96 CSF 86| IFO 60-0 80:0 FL-0 81-0 16-0 G9-0 86S | G6-T 00-0§ 00°G 6L-GG 16:S1 GOL-F8 “" -OF0E CLE L8:1 OT-0 80:0 F6'T 1S-0 Pel P91 €¢-G1 8S-G G9-1G 0G-G | ¥6-9F VG'66 9L-OL “$793 OS-L 60-T IL-0 F1-0 | 68-0 GS-0) 1@-1 68°C OF-S 1@1 LO-€€ OL LL-¥¥ 9F-O0G FG-6L “" GP0E 00:6 Ge €L-0 | 61-0 | S81 99-0 86-4 9€-S eel GE-9 9€°8 €6-1 GL-S§ 96-€S F0-9F “" GP9SGM ‘syed | arey Jo “apIxO ‘ayeydmng | ‘apmoryg |‘apipAquy| ‘apixo *OpIxO ‘OpIXQ | SOpIxQ pue, . uoq.rep ree *suioqivo ‘ay eu0q aIposg aTpos aIpog omoydsoyg | oIssejog | olsouseyy Bye)i38) “eur y vOS, poXT USAOTFENI -OIpAH, “USy *I9}}C 1 ‘ON o1oreo | soprxo-woxy | won | Veog | St09B0qe7 “(UJR-WIRA}S TO FYSIAA\ FUeASUOD 09 palIp [fog uo) “WH Aq sisAeUY [e40,, | “PpantyjuogIi—ANVIVAZ MAN LO SGNVIS]T OILOUVINVANG WHEL FO STIOG FO SEINDTALILSNO/) TV.LOT, 776 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Soils and Soil-formers. OO “1S Or OO DO BIBLIOGRAPHY. . Storer. “ Agriculture in Some of its Relations with Chemistry.” . Hilgard. “ Soils.” . Ann. de la Science Agronomique Journ., vol. 1. 1887. . Detmer. Landw. Versch. Stat., vol. xiv (1871), p. 271. Miller. Quart. Journ. Geo. Soc., vol. lix, 1903, p. 136. Nystrom. “Peat and Lignite.” Department of Mines, Ottawa, 1908. Mendeléefi. “ Principles of Chemistry.” English edition, 1905. . Bjorling and Gissing. “ Peat: its Uses and Manufacture.” Ryan. “ Report on the Irish Peat Countries.” “‘ Economic Proceedings of the Royal Dublin Society,” vol. i, pt. 10, July, 1907. . Cockayne. “ Botanical Excursion to the Southern Islands.” Trans. N.Z. Inst., WOl, seoqva, IGOR, j0, 275. . Buchanan. “ Campbell Island aad its Flora.” Trans. N.Z. Inst., vol. xvi, 1883, p- 398. . Kostyschoff. Quoted by Wiley, “ Principles and Practice of Agricultural Analysis,” vol. 1, Soils, 1906, p. 51. » lela, °> Wag Soil,” jo. ISS, GCs, . Prior, G. T. “ Petrographical Notes on Rock Specimens collected in Antarctic Regions during the Voyage of H.M.S. * Erebus’ and ‘ Terror,’ 1839-43.” Mineralogical Mag., vol. xu, p. 69. June, 1898. . Ganong. “On Raised Peat Bogs in the Province of New Brunswick.” Trans. Roy. Soc. Canada, vol. i, 1897, p. 131. . Von Feilitzen. Journ. Landw., vol. xlvi, 1809, p. 271. Ber. Deutsch. Chem. Gesselsch., vol. xxxil, 1897, p. 2571. . Johnson. “The Irish Peat Question.” KHcon. Proc. Royal Dublin Soc., vol. i, pt. 1, 1899. . Clarke. “Data of Geo-chemistry.” Bulletin 330, Geological Survey, U.S.A., 1907. . Friih and Schroter. ‘‘ Die Moore der Schweize.” Bern, 1904. . Persons. Florida Experiment Station Bulletin No. 43, 1897. . Hector. “‘ Notes on Geology of Outlying Islands of New Zealand.” Trans. N.Z. Inst., vol. 11, 1869, p. 176. . Cox. “On the Mineralogy of New Zealand.” Trans. N.Z. Inst., vol. xv, 1882, p. 382. . Park. ‘‘ On Occurrence of Rare Minerals in New Zealand.” ‘Trans. N.Z. Inst., vol. xxvi, 1893, p. 367. . Maclaurin. 36th Col. Lab. Report, 1903, p. 10. . Merrill. “‘ Rocks, Rock-weathering, and Soils.” New York, 1904. . Geikie. “ Geology,” p. 83. London, 1893. . Hutton. Quart. Journ. Geo. Soc., vol. xli, p. 213, 1885. . Booth. Trans. N.Z. Inst., vol. vu, 1874, p. 123. . Paturel. Expt. Station Record, 8, xix, April, 1908. . Prior. Nat. Antarctic Expedition. British Museum Nat. Hist. Rep., vol. i, p. 105, 1907. . Taylor. Expt. Station Record, 8, xix, April, 1908. Soils and Soil-formers.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. Maid . Vogeler. Expt. Station Record, 5, xvu, p. 446, January, 1906. . Muntz and Girard. Journ. Soc. Chem. Ind., 1907, vol. xxvi, p. 1100. . King. Wisconsin Station Report, 1897, p. 232. . Whitson. Wisconsin Station Report, 1902, p. 210. . Groves and Thorpe. “Chemical Technology,” art. “ Fuel.” . Kream. Roy. Ag. Soc. Journ., 1893, p. 780. . Drude. “ Botanisches Centralblatt,” Ixvii, 173. . Wood and Berry. Cambridge Journ. Ag. Science, 1905-6, p. 117. . Darwin. “ Voyage of the ‘ Beagle,’”’ p. 188. . Quart. Journ. Geo. Soc., vol. xxxv, proc. 96, and vol. xliii, proc. 2. . Ryan. ‘The Peat Fuel Industries,” Econ. Proc. of the Dublin Royal Society, vol. i, August, 1908, pt. 13, gives a comprehensive bibliography on peat. ARTICLE XXXTI.—ON THE RADIUM CONTENT OF CERTAIN [GNKOUS ROCKS FROM THE SUBANTAROTIC ISLANDS OF NEW ZEALAND. By C. Coteripcr Farr, D.Sc., and D. C. H. Frorancer, M.A., M.Sc. RuTHERFORD (‘“ Radio-active Transformations’) has calculated that 4-6 x 10~-“ oramme of Radium per unit mass of the earth would generate an amount of heat equal to that lost by the earth by conduction through its crust. Strutt (Proc. Roy. Soc., 1907) has examined the radium content of a number of igneous and sedimentary rocks, and his results, corrected by Eve and McIntosh for an error of standardisation, give for mean values, igneous rocks, 1-7 x10"; sedimentary rocks, 1:1x10-"’—a result which is about twenty-eight times as large as the theoretical quantity. Eve and McIntosh (Phil. Mag., Aug., 1907) find a mean for igneous and sedimentary rocks of 1-1 x 10-¥ per gramme of rock. Joly ( “ Nature,” Sept., 1908) finds values much higher than these, his mean for igneous rocks being 6-1 x10-"”, whilst the mean content of the basalts examined by him was 5:0 x10-" gramme per gramme of rock, which is much higher than that found by other investigators for similar rocks. METHOD OF TESTING. The rocks examined by us were ground to a very fine powder, and 20 grammes were fused in a platinum crucible with ordinary fusion mixture, 1 gramme of the powdered rock being mixed with 6 grammes of the mixed carbonates, which had been previously tested for radium. The fused mass was dissolved partly in dis- tilled water and partly in HCl, both of which were separately examined. The acid and alkaline solutions were then separately corked, with glass tubes and clipped rubber connections in position, the cork coated with sealing-wax to prevent pos- sible leakage of the emanation, and the flasks were then set aside for three weeks to mature. The apparatus was similar to that described by Strutt (/oc. cit.). The solution was boiled for one hour, the steam being condensed in a Liebig condenser, and the gases evolved being collected over freshly distilled water. Before the boiling was stopped the water in the jacket of the condenser was run off, and steam was passed through as far as the collecting-vessel to drive over the last traces of emana- tion. The electroscope was similar to that used by Boltwood (“ American Journal of Science,” 1904), and it was exhausted to half an atmosphere by means of an oil- pump before any of the collected gases were passed in. These gases’ were allowed to enter very slowly through CaCl, and H,SO, drying-tubes, and the, pressure was then brought up to atmospheric by the admission of air, the time occupied over this . 27 Radium.]| SUBANTAROTIC ISLANDS OF NEW ZEALAND. 779 transference through the drying-tubes from the stoppage of the boiling to the com- mencement of observations in the electroscope being in every case fifteen minutes. Readings of the position of the leaf were taken every two minutes for half an hour, and the instrument was then recharged and the leak for the next half-hour similarly determined. This was sufficient indication that the effect was that due to Ra emana- tion. In some cases, however, the gas was studied for three or four hours; but although the leaks increased and the certainty was confirmed of the presence ol Ra, yet, owing to the active deposit, there was considerable time lost in bringing the electroscope back to a normal condition of natural leak, and it was thus de- cided to pump the gases out after an hour. The electroscope was charged to a potential of 220 volts, and the movement of the leaf was read by a microscope and a micrometer eye-piece, so that the large scale divisions could be subdivided into hundredths. For the means of its standard- isation we are indebted to Professor Rutherford, who kindly sent us a solution con- taining 3:14 x10-° gramme of Ra. Two separate eighths of this, each contain- ing 3-925 x10-1° gramme of Ra, were used as standards. Hither of these treated in the same way as the rock-solutions produced an initial leak of 64 micrometer divisions per minute; hence one micrometer division per minute corresponds to 6:13 x 10-” gramme of Ra. “NATURAL” LEAK. The determination of this is an important matter in an attempt to evaluate with accuracy the amounts of Ra contained in a rock. In our case the electroscope was always charged overnight—the insulation was sufficiently good for 1t to retain a considerable portion of its charge for forty-eight hours—and the amount of the leak determined. In general it decreased from the time of its initial charging the night before to the introduction of the rock-gases. From a value of 1:5 per minute the night before it became about 0-8 micrometer divisions per minute at the time of introduction of the emanation. We consider that this latter number cannot be more than 0-2 from the actual value during the examination of any of the rocks. Such an error would lead to a wrong estimation of the amount of Ra in 20 grammes of rock of 1:23 x10~”, or an error of 0:10 x10~-” per gramme of rock. We therefore consider that for the specimens we have examined the results may be relied upon to about this extent. RESULTS. Bounty Island— No. 1. Granite Pa: Me AY no 2D Se NO Auckland Island— No. 2. Trachyte, Musgrave Peninsula .. J. BAO No. 3. Granite, oF Mt a AO No. 4. Pitchstone, mon, PLO) No. 5. Basic porphyrite, Adams Island. . OS) No. 6. Basalt, a (top) Pe Orsil No. 7. ‘Delete. - Pn Ose No. 8. Diabase, Musgrave Peninsula. a ORES} No. 9. Gabbro, McClure Head me fo) OSS 780 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Radium. Campbell Island rocks— No. 10. Porphyry, Perseverance Harbour 280) No. 11. Trachyte, Filhol Point ay e264 INonm2. 3 Mount Lyell oe oe eas No. 18. Melilite basalt, Mount Lyell .. oe Gil No. 14. Dolerite, Mount Honey ey = 10519 No. 15. Limestone, Garden Cove os on O4G No. 16. Gabbro, North-west Bay Ms . O84 No. 17. Marble, a ae .. 0:28 CONCLUSIONS. Since the rocks examined are derived from islands to the south of New Zealand which are mostly igneous in character, the mdication appears to be that the distribution of Ra in the rocks of this region is in excess of that required to maintain the constancy of the heat of the earth, and the necessity of an examination of rocks from all places and of all age is emphasized. The mean Ra content of these rocks approximates very closely to that found for others by Strutt and by Eve and McIntosh, but differs considerably from Joly’s values. Further investiga- tions are being made on the rocks of these antipodeal regions, and it is hoped to examine the distribution of Ra in the material encountered in piercing the Lyttelton Tunnel, which will give a range of time from the earliest sedimentary rocks to the last flow of the volcano through the wall of whose crater the tunnel is bored. NOTES ON GEOLOGY OF ROCKS. The following notes on the geological properties of these rocks, by Mr. R. Speight, F.G.S8., may be of interest :— The two granites are both biotite-granites of ordinary type and of uncertain age. The other plutonic rocks—the gabbros—though coming from widely sepa- rated localities of the same region, exhibit marked similarities: both are olivine gabbros, but that from Auckland Island is very coarse-grained at times. ‘The co- incidence of their Ra content is very striking. The porphyry from Campbell Island contains large numbers of zircon and anorthoclase phenocrysts. The trachytes are all of highly alkaline type. Those from Auckland Island contain a very high} percentage of soda and a very low percentage of alumina (11 per cent.) and a small quantity of free quartz: they are decidedly acid. Anortho- clase is a common constituent, as well as an alkaline hornblende. The pitchstone belongs to the same group, but is somewhat more basic in composition, and con- tains abundant aegerine-augite microlites. The Campbell Island trachytes, accord- ing to Dr. Marshall, are slightly more alkaline, and contain less iron. The rock from Filhol Point contains some zirconium minerals. They are all augite-trachyte with anorthoclase and occasional riebeckite and cossyrite. These rocks, together with the melilite basalt, all appear to have similar amounts of Ra present. The other rocks from Auckland Island are basic in character, and include flows and dykes. They are generally marked by a high percentage of titaniferous magnetite. No. 8 is a diabase of greater age than the rest, and No. 5 a basic Radium.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 781 porphyrite approaching an augite-camptonite in character and composition, found penetrating the more recent basalts. The age of these basalts is probably from middle to late Tertiary. There is no marked difference between the earlier and later basaltic flows, but the one from sea-level at Adams Island is fine-grained and contains hornblende, while the other is very coarse-grained, with large phenocrysts of olivine and augite. The Campbell Island limestone is of Tertiary age and markedly foraminiferal. The marble is derived from it as a result of basic intrusion. A general conspectus of these results shows that the Ra content corresponds roughly with the basicity and not with the age of the rocks. ARTICLE XXXIT.—MAGNETIC OBSERVATIONS MADE DURING THE EXPEDITION TO THE SUBANTARCTIC ISLANDS OF NEW ZEALAND. By H. F. Sxery, B.Sc., Magnetic Observatory, Christchurch, and C. CoLERIDGE Farr, D.Sc., Canterbury College. THE magnetic observers accompanying the expedition were enabled to make a series of observations at thirteen stations on the islands visited, distributed as follows: Stewart Island, one station; Snares, one station; Auckland Islands, eight stations ; and Campbell Island, three stations. With the exception of two stations, the observations were complete, and included observations of magnetic declination, magnetic horizontal force, and magnetic inclination or dip. At Station B, Campbell Island, and at Enderby Island, the prevalence of cloud prevented observation of the sun, and no declinations were observed. The magnetic bearings of permanent marks were, however, observed at these stations, and a future occasion may make these valuable. The observations at Stations I to V, Auckland Island, were made by Dr. C. C. Farr and Mr. H. D. Cook, M.Sc., of Canterbury College. The observations at Camp- bell Island were made by Mr. H. F. Skey, B.Sc., of the Christchurch Magnetic Observatory, and Mr. EK. Kidson, M.Sce., of Canterbury College, with the very willing assistance of Mr. C. A’C. Opie. At the other stations the two parties combined to make the observations in the shortest possible time. The instruments used were a magnetometer (No. 1) by the Cambridge Scientific Instrument Company, and dip circle (147) by the same makers. These belonged to the Observatory. A second magnetometer, by Jones (also Kew pattern), and a dip circle were on loan to the Observatory by the committee of the British National Physical Laboratory. For observations of time and latitude a 10in. reflecting- circle by Troughton and Sims was used—a very efficient and accurate instrument. On Campbell Island time-observations were made by an excellent 5 in. transit theodolite belonging to the Survey Department, which was also used to throw bearings from station to station. All but two of the stations were permanently marked by a brass pipe 14 in. internal diameter and 18 in. long, with foot-plate 6m. square; and at most of them an iron pipe of similar dimensions was placed at a sufficient distance to serve as a permanent referring-mark of known bearing for future use. On Mount Honey, Campbell Island, an iron pipe was also placed, and the bearings from it of various landmarks and the two magnetic stations A and C definitely determined. Both the Auckland Islands and Campbell Island were found extremely difficult to travel over, the former on account of the dense growth and the mountainous character of the country, and the latter on account of the soft peaty surface. This accounts for the observations being made on the sea-coast, with the exception of one Magnetic Observations.| _SUBANTARCTIC ISLANDS OF NEW ZEALAND. 783 station in the north of Campbell Island. In this last case the porterage of the instru- ments a distance of nearly five miles and back was the work of nine men. At all stations sufficient data were collected and sketches made to insure the subsequent refinding of the stations, in some cases supplemented by photographs. It would be of no use to publish these details, since any subsequent observations 169°10 Longitude East of Greenwich | Oo Courrejolles Paint © Bull Rock §2°30' South Latitude Ras Se xe Fi eae et Lips eHOS Mille). <= Se North-east Harb. | NI ‘pss Ml. Fizeau Dent | & Mt. Lyell North-west Bay Mist yun, Smoolth-water Bay 2 cay Transit of Venus Observation Point melas at Perseverance Harbour oO ce Narielessiistand TSS DP ; PS B) = id So South Point whee mea Puiseux Pk Sketch Map of CAMPBELL ISLAND. ———— es Scale of Nautical Miles. Based on Admiralty Chart. will almost certainly be undertaken from Christchurch Observatory ; and, besides, full details will be given in the published account of the Magnetic Survey of New Zealand. The Observatory, however, will be very glad to furnish the information to any one who may desire in the future to reoccupy these stations. Cs Ss 3S S iS} S) S o w sS S & S$ v S S 8 a OF NEW ZEALAND. SUBANTARCTIC ISLANDS 784 *A[UO T “ON 9[P99N (a) *U9ZT WO ATUO SuoT4eUTTDeq (p) “ATWO Z “ON 21P29ON (0) *A[WO SUOTZRIQTA (q) *SUOIJZOOHOP IOF | pure ‘suoyVvIqiA Ioy spurys A (z) | (9 WW odid ssevitq) (ArojeAtesqgQ snueA fo (ep pue be JISULIT, YOUSIT JO 9418S) 9A0D “1 H) &% “AON TP 91 OF FP_ST | LPL 6-€F FL | A8E LI 0} FS LT a IG891-0 jf LT 9 9G 9T | 6-2S 02 |-6 G69L | LEE ZG | snueA ‘spurs, jeqdurep ‘pH | ‘(dip) gz “AON (WwW edid =| (drp) 6T “Aon GO 81 92 LI SSVIq) Je[Uy sduUeIEADSIEg ‘(‘.9p pue I€ &1 9% 6T GI | LPT 9-G¢ PL | A€L FL 0} 6E EL ie LGLOT-0 |9€ GI °F 8G GI | G-6E GI | L:8 69T | G-E Ze | Jo proy ‘spurisy Tfeqdmrp ‘y | ‘| H) SI “AON (OT W edid sseaq) ohvwy G@8r 9L 0} IL 9T junoyy fo sedojs uleqsea (WH) 12 “AON 8G GT 09 FE GI | LPT |()%-1€ GL |A6 91 9 92 ST] TON | ZTE9T-0 c i 9-GI 691 | 8:66 GG | UO ‘spurysT tee ‘q | (dtp) 06 “AON &W 9p GI 9% GT GI O€ 91 | LF 81 adid ssvaq) dnory puryst by GI 9 6 FT LTS PL | A9G OL 99 TF OL fe ISTLT-O j6f OL % G OL /9-€ 81/20 991 | 8-1 0G | pueppony “purisy suepy TIT : Giles GE GL | F-9F LT OF II | 6-€F LT (FW edid sseaq) O€ OL | &-GF LT Sspue[s— purppny <“inoq ST 91 99 OF FI "S ¥:99 PL OS €1 °F O&€ €1 “ GLS9T-0 (JST SL | F-€S LT |G-0 99T | F-0¢ 0G | -zeFT AoqureD ‘erog dup “7 vo LT (LW edid sseiq) spurt VG 91 9 F YT | 6-66 YT “SJ pueppony “moqrey Asy . d@6P It °F 6I II ¢ II | &-T@ 91 “ule ‘eAo) durey Jo yy10u GG ST 9% OL FI 7 GEE FL | ATH OL 94 9 OT is OOTLT-O \61 OL °F TL OL | F-0G 91 | 2-0 991 | G-6F OG | F[tu oUO 4oTUT [Tes uO ‘TT SLs Go LT | €-9¢ L (6 W edid ssvaq) spurl a9 €&1 °F 9€ GI cE IL | 6-6F L “sf purppny “moqrey Aer GY OL °F FE FI ES 6-LE FL | AGE IL %F OT IL GEOLT-‘0 |S OT 99 OF OL | G-9F L | 8-0 99T | S-8F OG | -ureD “moqaieAy eae “AI 1% ~ TW LE ST | G-OL 0@ adid ssviq) spurj[sy purl G STI °F O€ FI ag€é OL 9 0 OT Gé 6 | G-LS 61 3pny ‘moqrey Aayurep OG €1 99 GE IL | PI GIG PL |AGSE6 1 066 | OA VN) LSTILI-O |G 8 0€ 8 |1-:0 02] TIT 991 | LFF OG | “Wy YRION Jo pro ‘A & G@s8éL % LL (¢W edid sseaq) spur[sy LT L 0% €1 ¢ LYL Poy SL |AG L 4 sE 9 ee POGLI-0 |6h G 91 6E GS | LE 8L\F8 991 | SFP OG | purppony “seruT Sener A LG SIN are ssviq) spurs, purppony GI OL °F OF 6 LoL \(o)8-01 FL | (a) IT 04 8F OF is 88EZ1-0 |0G OL F LI OL | 6-9 81 | 9-ZI 99T | G-ZE OG | ‘Ssoy 410g Se ay IA 66 GL IN 9ct OE 91 99 9G ST sseiq) dnorg pursy purl Gy ST 99 GE FL | LPT LG PL | AIP ST 4 66 ST 3 OGLLT-0 Be a 91 991} T§ 0¢ | ony ‘purjsy Aqropug “TTA Le APE 1s 4 GZ 1Z (sed OL If 99 Tf OL | LFT €1 GL | A OG 0G 93 O€ 0G = SZI61-0 |IT 8 99 8 |G-1E LI | 8-SE 991 | 4:0 8F | Aoams a0A0) purjsy sereug Gas Gg LI 04 gg OF (yoveq Apues wo) ey ZT 94 16 OL | LPL GST TL | AGZ 91 94 GI 9T| LT ON | 9L661-0 |Z OT 99 LG GIT |S-8 81) Z-eh L91|9-6 Lh | puUrlsy qremojzg ‘snseseg 410d FI “AON “UL “AL “Ul “AL 4 6 “Ul “H “UL “AL fro) 0) TW A “td ° AL 4 }, y a 4 6 “LO6L ‘ESZN oun | NOC) -3 aa @) TSZN OMT | go eats | patomon | ISZN ‘oun ae | ween | Sen WOTHRYS oye ‘SaNVIST NYWHLAOG AHL OL NOIWIGGaxy s MLALILSN] TIVOIHdOSOTIHG AYNAUAINVO FHL ONINNG NAV, SNOILVAYHSAO DILHANOVIN Magnetic Observations.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 785 Enderby Island E. of Greenwich 50°30’South Latitude a North-east Cape Column is orth-easlt Cape North- wanes Se Sy Sout / Z VA f : RE ie Ewing |. ° “Yule I. Crozier Pl. 9 Green I. 3 Dundas |. Sv Eden Kekeno Pt. o Disappointment |. Blanche Rk. Chambres Inlet Granger Inlet Griffith Inlet Beehive Rks. 3 S Deep Bight Wat ment A yt Mi “eS Ine = ny 04), Dre ete aie WY mm Hs «i YY ‘ NBG Flat-topp McLennan Inlet Iravantle ge ee % tn Waterfall Inlet Sass! «Ve Sv SAN C. Bennell 50°50’ South Latitude Sketch Map of the AUCKLAND ISLANDS. Based on Admiralty Chart. SE aEEEIIETInEIIEEEEEE Ee 4 6 ——_ Scale of Nautical Miles. 786 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Magnetic Observations. NOTES. At Enderby Island and Mount Faye it was impossible to determine the azimuth, as the sky was obscured. These declinations could, however, be calculated at a future date from the observations of the magnetic axis and reference marks (perma- nent) which were taken. At the stations where only one needle was used to determine the dip a correction has been applied. The horizontal intensity as determined by N.A.B.C. magnetometer requires 2 correction of —0-00013 C.G.S. to be applied to reduce to observatory standard (magnetometer No. 1). Dips observed with “ field circle ” require a correction of + 08 to bring them to the observatory standard (circle 147 ; three needles). New Zealand standard time is 11 hours 30 minutes fast of Greenwich mean time. The times as given are counted continuously from midnight (0h.) to midnight (24h.). The latitudes at Stations I and V were obtained from circum-meridian observations. At Stations II and IV the latitude is from a meridian altitude. The other latitudes are scaled off from Admiralty charts, as also are the longitudes, the longitude as determined at Station I being taken as a standard for stations im Carnley Harbour, and that given for the Transit of Venus station in Terror Cove for the stations about Port Ross. The dips observed with circle 147 (two needles), as given above, require a correc- tion of —0’-5 to bring them to the observatory standard (circle 147; three needles). The results of an accurate comparison of the Christchurch Observatory standards with the adopted international magnetic standards has recently been published by the Department of Terrestrial Magnetism, Carnegie Institute, Washington. With regard to the character of the rock in the immediate neighbourhood of the stations, it will be useful to note the following :— Station I.—On a sandy beach; granite rocks in the neighbourhood. Station IJ.—On granite rocks. Station III.—On a beach of shelly sand; basaltic rocks not far distant. Station IV.—On a beach of basaltic boulders. Stations XI, XII.—The rocks chiefly limestone, with dykes of basalt. Station XIII.—Rock lavas of intermediate character. The nature of the rocks at other stations may be gathered from the report of the geologists. It is significant that the extremely low value of between 7° and 8° K. for the magnetic declination at Station V, Auckland Island, accompanies the peculiar geological nature of the rock as seen in the geological map. We are indebted to the kindness of Dr. Chree for bringing to our notice the following magnetic observations made by the “ Erebus” observers in 1840-41. They are taken from General Sabine’s “ Contributions to Terrestrial Magnetism, No. XI” (Phil. Trans. for 1868, pp. 371, 388, 397, 408). Between November, 1840, and April, 1841. On Shore at Lat. S. Long. E. aa i a ee Auckland Island .. a gh 50 32 166 12 8 17 44 Campbell Island .. Poe sf 52 33 169 9 9) 17 40 Magnetic Observations.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 787 Between September, 1840, and April, 1841. 7 : i Total Force. On Shore at | Lat. S. Long. E. Berean aa == = ce ers Units. | C.G.S Unit. . oe a [ee Bee ats | fo} 4 | io} é 1 Auckland Island .. eee || 50 33 | 166 19 20 13:79 0:63577 Campbell Island .. sisal Oeuodee ih lOO 10 10 14-08 0-649 14 Between 13th November, 1840, and April, 1841. | Number of Ob- Magnetic Deduced Value of On Shore at Lat. S. Long. E. servations. Dip S. H., 0.G.S. Unit. a ot Oe : Bunt eect a ¢ Auckland Island .. ae 50033 166 19 3 73 24 0-18586 Campbell Island .. rout 52 34 169 10 8 TA 20 0-18617 Average Annual Change of Declination over the Sixty-seven Years, 1840-1907. At Terror Cove, Auckland Island .. a L$ 0°34 At Head of Perseverance Inlet, Station C, Campbell Islend) oe eB Mean a “8 ie an 2. G0 Average Annual Change of Horizontal Force and Dip over the Sixty-seven Years, 1840-1907. H. Dip. - Entrance to Port Ross, Auckland Island .. —12-9y + 07-58 At Head of Perseverance Inlet, Campbell Island .. — 26-8y + 0/:36 Means of , .. — 19-8y + 0°47 At the entrance to Port Ross the 1840-41 observations appear, from the position assigned, to have been made near the south side of the entrance, some distance south of the 1907 station. The observations at Terror Cove and at Station C, Campbell Island, appear to have been made almost, if not quite, upon the spots at which the “ Erebus ” observers worked. In New Zealand at the present time the annual rate of change of declination is + 3’:1; of horizontal force —20y; and of dip + 179. At Station C, Campbell Island, the observations show that the average change per annum in the total magnetic force over the sixty-seven years has been —15y, whereas at Christchurch, New Zealand, the average annual change at present over SIX years 1s + 6y. Comparing Station VII with the station at the entrance to Port Ross of the “ Hrebus ” observers gives an average increase of total force per annum of + 13y. Accurate determinations of latitude were made at Stations V and I— Station V, eight circum-meridian altitudes of the sun; latitude deduced, 50” 44° 437 S. Station I, seven circum-meridian altitudes of the sun; latitude deduced, 50° 50’ 26” S. The longitude determined at the depot, Camp Cove, Carnley Harbour, was 166° 0’ 42” EK. 788 SUBANTAROTIC ISLANDS OF NEW ZEALAND. [Magnetic Observations. There have been no less than eight vessels wrecked on the Auckland Islands, and the present charts of those islands are by no means accurate in details. It would seem, therefore, that in view of the recent horror of the wreck of the “ Dun- donald,” with considerable loss of life and great hardship to the survivors, an exact survey of the islands should be made, in order to correct existing accuracies in the chart. When the Government of New Zealand, some years ago, instituted a survey of the Snares Islands, with a view to the selection of a site for a possible lighthouse, it was found that the resulting plan of the islands in no wise resembled that shown on the old Admiralty chart. The latter has, of course, since been amended, but we are able to state that a careful survey of the Auckland Islands would reveal similar discrepancies. ARTICLE XXXIV—THE METEOROLOGY OF CAMPBELL ISLAND. By Professor P. Marsuaiz, M.A., D.Sc., F.G.S., School of Mines, Otago University. THE arrival of the expedition on the 17th November, 1907, was marked by a dark, lowering sky, with a fresh breeze from the north-north-west. Mist occasionally shrouded the tops of the hills, and a few drops of rain were falling. Generally, meteorological conditions were quite similar to those close to the eastern side of the Southern Alps on the approach of a cyclone. During the night the wind changed to south-west, and blew a moderate gale. The next day was cold, and punctuated by showers of rain, hail, sleet, and snow, which continued into the night. But little snow lay on the ground, and this soon melted on the next day, which broke fine; but the wind soon backed into the west, fog covered the hills, and rain fell. A change to the south-west brought a fine day and bright sky on the 19th November. The wind dropped, and after- wards sprang up from the north-east and veered to the north-west, where it remained, blowing lightly till our departure, at which time fog again covered the hill-tops. These weather-changes are exactly those which are experienced in the south of New Zealand during the passage of a small double-centred depression. Until the publication of the “New Zealand Year-book” of 1907 the only in- formation obtainable in regard to meteorological conditions in Campbell Island were to be found in the charts issued by the Admiralty, the “ New Zealand Pilot,” the “ Deutsche Seewarte,” and similar publications. In 1904 some meteoro- logical instruments were installed, and the observations made in 1906 by Mr. Nicholson, the manager of Captain Tucker’s sheep-stations, were recorded in the Year-book for 1907. In addition to these, I have been able to copy the records for 1905. The results, which are summarised in tabular form, give information of a fairly accurate nature as to the most important climatic features. In connection with these records, the following observations may be made, as showing the relations between the climate of the island and that of the south of New Zealand, and the advance that these data have made upon our previous know- ledge of the meteorology of the island. WINDS. The tables show clearly that the prevailing winds blow from the quarter between north and west. In 1905 these winds blew for 248 days; in 1906, for 278 days ; and for seven months of 1907, for 164 days. 790 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Meteorology. In the second edition of the meteorological charts of the Southern Ocean a com- plete graphical statement is made as to the prevalence and force of the winds in the different months, from observations which have been made by passing vessels. For April, July, and September no records are available; but for the remaining months, especially from January to September, with the exception of June, north-westerly winds are given as prevalent. In June south-west winds are given as most frequent, and in October north-easterly. Of 368 observations referred to, only eight were made in calm weather. The gales mentioned in the charts are mostly from the north and north-west, whereas the records made by Mr. Nicholson for seven months of 1907, for example, give five from south-west, two from north-east, and one each from north, south, east, west, and north-west. Analysis of the records does not show that any months differ noticeably from others in the direction or force of winds. In all months westerly to northerly winds prevail, and gales may blow at any time. In the “ Deutsche Seewarte Stiller Ocean Atlas” the January and February winds are represented as west-north-west, blowing with a velocity of 9-12 metres per second. In July and August the winds are from the same direction, with a velocity of over 12 metres per second. . The “ Meteorological Atlas’ gives records for every month except April, July, and September. In every other month gales are recorded almost always between north-west and north-east. Winds are recorded from all directions in nearly every month, but from south to east they are not usual. The charts give an analysis of 368 observations. If these winds are compared with those of Dunedin there is an important differ- ence. In Dunedin, in 1906, for instance, there were 123 days when the wind blew from south-west, and 116 when it blew from north to west. At Campbell Island the corresponding numbers were 26 and 278. The prevalence of south-west winds is maintained throughout the south of New Zealand, and apparently as far south as Auckland Islands, so far as can be judged by Musgrave’s statements. BAROMETER. The only records have been taken with a barograph, and considerable in- accuracies result when such graphic records are transferred to figures. Still, the main result—an average of 29°58 in.—agrees closely with that given in “ Berghau’s Atlas,’’? 29°53 in. The highest monthly average in the seven months of 1907 was 29:74 in., in August; and the lowest, 29-44in., in September. The highest barometer-reading was 30-2in., and the lowest 28-6in., a range of 1-6in. The records for seven months only of 1907 are available. The movements of the barometer appear to have much the same relations to the weather as in New Zealand. TEMPERATURE. The most striking feature is, as would be expected, the small diurnal and annual variation. In 1905 the extreme range was 47° F., and in 1906 it was 38° F. In =: va ee _ Meteorology. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. TON 1906 the mean daily range varied from. 14-4° Ff. m January to 7:8° F. in April, having an average for the year of 9:7° F. The mean annual temperature ap- pears almost constant—44°5° F. for 1905, and 44° F. for 1906. The means for the months in 1905 vary only from 51° F. in February to 39° F. in June, and in 1906 from 49° F. in February to 39° F. in June. These mean temperatures are, of course, mainly dependent upon the temperature of the ocean, which in “ Berghau’s Atlas” varies from 48° F. to 42° F. in different months, with an average of 44°8° F. The meteorological charts of the Southern Ocean show surface temperatures ranging from 47°5° F. in January and February to 43° F. in June and July, and the average 1s 45° F. It appears, therefore, that the atmosphere observations recorded in such a sheltered spot as Tucker Cove give a slightly lower average than those obtained from surface temperature, but the difference is very slight. In 1906 the average temperature in Dunedin was 49:4° F., and the sea-temperature in “ Berghau’s Atlas” is 52:9° F., and an identical result is obtained from the “ Meteorological Atlas.” In the German and English charts, as well as in “ Berghau’s Atlas,” the oceanic current 1s given as flowing from the south-west throughout the year. The maximum set in the neighbourhood of the island is 25 nautical miles in twenty-four hours. The current apparently brings icebergs past the island occasionally, for in the months January, March, May, September, October, November, and December much ice has been seen to the north-east, especially in the later months of the years 1892 and 1897. The map of isanomalies in “ Berghau’s Atlas” shows that the temperature of Campbell Island is distinctly lower than that proper to the latitude. WEHATHER. Sunshine does not prevail for many hours in the year. In 1905, 99 days were fogey and 168 were cloudy, while only 98 were clear, and these results agree with those of 1907 and with those stated in the “ Meteorological Atlas.” The atmosphere is apparently nearly at saturation-point. This is what would be anticipated, seeing that the prevailmg winds have blown across thousands of miles of ocean-surface, and are constantly entermg colder latitudes. The rainfall also is rather heavy, though higher land would probably increase it considerably. In south New Zealand the record is always higher on the western side or in the immediate vicinity of the mountains on the eastern slopes. The number of days on which rain falls is exceedingly high—about three out of four, on the average. No continuous series of observations have yet been taken at any point in the Auckland Islands. Those taken by Musgrave cover a long period, but they are of a general nature, and usually only record marked weather-charac- teristics. The Snares Islands climate is still less known from continuous observations. When the expedition visited them (15th November) the peat was quite dry on the suriace, and it was evident that rain had not fallen for some days. A fall, however, began before the expedition left. 792 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Meteorology. TABLE SHOWING SOME CLIMATIC FEATURES OF CAMPBELL ISLAND. ne Rain. Direction of Wind (Days). Weather (Days). | Month. 3 | | ; E : = |e lee \ereeann er eal eS elal s | &isz Bal ie ee eee ys | PIS e Iie ie bale Sle le je) es 1905. In. January aa er | Dil 43) 0), 20) a OF SB ZOO) MOP By | a, GCI February .. ea | 58) 40) oll ii ESS Ss Ol Ol 4) 9) 16 eae eae March “a ge) DO Ae TO 2B BS 7 Sh ©) Op Or a I ae, SD Apuil c: Be eel ks a OS OW) GAM ef aa AG |S May aa Sar (<3) (e374 (co) oD 0] GO) US TL) ks fl Ks J) June Me .. | 42) 36} 39) 20) 4:8 7 1) OF LT] OF 4 5) 12) 4) 11) 13) 6 July — oo | 28 BS 0 Beal | SE SO, 2) ll WSO g iA 8 August rs so | ek eis) © CNN PA bs OP SB Ol Oh I 2 ay MG gl ep ub September .. -, | 44)°38) 40) WO) 4e7y 2) 3) 5 Ol 2) OF 2h 6) 6) ae October mt a2 | HOP ae, ey IO OT Bl Ol BS) A I Sy Oh TO, BD) November .. So |) a) AKO aus NN ES SI DE Ol © aI) 1| 8} 14) 8 December .. .. 7 04) 44, 49) 18) 5-0) 3] -4) 1] 0} 0} 3) 5) 15 0 2 ee a Average a: Beene eli: 0) fees: ao enem Wate rci ta.a|| coed nyaees | 5 tol) aaall ite Moros Nose Total Se eel Sel eer a) 2OOND S28) Ro Ole 26) feito |G) e528 | ae 28) 99, 168) 98 — a — = ) = 1906. | January .. oo} On| a ae} ILL ©) ©} O} BS] Ul BP OB. Helnuiaiyammeree se || OL ey AS ae ap OTE, TM Tab ae) ali March a oe 49 A AD 26) 5 95:9) OS re Sai ear Ole April Pe oo |) eS ep A Ge hy Sy ©) Gl a A oO Gls May oe oe | Sa) eS A ae SO) LT June ae ee 2436) 4.0) 22.5) AS 5) eh SO 4 Oi | SNGi ae July ae 5a 2 eo) BS AO GHG) Sy BT OP SE A BN August rae .. | 43] 35) 39) 24) 4-1) 12) OF 1} 0} OF 2) G6 10} September .. Be GSS) Az ON 3:2) O20) 0) ets eG) October te os Aa) ae eh a) BS) a) Oy, a) Op hh November .. oe | Ol ci) aly) 2 eet 13) OO; O| S i sp se December .. So || OO) Ge Ae Sl Gr), I) ©! Be Oy © O14. lo Average Bs .. | 49, 39) 44 ssi ne Total 55 fool cal co | oO) Dow) Ju) WG) 24 BO) AS) Wai, te UXO | | March a oe PCO Ze) will Ga) eg) 18) ©} 4) O| G| 1) Bi WO 2) 1G) 6) 9, 29-52 April 4 a | EA S| BO Ol Or Ay | i i) e) Gs 29).62 May | 46) 35) 40) 23)" 5:7) 6) 2) 3 2) 0) By 2) Bly 9) 12) 02559 June x | ars) 20) la) ley, Gy Bl ©) ©; OF BS) OB) 2p Me) sul) 7) 28-60 July “4 Se) A Ol A 20 6-4) 5G OO el a7) ee) SS eo 2S oM August bs .. | 45] 35] 40/23) 3:9 2) 70) VO} 20) 4) 2) 50) 4 eal SS 9eiee September .. Se LOS ZO Cen Ge a UW BW Ble Bl Bl OY) Tey Saved October Ne .. | 48) 36) 42) 25) 4-1) 12) OF J) Ly 1 9 4) 63) 50) 18) 18) 5) 29-45 ARTICLE XXXV.—THE BIOLOGICAL RELATIONS OF THE SUB- ANTARCTIC ISLANDS OF NEW ZEALAND; SUMMARY OF RESULTS.* By Cuarues Cuitton, M.A., D.Sc., F.L.8., Professor of Biology, Canterbury College, University of New Zealand, Editor. In the foregoing pages the different contributors have given as full an account of the subjects treated by them as the material and time at their command allowed, and the result 1s a very considerable addition to our knowledge of these islands and of their fauna and flora, and also some advance towards the solution of one or two problems of wider interest connected with them. Before discussing the latter, however, it will be desirable to indicate the sub- jects that have not yet been satisfactorily investigated. The primary object of the expedition was the extension of the magnetic survey of New Zealand to the outlying islands, and this has been accomplished so far as the Snares, Auckland Islands, and Campbell Island are concerned ; but no observa- tions have yet been taken on the Antipodes, Bounty, and Macquarie Islands. The Antipodes could usually be reached during any of the periodical visits of the Govern- ment steamer, and so could the Bounties if the weather be sufficiently favourable to allow of the landing upon them. There is no regular visit of a suitable steamer to Macquarie Island, and special arrangements would therefore have to be made for extending the magnetic survey to that island. It is very desirable that this should be done, and that the fauna and flora of the island should at the same time be more completely examined, since, owing to its position between New Zealand and the Antarctic Continent, its biology is of especial interest. In the report on the magnetic observations made during the expedition it is pointed out (see p. 788) that the present charts of the Auckland Islands are by no means accurate in details, and it 1s suggested that this may possibly have some con- nection with the large number of wrecks that have occurred on these islands. In view of these facts it 1s highly desirable, both in the interest of science and of humanity, that a careful re-survey of the Auckland Islands should be made, in order to correct any imaccuracies in the Admiralty charts that may exist, similar to those that were found some years ago in the case of the Snares Islands. The main outlines of the geology of the Snares, Auckland Islands, and Camp- bell Island are now known, and something also of the Antipodes Islands and Bounties ; but there is no report on Macquarie Island by a geologist, and naturally much still remains to be done both in Campbell Island and in Auckland Island, particularly * T am greatly indebted to Professor W. B. Benham, Dr. C. C. Farr, Dr. L. Cockayne, Mr. H. Suter, Mr. R. Speight, and Mr. E. R. Waite for information on special points and for much general assistance 1 in the preparation of this summary.—C. C. 194 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Summary of Results. in the northern end of the latter. The surface of the Auckland Islands is so thickly covered with vegetation, and the difficulties of locomotion are so great, that the time of the geologist cannot be used to anything like the full advantage. In connection with the discussion of the geological and biological questions, 1t will be desirable to have a much more complete series of soundings between New Zealand and the various islands, and particularly between Auckland and Campbell Islands and Macquarie Island, and between the latter and the Antarctic Continent. In response to a request made through the Premier, Sir J. G. Ward, by the Philo- sophical Institute of Canterbury, the Vice-Admiral and Commander-in-Chief of the Australian Station has given instructions that His Majesty’s ships visiting the southern islands are to take soundings between New Zealand and the Auckland and Campbell Islands on every opportunity. The rich results, especially in the Foraminifera, of the few dredgings that were taken during the expedition show how greatly our knowledge would be increased by further and more systematic dredging in the New Zealand region. Much also still remains to be done in the collection of the littoral marine fauna and flora, for during the visit of the expedition attention was primarily devoted to the terrestrial forms, and the marine animals and plants were only more or less incidentally col- lected as opportunity offered. Several of the marine groups—such as the Polyzoa, the Hydrozoa, the Currupedia, and the Tunicata—have not been collected at all, and even in the others that are dealt with somewhat fully in this work much still remains to be done. The surface of the islands, and of the Auckland Group particularly, is so diversified, and includes so many different islands, that it is not to be expected that more than a small proportion of the commonest of the smaller animals was obtained during the expedition. The visit was, moreover, made too early in the season to allow of a satisfactory collection of the insects bemg made. No doubt further re- search will bring to hght many other forms, especially among the insects, the spiders, myriapods, and probably among the land-shells. Owing partly to unfavourable weather and to difficulties of collecting, the fresh-water streams have been only very superficially examined, and a more complete biological survey of the lakelets in Auckland Island and of the whole of the fresh waters of the islands is very much to be desired—e.g., the Hntomostraca have not been collected at all, but would pro- bably yield important results. . A few Myriapoda and fresh-water pianarians were collected, but time has not allowed of a report being prepared upon them, and the reports on the land planarians and nemerteans, and on the Hvrudimea are only preliminary. The flowering-plants and ferns have been pretty thoroughly collected more than once by competent botanists, and probably most of the species on these islands are now known; but much more requires to be done regarding their local distribution and ecology. In several cases uncertainties still exist with regard to the species, due partly to the tact that, owing to the time of the year at which the visit had to be made, many of the plants were not in full flower. On the other hand, the cryp- togamic plants—mosses, liverworts, and fungi—which appear to be extremely varied and abundant, are only imperfectly known. The fresh-water Algae have as yet received no attention, and the marine Algae have been collected at only a small number of stations. Summary of Results.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 795 In the biological relations of these islands there are two points to be considered —viz., (1) the connection of the islands with New Zealand, and (2) their connec- tion with other subantarctic lands and with the Antarctic Continent. The facts bearing on the question of the connection of the islands with New Zealand appear to be as follows :— In the first place, the sea between them and New Zealand is probably not more than 1,000 fathoms deep; between New Zealand and the Auckland Islands it is doubtless very much less—a sounding twenty miles north of Auckland Island gave 95 fathoms; and the 1,000-fathom line extends far to the east and south of New Zealand, and, so far as is known, includes Chatham Island, Bounty Islands, Antipodes, and Campbell Island, as well as the Aucklands. There is probably a greater depth south of Campbell and Auckland Islands and between them and Macquarie, but, unfortunately, definite evidence as to the depth of the sea in this locality does not appear to be obtainable. From the appearance of the islands it is quite evident that they have been of sreater extent than they are at present, and that, as at Auckland Island for example, the west coast is gradually being worn away by the action of the strong westerly waves. The reefs lying to the north and east of Auckland Island also pot to a greater extension of the island. From other physiographical considerations it appears that Auckland Island, at any rate, was in recent times higher than at pre- sent; and the same thing is probably true of Campbell Island. A greater eleva- tion of the land would, to some extent, account for the signs of recent glaciation that were observed. The occurrence of fragments of granites, gneisses, contorted schists, and gab- bros in a conglomerate at Carnley Harbour, in Auckland Islands, indicates the pre- sence of a continental area in that neighbourhood, and the large masses of plutonic rocks at the Auckland Islands, Snares, Bounty Islands, and Campbell Island point to the fact that all these islands probably formed part of a continental area connecting them with New Zealand. In Campbell Island a number of fossils have been found in limestone beds. The species are for the most part the same as some found in similar beds in New Zealand which have been classified by Hutton as Oligocene, by Hector as Cretaceo-tertiary, and by Park as Miocene. In his report (p. 703) Professor Marshall is inclined to think that the Miocene is more likely to be the Kuropean equivalent of these beds, though exact correlation is impossible in the present state of our knowledge. These fossils indicate shallow water when they were deposited, and consequently the contemporaneous existence of land-areas in the neighbourhood. Ii we turn to the biological evidence we find that the connection between New Zealand and its subantarctic islands is referred to by most of the contributors to this work. Mr. Suter poimts out (p. 54) that “a former land connection between New Zealand and the subantarctic islands seems to be proved by the occurrence in the islands of Allodiscus planulatus, Flammulina phlogophora, Endodonta anguiculus, Sphaerium novae-zealandiae, and also of the genus Athoracophorus.” Professor Benham (p. 254) speaks of the existence of the genera Rhododrilus and Plagiochaeta in New Zealand and also in the Auckland and Campbell Islands, and of the occurrence in the Auckland Islands of Leptodrilus, an endemic genus which is closely related to the purely New Zealand genus Rhododrilus, and says, “The occurrence, then, of 796 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Summary of Results. these three genera evidently indicates a former land continuity between the main- land and these subantarctic islands.” I have pointed out (p. 602) that two fresh- water species of crustaceans (Idotea lacustris and Chiltonia mihiwaka) which occur in New Zealand are found in Campbell Island, and the latter also in Auckland Island, where the former probably also exists, though it has not yet been recorded from that locality; also that there are three species of terrestrial Isopods (Trichoniscus thomson, Oniscus punctatus, and Cubaris rugulosus) common to New Zealand and the Auckland Islands; that the last one also extends as far as Campbell Island; and that the evidence thus afforded for former land connection is strengthened by the distribution of the land and shore hoppers, and of the genera Deto and Scyphoniscus, which have closely allied species in New Zealand and in the islands. Among the insects there are many wingless forms and forms with reduced wings, as might naturally be expected on islands so much exposed to violent gales. Wing- less forms are described among the Coleoptera, the Diptera, and the Hymenoptera. Some of the birds, too, such as the snipe, are poor fliers. In his general remarks on the Coleoptera Major Broun points out (p. 80) that the bulk, ornamentation, and structure of a large proportion of species lead to the belief that their progenitors originated and flourished in some large, 1f not con- tinental, area enjoying a climate very different from the mclement one that now prevails. The large number of genera confined to the islands seems to point to their isolation during a considerable period ; but the existence of five genera, all but one apterous, confined to New Zealand and the islands, and the relationship of a number of others to New Zealand forms, seem to necessitate previous land connection. Several of the other insects are identical with species from New Zealand—e.q., the weta on the Snares, &c.; fresh-water fishes are found both in the islands and in New Zealand; the spiders show relationship to those of New Zealand; land planarians and nemerteans of the same genera as those in New Zealand are present, and one of the Auckland Island land planarians is a variety of a widespread and variable New Zealand species. Further evidence in the same direction could readily be obtained from other groups of animals and from the plants, but the facts thus briefly referred to will be sufficient to prove that these islands are not oceanic islands, but once formed part of a continental area connecting them with the present main- land of New Zealand. With regard to Macquarie Island much less is known, and the evidence is not quite so clear; but, although it is probably separated from Auckland Island by a greater depth of the ocean, there is considerable evidence in favour of its having formed part of this continent; thus, the land-slug Athoracophorus huttom is found at the Snares and also on Macquarie Island, and A. martensi is found on Auckland Islands and on Macquarie Island. Among the earthworms the genus Notiodrilus is represented by several species—two on the Snares, another one widely distributed on the Auckland Island group, a third in Campbell Island, and a fourth in Macquarie Island—and the differences between the Macquarie Island species and some of the others, though apparently constant, seem to be slight. Similarly among the spiders we have the genus Myro represented by one species from Macquarie Island and two others from the Snares; and various plants, such as Stilbocarpa polaris, Cotula plumosa, Pleurophyllum Hookeri, Colobanthus muscoides, Coprosma repens, Poa et ns od Summary of Results.) SUBANTARCTIC ISLANDS OF NEW ZEALAND. 197 foliosa, &c., are common to Macquarie Island and Auckland or Campbell Islands : so that there is considerable evidence for assuming that Macquarie Island once formed part of the New Zealand continent. Further corroborative evidence is derived from the former existence on the island of a land-rail (Hypotaenidia mac- quariensis), and of a parrakeet (Cyanorhampus erythrotis) which is found also on the Antipodes ; the latter, moreover, does not seem able to fly far. The existence of certain plants and animals in New Zealand which are identical with or closely allied to others in South America and islands adjacent thereto was long ago referred to by Hutton (1884 and 1885) and Wallace (1902), and has been more recently dealt with by Benham (1902) and Cockayne (1904), who have shown that, particularly in the islands to the south of New Zealand, there is an element common to other subantarctic lands. Many additional facts bearing on this question will be found in the reports that constitute this work. It will be best to deal first with the terrestrial forms, since these are naturally the most important from the point of view of geographical distribution. As regards the plants, Mr. Cheeseman gives a most complete and exhaustive analysis of the flora of these islands (pp. 453-471), and of the points that it has in common with those of Fuegia and of the Kerguelen -South Georgia region ; and to this he adds a full discussion of the causes that may be considered to account for it. For full details reference should be made to his paper; but it may be stated here that, out of the thirty-seven flowering-plants and ferns in the Kerguelen and South Georgia groups of islands, twenty—z.e., 54:1 per cent.—are found in the sub- antarctic islands of New Zealand, while twenty-seven are found in Fuegia and the Falkland Islands. The total number of Fuegian plants found in the subant- arctic islands of New Zealand is twenty-nine, fourteen of these extending also to the Kerguelen and South Georgia groups of islands. These figures deal only with specific identity ; 1f we consider the genera, we find that out of eighty-eight genera found in the subantarctic islands of New Zealand there are no less than fifty-six with representatives in Fuegia. There appears, then, to a be a large element common to the floras of Fuegia and of the subantarctic islands of New Zealand. Against this, however, Mr. Cheeseman points out many important differences between the flora of Fuegia and that of New Zealand—e.g., the presence in Fuegia of thirteen families not found in New Zealand ; twice that number of families represented in New Zealand but not in Fuegia proper; the differences in the Compositae, &c. (see p- 460).* It is worth noting that eleven species of plants found in the subantarctic islands of New Zealand are also found either in the Tristan da Cunha group in the South Atlantic or in the Amsterdam Island group in the Indian Ocean, the flora of these two groups possessing many points of agreement notwithstanding their wide separation, and showing also undoubted traces of affinity with those of Fuegia and Kerguelen. Two of these eleven species, however, do not occur in Fuegia or the Kerguelen - South Georgia group of islands. * The relations between the floras of Tierra del Fuego and New Zealand were very fully discussed by the late Nicholas Alboff in his ‘“‘ Flore Raisonnée de la Terre de Feu,” written in 1897, though not published till 1902, three years after the death of the author. 798 SUBANTARCTIC ISLANDS OF NEW ZHALAND. — [Summary of Results. However striking these resemblances may appear, it must be remembered that nearly all these plants have seeds adapted for wide dispersal by birds, wind, or ocean-currents, and attention must therefore be directed to the terrestrial animals, which are much less likely to be carried across the ocean by means of this kind. The fresh-water fish found on the Auckland and Campbell Islands belong to the Galaxudae, a family that has representatives in New Zealand, South America, Falk- land Islands, Australia, and South Africa, one species (Galaxias attenuatus) having a very wide range, being found in New Zealand, Tasmania, southern Australia, Falk- land Islands, and the extremity of South America. This species is known to breed in the sea, and it has been supposed that this accounts for its wide distribution ; and as another species (G. bollansz) was originally described as a marine one Dr. G. A. Boulenger (1902) has used this as corroborative evidence, and appears to consider that the Galaxudae can no longer be used as evidence of former land connection. A reference to Mr. Waite’s paper (p. 586) shows that the description of G. bollansi as marine 1s probably erroneous, and, as Mr. Geoffrey Smith (1909a, p. 138) has pointed out, the fact that species of Galaxudae breed in the sea by no means does away with all the value of the group in favour of land connection, or proves that they can readily cross the wide oceans. Among the spiders Mr. Hogg (p. 156) calls attention to the preponderating number of specimens, as well as to the area of distribution of the genera Myro and Rubrius, and points out that this group stands out as quite an antarctic type. Of the genus Myro, one species is found at Macquarie Island, two others at the Snares, a fourth at Kerguelen Island, and another at the Cape of Good Hope; of Rubrius, Mr. Hogg describes three species, one of which (R. falazatus) is said to approach more nearly to a South American species than to the Tasmanian forms of the genus. The species Pacoficana cockayni, from Bounty Island, though referred to a genus of its own, shows affinities to other genera recorded from Cape Horn and from Tasmania. Among the earthworms, Notiodrilus (s.l.) is a genus found in New Zealand and its outlying islands, in Kerguelen, Marion Island, Crozets, South Georgia, Falkland Islands, Patagonia, and the Cape of Good Hope, and is therefore distinctly circum- austral, though some species extend to Mexico, Madagascar, north-west Aus- tralia, Ge. The genus Phreodrilus is also characteristically subantarctic, and the family Haplotaxidae is largely represented in the Southern Hemisphere. The new species of Lumbricillus is closely related to two other species which occur in sub- antarctic regions, though the genus is found all over the Northern Hemisphere. The genera Marionina and EHnchytraeus, though of northern origin, have species in subantarctic regions. The subfamily Megascoleconae is Australian in origin, but has representatives in the North Island of New Zealand, while it is represented in the New Zealand subantarctic islands by four species of Dzporochaeta and one species of Plutellus, a distinctly Australian genus; Professor Benham, however, is inclined to think that this species (P. aucklandicus) may have been introduced into the Auck- land Islands by man. The earthworms thus appear to show close connection with those of other sub- antarctic lands, and it has been supposed by Beddard and Benham that this could only be accounted for by the existence of former land connection, such as might be supplied by an antarctic continent. Dr. Michaelsen (1902, p. 158) has poimted Summary of Results.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. woo out, on the other hand, that some of these earthworms live on the sea-shore, and their cocoons might therefore be carried across the sea from island to island on sea- weed, &c. It does not appear, however, that this could apply to all the earthworms found in these islands, and very weighty objections are raised by Benham against the supposition that it applies even to those living on the sea-shore. Moreover, this means of dispersal is not applicable to many of the other land-animals that have to be mentioned. Of the terrestrial crustaceans found on the New Zealand subantarctic islands, two species belong to a special section of the genus Trichoniscus, which Dr. Budde- Lund (1906) has pointed out is subantarctic in distribution, and one of these appears to be identical with the species found in Tierra del Fuego and on the Falkland Islands, and is very closely related to, if not actually identical with, another one found on the Crozets. Another terrestrial Isopod (Oniscus punctatus) is found also in Tasmania and Victoria, while the fourth (Cubaris rugulosus) belongs to a genus that appears to be Polynesian in distribution. The Isopods already mentioned are true terrestrial forms, and, as the young are hatched out in a pouch below the body of the female, it does not appear hkely that they could readily be carried across wide stretches of sea. The same thing is true of Deto aucklandiae, which, however, is found on or near the sea-shore. This genus is distinctly subantarctic, bemg known only from Auckland Islands, New Zealand and Chatham Islands, South America (Chih), Cape of Good Hope, St. Paul (in the Indian Ocean), and Australia, and the species found on the mainland of New Zealand is very closely allied to the Chilian one, if not identical with it. Another terrestrial Isopod (Scyphoniscus magnus)— found, like the last, on the sea-shore—belongs to a genus which contains only one other species, a closely allied one found in New Zealand. Two fresh-water Am- phipods are known from the islands, one (Chiltonia mihiwaka) being found in New Zealand as well as on the Auckland and Campbell Islands, and belonging to a genus which is represented by several species in Tasmania and Victoria and is very nearly related to the genus Hyalella from South America. The second fresh-water Am- phipod belongs to the genus Atyloides, of which two fresh-water species are known from Victoria; the genus as at present defined includes some marine species also. The fresh-water Isopod ([dotea lacustris) 1s known from New Zealand and Campbell Island, and is identical with a species from the Straits of Magellan. Of the land and terrestrial Mollusca no species is found outside the New Zea- land region. The family Phenacohelicidae is represented in the islands by six genera, of which four are not known beyond the New Zealand region, one being confined to the Auckland Islands. Of the others, Allodiscus is known from New Zealand, Tasmania, and Australia; and Flammulina from New Zealand, Tasmania, Lord Howe Island, Norfolk Island, and Ponapé (Caroline Islands). These land-shells are related to the South American genera Amphidoxa and Stephanoda, and two species of Amphidoxa found in Juan Fernandez are nearly allied to a New Zealand species of Flammulina. The land-shell from Kerguelen Island (Helix hookeri, Reeve) is placed under Stephanoda by Pilsbry, and the New Zealand Phenacohelicidae are allied also to Phasis and Trachycystis from South Africa. The EHndodontidae are represented in the islands by the genus Hndodonta, which is widely distributed in Polynesia, Carolines, Philippines, New Caledonia, Australia, South Africa, and St. Helena, and by some authorities is considered an antarctic genus. 800 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Summary of Results. Of the beetles, forty-six species, contained in twenty-six genera, are now known from the islands. Seven of these genera appear to be cosmopolitan, and the majority of the others are more or less confined to New Zealand. Loxomerus, however, is a purely antarctic form, and points to connection with South America, being the only genus that points directly in this direction, though Pseudhelops and Oopterus appear to be also southern forms. Of the other insects, the distribution does not appear to be sufficiently known to render us much assistance in the present discussion. The genera of the Doptera are, according to Mr. Lamb, very cosmopolitan in their distribution ; the new genus Zaluscodes, formed for a wingless Limnobiid from the Auckland Islands, seems, however, to come very close to Zalusa from Kerguelen. In the Collembola, Professor Carpenter has pointed out that Triacanthella alba, from Campbell Island, belongs to a genus which contains other species from Tierra del Fuego, and is characteristic- ally subantarctic, and suggests a former wide extension of the Antarctic Continent. I have said nothing of the seals, nor of the penguins and other birds found on the islands, because, although they are widely distributed in other subantarctic lands, and mainly confined to southern regions, they can readily mes their way across wide stretches of ocean. The marine forms of life found at these islands are, of course, not so important in questions of geographical distribution as terrestrial and fresh-water species, but many of them are littoral or shallow-water forms, which could not easily pass over any wide stretch of deep ocean, and it is important, therefore, to consider briefly their distribution. Among the crustaceans there are fifty marine species recorded from the shores of the subantarctic islands of New Zealand, and of these about thirty extend either to South America or some of the islands lying near to that continent, some of them reaching as far as Kerguelen, or they are found on the shores of the Antarctic Con- tinent itself. Among these crustaceans there are several which appear to be exclusively confined to the coast, such as Halicarconus planatus, the species of Hyale, Paramoera, Sphaeroma, Las, &c. Halicarcinus planatus has a free-swimming zoaea which might be carried considerable distances by ocean-currents ; but the Isopods and Amphipods all hatch out their young in an incubatory pouch under their own bodies, and 1t does not seem likely that these could readily be distributed from one place to far-distant localities without a more or less continuous sea-shore stretching between them. Of the marine Mollusca, Mr. Suter mentions nine species which are more or less circumaustral in their distribution, and to these there is to be added a tenth from Macquarie Island, which, however, has not yet been found atjjthe other, islands to the north-east ; in addition to this, several of the genera represented appear to be circumaustral—e.g., Plaxiphora, Nacella, Photinula, Laevilitorina, Huthria, Cominella, and Modiolarca. Of the Holothurians only three species were obtained, two of which occur also in the South American fauna, the third one ‘being. confinedijto’ the ‘Auckland Islands, but nearly related to a New Zealand species. Of the four Stellerids'known from the islands two are widely distributed in other subantarctic seas. Of the Polychaeta, out of thirteen species, only two are endemic to New Zealand, eight are found in South America or the Falkland Islands, and two extend to Summary of Results.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 801 Kerguelen, thus confirming the statement made by Ehlers that many of the commonest New Zealand Polychaeta are identical with those of the Magellan Strait, Fuegia, and Chili. The relationship of the crustacean fauna of these islands and of New Zealand to that of Tasmania and Australia seems perhaps a little closer than was previously imagined. The fresh-water genus Chiltonia, which is found in New Zealand and its subantarctic islands, is represented by several species in Tasmania and Victoria ; and the fresh-water Atyloides aucklandicus from Auckland Island belongs to a genus of which two species are found in the fresh waters of Victoria, though, as stated above, the genus also contains marine species, and it is not very clearly marked off from other closely allied genera. The terrestrial Isopod Oniscus punctatus is also found in Tasmania and Australia. In the south of New Zealand occur a shore Isopod, Actoecia euchroa, which is also found in Tasmania, and species of Phreatoicus, a peculiar genus of fresh-water Isopods which has several species in Tasmania and Victoria. Harthworms belonging to the subfamily Megascolicinae, which is so widely distributed in Australia, are also found in these islands, including a species of the characteristically Australian genus Plutellus, though, as mentioned above, this may have been accidentally introduced into the islands by man. It is now well known that several groups of animals are exclusively confined to south temperate and subantarctic seas. The case of the penguins has been referred to by Beddard, Lydekker, Forbes, &c. This group contains numerous species which at the present time are found on the Antarctic Continent, on the various subantarctic islands, in New Zealand, Australia, South America, South Africa, and St. Paul (in the Indian Ocean’. They are thus circumaustral and are confined to southern seas, and have consequently been usually looked upon as a group that has arisen in the south, probably on the shores of the Antarctic Continent. Their geological history, so far as is known, fully bears this out, for fossil penguins are known from New Zealand and from Patagonia, while recently the Swedish Antarctic Expedition found on Seymour Island fossil remains of no less than six new genera of penguins in beds of early Tertiary age, so that the group was numerous and varied at that time. Among the fishes, the family Galaxudae, which is widely distributed in the southern regions and exclusively confined to them, has been already referred to, and there are also marine fishes equally characteristic of southern seas. Of the earthworms we have the Notiodrilidae, with a distribution closely parallel to that of the penguins; but here, unfortunately, no palaeontological evidence of the past history of the family is possible. Among the Crustacea the Parastacidae, a family of fresh-water crayfishes, is confined to the Southern Hemisphere, and is represented by different genera in South America, Madagascar, Australia, and New Zealand. These genera are already pretty well marked off from one another, and doubtless show that they have long been isolated, as would naturally be the case owing to their occupying different river- basins, even if they were all on the same great continental area. The group, how- ever, 18 well marked off from the Northern Hemisphere crayfishes, and probably had its origin in the south. The Crustacea supply another example in the genus Deto, which has been already referred to; this genus is so distinct that it is 52—S. 802 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Summary of Results. considered by Budde-Lund as forming a subfamily, and it is found only in the sub- antarctic regions and the southern extremities of the great continents. A still more striking example is afforded by the Serolidae, a marine family of Isopods. Not- withstanding their marine habitat, and the fact that some of them extend into deep water, they are almost entirely confined to subantarctic regions, being represented at New Zealand, Patagonia, Chili, South Shetlands, South Georgia, Marion and Prince Edward Islands, Crozets, Kerguelen, and Australia, while one species, which is perhaps doubtful, is recorded from California. Another group of the Crustacea that 1s widely distributed in subantarctic regions is that section of the fresh-water Copepoda that includes the genus Boeckella and allied genera. Boeckella was first established for a New Zealand species, B. triarti- culata (G. M. Thomson), and Sars has described two other New Zealand species of this genus, although it is doubtful 1f they are more than varieties of the first. Dr. Al. Mrazek, in his account of the fresh-water Copepoda collected by the Hamburg Magellan Expedition, points out that this group consists of comparatively large forms readily secured with the surface-net in open sheets of fresh water, and they are therefore likely to be more easily captured than the smaller Copepoda living among weeds on the edges of small ponds, and that consequently, although our knowledge of the group is not perfect, a fair idea of its distribution 1s already obtain- able. In South America, the Falklands, and South Georgia the group is represented by several species for which Mrazek establishes new genera closely allied to the original Boeckella ; one species of these, Paraboeckella brevi-caudata, is also found in Kerguelen. A species belonging to the same group, Deguernea antarctica, has been described by Thomson from Macquarie Island ; Mrazek considers this a young form of a species of uncertain position, but undoubtedly a Boeckellid. The genus Boeckella is also found in the alpine lakes and tarns of Tasmania, where it apparently dis- places the northern genus Diaptomus, and Geoffrey Smith (1909a, p. 137) has recently called attention to the importance of the distribution of the genus. The group is, therefore, widely distributed in all subantarctic lands, but it nowhere extends into the tropics, and may be looked upon as a group that has arisen, perhaps, in the Antarctic Continent. This appears to be confirmed by this discovery of a species in fresh-water pools on the Antarctic Continent itseli—viz., at Hofinungs- bucht, in Louis Phillipe Land, during the Swedish Antarctic Expedition. Dr. Gunnar Andersson was forced to winter there with two companions, and in a small pool filled with fresh water from the melting ice, near the winter station, he found specimens which he collected with difficulty and preserved with care throughout the long and dreary winter spent at that place. I have not yet seen the full account of this species, but in Andersson’s account of this part of the expedition (Nordenskjold, 1904, p. 164) it is stated that it was examined by Dr. Sven Ekman, and found to be a species of Boeckella closely allied to the Patagonian species B. entzw; it will perhaps be assignable to one of the genera established by Mrazek. In the Mollusca certain marine genera that are distinctly confined to subant- arctic regions have been already mentioned; and the Phenacohelicidae, a family of land-snails, has a similar distribution. To this it may be added that the fresh-water mussels of the genus Diplodon are found over the whole of South America, in New Zealand, the northern rivers of Tasmania, Australia, and New Guinea. Summary of Results.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 803 Numerous additions to this list of groups of probably southern origin could readily be made from other divisions of the animal kingdom, from the terrestrial plants, and the marine Algae. From the facts above recorded it is clear that many forms both of animals and of plants are circumaustral, being found on nearly all the suitable subantarctic situations ; that this applies particularly to several groups of land-animals as well as to plants; and that the community thus demonstrated between the terrestrial fauna and flora of the subantarctic islands is accompanied by similar resemblance in the marine littoral fauna of these places. Two explanations of this distribution have been put forward. According to one, which has recently been called by Dahl (1908, p. 669) the “relict theory,” the inhabitants of these subantarctic islands are the remnants of groups of animals which probably developed in the great land-masses further north, but, being driven south by forms subsequently developed, have been preserved only in these scattered islands and in the most southerly parts of the great continents. According to this view, these subantarctic islands have always been widely separated from the great continents, and their inhabitants must therefore have been able to reach them by crossing over wide stretches of sea. It is this explanation that has been accepted by Schenck in his valuable and exhaustive study of the flora of the subantarctic islands, and in this he has been followed by Cheeseman in his article in the present work. The other theory, usually known as the “ Antarctic Continent theory,” accounts for the similarity of the fauna and flora of subantarctic lands by supposing that these lands were previously more or less closely connected with the Antarctic Continent, from which outliers reached, perhaps at different times, to South America, towards New Zealand and Australia, and perhaps towards South Africa, and that the existence of similar forms in the most southerly parts of the great continents is due to the fact that they have been driven northward from the more southerly lands by a sub- sequent colder period. Some of the points in this second theory are no longer theoretical, but are demonstrated facts. We know that an Antarctic Continent does exist, much smaller, indeed, than the great Antarctic Continent that was believed in before the time of Cook’s voyages, but still a continent of great size. The National Antarctic Expe- dition has proved that in South Victoria Land a chain of high mountains extends from the north, near:Cape Adare, more or less continuously far beyond McMurdo Strait, and that to the west of this there lies a great elevated plateau. Lieutenant (now Sir EK. H.) Shackleton was afterwards able to follow this chain almost to the South Pole, and from his observations it is practically certain that high land extends beyond the Pole itself. High land was also seen by Captain Scott east of the great ice-barrier, and was named by him King Edward VII Land. The various expeditions—Swedish, Scotch, and French—have demonstrated the existence of large masses of land south from South America—+z.e., Louis Phillipe Land, King Oscar II Land, Graham Land, Alexander Land. Bruce (1905, p. 411) gives various reasons for the belief that the various portions of land found here are not merely {parts of an archipelago, but portions of a continental area, and the existence of sedimentary rocks in Seymour Island, in the South Orkneys, and in South Georgia points in the same direction, and indicates a former closer 804 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Summary of Results. land connection with South America. Coats Land, in lat. 72° S., long. 17° W., discovered by the Scottish National Expedition under Bruce, and Kaiser Wil- helm II Land, investigated by the German Antarctic Expedition, are probably also parts of the Antarctic Continent, other parts of which have been named Enderby Land, Sabrina Land, Clarie Land, &c., though the existence of these lands does not rest on such good authority. At present the northern shore of this continent is situated at some considerable distance from the subant- arctic islands, and in places there are great depths of ocean lying to the north of the Antarctic Continent. Recent explorations, however, certainly tend to show that this separation was not always so great. South America comes, of course, nearest, for the direct distance from Cape Horn to Graham Land is only 550 miles, and a shallow area of sea extends east past Falkland Islands and on towards South Georgia, thence south past the Sandwich Group, and then west past the South Orkneys and other islands towards Louis Phillipe Land. An elevation of the land of only a moderate amount would establish land connection between the parts just named, and the existence of the sedimentary rocks referred to above is evidence that areas of land existed in the neighbourhood in past geological ages. Macquarie Island is about 970 miles from Adelie Land, and somewhat less than this from the Balleny Islands. Thirty miles to the south of Macquarie Island there is a group of rocks known as the Bishop and Clerk, and from the evidence given in an earlier part of this paper it is pretty certain that Macquarie Island at one time formed part of a great area of land connecting it with the Auckland and Campbell Islands, and in any case it must certainly have extended beyond the Bishop and Clerk rocks, so that its distance from the Balleny Islands would be to this extent reduced. Unfortunately, our information as to the depths of the sea in this region is very scanty; but it appears probable from the existence of Scott Island at a considerable distance east from the Balleny Islands that there is a wide extent of shallow sea in that region. It is not known definitely how far north this shallow area extends, but a moderate elevation of the Antarctic Continent would bring it much closer to Macquarie and the other islands, if it did not actually connect it with them. In the Indian Ocean, Kerguelen Island is about 1,225 miles from Hnderby Land, and much the same distance from Kaiser Wilhelm II Land; but about 300 miles to the south of Kerguelen Island lies the McDonald and Heard Islands group, and from the similarity of its flora to that of Kerguelen Island, and from the existence of shallow water between the two, it 1s most probable that a greater mass of land once existed in this region, and its distance from the Antarctic Con- tinent at that time would be proportionately reduced. It seems likely that the Crozets and Prince Edward and Marion Islands are in the same way remnants of a former larger land-area, and that they were then brought into fairly close connection with Kerguelen Island, though a narrow strait of water more than 2,000 fathoms in depth appears to lie between Kerguelen and the Crozets at the present time. Further soundings in the subantarctic seas are urgently required for the fuller investigation of these points. The ‘“‘ Scotia’ has done much valuable work in this direction, and it is worth while pointing out that, from the soundings obtained, Bruce is of opinion that a shallow area of sea less than 2,000 fathoms, and 300 miles in breadth, extends north-east from the Sandwich Group, that one arm of this then Summary of Results.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 805 runs northward towards Gough Island and Tristan da Cunha, and another arm extends east past Bouvet Island and the South of Africa towards Madagascar. The Prince Edward and Marion Islands and the Crozets would apparently be also con- nected with this ridge. It is important to note that these soundings seem to show how connections could be made between the Antarctic Continent and South America and the sub- antarctic islands to the south of the Indian Ocean by local elevations of the land of comparatively small extent and without assuming any very great change in the relative positions of the great continents and oceans. These would supply the connections required by Forbes without assuming such a huge subantarctic con- tinent as appeared to be necessary at the time when he wrote. At the present time the Antarctic Continent proper is almost devoid of terrestrial hfe. The plants are represented by one species of flowering-plants (Deschampsia antarctica), and by a few minute mosses and lichens; while the land-animals are confined to a small number of wingless insects and spiders. Evidently the Antarctic Continent theory requires a warmer climate for antarctic land in past ages, and evi- dence for this is now available. The National Antarctic Expedition found fossil leaves near the winter quarters of the “ Discovery,” though these were unfortunately not sufficiently well preserved for exact identification ; seams of coal were afterwards found by Lieutenant Shackleton still further south. The most important discovery, however, was that made by the Swedish Antarctic Expedition, of abundant fossil plants in a sandstone formation of Tertiary age on Seymour Island. These have since been investigated by P. Dusén, and, though his report is not available in New Zealand at the present time, abstracts of it which have appeared are sufficient to show the very important bearings that this discovery has on the question now under discussion. Among the fossils are specimens referred to a large number of genera, such as Laurelia, Drumys, Ilicuphyllum, Lomatia, Knghtia, Fagus, Nothofaqus, Araucaria, &c. This Tertiary flora of Seymour Island appears to be a mixture of temperate and subtropical species, and is distinctly related to those of two present South American floral regions —namely, the temperate flora of south Chill, and the subtropical plant- world of south Brazil. Its connection with Australia and New Zealand is evidently greater than Dusén appears to think it. The presence of Knightia Andreae, a close relative to K. excelsa of New Zealand, and of such genera as Nothofagus, Drimys, and Laurelia shows very considerable connection with the present plant covering of New Zealand. In this connection it should be mentioned that Ortmann (1904) has demon- strated the existence of fossils in the Lower Miocene beds of Patagonia, which contain Mollusca alhed to those found in Chili, New Zealand, and Australia, and has pointed out that these species of Mollusca belong to a littoral fauna. Unfortunately, Ort- mann’s paper is not here available, and I do not know what species are represented in these Patagonian beds; but the existence of the same Tertiary fossils in beds in South America and New Zealand was already to some extent known, and Mr. Suter has given me a list of six species common to New Zealand and South America.* The existence of this littoral fauna in Lower Miocene times implies the existence * Hpitonum rugulosum lyratum, Zittel; Crepidula gregaria, Sow. ; Turritella ambulacrum, Sow. ; Cucullaea alta, Sow.; Venericardia patagonca, Sow.; Brachydontes magellanica, Lam. (= Mytilus). 806 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Summary of Results. at the same time of land in the neighbourhood, and it seems probable that the Antarctic Continent was at that time at a higher level, and that there was a more or less continuous coast-line connecting South America on the one hand with the New Zealand continent on the other. The same question is dealt with by von Jhering, who in his “ Mollusques fossiles de l Argentine” (1907) devotes several pages to a consideration of the “ Relations de la faune pan-patagonienne avec celle de la Nouvelle-Zélande”; and A. Gaudry (1907, p. 351), after a review of the fossil mam- malian fauna of Patagonia, comes to the conclusion that it cannot be explaimed unless Patagonia is the remnant of a vast antarctic continent. The facts given in the preceding pages show that in the subantarctic islands of New Zealand we have some forms of hfe showing close connection with New Zealand, and evidently derived from that land, many of them doubtless having come originally from the north; and that with these there is mixed a pretty large ant- arctic element showing affinities particularly with South America, and other more remote affinities with the Kerguelen-Crozet groups and with South Africa. The evi- dence pointing to former extensions of land from the Antarctic Continent northwards, and to the warmer climate that was enjoyed by this continent in early Tertiary times, seems to offer a fairly satisfactory explanation of the facts before us; and it accounts also for the occurrence of special endemic genera, such as Pringlea and Lyalia on Kerguelen, and Stiulbocarpa* and Pleurophyllum on the subantarctic islands of New Zealand. These are admittedly antarctic forms, and have doubtless developed in the northern extensions of the former Antarctic Continent, which were naturally more isolated from one another than the regions further south. The forms both of plants and of animals that are now common to the whole of these subantarctic islands must at one time have been spread over practically the whole of the Antarctic Continent, and since they no longer exist on that continent in its present form it is evident they must have continued more or less unchanged throughout probably the greater part of the Tertiary period, so that a species such as the land Isopod Trichoniscus magellanicus, now found in the New Zealand region, in South America, and the Falklands, and probably also on the Crozets, 1s a form of very considerable antiquity. That species do actually continue unchanged for long periods in this way there is, of course, abundant evidence, though, unfor- tunately, that evidence is not often obtainable in the case of small crustaceans owing to their unsuitability for preservation as fossils. Mr. Suter has recently drawn up a list of 219 shells still existmg in New Zealand which are also found fossil, some of them dating back to the Oamaru series, which is probably to be considered of Oligocene or of Hocene age. In the above discussion I have confined myself to the points that naturally arose in connection with the fauna and flora of the islands under consideration, and have not attempted a general revision of the evidence for and against the sub- antarctic continent. This question has been discussed by many writers in the past, and more recently by Geoffrey Smith (1908, 1909a, and 19096), Dahl (1908), and Kolbe (1909), &c., and it will doubtless be still further discussed as additional results of the recent antarctic expeditions are published. Many of those that have already appeared are not yet available in New Zealand, and the conclusions given above * Stilbocarpa is also found on the extreme south of New Zealand. Summary of Results.| | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 807 have had to be arrived at without the full consultation of all the literature. But other confirmatory evidence is already forthcoming, for since the above was written I have received the volume of the Scottish National Antarctic Expedition con- taining Professor Carpenter’s article on the Collembola, in which he points out that this group affords further support for the ancient extension of Antarctica, shows how this might have occurred by means of the connections between that continent and South America and the Kerguelen Group, and further points out that these insects must therefore be of considerable geological age, just as I have done in con- nection with the terrestrial Zsopoda and other groups.* The abstract in the Journal of the Royal Microscopical Society for June, 1909, of J. Cardot’s report of the bryological flora of Magellan, South Georgia, and Ant- arctica collected by the Swedish South Polar Expedition states that he has shown conclusively. thatZthej|Magellanic’.moss flora has closer affinity to that of Australasia than to that of Patagonia, the former land connection being probably indicated by the Islands of Falkland, South Georgia, Kerguelen, Auckland, and New Zealand. This is unexpected confirmation derived from plants which are usually supposed to possess every facility for crossing over the oceans from land to land. The exact date at which the Antarctic Continent had its great extension north- wards can perhaps hardly be definitely decided upon at present. Hutton (1905) argues for the Jurassic period as the period of greatest extension; but since he wrote much further evidence has arisen, and it seems probable that the date should be placed later—perhaps in Eocene times. Ortmann, discussing the matter from a somewhat different aspect, considers that it probably occurred in the Cretaceous and Kocene periods. The facts given in this work, and thus briefly and impertectly recapitulated in this article, will be sufficient, whatever conclusion may be drawn from them, to show that they have an important bearing on these general questions, and that it is from the study of the fauna and flora of subantarctic islands that we must look for the facts that will enable a final decision to be arrived at. We may therefore appropriately conclude with the prophetic words of Hooker, which have already been placed at the beginning of the work—viz., “ Hence it will appear that islands so situated furnish the best materials for a rigid comparison of the effects of geographical position and the various meteorological phaenomena on vegetation, and for acquiring a knowledge of the great laws according to which plants are dis- tributed over the face of the globe.” * Since this was written, I find from the review of his reports in the Naturwiss. Wochenschrift for 17th October, 1909, that Enderlein (1909a) has come to conclusions very similar to those expressed above, and has supported them by facts derived from the distribution of the insects and spiders of the Kerguelen region. Goa Newb JSIUS JK OGdtwe leet RELATING TO THE SUBANTARCTIC ISLANDS OF NEW ZEALAND. Tuis list contains the names of works dealing with the subantarctic islands of New Zealand. A number of general works that contain references to the islands or to the questions concerning them discussed in the Historical Introduction and in the concluding article are also included. Articles or papers on special groups of animals and plants from the islands are not included unless of general interest, but will be found in their appropriate places in the bibliographies at the ends of the articles dealing with these groups. ALBOoFF, N. 1902. Hssai de Flore Raisonnée de la Terre de Feu. Ann. Mus. d. 1. Plata. [Contains a chapter dealing with the relations between the flora of Fuegia and that of New Zealand, and another on the origin of the flora of Fuegia. | ALLEN, J. A. 1896. The Origin and Relations of the Floras and Faunas of the Antarctic and Adjacent Regions. Vertebrata of the Land: Birds and Mam- mals. Scrence, New York, in (1896), pp. 316-319. ANTARCTIC. 1897. Antarctic Lands, Islands, and Ice: im Findlay’s Directory for the Navigation of the Indian Ocean, Red Sea, Bay of Bengal, &c. ; with Supplement, 1897; pp. 84-92. London: R. H. Laurie. 1899. Antarctic Islands: in Findlay’s Sailing Directory for the South Atlantic Ocean ; with Supplement, 1899; pp. 127-138 and 295-310. Lon- don: R. H. Laurie. 1900. Antarctic Land, Islands, and Ice: in Findlay’s Directory for the Navigation of the South Pacific Ocean; with Supplement, 1900 ; pp. 306-344 and 1144-1153. London: R. H. Laurie. 1901. Antarctic Manual. [See Murray, G., 1901.] 1905. Deutsche Sudpolar Expedition, 1901-3. Edited by E. von Drygalsk. [In course of publication. | 1905. Wissenschaftliche Ergebnisse der Schwedischen Siidpolar Expedition, 1901-3, unter Leitung vom Dr. Otto Nordenskjold. [In course of publication. | Bibliography.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 809 ANTARCTIC—continued. 1907a. National Antarctic Expedition, 1901-1904. Natural History: Vol. i, Geology (Field Geology, Petrography) ; Vol. 11, Zoology (Vertebrata, Mollusca, Crustacea); Vol. in, Zoology and Botany (Invertebrata, Marme Algae, Musci). British Museum (Nat. Hist.), London, 1907. [ Vol. iv not yet issued. j 1907b. Scottish National Antarctic Expedition. Report on the Scientific Results of the 8.Y. “Scotia”? during the Years 1902, 1903, 1904, under the Leadership of W. 8. Bruce. {In course of publication. | ARMSTRONG, H. 1868. Cruise of the Brig ‘“ Amherst.” New Zealand Government Gazette, Province of Southland, 1868. AUCKLAND ISLANDS Map. 1870. Map of Auckland Islands. [Prepared by H.M.S. “Blanche” while searching for castaways in July, 1870. | Baker, J. H. 1865. The Physical Geography, Geology, and Resources of the Auckland Islands. By J. H. Baker, Chief Surveyor. New Zealand Govern- ment Gazette, Province of Southland, 1865, p. 117 et seq. BALLENY, J. 1839a. Discoveries in the Antarctic Ocean in February, 1839. Journ. Roy. Geog. Soc., London (1839), pp. 517-526. [Reprinted in Antarctic Manual, 1901, pp. 336-347. ] 1839b. Extrait du Journal du navire |’Elisa Scott allant de Vile Campbell vers le Pole sud. Bull. Soc. Geog., Paris, 11, ser., 12 (1839), pp. 84-88. BEDDARD, F. E. 1895a. The Former Northward Extension of the Antarctic Continent. Nature, London, li (1895), pp. 750-752. 1895b. A Text-book of Zoogeography. BELLINGSHAUSEN, F. G. 1822. Voyage de découvertes du Capt. Bellingshausen dans lOcean austral. Journal des Voyages, xiv (1822), 126-128. 1831. Two Voyages of Exploration in the Antarctic Ocean and a Circum- navigation of the World in the Years 1819, 1820, 1821, carried out in the Corvettes ‘ Vostok” and “‘ Mirny,” under Captain Bellings- hausen as Leader of the Expedition and Commander of the “ Vostok,” and Lieutenant Lazareff as Commander of the “ Mirny.” 2 vols. St. Petersburg, 1831. [In Russian. | 810 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Bibliography. Benuam, W. B. 1901. On some Earthworms from the Islands around New Zealand. Trans. ING e nS, COM, JO. 1ZY, 1902. Harthworms and Palaeogeography. Report Aust. Ass. Adv. Sci., Hobart, 1x, pp. 319-343. [Presidential address to Section D, Biology. | Bernaccui, L. 1901. To the South Polar Regions. Expedition of 1898-1900. London, 1901. Biscogr, J. 1832. Journal of a Voyage towards the South Pole, on Board the Brig “ Tula,” under the Command of John Biscoe, with Cutter “ Lively” in Company, 1830-32. 1832. [Manuscript in Library of Roy. Geog. Soc., London. | BLANCHARD, KH. 1878. La Nouvelle Zelande et les petites iles australes adjacentes. [Seven articles in the Revue des Deux Mondes, from 1878 to 1884. ] 1882. Proofs of the Subsidence of a Southern Continent durimg Recent Geo- logical Epochs. N.Z. Journ. Sci., vol. 1, p. 251. ‘Abstract from the Comptes Rendus. | BorcHGREVINK, C. E. 1895. The “ Antarctic’s”’ Voyage to the Antarctic. Geog. Jour., London, v (1895), pp. 375-377. 1896. The Voyage of the “ Antarctic’ to Victoria Land. Report Sixth In- ternat. Geog. Congress, London, 1895 (1896), pp. 169-176. 1900. The “Southern Cross” Expedition to the Antarctic, 1899-1900. Geog. Jour., London, xvi (1900), pp. 381-414. [Maps and illustrations. | BouLENGER, G. A. 1902. The Explanation of a Remarkable Case of Geographical Distribution among Fishes. Nature, lxvu, p. 84. Britton, N. L. 1896. The Origins and Relations of the Floras and Faunas of the Antarctic and Adjacent Regions. Botany. Science, New York, i (1896), pp. 310-311. Brown, R. N. R. 1906. Antarctic Botany: Its Present State and Future Problems. Scot. Geog. Mag., Sept., 1906. Bibliography.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 811 Bruce, W. S. 1905. Some Results of the Scottish National Antarctic Expedition: 1. Bathy- metrical Survey of the South Atlantic Ocean and Weddell Sea. Scot- tish Geog. Jour., August, 1905. 1907. Reports Scottish National Antarctic Expedition. [See under “ Antarctic,” 19070. ] BucHANAN, J. 1870. Sketches of Headlands and Harbours of the Auckland Islands: to accompany a Chart by Officers of H.M.S. “ Blanche.” 1884, Campbell Island and its Flora. Trans. N.Z. Inst., xvi, p. 398. BuppgE-Lunp, G. 1906. Die Landisopoden der deutschen Siidpolar Expedition, 1901-1903. IX, Zoologie, I, pp. 71-92. Bute EJ. 1895. Whaling in the Southern Seas: the Cruise of the “ Antarctic.” 1895. [ Illustrations. | 1896. The Cruise of the “ Antarctic”: a Voyage to the South Polar Regions in 1895. London and New York: Edward Arnold, 1896. Pp. 256. [Map and illustrations. | ? Care Horn. 1891. Mission Scientifique du Cap Horn. Paris: Ministére de la Marine, 1885-91. [7 vols. Maps and plates. | Carrick, R. 1892. New Zealand’s Lone Lands: being Brief Notes of a Visit to the Out- lying Islands of New Zealand. Wellington, 1892. CHApMaAN, F. R. 1891. The Outlying Islands South of New Zealand. Trans. N.Z. Inst., xxiii, p- 491. CHEESEMAN, T. F. 1906. Manual of the New Zealand Flora. Cuun, C. 1900. Aus den Tiefen des Weltmeeres. Jena, 1900. [General account of the “ Valdivia” Expedition. | CockaYNg, L. 1904. A Botanical Excursion during Winter to the Southern Islands of New Zealand. Trans. N.Z. Inst., xxxvi, pp. 225-333. [With maps and illustrations. | 812 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Bibhiography. CoLBEcK, W. 1900. Map of the Route of the “Southern Cross” Antarctic Expedition. Geog. Jour., London, xvi (1900), pp. 500, 569. IDARI, —=. 1909. Ergebnisse der Hamburger Magalhaensischen Sammelreise 1892 bis 1893. Naturwissenschaftliche Wochenschriit, vu, October, 1909, pp. 668-670. DELANO, AMASA. 1817. A Narrative of Voyages and Travels in the Northern and Southern Hemispheres. Boston, 1817. [Slight description of the Snares, Bounty Islands, New Holland, and Van Diemen’s Land. | * DISCOVERY.” 1902. The Voyage Southward of the pecoreny =. Geog. Jour., xix, 1902. DouGALL, W. 1888. Far South: Stewart Island, the Snares, Auckland, Comore Antipodes, and Bounty Islands. Notes by Dougall, Photo. Invercargill, 1888. [Mr. Dougall accompanied the C.G.S. “ Stella,” Captain Fair. child, to these islands, and took a series of 51 views, of which this is the descriptive letterpress. | Drupe, O. 1897. Manuel de Geographie Botanique. ‘Translated from the German by G. Poirault. [Pp. 503-511 deal with the flora of the antarctic islands. | DryYGALSKI, HK. VON. 1905. Reports German South Polar Expedition. [See under “ Antarctic,” 1905. ] D’Urviute, J. 8. C. Dumont. 1834. Voyage pittoresque autour du Monde. Paris, 1834. [An imaginary voyage, in which D’Urville weaves into his own narrative extracts from the voyages of other circumnavigators, the whole forming an admirable résumé. The story of New Zealand is given in 76 pages and 30 illustrations, and includes the outlying Macquarie, Campbell, Auckland, Antipodes, Bounty, Chatham, and Kermadec Islands. ] {A German translation in 2 vols., Leipsic, 1834-39. | 1854. Voyage au Pole sud dans |’Océanie sur les corvettes “Astrolabe” et la “ Zélée”” exécuté par ordre de Roi pendant 1837-1840 sous le commandement de M. J. Dumont D’Urville; publié par ordon- nance de §.M. sous la direction supérieure de M. Jacquinot. 32 vois., avec un atias en 6 vols. Paris, 1842-54. Bibliography.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 813 ENDERBY, C. 1847. Proposal for Re-establishing the British Southern Whale-fishery through the Medium of a Chartered Company and in Combination with the Colonisation of the Auckland Islands, as the Site of the Company’s Whaling-station. 3rd edition. London, 1847. [With maps.] 1849a. The Auckland Islands: a Short Account of their Climate, Soil, and Productions; and the Advantages of establishing there a Settlement at Port Ross for carrying on the Southern Whale-fisheries. London, 1849. [With a panoramic view of Port Ross, and a map of the islands. | 1849b. Proceedings at a Public Dinner given to Charles Enderby, Esq., F.R.S., at the London Tavern, Bishopsgate Street, on Wednesday, the 18th of April, 1849. London, 1849. [A farewell on his departure for the Auckland Islands. Mr. Enderby’s speech gives some account of the southern whale-fishery. | ENDERLEIN, G. 1909a. Die biologische Bedentung d. Antarktis. 19095. Die Inseckten des antarktischen Gebietes. 1909c. Die Spinnen der Crozet-Inseln und von Kerguelen. Deutsche Siidpolar Expedition, 1901-3, Band x. [Review by Lohrmann in Naturwiss. Wochenschrift, viii, p. 667. | ENGLER, A. 1882. Versuch einer Entwickelungsgesichte der Pflanzenwelt. Teil 2. Escott-Inman, H. 1908. The Castaways of Disappomtment Island: being an Account of their Sufferings. Written by H. Escott-Inman, from the Description supplied to him by Mr. Charles Eyre, of Dulwich, London, one of the Survivors. “Sunday Companion” Library, Amalgamated Press (Limited), London. [An account of the “ Dundonald” wreck. ] FAIRCHILD, J. 1886. Sketch-map of the Antipodes Islands. FintHoi, H. 1882. Geological and Zoological Relations of Campbell Island and the Neigh- bouring Southern Islands. N.Z. Journ. Sci., 1, p. 259. 1885. Mission de Vile Campbell: Recueil de Mémoires, &c., rel. & observation du Passage de Venus sur le Soleil. T. iu, 2° partie. 814 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Babliography. Forsess, H. O. 1893a. Antarctica. Natural Science, 11, p. 54; and Fortnightly Review tor May, 1893: 1893b. The Chatham Islands and an Antarctic Continent. Nature, xlvii, p. 474. 1893c. The Chatham Islands. Supp. Papers, Roy. Geo. Soc., iti, p. 607. 1894. Antarctica: a Vanished Austral Land. Fortnightly Review, London, lv (1894), pp. 194-214. : 1896. Antarctica: a Vanished Austral Land. Annual Rep. Smithsonian Inst., 1894, pp. 297-316. Washington, 1896. [Reprinted from Fortnightly Review for 1894. ] FRICKER, K. 1898. Antarktis. Bibliothek der Landerkunde, herausgegeben von Alfred Kirchoff und Rudolf Fitzner. Bandi. Berlin: Schall und Grund, 1898. Pp. vi, 230. [Maps and illustrations. ] 1900. The Antarctic Regions. Pp. xi, 292. London: Sonnenschein and Co., 1900. [Maps and illustrations. Translated by A. Sonnenschein. | Gaupry, A. 1907. Patagonia and Antarctica. Science, n.s., xxvi, p. 350. ‘* GAZELLE.” 1874. Die Deutsche Wissenschaftliche Forschungsreise durch 8.M.S. “ Gazelle ” und ihre Aufgaben. Hydrogr. M., Berlin, 11 (1874), pp. 123-127. Gaibih, JE 1893. A Comparison of Antipodal Faunas. Mem. Nat. Acad. Sci., Washington, Waly [Oe WIE 1896. The Origins and Relations of the Floras and Faunas of the Antarctic and Adjacent Regions. Vertebrata of the Land: Fishes, Batrachia, and Reptiles. Vertebrata of the Sea. Scvence, New York, i (1896), pp. 314-317, 319-320. GRAVELIUS, H. 1902. F. von Bellingshausen’s Forschungstahrten im Siidlichen Hismeer 1819- 1821. Auf Grund des russischen Originalwerks herausgegeben vom Verein fur Erdkunde zu Dresden. Leipzig, 1902. [An abridged translation into German from the Russian of Bel- lingshausen’s Voyage to the Southern Frozen Sea in 1819-21.] Gray, J. EK. [See Richardson, 1845. | GREGORY, J. W. 1908. Some Scientific Results of the Antarctic Expeditions, 1901-1904. Geog. JOur., XXXll, p. 24. Bibliography. | SUBANTARCTIC ISLANDS OF NEW ZEALAND. 815 Guppy, H. B. 1888. Flora of the Antarctic Islands. Nature, xxxvili (1888), p. 40. [With note by W. T. Thiselton Dyer. | Hamiuton, A. 1895. Notes on a Visit to Macquarie Island. Trans. N.Z. Inst., xxvu, p. 559. Hector, J. 1870. Notes on the Geology of the Outlying Islands of New Zealand; with Extracts from Official Reports. Trans. N.Z. Inst., i, p. 176. [Contains extracts from the reports by Baker (1865) and Arm- strong (1868), and notes on specimens of rocks collected by Arm- strong. | 1896. Note on the Geology of the Outlying Islands of New Zealand. Trans. INS7e instep xecvililis psayo O. HEDLEY, C. 1893. On the Relation of the Fauna and Flora of Australia to those of New Zealand. Natural Science, ui, p. 187. 1895. Considerations of the Surviving Refugees in Austral Lands of Ancient Antarctic Life. Proc. Roy. Soc. N.S.W., xxix, p. 278. HEILpRIN, A. 1896. The Origin and Relations of the Faunas and Floras of the Antarctic and Adjacent Regions. The Geology of the Antarctic Regions. Scrence, New York, i (1896), pp. 305-307. 1897. Les Regions antarctiques. Rev. Scient., vil, p. 329. Hemstey, W. B. 1895. The Flora of Macquarie Island, in Report on Present State of Know- ledge of Various Insular Floras, p. 62. “Challenger”? Report, Botany. London, 1895. Hocxen, T. M. 1909. A Bibliography of the Literature relating to New Zealand. Wellington, 1 [A valuable bibliography, that has been of much use in the preparation of the present list. | HoMBRON AND JACQUINOT. 1852. Voyage au Pole sud, Botanique. Paris, 1852. [See D’Urville, 1854.] Hooker, J. D. 1847a. The Botany of the Antarctic Voyage of H.M. Discovery Ships “ Erebus” and “Terror,” in the Years 1839-1843, &c. Map and plates. 2 vols. 1847. I. Flora Antarctica. Part i, Botany of Lord Auckland’s Group and Campbell Island; Part u, Botany of Fuegia, the Falklands, Kerguelen’s Land, &c., 816 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Bibliography. Hooxer, J. D.—continued. 18476. The Cryptogamic Botany of the Antarctic Voyage, &c. London: Reeve, 1845-1847. 1853-55. Flora Novae Zealandiae. Part i: Botany of Antarctic Voyage. London. 1864. Handbook of the New Zealand Flora. [In two parts; Part 1, 1864; Part ii, 1867.] Hutton, F. W. 1873. On the Geographical Relations of the New Zealand Fauna. ‘Trans. N.Z. Inst., v, p. 227; and Ann. Mag. Nat. Hist., ser. 4, xii, p. 25. 1884. On the Origin of the Fauna and Flora of New Zealand: Parti. N.Z. Journ. Sci., u, p. 1; and Ann. Mag. Nat. Hist., ser. 5, xiii, p. 425. 1885. On the Origin of the Fauna and Flora of New Zealand: Part u. N.Z. Journ. Sci., u, p. 249; and Ann. Mag. Nat. Hist., ser. 5, xv, p. 77. 1896. Theoretical Explanations of the Distribution of Southern Faunas. Proc. Linn. Soc. N.S.W., xxi, p. 36. 1902a. On a Small Collection of Diptera from the Southern Islands of New Zealand. Trans. N.Z. Inst., xxxiv, p. 169. 1902b. The Beetles of the Auckland Islands. Trans. N.Z. Inst., xxxiv, p. 175. 1904. Index Faunae Novae Zealandiae. Dulau and Co., London, 1904. [The Introduction deals with the origin of the fauna. | 1905. Ancient Antarctica. Nature, |xxu, p. 244. JHERING, H. VON. 1892. On the Ancient Relations between New Zealand and South America. Trans. N.Z. Inst., xxiv, p. 431. 1907a. Archelenis und Archinotis. Gesammelte Beitrage zur Geschichte der Neotropischen Region. Leipzig, 1907. 19076. Les Mollusques fossiles du Tertiaire et du Crétacé supérieur de l Argen- tine. Anales del Museo Nacional Buenos Aires, Serie i, Tomo vii, pp. E611. Kirk, T. 189la. On the Botany of the Antarctic Islands. Rep. Aust. Assoc., 11, p. 213. 1891b. Report on a Botanical Visit to Lord Auckland, Campbell, Antipodes, and other Antarctic Islands. Journ. Linn. Soc. London (Bot.), XXvill (1891), pp. 327-330. Kose, H. 1909. Die Sudpolarkontinenttheorie uber tiergeographische Verhaltnisse auf der Stidhemisphare. Naturwissenschatftliche Wochenschrift, viii, July, 1909, p. 449, Bibliography.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 817 Kurtz, F. 1875. Flora der Aucklands-Inseln. Verh. d. Bot. Vereins der Provinz Bran- denburg. Bd. xviii, s. 3. 1877. Flora der Aucklands-Inseln. Nachtrag. Jbid., Bd. xix, s. 168. Luxins, H. 1896. Round Southern Isles. Nelson, 1896. McNap, R. 1907. Murihiku and the Southern Islands. A History of the West Coast Sounds, Foveaux Strait, Stewart Island, the Snares, Bounty, Anti- podes, Auckland, Campbell, and Macquarie Islands, from 1770 to 1829. Invercargill, New Zealand, 1907. 1909. Murihiku. A History of the South Island of New Zealand and the Islands Adjacent and lying to the South, from 1642-1835. Christ- church, 1909. [A new edition of the work mentioned above, with some altera- tions and additions. |] Macquarie IsuanpD Map. 1850. Macquarie Island, South Pacific. Carte des [les Macquarie, Campbell, Auckland, Chatham, et de la partie méridionale de la Nouvelle Zelande. Dressée par M. ©. A. Vincendon-Doumoulin. 1847. Paris, 1850. [Cat. of Maps, Plans and Charts, British Museum, u, p. 2566. ] Matong, R. E. 1854. Three Years’ Cruise in the Australasian Colonies. London, 1854. [A good account of the Auckland Islands and the Hnderby Settlement, with weather-tables for May, June, and July. ] Martin, R. M. 1851(?). History of the British Colonies. [An account of the Auckland Islands is given in vol. iii of the edition published in 1851 (?).] MICHAELSEN, W. 1902. Die Terricolen fauna der Inseln des Subantarktischen Meeres, in Die Oligochaeten der deutschen Tiefsee Expedition. Wiss. Ergebnisse der deutschen Tieisee Expedition, i, p. 158. 1903. Die geographische Verbreitung der Oligochaeten. Berlin, 1903. Mitnre-Epwarps, A. 1874. Recherches sur la Faune Ancienne des Iles Mascareignes. Ann. Sci. Nat., BEE a.) xbxen 0. le Montaene, C. 1845. Voyage au Pole sud: Botanique. Plantes Cellulaires, d3—S. 818 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Babliography. Morre tt, B. 1832. A Narrative of Four Voyages. United States, 1832. [Gives a glowing description of “ Auckland’s Island,” which “is one of the finest places for a small settlement that can be found on any island in the Southern Hemisphere above lat. 35°.” In 1823 the schooner “ Henry ” took from these islands 13,000 “of as good fur-seal skins as were ever brought to the New York market.” “ Scarcely one to be found now.” Same at the Snares. | Mosetey, H. N. 1876. Botany of the Antarctic Islands. Journ. Linn. Soc., Botany, xv, p. 485. Murray, G. 1901. The Antarctic Manual for the Use of the Expedition of 1901. Edited by George Murray, F.R.S. [Contains chapters on terrestrial magnetism, antarctic climate, geology, instructions for collecting, notes on antarctic zoology (including a chapter on “ Kerguelen Island: an Introduction to Antarctic Zoology”), and botany, and extracts from the journals of Biscoe, Balleny, Wilkes, D’Urville, Arctowski, &c., and a valuable bibliography of Antarctic literature (compiled by H. R. Muill).] MuserRaAvE, T. 1866. Cast away on the Auckland Isles: a Narrative of the Wreck of the “Grafton” and of the Escape of the Crew after Twenty Months’ Suffermg. Lockwood and Co., London, 1866. Hdited by John J. Shillinglaw, F.R.G.S. [With map and appendices, including a short account of the Auckland Islands and the loss of the “ Invercauld.”’] Newton, A. 1893. Mr. H. O. Forbes’s Discoveries in the Chatham Islands. Nature, 1893, xlviil, pp. 101, 150 (27th May, 1893). New ZEALAND PILotT. 1901. New Zealand Pilot. Edition 1901. [Contains information as to position, size, &c., of the islands. Supplementary information to date is given in the “ New Zealand Nautical Almanac and Tide-tables,”’ published annually. | NORDENSKJOLD, O. 1904. “ Antarctic.” Zwei Jahre in Schnee und His am Siidpol. 2 vols. Berlin, 1904. [A translation from the Swedish original by Mathilde Mann. ] 1905, Reports Swedish Antarctic Expedition. [See under “ Antarctic,” 1905. ] Bibliography-] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 819 OrtTMANN, A. EH. 1901. The Theories of the Origin of the Antarctic Faunas and Floras. Amer. Nat., xxxv, p. 139. 1904. Die patagonische Formation. Rep. Princeton Univ. Exped. to Pata- gonia. Palaeontology, Part u. Tertiary Invertebrates, pp. 45-332, pl. x1-xxxix. OuTLyInG IsLANDs. 1868. Description of the Outlying Islands South and Hast of New Zealand —viz., Auckland, Campbell, Antipodes, Bounty, and the Chatham Islands. Published by Order of the Lords Commissioners of the Admiralty. London, 1868. [Interesting notes of their discovery and productions, which were afterwards incorporated with the “ Official Sailing Directions.”’| PacKarp, A. 8S. 1896. The Origin and Relation of the Floras and Faunas of the Antarctic and Adjacent Regions. The Terrestrial Invertebrata. Science, New York, iii (1896), pp. 311-14. Rayna, F. E. 1873. Les Naufrages; ou, Vingt Mois sur un Recif des Iles Auckland. 3"° ed. Paris. [With 40 woodcuts and a map, and appendices on the wrecks of the “ Invercauld ” and “ General Grant,” and on the sea-lions. | 1885. Wrecked on a Reef; or, Twenty Months in the Auckland Isles. Nelson and Sons, London, 1885. [A translation of “ Les Naufrages,” with the same appendices. | REISCHEK, A. 1889. Notes on the Islands to the South of New Zealand. Trans. N.Z. Inst., OME Oe Vee RicHARDSON, J., and Gray, J. EH. 1845. The Zoology of the Voyage of H.M.SS. “ Erebus” and “ Terror,’ under the command of Captain Sir James Clark Ross, 1839, 1840, 1841, 1842, and 1843. In 15 parts. Parts i-x. Coloured plates. London : Longmans and Co., 1844-1845. Ross, J. C. 1842. Antarctic Discoveries. Notice of the Magnetometric, Geographical, Hydrographical, and Geological Discoveries ; on Observations made by the Expedition under command of Captain James C. Ross, R.N., F.R.S.; being extracts from a Despatch addressed to the Secretary of the Admiralty. Tasmanian Journ. Nat. Sci., i (1841-42), pp. 409-414. 1847. A Voyage of Discovery and Research in the Southern and Antarctic Regions during the Years 1839-43. 2 vols. London, 1847. 820 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [ Bibliography. RusseEwy, H. C. 1897. Icebergs in the Southern Ocean. No.2. Pp. 32. Map and diagram. Also in Proc. Roy. Soc. N.S.W., Sydney, xxxi (1897), pp. 221-251. Map and illustrations. Sydney, 1897. ScHEeNcK, H. 1905a. Vergleichende Darstellung der Pflanzengeographie der subantark- tischen Inseln inbesondere tiber Flora und Vegetation von Ker- guelen. 19056. Ueber Flora und Vegetation von St. Paul und New Amsterdam. Wissensch. Hrgebnisse der deutschen Tiefsee-Expedition auf dem Dampfer “ Valdivia,” 1898-1899. Band u. Scomma deals 1883. Macquarie Island. Trans. N.Z. Inst., xv, p. 484. Scort, R. F. ; 1905. The Voyage of the “ Discovery.” 2vols. London, 1905. Scott-Enuiot, G. F. 1909. Botany of To-day. [Chapter ix, “The Antarctic,” deals with the distribution of subantarctic plants. | SHACKLETON, HE. H. 1909. The Heart of the Antarctic; being the Story of the British Antarctic Expedition, 1907-1909. 2 vols. London, 1909. [With an Introduction, by H. R. Mill, on “South Polar Expedi- tions in the Last Hundred Years,” and Appendices on Geology, Biology, &c., by Professor T. W. E. David, J. Murray, and others. | SHILLINGLAW, J. J. [See under “ Musgrave,” 1866.] SIMONOFF, J. 1824. Précis Voyage de découvertes fait en 1819-1821 par Bellingshausen dans Pocéan Pacifique et les mers Australes. Journal. des Voyages, xxi, (1824), p. 543. SKOTTSBERG, C. 1904. On the Zonal Distribution of South Atlantic and Antarctic Vegetation. Geog. Journ., Dec., 1904. 1906. Some Remarks upon the Geographical Distribution of Vegetation in the Colder Southern Hemisphere. Ymer; Tidskrift utgifven af Svenska Sallskapet for Anthropologi och Geografi. Arg. 1905. H. 2, p. 402. Stockholm, 1906, Bibliography.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 821 SmitH, EK. A. 1902. On the Supposed Similarity between the Mollusca of the Arctic and Antarctic Regions. [ Presidential address to the Malacological Society of London. |] Smitu, G. W. 1908. Proc. Roy. Soc. London, B, Ixxx (1908), p. 385. 1909a. A Naturalist in Tasmania. Oxford, 1909. [Chapter v deals with the origin of the fauna of Tasmania and with the Antarctic Continent theory. | 19096. The Fresh-water Crustacea of Tasmania, with remarks on their geo- graphical distribution. Trans. Linn. Soc., 2nd ser., Zool., xi, p. 61. SOUTHERN IsLanpDs EXPEDITION. 1908. Southern Islands Expedition. Kew Bulletin for 1908, p. 237. [Compiled from letters by Captain Dorrien-Smith and news- paper articles by Dr. Cockayne. ] Tompson, D’A. W. 7 1898. On Supposed Resemblances between Arctic and Antarctic Faunas. Proc. Roy. Soc. Edin., xxu, p. 311. 1901. Kerguelen Island: An Introduction to Antarctic Zoology. [In Antarctic Manual, 1901, pp. 276-287. ] THomson, G. M. 1882. On the Origin of the New Zealand. Flora. Trans. N.Z. Inst., xiv, p. 485. TURNBULL, J. A. 1813. A Voyage round the World, in the Years 1800, 1801, 1802, 1803, and 1804. London, 1813. [The appendix gives some account of the discovery of the Mac- quarie and Campbell Islands in 1810.] VoyacEe au POLE Sup. [See D’Urville, 1854.] 1845. C. Montagne. Plantes celluiaires. 1852. J. Decaisne. Plantes vasculaires. Watiaceg, A. R. 1893. Mr. H. O. Forbes’s Discoveries in the Chatham Islands. Nature, 1893. xlvi, p. 27 (27th May, 1893). 1902. Island Life. 3rd edition. Chapters xxi-xxiv. Warmine, EH. 1909. Occology of Plants. [Pp. 199-200 deal with the “ Tussock Formation” of the sub- antarctic. | 822 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Bibliography. W[atson], W. 1909. New Zealand Plants. Gardeners’ Chronicle, vol. xlv, p.1 (2nd January, 1909). [Contains a list of New Zealand plants (including many from subantarctic islands) presented to Kew Gardens by Captain Dorrien- Smith. | WILKES, C. 1845. Narrative of the Exploring Expedition, by Authority of Congress, during the Years 1838-1842. Philadelphia, 1845. 4 vols., and atlas. [Auckland Islands, pp. 351-53; Macquarie Island, p. 354.] Woopwarp, H. 1903. The Distribution of Life in Antarctic Islands. Geolog. Mag. x, No. ix, 1903. INDEX. No attempt has been made to index the synonyms. The species in the systematic papers are indexed under their specific names, the name of the genus to which they are referred in this work being placed after the specific name, thus: ‘‘aucklandica, Plaxiphora.” The genera are also indexed, and the species of each genus will be found in order after the description of the genus. Species referred to in other papers are indexed under their generic names, thus: ‘‘ Stilbocarpa polaris.” The concluding article, on the ‘‘ Biological Relations of the Subantarctic Islands of New Zealand,” has not been indexed, as it is itself a summary of the results of the preceding articles. Volume II begins on page 389. A. Abrotanella, 417 Acaena, 403 Acaena a potential weed, 234 Acaena, its spreading by the albatros, 235 Acalles, 120 acanthocarpa, Rhodophyllis, 508 Acanthochites, 3 Acanthochitidae, 3 Acanthodoris, 33 Acanthodrilinae, 252, 255, 268 acaulis, Ranunculus, 398 Achorutes, 377 Achorutinae, 377 acicula, Daphnella, 32 aciculare, Ptychomnium, 536 Acid, humous, 191 Aciphylla antipoda, description of, 188 Aciphylla latifolia,— Description of, 188 Variation in colour, 203 Aclis, 23 Acmaea, 4 Acmaeidae, 4 Acrocladium, 538 Actiniaria; Description of two species from Campbell Island, by H. B. Kirk, 384 Actinidae, 384 acuticosta, Lagena, 334 acutifolium, Ligusticum, 409 Adams Island,— Basaltie series, 731 Description, 706 Dykes, 734 Moraines, 714 Probable lake, 713 Rata forest, 216 Adenocystis, 498 Admete, 28 adscendens, Acaena, 403 Aegerine-augite in pitchstone, 725 Aegerine-augite in trachyte, 691 Aegerine in phonolite, 694 aenea, Marphysa, 244 aeolodes, Platyptilia, 63, 73 aequilateralis, Globigerina, 351 aequinoctialis, Majaqueus, 564 aereus, Chiton, 4 aethiops, Monodonta (Neodiloma), 10 affine, Ditrichwm, 535 affinis, Palaemon, 604, 614 affinis, Trichia, 529 Affinities of mountain scrub, 217 African flora, connection with New Zealand, 461 Agalenidae, 156, 165 agglutinans, Teatularia, 328 aggregata, Cladonia, 531 Agrostis, 473 Air-chambers in Stilbocarpa polaris, 206 alba, Triacanthella, 378 Albatros, — Effect on vegetation, 234 Royal, 572 Sooty, 575 Spreading of Acaena by, 235 Wandering, 566 albescens, Marginella (Volvarina), 29 albida, Myosotis, 423 albidus, Enchytraeus, 252, 262 albiflorum, Gewm, 403 albo-viridis, Sagartia, 385 albula, Bathytoma, 30 Alcira, 28 Alcyonaria, Campbell Island, 702 Aleocharidae, 96 Algae, marine: Report by Laing, 493 Alkaline trachytes, 691, 723 Allodiscus, 34. allomorphinoides, Discorbina, 357 Allophylopsis, 127, 142 - Allorchestes, 605, 645 alpina, Carpha, 433 alsinoides, Epilobiuwm, 406 Amaurobioides, 156, 162 Amaurobius, 156, 159 ambiguus, Loxomerus, 60, 92 ambiguus, Trophon, 26 ** Amherst,” cruise of the, xxvii amoena, Marginella (Glabella), 29 amphialus, Polinices (Lunatia), 22 Amplipoda, 604, 615 | Amphipoda, fresh-water, distribution of, 602 | amphipsila, Daphnella, 32 Amphiura, 303 | Amphiuridae, 303 824 SUBANTARCTIC ISLANDS OF NEW ZEALAND. Index. Volume II begins on page 389. Amsterdam Island, flora of, 462 Antarctic continent, 453 Anabathion, 18 Connection with South America, 461 Analyses,— Distances from subantarctic islands, 454 American peat, 748 Former milder climate of, 454 Antipodes Island rocks, 763 Fossil plants from, 454 Auckland Island basalt, 768 Antarctic expedition (Sir James Ross), xxiii, 390, 391, 392 Basalt, Adams Island, 733 Antarctic types of plants, 468 Basalt, Beeman Hill, 696 antarctica, Callitriche, 404 Basalt, Mount Honey, 696 antarctica, Catharacta, 556 Basic dyke, Adams Island, 736 antarctica, Danthonia, 475 Bounty Island granite, 770 | antarctica, Deguernea, 607, 671 Campbell Island argillite, 767 antarctica, Durvillaea, 499 Campbell Island basalt, 767 antarctica, Gentiana, 422 Campbell Island carbonaceous rock, 766 antarctica, Melobesia, 525 Diabase, Musgrave Peninsula, 727 antarctica, Myosotis, 422 Dolerite, Adams Island, 733 antarctica, Planispirina, 324 Enderby Island basalt, 769 antarctica, Pontogeneia, 605, 624 Gabbro, Campbell Island, 690 antarcticus, Juncus, 431 Gabbro, Carnley Harbour, 722 antarcticus, Lyperanthus, 427 Granite, Bounty Island, 739 antarcticus, Nectocarcinus, 604, 608 Granite, Musgrave Peninsula, 720 antarcticus, Scirpus, 433 Trish peat, 747 Antennaria, 529 Tron pan, 767 Anthomedusae, 306 Melilite basalt, Lyall Pyramid, 697 Anthomyudae, 126, 134 Olivine diabase, Musgrave Peninsula, 728 Anthornis, 554. Phonolite, Perseverance Harbour, 695 Anthracite, Auckland Island, 727 Pitchstone, Musgrave Peninsula, 724 Anthus, 553 Porphyrite, Carnley Harbour, 736 antipoda, Atropis, 480 Porphyry, Campbell Island, 691 antupoda, Gentiana, 422 Russian peat, 749 antvpoda, Photinula, 11 Sea-birds’ dung, 771 antipoda, Poa, 478 Sea-lions’ dung, 770 antupoda, Therasia (?), 35 Snares granite, 769 antipoda, var. rosea, Photinula, 11 Soils, methods, 756 antupodea, Calliphora, 125 Trachyte,— Antipodes Island,— Circular Head, Auckland Islands, 724 Basalt, occurrence of, 743 Filhol’s analysis of, 692 Bog, 229 Filhol Point, Campbell Islands, 694 Coal, occurrence of, 743 Mount Honey, Campbell Island, 692 Coastal rocks and clifis, 227 Mount Lyall, Campbell Island, 692 Description, 740 Musgrave Peninsula, Auckland Islands, 724 Discovery, xi Unweathered rocks of Campbell Island, 765 Forms of Coprosma cuneata, 201 Weathered rocks of Campbell Island, 765 Inland rocks, 230 Anas, 583 Lichens, 228 Anatidae, 583 Plant formations, 227 Anchomenidae, 95 Position, size, &c., xiv, 390 anchylochaeta, Timarete, 247 Serub, 229 Anemonia, 384. Soil, 763 anguiculus, Endodonta (Charopa), 37 Structure, 740 angulata, Lima, 44. Swamp, 229 angulata, Miliolina, 321 Tussock meadow, 227 angulosa, Uvigerina, 349 antipodum, Laevilitorina, 15 Animals, effect on distribution of plants, 191 antipodum, Ligusticum, 408 Animals, effect on vegetation, 232 antupodum, Marionina, 252, 262 Animals, introduced, xxviii, 235, 598 antipodum, Megadyptes, 580 Anisodactylidae, 91 antipodum, Myodora, 48 annaskala, Cyanea, 309 antupodum, Neolepton, 45 Annelid fauna, facies of, 236 antipodum, Trichocera, 124 Annelids, marine, 236 antipodus, Senecio, 417 annosus, Fomes, 529 Antiseptic action of humus in Thuringia, 755 Annuals, 195 Aora, 605, 645 annulata, Dicranomyia, 124, 129 Aoridae, 645 Anomalina, 359 Apanteles, 77 Anomia, 42 Aphroditidae, 237 Anomidae, 42 Aphrosylopsis, 125, 132 Anomura, 604 apiculatum, Ceramium, 521 Anorthoclase in trachyte, 690, 691, 723 Apium, 407 “ Antarctic,” visits Campbell Island, xxx Apophloea, 525 Indezx.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 825 Volume II begins on page 889. Appendix to enumeration of plants, 446 appressa, Carex, 434 Aptenodytes, 576 Aquatic structure in species of Cotula, 206 aquilina, var. esculenta, Pleris, 438 Aracoda, 245 Araliaceae, 409 Araneae, 156, 157 Araneus, 156, 162 araucana, Discorbina, 356 Arca, 42 Arcidae, 42 arctica, Saxicava, 48 Arctocephalus, 542 arenaria, Pratia, 419 Arenicola, 246 Arfvedsonite in trachyte, 692 Argillite, analysis of, 767 Argiopidae, 156, 162 Argiopinae, 156, 162 Argobuccinum, 22 argodes, Stereocaulon, 532 argus, Argobuccinum, 22 argus, Sphaerophorus, 530 Armstrong on geology of islands, 718 Armstrong on cruise of ‘‘ Amherst,”’ xxvii articulata, Cristellaria, 344 articulata, Usnea, 531 Articulina, 323 Asellota, 606, 648 Ash content in Irish soils, 753 Ash content in subantarctic soils, 753 asperifolia, Lepidozia, 524 Asperococcus, 498 asper, Sonchus, 418 asperula, Spiroculina, 318 Aspidium, 441 Aspleniwm, 440 assimilis, var. affinis, Arenicola, 246 Associations, plant, of Campbell Island, 482 Astelia, 429 Astelia pumila a peat-former, 751 Asterina, 295 Asterinidae, 295 Asteroidea, 295 Astoni, Poa, 477 “ Astrolabe,” 391 Astrorhizidae, 325 ater, Modiolus, 44 Athoracophoridae, 39 Athoracophorus, 39 atkinsoni, Schismope, 7 atomus, Rissoa (Setia), 18 Atropis, 480 attenuatus, Galaxias, 585 Atyloides, 605, 627 Auckland Island,— Anthracite, occurrence of, 727 Bogs, 226 Camp Cove conglomerate, 725 Coast-features, 706 Depth of sea near, 707 Easterly drainage of, 713 Radium content of rocks from, 779 Schists, occurrence of, 726 Swamps, 225 Auckland Islands,— Coleoptera of, 60 Described by Morrell, xxii Discovery, xii Diabases, occurrence of, 728 Diptera of, 64 Dunes, 211 Gabbro, occurrence of, 720 Gabbro, magnetic disturbance associated with, 722 General remarks re glaciation, 714 General remarks ve vegetation, 210 Gneisses, occurrence of, 726 Granite, occurrence of, 718 Harbours, 708 Hartmann’s account of basalt, 734 Tnaccuracies in chart of, 788 Insects of, 59 Lakes, 713 Lepidoptera from, 61, 70 Map showing magnetic stations, 785 Marine erosion, 708 Older basic series, 726 Peat-forming plants, 752 Physiography, 706 Pitchstone, occurrence of, 724 Plant formations, 211 Plutonic rocks, occurrence of, 718 Porphyrites, occurrence of, 727, 734 Position, size, &c., xiv, 389, 391 Recent elevation, 705 Shipwrecks on, xxiv, 788 Species of limited distribution, 210 Stratigraphy and petrology, 718 Surface-covering, 714 Surface-features, 712 Terraces, Circular Head, 714 Trachyte, occurrence of, 723 West coast cliffs, 707 Zonal distribution of vegetation, 210 Aucklandella, '75 aucklandensis, Apanteles, 77 aucklandiae, Deto, 606, 666 aucklandiae, Orchestia, 605, 634 aucklandica, Arca (Lissarca), 42 aucklandica, Huluma, 24 aucklandica, Gallinago, 556 aucklandica, Geoplana, 673 aucklandica, Limnophora, 126 aucklandica, Nesonetta, 583 aucklandica, Plantago, 426 aucklandica, Plaxiphora, 2 aucklandica, Poa, 478 aucklandica, Urtica, 426 aucklandicum, Calliostoma, 12 aucklandicum, Ceramiuwm, 520 aucklandicus, Atyloides, 605, 628 aucklandicus, Anthus, 553 aucklandicus, Cyanorhamphus, 555 aucklandicus, Notiodrilus, 253, 271 aucklandicus, Pelodrilus, 252, 265 aucklandicus, Plutellus, 253, 278 aucklandicus, Ranunculus, 398 aucklandicus, Rhizodrilus, 252, 258 aucklandicus, Scirpus, 433 aucklandicus, Trophon (Kalydon), 27 aucklandicus, var. bollonsi, Notiodrilus, 272 aucklandicus, var. pallidus, Notiodrilus, 273 826 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Index. Volume II begins on page 889. aucklandicus, Zaluscodes, 64, 130 benhami, Catadryobius, 60, 110 auckland, Anguilla, 585 benhami, Chirodota, 151 Aurelia, 309 benhami, Endodonta (Charopa), 37 australe aucklandicum, Mesodesma, 46 Benthami, Veronica, 425 australe, Polypodiwm, 442 Beris, 124 Australian flora, connection with New Zealand, 461 Berkleyi, Heterosiphonia, 512 australis, Balaena, 549 bertheloti, Discorbina, 356 australis, Brogniartella, 514 bertheloti, var. baconica, Discorbina, 356 australis, Cotula, 416 “< Betsy,” voyage of, xx australis, Haliotis, 8 biancae, Lagena, 337 australis, Haplophthalmus, 606, 662 Bibliography, general, 808 australis, Leptomithrax, 604, 607 bicarinata, Omalogyra, 19 australis, Merganser, 584. biconcava, Discorbina, 357 australis, Modiolus, 43 bifarvum, Rhizogonium, 536 australis, Mopalia, 2 bifida, Rhodophyllis, 508 australis, Natica, 21 bifolia, Caladenia, 428 australis, Nereis, 236, 238 Billardiert, Colobanthus, 401 australis, Nesteraz, 555 Billardiert, Dicranoloma, 535 australis, Sphaerophorus, 530 Billardiert, Polypodium, 443 australis, Urtica, 426 Biloculina, 313 australis, Venericardia, Campbell Island, 701 Biotite in gabbro, 722 austrina, Paramoera, 605, 625 Birds, effect on vegetation, 232 avenacea, Microlaena, 472 Birds transporting seeds, 459 Aves, 551 bispinatus Hwprotomicrus, 585. avivorus, Catops, 60, 101 biternatus, Ranunculus, 398 Azimuth, Mount, trachyte, 691 Bittium, 19 Azorella, 407 bivalve, Hymenophyllum, 437 Azorella Selago, its station, 230 Blackbird, 584 Blackfish, 551 blainnillianus, Syngnathus, 588 B. Blennidae, 597 Bligh, W., discovers Bounty Islands, xi Bacteria, effect of humus acid on, 755 Blight-bird, 553 Badumna, 156, 160 Blooming of herbaceous plants, 204 Baeopterus, 127, 141 Bog, Antipodes Island, 229 Baeostethus, 79, 96 Bog-plants, xerophily of, 191 Baker: Report on Auckland Islands, xxvii Bogs, Auckland Island, 226 Balaena, 549 Bog-slides, 753 Balaenidae, 549 Bog-slides in Falkland Islands, 751 Balaenoptera, 550 Bolivina, 331 Balanidae, 607, 670 bollansi, Galaxias, 586 Balanus, 607, 670 Bollons Islands, description, 740 Ballia, 520 bollonsi, Orchestia, 605, 635 Banguaceae, 503 Boraginaceae, 422 bankswi, Prion, 565 Borengrevinek, C. KE , xxx Barometric records of Campbell Island, 790 Boryana, Marginaria, 502 barresis, Clynotis, 157, 176 Botanical collections, 396 Bartramia, 536 Botanical exploration, history of, 390 Basalt,— Botanical investigations of subantarctic islands, 184 Adams Island, 731 Botany, ecological, of subantarctic islands, 182 Antipodes Island, 743 Botany, systematic, by Cheeseman, 389 Auckland Island, analysis of, 768 botryocarpa, Microcolax, 516 Campbell Island, 695 boueana, Nonionina, 364 Enderby Island, analysis of, 769 Bounty Islands,— Hartmann’s account of, 734 Discovery by Bligh, xi North Arm, Carnley Harbour, 733 General description, 738 Basidiomycetes, 528 Granite, 739 Bass: Proposals for fishing monopoly, xii Position, size, &c., xiv, 390 Bathytoma, 30 Radium content of rocks from, 779 beddomei, Schismope, 7 Weta from, 59 Bell-bird, 554 Bouvet Island, 456 bellidioides, var. prostratum, Helichrysum, 415 Bovallia, 605, 622 Bellingshausen visits Macquarie Island, xxi Bovichtus, 595 Benham, W. B.,— Bowdleria, 552 ' Note on Myriapoda, 376 brachiocarpum, Plocamium, 511 On echinoderms, 295 Brachiopoda, Campbell Island, 700 On Hirudinea, 272 brachysoma, Diporochaeta, 253, 284 On Hydromedusae and Scyphomedusae, 306 Brachyura, 604, 607 Report on Oligochaeta, 251 Braconidae, 77 Report on Polychaeta, 236 | brad, Biloculina, 314 ee a ne ae eS a Se Index.} SUBANTARCTIC ISLANDS OF NEW ZEALAND. 827 Volume II begins on page 389. bradii, var. denticulata, Biloculina, 315 Branchiopoda, 607, 671 Breutelia, 536 brevidentis, var. carnleyensis, Cucumaria, 149 brevipinnis, Galaxias, 586 brevis, Pristancylus, 95 brevis, Schismope, 8 Bristow, Cape, 707 Bristow, Captain, discovers Auckland Islands, xii British birds, 584 Brogniartella, 514 Brotherus, W. F., on Musci, 535 Broun, T., on Coleoptera, 78 Brownit, Caulerpa, 496 Brownti, Plantago, 426 Brunonis, Hierochloe, 473 Bryaceae, 536 bryobia, Limnophila, 124 Bryophytes, peat-forming, 215 Buccinidae, 26 bucculenta, Planispirina, 324 bucculenta, var. placentiformis, Planispirina, 324 Buchanan, John, visits Campbell Island, xxix, 395 bufonius, Juncus, 430 bulbiferum, Aspleniwm, 441 Bulbinella, 430, 599 Bulbinella Rossii, description, 188 bulbosa, Seila, 20 bulbosum, Cyathophorum, 537 Bulimina, 330 Bulimininae, 330 Bull, H. J., on cruise of ‘‘ Antarctic,’ xxx bulleri, Diomedea, 575 bulloides, Globigerina, 350 bulloides, Sphaeroidina, 351 Bush-hawk, 555 Byrrhidae, 102 byrsopola, Proterodesma, 64, 74 C. Cacoecia, 73 Cadulus, 42 caeca, Geonemertes, 675 caerulescens, Zosterops, 553 caespitosa, Dechampsia, 481 Caladenia, 428 Calanidae, 671 Calathus, 95 Calcareous humus soil, 762 calliblepharoides, Callophyllis, 507 Calliopiidae, 621 Calliostoma, 12 Calliphora, 125, 134 Callithamnion, 517 callitricha, Ballia, 520 Callitriche, 404 Callochiton, 2 Callocolax, 507 Callophyllis, 507 Cameron, P., on Hymenoptera from Auckland Island, 75 Campanulaceae, 419 Campbell Island,— Absence of rata, xxvii, 496 Absence of trees, 483 Alcyonaria, 702 Barometric records, 790 Basalt of, 695 Botanical investigations, 482 Campbell Island—continued. Brachiopoda, 700 Coleoptera of, 66 Coprosma scrub, 488 Crinoidea, 700 Diptera of, 66 Discovery, xii Dykes, 696, 699 Eehinoidea, 700 Faults, 699 Filhol’s account of, 680, 685, 688, 699 Flint, 698 Foraminifera, 702 Gabbro, 690 Gasteropoda, 700 Geological iistory, 688 Geological structure, 685 Geology, 680 Glaciation, 689 Hemiptera of, 66 Insects of, 65 Introduced plants, 492 Lepidoptera of, 65 Limestone, 764 Limestone, age of, 702 Limestone, occurrence of, 697, 699 Littoral vegetation, 488 Map showing magnetic stations, 783 Marble, occurrence of, 698 Melilite basalt, occurrence of, 697 Meteorology of, 789 Mudstone, occurrence of, 698 Neuroptera of, 66 Palaeontology, 700 Pelecypoda, 701 Petrography, 690 Phonolite, 694 Plant formations, 482 Plants erroneously recorded from, 486 Pleurophyllum Hooker formation, 489 Porphyry, 690 Position, size, &c., xiv, 389, 391 Quartz, occurrence of, 698 Radium content of rocks from, 779 Sequence of volcanic rocks, 698 Soil, 762 Subalpine tussock meadows, 489 Surface, 683 Table of climatic features, 792 Trachyte, 683, 691 Uniformity of plant covering, 487 Weather on, 789, 791 Winds, 789, 790 | campbellensis, Celmisia, 414 | campbellensis, Pseudosphaeroma, 606, 654 campbelli, Acmaea, 5 campbelli, Balanus, 607, 670 campbelli, Cominella, 26 campbelli, Bupagurus, 604, 612 campbelli, Galaxias, 585 campbelli, Onchidella, 41 campbelli, Phalacrocorax, 580 campbelli, Polypus, 48 campbellianus, Notiodrilus, 253, 273 | campbellianus, Phreodrilus, 252, 256 | campbellica, Hyale, 605, 644 campbellica, Laoma (Phrixgnathus), 38 | campbellicus, Helophilus, 125 Camp Cove conglomerate, 725 828 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Index. Volume II begins on page 389. campestris, var. crimta, Luzula, 432 campestris, var. floribunda, Luzula, 432 Camptochaete, 537 Camptonites, Auckland Island, 736 Campylium, 537 Campylopus, 535 canaliculata, Cerithiopsis, 20 canariense, Haplophragmium, 327 canariensis, Pulvinulina, 361 Cancellariidae, 28 Cancer, 604, 608 Candeina, 352 candeina, Textularia, 329 cannabina, Linota, 584 Cantharidus, 10 cantharus, Acmaea, 5 Cape hen, 564 Cape pigeon, 564 capensis, Daption, 564 capensis, Lomaria, 440 capitata, Myosotis, 423 capito, Kenodactylus, 91 Caprellidae, 648 Caprellidea, 605, 648 Caprellinopsis, 605, 648 Capulidae, 21 Carbonaceous rock, Campbell Island, 766 Cardamine, 398 Cardiudae, 47 Carditidae, 45 Carex, 434 Carmivora, 542 Carnley Harbour,— Depth of, 710 General description, 708 Structure of surrounding country, 729 Western entrance, 708 Carolobatea, 605, 620 Carpenter, G. H., on Collembola, 377 Carpha, 433 Carpophyllum, 502 Caryophyllaceae, 401 Cascade Inlet,— Hanging valley, 715 Map of, 716 Cassidulina, 332 Cassidulininae, 332 Cassinia, 415 Cassinia Vawvilliersii,— Life-form, 195 Not found in Campbell Island, 486 Variation according to environment, 201 castaneus, Pristancylus, 95 castigatum, Diglymma, 58, 84 Catarractes, 576 Catharacta, 556 Catodryobius, 79, 108 Catops, 101 Cattle on Auckland Islands, 598 Caulerpa, 496 Caulerpaceae, 496 celata, Sigmoilina, 323 Celmisia, 414 Celmisia vernicosa,— Description of, 188 Variation in colour, 203 Variation of, 202 Centrolepidaceae, 432 cephaloscypha, Marchantia, 532 Ceramiaceae, 517 Ceramium, 520 | ceratoclada, Herposiphonia, 516 cerina, Gentiana, 421 Cerithiidae, 19 Cerithiopsidae, 19 Cerithiopsis, 19 cernuus, Scirpus, 433 cessicus, Cerithiopsis, 20 Cetacea, 549 Chaetangiaceae, 505 Chaetangium, 505 chambersi, Glycumeris, Campbell Island, 701 Chapman, F., on Foraminifera, 312 Chapman, Mr. Justice,— Account of Fly Harbour, 708 Paper on subantarctic islands, xxix, 184, 395 Chapman, Deschampsia, 474 charidema, Venusia, 61, 65, 70 Charidriidae, 556 chariessa, Daphnella, 32 chathamensis, Trochus (Infundibulum), 9 Chatham Island, discovery of, xi Cheeseman, T. F., on systematic botany, 389 Cheiloscyphus, 534 cheilostoma, Rissoa, 16 Chiloglottis, 428 chiloscyphoides, Leiocyphus, 533 Chilton, C.,— On biological relations of islands, 793 On Crustacea, 601 On scientific investigation of islands, xiv Chiltonia, 605, 644. chiltoni, Baeostethus, 97 chilton, Rissoina (Hatoniella), 18 Chirodota, 151 Chione, 46 Chiton, 4 Chitonidae, 4 chlamydophyllum, Acrocladium, 538 Chloroclystis, 70 Chloroclystis sp., 65 Chlorophyceae, 493 Choleva, 101 Chonos Archipelago, peat-forming plants, 751 Chorda, 498 Chordaria, 497 Chordariaceae, 497 chrysocome, Catarractes, 576 Chrysodomidae, 25 chrysogaster, Stereodon, 538 chrysostoma, Miliolina, 322 Chydorus, 607, 671 ciliata, Coprosma, 411 ciliata, Gaimardia, 432 ciliatus, Cheiloscyphus, 534 cilicollis, Loxomerus, 94. cincta, Litorina (Melaraphe), 15 cinerea, Otaria, 548 Circular Head,— Gabbro, 720 Trachyte, 723 circularis, Miliolina, 318 circularis, var. sublineata, Miliolina, 319 Circulus, 13 Cirolana, 606, 651 Cirques,— Auckland Islands, 715 Campbell Island, 684 Index.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 829 Volume II begins on page 389. Cirratulidae, 247 Cirripedia, 607, 670 Cirsonella, 13 Cladocera, 607, 671 Cladonia, 531 Oladoniaceae, 531 Cladophora, 494 Cladophoraceae, 494 clathrata, Enteromorpha, 494 clathrata, Rotalia, 362 clathrata, Trichotropis, 22 clavigera, Phyllachne, 419 Climate,— The subantarctic, 190 With regard to formation of peat, 191 Climatic features of Campbell Island, 792 Climatic physiognomy, 185, 186 clivinoides, Oopterus, 86 closterbranchia, Syllis, 237 Clubioneae, 156, 162 Clubionidae, 156, 162 Clubioninae, 156, 162 Clynotis, 157, 176 clypidellaeformis, Cocculina, 15 Coastal cliffs, vegetation of, 212 Coastal formations,— Auckland Islands, 706 Campbell Island, 680 Snares, 209 Coastal-rock formation, 211 Coastal rocks, Auckland Islands, 211 Coastal rocks and cliffs, Antipodes Island, 227 coccineum, Plocamium, 511 Cocculina, 14 Cocculinidae, 14 cochlearifoliwum, Lembophyllum, 537 Cockayne, L.,— Botanical survey of Stewart Island, 717 On ecological botany, 182 On soil of Auckland Islands, 752 Winter visit to subantarctic islands, 185, 396 cockaynet, Calliphora, 125 cockaynei, Hycanus, 116 cockayni, Pacificana, 157, 176 cockayni, Rhododrilus, 253, 277 cockayni, var. waterfieldi, Rhododrilus, 278 Codiaceae, 496 Codium, 496 Coelopa, 126, 139 Coenosia, 126, 138 coerulea, Aurelia, 309 cognata, Laoma (Phrixgnathus), 38 colbecki, Notothenia, 594 colensot, Phalacrocorax, 581 Colensoi, Prasophyllum, 427 Coleoptera,— Distribution of, 78, 80 From the subantarctic islands: Descriptions by Broun, | 78 Of Auckland Islands, 60 Of Campbell Island, 66 Of Snares, 58 Coleridge Bay,— Cirque, 715 Lake, 713 Moraines, 714 Trachytes, 725 Collections, botanical, 396 Collema, 529 Collembola, 377 Collembola, report on, by Carpenter, 377 Oolobanthus, 401 Colobanthus muscoides, size of cushions, 209 colorata, Lima, Campbell Island, 701 Colours of flowers, 199 Colour-variation,— Aciphylla latifolia, 203 Celmisia vernicosa, 203 Gentiana cerina, 202 Veronica benthami, 203 columbina, Porphyra, 505 Columnar structure in basalts, 731 Cominella, 26 comma, Lepidasthenia, 237 communis, Nodosaria, 341 communis, Polymorphina, 348 comosa, Breutelia, 536 comosum, Mniodendron, 538 Compositae, 412 Composition of humus, 747 Composition of peat-ash, 747 Composition of Pleurophyllum meadow, 221 Composition of rata forest, 213 compressa, Cristellaria, 343 compressa, Hnteromorpha, 494 compressa, Polymorphina, 348 concava, Textularia, 328 concentrica, Pulvinulina, 360 Conchothyra, Campbell Island, 700 concinna, Gentiana, 422 confertifolium, Epilobium, 406 Conglomerate, Camp Cove, 725 conica, Stylohalina, 539 connivens, Pallavicinia, 533 consobrina, Nodosaria, 341 consobrina, var. emaciata, Nodosaria, 341 constanceae, Ranfurlya, 37 constrictus, Helodrilus, 253, 289 Continental connection of Auckland Island, evidence tor, 705 contracta, Festuca, 479 contractus, Ischnochiton, 1 contractus, Zaus, 607, 671 contraria, Bulimina, 331 Copepoda, 607, 671 cophinodes, Mangilia, 31 Cook, Captain, xi Coprosma, 411 Coprosma ciliata, reversion shoots, 204 Coprosma cuneata, form on Antipodes Island, 201 Coprosma foetidissima not found in Campbell Island, 486 Coprosma foetidissima, variation according to environ- ment, 200 Coprosma parviflora, variation according to environment, 201 Coprosma scrub of Campbell Island, 488 coracina, Monodonta (Diloma), 10 Corallina, 524 corallina, Rhodymenia, 510 Corallinaceae, 524 corbis, Venericardia (Miodon), 45 cordiformis, Pinnoctopus, 49 Coriaceous leaves, 199 corticeps, Notothenia, 585 Cornuspira, 325 cornuta, Chiloglottis, 428 830 SUBANTAROCTIC ISLANDS OF NEW ZEALAND. [Indec. Volume II begins on page 389. cornuta, Cladonia, 532 coronata, Anomalina, 360 corrugata, Oouthouyia, 15 corrugata, Patellina, 354 Corysanthes, 428 Cossyrite in phonolite, 694 costata, Marginulina, 342 costata, Philobrya, 43 Ootula, 416 Cotula species, their aquatic structure, 206 Couthouyia, 15 Crambidae, 71 crassa, Chione, 47 crassicostata, Leptothyra, 14 crassilirata, Daphnella, 32 crassinervium, Hypoglossum, 512 crassipes, Parapherusa, 605, 630 Crassulaceae, 404 Creeping form, 198 crenistria, Oerithiopsis, 19 crepidula, Oristellaria, 343 criniferum, Pleurophyllum, 413 Crinoidea, Campbell Island, 700 crispa, Polystomella, 364 crispatum, Platyclina, 511 crispulatus, Trophon (I'rophonopsis), 27 Cristellaria, 342 Cristellaria, Campbell Island, 702 Crossea, 23 Crozets, 455 Cruciferae, 398 cruciferum, Plocamium, 511 cruciferus, Rubrius, 157, 167 Crustacea,— Distribution of, 602 Fresh-water and terrestrial, 602 Report on, by Chilton, 601 Cryptogams, list of, 528 cryptopterum, Callithamnion, 517 Cryptorhynchidae, 120 Ctenodoris, 33 Cubaris, 606, 668 Cucumaria, 146 cultrata, Cristellaria, 345 Cumacea, 669 cuneata, Coprosma, 411 cunetfolia, Huzoniella, 514 curtus, Trophon ( Kalydon), 27 curvipes, Coelopa, 126 curvipes, Trichopticus, 126, 138 Cushion form, 196 Cushion form, variation according to environment, 201 Cushion plants, water-storage, 197 Cushions, size in Colobanthus muscoides, 209 Cyanea, 309 Cyaneidae, 309 Cyanorhamphus, 554- Cyathodes, 419 Cyathophorum, 537 Cyboeeae, 156, 165 Cyboeinae, 156, 165 Cyclostrema, 13 Cyclostrematidae, 13 Cyclostremella, 13 Cymodocella, 606, 657 Cymothoidae, 651 Cyperaceae, 433 Cystophora, 502 cystostegia, Aspidium, 442 D. Dachnowski on xerophily in bog-plants, 191 dannefordi, Muro, 552 Danthonia, 475 Danthonia antarctica, description of, 186 Danthonia meadow, 218 Danthonia meadow, scrubby undergrowth, 219 Daphnella, 32 Daphniidae, 671 Daption, 564 Darbishire, A. D., on land nemerteans, 674 debilis, Coelopa, 127, 140 Decapoda, 604 decipiens, Polysiphonia, 513 decipiens, Stellaria, 401 Declination, annual rate of change, 787 decorata, Spirillina, 353 decorus, Balanus, 607, 670 decumanus, Mus, 600 deflexa, Camptochaete, 537 Degquernea, 607, 671 Delano, Captain, sights the Snares, xii, xxii Delesseriaceae, 511 deliculata, Pecten, Campbell Island, 701 Delphinidae, 551 demissa, Puncturella, 9 demissum, Hymenophyllum, 436 Dendy, A., On holothurians, 146 On land planarians, 672 densilirata, Cirsonella, 13 Density of mountain scrub, 216 Dentaliidae, 41 Dentalium, 41 denticulo-granulata, Spurillina, 354 Depots tor castaways, xxvii Depranodes, 70 depressa, Biloculina, 313 depressa, Cardamine, 400 depressa, Nertera, 412 depressa, Paralimnophora, 126, 136 depressa, vax. stellata, Cardamine, 400 depressula, Nonionina, 363 depressum, Hymenosoma, 604, 610 Deschampsia, 474 Deschampsia caespitosa, 481 Desmarestia, 496 Desmurestiaceae, 496 desolatus, Prion, 566 Deto, 606, 666 devia, Mangilia, 31 Dexaminidae, 632 Deyeuxia, 473 Diabase, Auckland Island, 727, 734 Diabase-porphyrite, Auckland Island, 727 diaphanum, Ceramium, 520 dichogama, Anemoma, 384 dichotoma, Rhodymenia, 510 dichotoma, Schizoneura, 512 Dicranaceae, 535 Dicranoloma, 535 Dicranomyia, 124, 129 Dictynidae, 156, 159 Dictyosiphonaceae, 497 dictyota, Mangilia, 31 Diels, L. : Ecology of New Zealand flowering-plants, 185 difficilis, Turritella, 21 difficilis, Venericardia, 45 — Ne ee ee ee ee Index.} SUBANTARCTIC ISLANDS OF NEW ZEALAND. 83] Volume II begins on page 389. Diglymma, 79, 83 dilatum, Hymenophyllum, 436 dioica, Cotula, 417 Diomedea, 566 Diomedeidae, 566 Diporochaeta, 253, 281 Diporochaeta perionychopsis, affinities and distribution of, 288 Diptera,— Aetiology and distribution of, 129 General remarks on, 128 Of Auckland Islands, 64 Of Campbell Island, 66 Of Snares, 59 Of the subantarctic islands of New Zealand: Report by Lamb, 124 Preservation of, 128 Dipterosiphonia, 516 Disappointment Island,— Description of, 737 Meadow formation, 222 Melaphyre, 738 Pleurophyllum meadow, 225 Rhyolite, 738 discolor, Lomaria, 439 Discomedusae, 309 Discorbina, 354 Discovery of the islands, xi ** Discovery ” visits Macquarie Island, xxxiv dissimilis, Seila, 20 Distichophyllum, 537 Distribution,— Effect of life-forms of plants on, 191 Of Mollusca, table showing, 49 Of Oligochaeta, 254 Of phanerogams, ferns, and Lycopodiaceae, 448 Of plants, effect of animals on, 191 Of plants of subantarctic islands, 457, 466 Of spiders and Opiliones, 155 Zonal, in Auckland Islands, 210 Ditrichaceae, 535 Ditrichum, 535 divaricata, Gigartina, 505 divaricata, Myrsine, 421 Divaricating form of shrubs, 194 Dolerite,— At Auckland Island, 732, 736 At Campbell Island, 696 Dolichopidae, 125, 132 Domestic poultry in Campbell Island, 600 domesticus, Passer, 584 dominicanus, Larus, 560 donaciformis, Rochefortis, 45 Dorididae, 33 Doris, 33 Dougherty Island, xx, 456 dracophyllae, Brirhinus, 61, 118 Dracophyllum, 419, 554 Dracophyllum form of shrubs, 195 Dracophyllum longifoliwm,— Heterophylly in, 203 Leaves, 194 Trunk, 194 Variation according to environment, 200 Drillia, 30 Drosera, 404 Droseraceae, 404 Drude on vegetation of New Zealand, 192 Dryness, physiological, 191 Dryomyzidae, 127, 143 dubia, Liljeborgia, 605, 619 Duck,— Blue, 584 Flightless, 583 Grey, 583 Dumas, Mount, trachyte, 691 Dumontia, 524 Dumontiaceae, 524 dumosa, Polysiphonia, 514 Dunes,— Absence of Scirpus frondosus, 211 Increase of Ruwmex neglectus on, 211 Of Auckland Islands, 211 Dung,— Sea-birds’, analysis of, 771 Sea-lions’, analysis of, 770 dura, Lomaria, 439 Durvillaea, 499 D’Urville, Admiral, xxii, 390, 393 dutertrei, Globigerina, 350 Dykes,— Basalt, at Campbell Island, 687, 696 Camptonitic, at Auckland Island, 736 Diabase, at Auckland Island, 734 Porphyrite, at Auckland Island, 734 Trachyte, at Auckland Island, 723 Dynamenella, 606, 657 E. Eboulé, Mount, trachyte, 691 Echinasteridae, 298 echinata, Leucosolenia, 539 echinatus, Asperococcus, 498 Echinidae, 304 Echinoderms,— Bibliography of, 305 Report by Benham, 295 Echinoidea, 304 Echinus, 304 Ecological botany of subantarctic islands of New Zea- land, 182 Keological factors, 190 Ecology of New Zealand flowering-plants (L. Diels), 185 Ecology, special, of the plants, 192 Edaphie physiognomy, 186 edentula, Ornithobdella, 374 edulis, Mytilus, 43 edwardsii, Porcellanopagurus, 604, 610 edwardsii, Prionorhynchus, 604, 608 effusus, Juncus, 430 Ehlers on Polychaeta, 236 Ehrenbergina, 332 elegans, Carduelis, 584 elegans, Pulvinulina, 362 elegantissima, Bulimina, 330 elegantissima, Polymorphina, 349 elegantissima, var. apiculata, Bulimina, 330 Elevation of land, Auckland Island, 705, 717 elluptica, Veronica, 424 elongata, Biloculina, 317 elongata, Idotea, 606, 658 elongellus, Oopterus, 66, 89 Emarginula, 9 Emerald Island, xviii empetrifolia, Cyathodes, 419 Empidae, 125 832 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Indeax. Volume II begins on page 389. Empis, 125 Enchytraeidae, 252, 261 Enchytraeus, 252, 256, 262 Encoeliaceae, 498 “Endeavour,” sealing trip of, xu Endemic species of plants in subantarctic islands, 465, 467 enderbei, Trioenonyx, 157, 158 enderbiensis, Lagena, 339 Enderby, Charles, xxii Enderby Island,— Columnar basalt, 731 Sand-dunes, 712 Enderby, Messrs., whaling by, xii Enderby Settlement, xxiv Endodonta, 37 Endodontidae, 37 EHnteromorpha, 494 Entomology, general notes on, by Hudson, 58 Enumeration of phanerogams, 397 Hpacridaceae, 419 epentroma, Mangilia, 31 epicomia, Scoparia, 62, 72 Epilobiwm, 406 Epitoniidae, 23 LEipitonvum, 23 Lipymenia, 510 erebia, Melanchra, 61, 68 “Hrebus,”’ 391 Maenetic observations by, 786 Hrechtites, 417 Hrechtites sp., 417 erectus, Trophon (Kalydon), 27 Hrirhinidae, 118 Hrirhinus, 118 erubescens, Catadryobius, 61, 111 erythrotis, Cyanorhamphus, 555 Hucopepoda, 607, 671 Hucopidae, 307 eudypti, Calliphora, 125 Hulima, 24 Hulimidae, 24 eumorpha, Cyclostrema, 13 Humeidae, 244 Hupagurus, 604 Huptilota, 518 HButhria, 518 Huzoniella, 514 Evergreen herbs, 196 Expedition,— French (D’Urville), xxii, 390, 393 French (Transit of Venus), xxix, 394 To New Zealand subantarctic islands, in Kew Bulletin, 185 exigua, Planispurina, 323 exigua, Pulvinulina, 361 Exosphaeroma, 606, 652 expeditionis, Thermia (7), 34 exserta, Rissoa (Alvania), 16 exsul, Pelecanoides, 566 exulans, Diomedea, 566 F. Factors, the ecological, 190 Fairchild’s Garden, 223 Falconidae, 555 Falkland Islands, peat in, 751 fallax, Notiodrilus, 253, 268 falaiatus, Rubrius, 157, 170 Families of plants in subantarctic islands and New Zea- land, 463 Karr, ©: ©: On magnetic observations, 782 On radium content of rocks, 778 fascelina, Triphora, 21 fasciata, Lagena, 338 fasciatus, Galaxias, 586 fasciculatus, Dictyosiphon, 497 fastigiatum, Lycopodiwm, 444 Faults, Campbell Island, 699 ~ Favorite.” sealing by, xii Feilitzen, Von, on carbon in peat, 753 Fern-bird, 552 Ferns,— Hnumeration of, 435 Tabular view of distribution, 452 ferussacti, Miliolina, 322 Festuca, 479, 481 Filhol, H., 394 - Filhol, H., “‘ Mission de l’Ile Campbell,” xxix, 680, 688, 699 Filhol Peaks trachyte, 691 Filhol Point pantellarite, 694 filholi, Notothema, 585 Filices, 435 filicina, Sticta, 531 filiformis, Deyeuxia, 474 filiformis, Dumontia, 524 filiformis, var. Lyallin, Deyeuxia, 474 filiformis, var. micranthera, Deyeuxia, 474 filipennis, Coenosia, 126, 138 fimbriata, Asterina, 295 fimbriata, vax. simplex, Cladonia, 532 Fire, reproduction of plant covering after, 231 firma, Cladoma, 531 Fissurellidae, 8 Fissuridea, 9 fistulosa, var. australis, Schizaea, 443 flabellatum, Hymenophyllum, 436 Plabellifera, 651 flabellifera, Doris (Ctendoris), 33 flabellifera, Huzoniella, 514 Flabellum, Campbell Islands, 702 flaccidum, Collema, 529 flagelliformis, Chordaria, 497 Flammulina, 36 flavipes, Calliphora, 125, 134 flavolineata, Aucklandella, 77 flawomaculata, Aucklandella, 76 flexicostata, Mangilia, 31 Flint, Campbell Island, occurrence of, 698 «‘ Flora Antarctica,” 184 Florance, D. C. H., on radium content of rocks, 778 Flora of Kerguelen, affinities of, 458 Flora of subantarctic islands, affinities, 453 Flora of subantarctic islands, history of, 453 Florida, humus soil of, 749 Flowers, 199 Flowers, colours of, 199 fluctuata, Leptothyra, 14 fluctuata, var. immaculata, Leplothyra, 14 fluviatilis, Lomaria, 440 foetidissima, Coprosma, 411 foliata, Rissoa (Onoba), 16 foliosa, Poa, 476 Fomes, 529 fontana, Montia, 402 Foraminifera,— Campbell Island, 702 Report on, by F. Chapman, 312 Index.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. Volume II begins on page 3889. Forest,— Geological history, Campbell Island, 688 Olearia Lyallii, 208 Rata, 213 Snares, 208 Formation,— Pleurophyllum, 223 Poa litorosa, 221 Swamp and bog, 225 formosissima, Hwuptilota, 518 forsteri, Arctocephalus, 548 Fossaridae, 15 fossulatus, Loxomerus, 60, 93 foveauxiana, Rissoa (Ceratia), 16 fragilis, Sphaerophorus, 530 fragillima, Sticta, 531 fraxinea, Homalomyia, 126 French expedition (D’ Urville), xxii, 390, 393 French expedition, Transit of Venus, xxix Fresh-water Crustacea, distribution of, 602 Freyconetii, Sticta, 531 frontalis, Hycanus, 61, 116 frontalis, Sterna, 560 Fucaceae, 499 Fuegian affinity of New Zealand flora, 459, 460 Fuegian plants in subantarctic islands of New Zealand. 460, 467 fuegiensis, Nacella (Patinigera), 6 fuliginosa, Homalomyia, 126 fuliginosa, Phoebetria, 575 Fulmar, mottled, 563 Fulmar, silver-grey, 563 fulva, Bowdleria, 552 fulva, Lophocolea, 533 fumata, Rissoa (Ceratia), 16 fumipennis, Coenosia, 126, 139 funalis, Articulina, 323 Fungi, 528, 529 funiculare, Stypocaulon, 496 furcata, var. racemosa, Cladonia, 531 Fur-seal, 548 G. Gabbro,— Association with magnetic disturbance, 722 Auckland Islands, 720 Gaimardia, 432 Galaxias, 586 Galaxtidae, 586 Gallinago, 556 Gammaridae, 630 Gammaridea, 604, 615 Garden Cove, Campbell Island, 697 Garden, Fairchild’s,"223 Garrodia, 562 | Gasteropoda, Campbell Island, 702 gavia, Puffinus, 563 Gaviae, 556 gemmata, Mitromorpha, 30 Gentiana, 421 Gentiana cerina, variation in colour, 202 Gentianaceae, 421 Geographical relations of Oligochaeta, 253 Geology,— Antipodes Island, 740 Auckland Islands, 705 Bounty Islands, 738 Campbell Islands. 680 Disappointment Island, 737 Snares, 703 54—S., Geonemertes, 674 Proboscideal nerves, 676 Geoplana, 672 _ Geraniaceae, 402 Geranium, 402 Geum, 403 gibba, Cristellaria, 344 gibba, Polymorphina, 348 gibba, var. racemosa, Polymorphina, 348 gibbosa, var. tuberosa, Textularia, 328 giganteus, Macronectes, 564 Gigartina, 505 Gigartinaceae, 505 gigas, Bxosphaeroma, 606, 652 glabella, Crossea, 23 glaberrima, Metzgeria, 532 glacialis, var. subcarnosa, Cardamine, 398 glacialoides, Priocella, 563 Glaciation,— Cause of, 689, 717 Evidence for former, Auckland Islands, 714 Evidence for former, Campbell Island, 689 In subantarctic islands of New Zealand, 469, 470 Park, J., Remarks on, 716 gladiata, Xiphophora, 500 Globigerina, 350, 702 Globigerinidae, 350 Globiocephalus, 551 globosa, Lagena, 333 globularis, Biloculina, 355 globulus, Discorbina, 317 Glyceridae, 246 Glycimeris, TO1 Gnaphalium, 415 Gnathobdellida, 374 Gneiss in conglomerate, Auckland Islands, 726 Goats, 599 Godwit, 556 Goldfinch, 584 Goniodorididae, 33 gonosemana, Heterocrassa, 64, 74 ** Governor Bligh,” sealing trips of, xxi gracile, Callithamnion, 517 gracile, Leptostomum, 536 gracilis, Lagena, 333 gracilis, Rostkovia, 430 gracilis, var. Campbelliana, Cladonia, 531 gracillima, Deschampsia, 475 gracillimus, Galaxias, 587 gradatum, Anabathron, 18 “Grafton,” wreck of. xxiv gramen, Textularia, 328 | gramen, var. jugosa, Textularia, 329 Gramina, report on, by Petrie, 472 Gramineae, 435 grammitidis, Polypodiwm, 443 grandis, Catadryobvus, 61, 112 Granite,— Bounty Island, 739 Bounty Island, analysis of, 770 Carnley Harbour, in conglomerate, 726 Musgrave Peninsula, Carnley Harbour, 718 Snares, 704 Snares, analysis of, 769 granosus, Turbo (Modelia), 14 Gratelowpiaceae, 524 Griffithsia, 517 834 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Index. Volume II begins on page 389. griseus, Puffinus, 563 grosserugosa, Anomalina, 359 guanicola, Thomosis, 100 gularis, Oestralata, 563 Gull,— Black-backed, 560 Mackerel, 561 Gypsina, 362 H. hadingert, Truncatulina, 359 Haliacris, 606, 650 Halicarcinus, 604, 609 Haliotidae, 8 Haliotis, 8 Hall on nitrogen in peat, 753 halopis, Scoparia, 63, 72 Haloragaceae, 404 EC Haloragis, 404 394 hamiltoni, Myro, 157, 172 Hamiltoni, Poa, 477 Hanging valleys,— Auckland Island, 715 Campbell Island, 685, haplochaeta, Leprea, 248 haplocystis, Notiodrilus, 252, 270 Haplophragmium, 327 Haplophthalmus, 606, 661 Haplotawidae, 252, 263 Haplotaxidae, distribution of, 255 Harbours,— Auckland Islands, 708 Campbell Island, 681 Harker: Tertiary igneous rocks of Skye, 722 Harpinia, 605, 619 ~ Hartmann: Basalts of Auckland Islands, 734 Hasselbourgh, Captain, discovers Auckland and Mae- quarie Islands, x11 Hauerininae, 323 Heard Island, 455 Hector, J.: Structure of Auckland Islands, xxxii, 729 Hedge-sparrow, 584 Helcioniscus, 6 Helichrysum, 415 Helichrysum bellidioides, leaf-anatomy, 207 Helodrilus, 253, 287 Helomyzidae, 127, 142 helophila, Diporochaeta, 253, 283 Helophilus, 125, 133 Helopidae, 106 Hemiplax, 604, 608 Hemipodus, 246 Hemiptera of Campbell Island, 66 Hemitelia, 437 Henricia, 298 Hepaticae (identified by F. Stephani), 532 Herbaceous plants, 195 ~ Effect of sheep on, 491 Leaves, 198 Roots, 199 Stems, 199 Time of blooming, 204 Vegetable parts, 198 Herbs,— Evergreen, 196 Summer-green, 195 Herposiphoma, 516 Herriott, Miss E. M.: Paper on leaf-anatomy, 185 Heterexis, 79, 113 heterochaeta, Diporochaeta, 253, 281 heteroclada, Dipterosiphonia, 516 Heterocrossa, "74. Heteromyza, 127 Heterophylly, 203 In Dracophyllum longifolium, 203 In Polypodium diversifolium, 203 In Nothopanax simplex, 203 Heterosiphonia, 512 hexagona, Lagena, 334 HMierochloe, 472 Hill humus soil, 757 Hippocrene, 306 lursuta, var. corymbosa, Cardamine, 389 hirsutum, Sterewm, 530 | hirtipalma, Hyale, 605, 643 | hirtipes, Hemiplax, 604, 608 Hamilton, A.: Exploration of Macquarie Island, xxix, 184. hirtum, Pleonosporium, 518 Hirudinea, report on, by Benham, 372 | Hirudinidae, 374 Hirudininae, 374 ; Historical introduction, by R. McNab, xi History,— Botanical exploration, 390 Discovery of the islands, xi Scientific investigation of islands, xiv Flora, 453 Hochmoor, 752 Hogg, H. R., on spiders and Opiliones, 155 Holothurians from the Auckland Islands, 146 Homalomyia, 126 Hombron, 393 Hombroniana, Callophyllis, 507 Hombronii, Corallina, 524. homomallus, Tylimanthus, 533 Honey, Mount,— Basalt, 695 Trachyte, 691 Hooker, Sir Joseph, xxiii, 390, 391 « Flora Antarctica,’ 184 On antarctic element in New Zealand flora, 468 hookert, Arctocephalus, 542 Hooker, Pleurophyllum, 414 Hookeri, Plocamium, 511 Hookeri, Rhixocloniwm, 495 Hookeriaceae, 537 Hornblende,— In porphyry, 690 In trachyte, 724 House-mouse, 600 House-sparrow, 584 Howeana, Plagiochila, 533 Hudson, G. V.,— General notes on entomology, 58 On Macro-lepidoptera, 67 hudsom, Mecodema, 58, 83 huegeli, Gallinago, 556 Humous acids, 115 Humus, 745 Antiseptie action, 754 Coarse sandy, 761 Composition, 747 Humus acid, effect on bacteria, 755 Humus soil,— Its amelioration, 750 Littoral,760 Index.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. Volume II begins on page 389. Hutton, F. W.,— On geology of islands, xxxii On glaciation of New Zealand, 689 On origin of fauna and flora of New Zealand, 718 huttoni, Athoracophorus (Pseudaneitea), 39 huttoni, Dentalium, 42 huttoni, Dynamenella, 606, 657 huttoni, Loxomerus, 94 huttoni, Rissoa, 15 huttoni, Triphora, 21 Hyale, 605, 643 Hycanus, 79, 115 Hydriomena simulata, 65 Hydriomenidae, 70 Hydromedusae, 306 Hydromedusae and Scyphomedusae : Report by Benham, 306 Hydrurga, 549 Hygrophytic conditions in rata forest, 215 Hymenolaimus, 584 Hymenophyllum, 435 Hymenophyllum multifidum, variation according to en- vironment, 201 Hymenoptera,— Descriptions of four species of, from Auckland Island, 75 Of Campbell Island, 66 Hymenosoma, 604, 610 Hyperammina, 325 Hypnaceae, 537 Hypoglossum, 512 Hypolepis, 438 Hypopterygiaceae, 537 Hypotaenidia, 555 Tais, 606, 649 Icaridion, 127, 140 icela, Calliphora, 125 Ice-sheet, absence of, 716 Ichneumonidae, 75 Ichthyodellidae, 372 Idotea, 606, 658 Idoteidae, 606, 658 Igneous rocks, radium content of, 778 iluminata, Nacella (Patinigera), 6 imbecilla, Poa, 480 Impennes, 576 improvisa, Parorchestia, 605, 641 inaequalis, Spirillina, 353 inaequistylis, Melita, 605, 630 incertus, Inocatoptes, 113 incisa, Pteris, 438 Incisura, 8 inclinans, Leptostomum, 536 incrassata, Poa, 479 indicataria, Chloroclystis, 61, 70 indistineta, Paralimnophora, 126, 137 indivisa, Rhizammina, 325 inflata, Globigerina, 350 inhaerens, Gypsina, 363 Inisidiatores, 157, 178 Inlets, east coast, Auckland Island, 710 Inocatoptes, 79, 112 inornata, Odostomia, 24 imsculpta, Rissoa (Ceratia), 17 Insects figured by Hudson, Plates ii, iii, iv insignis, Miliolina, 318 Insolubility of titanium-phosphate, 763 insularis, Dicranomyia, 124 insularis, Melanochila, 126, 137 insularis, Parorchestia, 605, 639 intermedia, Acmaea, 4 intermedia, Paphia (Ruditapes), 47 intermedius, Lumbricillus, 252, 261 interruptus, Helophilus, 125, 133 interruptus, Pseudhelops, 108 Introduced animals, xxviii, 235, 598 Introduced plants, 231 : Absence in primitive formations, 231 In Campbell Island, 492 introflecus, Campylopus, 535 involutum, Mastigobrywm, 534 involvens, Cornuspira, 325 tricolor, var. coerulea, Aracoda, 245 Tridaea, 506 iris, Haliotis, 8 Iron pan, analysis of, 767 Ironsand, titaniferous, 765 trregularis, Biloculina, 317 Ischnochiton, 1 Ischnochitonidae, 1 Ischyroplectron, 59 isolatum, Ischyroplectron, 59 Isopoda, 648 Isopoda, fresh-water, distribution of, 602 J. Jacquinot, 393 Janira, 606, 648 Jassa, 605, 647 Jassidae, 647 Juncaceae, 430 Juncus, 430 Jungermanniaceae, 532 K. Kenodactylus, 79, 90 Kerguelen Island, 455 Kerguelen Island, phanerogamic plants of, 457, 458 kerguelent, Lembos, 605, 646 kergueleni, Phoxocephalus, 604, 618 kerqueleni, Tryphosa, 604, 617 Kew Bulletin : Popular account of work of expedition, 185 kidderi, Nannonyx, 604, 615 Kirk, H. B.,— On Actiniaria, 384 On sponges, 539 Kirk, T.: Botanical investigation of subantarctic islands, xxix, 184, 395, 740 kirki, Myro, 157, 173 Kroneana, Gigartina, 506 Kroneanum, Mniodendron, 538 kronei, Dicranomyia, 124 L. Lachnea, 528 lactea, Polymorphina, 346 lactea, var. diffusa, Polymorphina, 346 lactea, var. racemosa, Polymorphina 346 lactuca, var. latissima, Ulva, 493 lacustris, Idotea, 606, 658 laevigata, Cassidulina, 332 836 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Index. Volume II begins on page 3889. Laevilitorina, 15 laevis parva, Drillia, 30 laeviusculus, Heterexis, 114 Lagena, 333 Lagenidae, 333 Lageninae, 333 lagenoides, Lagena, 338 Lagenophora, 412 Laing, R. M.,— On marine Algae, 493 _ Plant-formations of Campbell Island, 482 Lakes, Auckland Islands, 713 Lamb, C. G., on Diptera, 124 Lamellaria, 22 Lamellarudae, 22 Lamellirostres, 583 Laminariaceae, 498 laminarioides, Iridaea, 506 lanata, Cotula, 416 Land connection,— Between New Zealand and subantarctic islands, 468 Of subantarctic islands, 468 Shown by Crustacea, 602 Shown by Mollusca, 54 Shown by Oligochaeta, 254 Land planarians, report on, by Dendy, 672 “Lanes” of vegetation, 220 laquet, Heteromyza, 127 Large-leaved form, 197 Laridae, 560 Larus, 560 Lasaea, 45 lasiophthalma, 126, 135 lata, Cristellaria, 343 lateralis, Siphonaria (Liriola), 33 latifolium, Ligusticwm, 408 latifrons, Serolis, 606, 652 latissimum, Rhodoglossum, 508 Laurencia, 513 Leaf-anatomy, 205 Of Helichrysum bellidioides, 207 Of Pleurophyllum speciosum, 207 Paper by Miss E. M. Herriott, 185 Leaves,— Coriaceous, 199 Dead persistent, 204 Margins incurved, 199 Of Dracophyllum longifolium, 194 Of herbaceous plants, 198 Of Olearia Lyall, 194 Of Senecio Stewartiae, 194 Of trees, 194 Stiff or rigid, 199 Succulent, 199 Tomentose, 199 Lecideaceae, 530 legumen, Vaginulina, 342 Lehmanmiana, Schistochila, 534 Leioscyphus, 533 Lembophyllaceae, 537 Lembophyllum, 537 Lembos, 605, 646 Lentinus, 528 leonina, Cucumaria, 146 leoninus, Macrorhinus, 548 Leopard seal, 549 Lepidasthenia, 237 Lepidium, 400 Lepidolaena, 534 Lepidoptera,— From Auckland Island, 70 Of Campbell Island, 65 Of Snares, 59 Lepidozia, 534 Leprea, 248 leptalea, Mitrella, 28 Leptamphopus, 605, 621 Leptodrilus, 253, 276 Leptomedusae, 307 leptomerus, Leptodrilus, 253, 276 Leptomithrax, 604, 607 Leptonidae, 45 leptonyx, Hydrurga, 549 Leptostomum, 536 Leptothrya, 14 lessont, Oestrelata, 563 Lessonia, 498 Leucania, 67 Leucosolenia, 539 levifoliatum, Epitoniwm (Acrillus), 23 Lichenes, 529, 530 Lichens, Antipodes Island, 228 Life-form,— Cassimia Vawulliersvi, 195 Creeping-plants, 198 Cushion plants, 196 Large-leaved plants, 197 Mat plants, 198 Rosette plants, 196 Styphelia empetrifolia, 195 Tufted plants, 198 Tussock plants, 198 Life-forms of plants. 192 Effect on distribution, 191 ligulata, Desmarestia, 496 Ligusticum, 408 Liliaceae, 429 Inljeborgia, 605, 619 Lnljeborgudae, 619 Tima, 44, 701 limbata, Spirilina, 353 limbata, var. denticulata, Spirillina, 354 Limestone of Campbell Island, 764 Occurrence of fossils, 700 Tnmicolae, 556 Tnimidae, 44 Limnobuidae, 124, 129 Limnophila, 124 Limnophora, 126, 134 Limosa, 556 Lindau, G.,— On fungi, 529 On Lichenes, 530 linea, Huthria, 25 linea, subsp. pertinax, Huthria, 25 linearis, Astelia, 429 linearis, Rhodymenia, 510 lineata, Lagena, 334 lineatus, Aphrosylopsis, 125, 132 linnaeoides, E'pilobium, 406 Linnet, 584 Liochoria, 79, 103 Inotia, 12 Tnotvidae, 12 Lnissospira, 13 Intorina, 15 Litorinidae, 15 litorosa, Poa, 477 . ee SS. en eee ee ee ee a ON OEE = - Index.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 837 Volume II begins on page 389. Littoral humus soil, 760 Littoral vegetation of Campbell Island, 488 littoralis, Coelopa, 126 littorinoides, Buthria, 25 Lituolidae, 327 Lituolinae, 327 Livoneca, 606, 651 lobatula, Truncatulina, 358 Lomaria, 438 lomentaria, Chorda, 498 longicollis, Caprellinopsis, 605, 648 longicornis, Nebalia, 607, 669 longifolia, Thelymitra, 427 longifolium, Dracophyllum, 419 longitarsis, Odontria, 58, 105 longula, Liochria, 60, 104 Lopadium, 530 Lophocolea, 533 Lord Auckland Islands : see Auckland Islands Lowland humus soil, 758 Loxomerus, 78, 91 lubrica, Rissoa (Setia), 17 lucernula, Biloculina, 315 lucida, Metrosideros, 405 lucidum, var. obliquum, Asplenium, 441 lucidum, var. scleroprium, Asplenium, 441 lucidus, Fomes, 529 lucifuga, Nubecularia, 313 Lumbricidae, 253, 289 Lumbricillus, 252, 256, 261 lurida, Marginella (Glabella), 29 lusoria, Pachymenia, 524 lutea, Triphora, 21 luteo-album, Gnaphalium, 415 luteoroseus. Ischnochiton, 2 Luzula, 432 Lyall, Mount, trachyte, 692 Lyall Pyramid, melitite basalt, 697 Lyallii, Caladenia, 428 Lyallii, Olearia, 413 Lyall, Polysiphonia, 514 Lyallii, var. robusta, Stilbocarpa, 409 Lycastis, 242 Lycopodium, 444, 555 Lycoridae, 238 Lyperanthus, 427 Lysiosquilla, 604, 615 Lysianassidae, 615 lytteltonensis, Incisura, 8 M. Macdonald Group, 455 macella, Polystomella, 364 McNab, R., on discovery of the islands, xi Macquarie Islands, xviii Discovery, xiii Exploration by A. Hamilton, 184 Its vegetation, 230 Paper by J. H. Scott, 184 Porphyrites, 727 Position, size, &c., 390 macquariensis, Lumbricillus, 252, 262 macquariensis, Notiodrilus, 253, 275 macrantha, Corysanthes, 429 macrocarpa, Gigartina, 506 macrocephala, Notothenia, 585 macrocephala, Petroica, 552 macrocephalus, Physeter, 551 Macrocystis, 499 Macro-lepidoptera: Descriptions of four new species from the southern islands, 67 Macronectes, 564 macrophyllum, Carpophyllum, 502 Macrorhinus, 548 macrorhynchus, Globiocephalus, 551 maculata, Cominella, 26 maculatus, Galaxias, 587 maculipennis, Plutella, 64 Madotheca, 534 magellanica, Agrostis, 473 magellanica, Atyloides, 605, 627 magellanica, Rostkovia, 430 magellanicus, Mytilus, 43 magellanicus, Trichoniscus, 606, 661 Magnetic force: Annual rate of change, 787 Magnetic observations,— By “ Erebus ” observers, 786 By H. F. Skey and C. C. Farr, 782 Notes on, 786 Table of results, 784 Magnetic stations, 782, 786 Magnetite in basalt, 696, 733 magnus, Scyphoniscus, 606, 665 Majaqueus, 564 malacorhynchus, Hymenolaimus, 584 Malacostraca, 604, 607 Malone, E.: ‘‘ Three Years’ Cruise,” xxiv Mammalia, 542 Mangilia, 31 Maps,— Antarctic and Subantarctic Regions, in pocket in cover of Vol. II Antipodes Islands, 741 Auckland Island, xxiv, 785 Campbell Island, 783. Geological,— Campbell Island, 681 Southern half of Auckland Island, 719 Marble, Campbell Island, 698 Marchantia, 532 Marchantiaceae, 532 margaritaceus, Hchinus, 304 Margelidae, 306 Marginaria, 502 marginata, Bulimina, 330 marginata, Cerithiopsis, 20 marginata, Lagena, 335 margimata, var. semi-marginata, Lagena, 337 Marginella, 29 Marginellidae, 29 Marginulina, 342 mariana, Pannaria, 530 marina, Pelagodroma, 562 Marine Algae, report on, by Laing, 493 Marine annelids, 236 Marine Crustacea, distribution of, 603 Marine erosion,— Auckland Island, 708 Campbell Island, 682 Marionina, 252, 256, 262 Marion Island, 455 marionis, Nauticaris, 604, 614 Marphysa, 244 Marpisseae, 157, 176 marrinert, Mynoglenes, 156, 165 marrineri, Oopterus, 66, 68 Marshall, P.,— Geology of Campbell Island and Snares, 680 Meteorology, 789 Some rocks from Macquarie Island, 727 838 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Indez. Volume Il begins on page 389. Marsippospermum gracile: Its station, 221 martenst, Athoracophorus (Amphiconophora), 40 Massee, G.,— On fungi, 528 On Lichenes, 529 Mastigobryum, 534 Mat form, 198 maynei, Parorchestia, 605, 637 Meadow,— Danthonia, 218 Formation, Disappointment Island, 222 Formations, Snares, 209 Pleurophyllum, its winter aspect, 204 Poa foliosa, 209 Poa litorosa, 209 Tussock, of Antipodes Island, 227 Mecodema, 79, 83 ' Mecostethi, 157, 178 Megadyptes, 580 S Megascolecidae, 252, 268 Megascolecinae, 253, 255, 278 Distribution of, 255 Melanchra, 68 Melanochila, 126, 137 melanophrys, Diomedea, 574 melanurus, Anthornis, 554 melas, Globiocephalus, 551 Meliphagidae, 553 Melita, 605, 630 Melitite basalt, Campbell Island, 697 Melobesia, 525 Melolonthidae, 104 © Menhir, Mount, Campbell Island, 683, 685, 690 Menzesn, Dicranoloma, 535 Menziesti, Dicranoloma, form nana, 535 Menzesn, Lepidolaena, 534 Merganettidae, 584. Merganser, 584 Merganser, southern, 584 Mergidae, 584 merula, Turdus, 584 Mesodesma, 46 mesodesma, Chione, 47 Mesodesmatidae, 46 Mesophytic structure in subantarctic plants, 206 Meteorology of Campbell Island, 789, 792 Metrosideros, 405 Metrosideros lucida,— Adventitious roots, 192 Description of, 186 In rata forest, 214 Not found in Campbell Island, 486 Metzgeria, 532 Meyrick, H., on Lepidoptera from Auckland Island, 70 Micaceous humus soils, 757 micans, Beris, 124 micans, Iridaea, 506 micans, Rissoa (Setia), 18 Mica-schist in conglomerate, 726 Michaelsen on distribution of Oligochaeta, 254 Michaelsen on genus Notiodrilus, 269 micra, Lissospira, 13 micrantha, Haloragis, 404 Microcladia, 523 Microcolax, 516 Microlaena, 472 microlepidota, Notothenia, 590 microphyllum, Geranium, 402 microstiriata, Rissoa (Setia), 17 mihiwaka, Chiltonia, 605, 644 miliaris, Lasaea, 46 Miliolidae, 313 Miliolina, 318 Miliolininae, 313 millefolium, Hypolepis, 438 Miller on nitrogen in peat, 753 minimum, Hymenophyllum, 436 minuta, Endodonta (Piychodon), 38 minutissimum, Stiphidion, 156, 157 minutus, Chydorus, 607, 671 Miocene submergence, Campbell Island, 689 Miro, 552 Mirounga, 549 Mitrella, 28 Mitromorpha, 30 Mnestia, 32 Mmodendron, 538 Modiolarca, 44. Modiolarcidae, 44. Modiolus, 43 modularis, Accentor, 584 mohrioides, Aspidiwm, 442 mollicella, Acanthodoris, 33 Mollusca,— Bibliography of, 55 Circumaustral species of, 54 General remarks on distribution of, 54 H. Suter’s report on, 1 Land connection proved by, 54 Summary on, 54 Table ot distribution of, 49 Mollymawk,— Common, 574 Grey-backed, 575 White-capped, 575 Momnilea, 12 monilifera, Fissuridea, 9 monoculoides, Bovallia, 605, 622 Monodonta, 9 Montagne, 393 montagui, Lagena, 335 Montia, 402 Mopalia, 2 Mopalidae, 2 Moraines, Auckland Island, 714 Morrell, Captain: Account of Auckland Islands, xxii Morychus, 102 Mosandrite in trachyte, 694 moschata, Tillaea, 404. Mould on dead mollymawks, 755 Motacillidae, 553 Mountain scrub, 216 Its affinity, 217 Its density, 216 Mudstone, Campbell Island, 698 muellert, Hypotaendia, 555 multifidum, Hymenophyllum, 437 Multistriata meretrix, Campbell Island, 702 Mumida, 604, 612 Muricidae, 26 Musci (identified by Brotherus), 535 Muscidae, 125, 134 muscoides, Colobanthus, 401 musculus, Balaenoptera, 550 Musgrave Peninsula,— Anthracite, 727 Diabase, 727 Granite, 718 CO ee a a Index.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. 839 Volume II begins on page 389. Musgrave Peninsula—continued. Older basic series, 726 Porphyrite, 728 Trachyte, 723 Musgrave, T., on wreck of ‘* Grafton,’ musicus, Turdus, 584 Musotima, 63 Mutton-bird, 563 Mynoglenes, 156, 165 Myochamidae, 48 Myodora, 48 Myosotis, 422 Myosotis, leaf-anatomy, 205 Myriapoda, note on, by Benham, 376 Myro, 157, 172 Myrsinaceae, 421 Myrsine, 421 Myrtaceae, 405 Mytilidae, 43 Mytilus, 43 Myxogastres, 529 Nacella, 6 Namostygnus, 79, 98 Nannonyx, 604, 615 nanum, Dentalium, 41 nassoides, Cominella, 26 nasutum, Icaridion, 127, 141 Natantia, 604, 614 Natica, 21 Naticidae, 21 National Antarctic Expedition, xxxiv Naturalised plants, 445 Nauticaris, 604, 614 Nebalia, 607, 669 Nebaliaceae, 607, 669 Nebaludae, 669 nebrioides, Loxomerus, 92 Nectocarcinus, 604, 608 neglecta, Janira, 606, 648 neglecta, Streblocladia, 516 neglectus, Callocolax, 507 neglectus, Rumex, 426 Nelly, 564 Neojanacus, 21 Neolepton, 45 Neonetus, 66 neoselena, Drepanodes, 62, 70 neozelanica, Cyclostremella, 13 neozelanica, Haliacris, 606, 650 neozelanica, Rissoa (Setia), 17 Nephelene in phonolite, 694 Nereis, 238 nereis, Garrodia, 562 Nertera, 412 Nesierax, 555 Nesonetta, 583 Neuroptera of Campbell Island, 66 mgerrima, Monodonta (Diloma), 10 nigrifrons, Coelopa, 127, 140 nigromaculata, Aucklandella, 76 Nimrod Islands, xviii. nitida, Candeina, 352 nitidalis, Musotina, 63 Nitrogen,— In subantaretie soils, 753 In European peats, 753 In top layer of soil, 753 > XXIV noblis, Porphyra, 504 Nodosaria, 340, 702 Nodosarvinae, 340 nodulosa, Reophax, 327 Nonionina, 363 : Norman’s Inlet, 710, 711, 715 North Arm, Carnley Harbour,— Basalts, 733 Marine erosion, 708 North-west Bay gabbro, 690 Nothopanax simplex,— Heterophylly, 203 Trunk, 194 Notiodrilus, 252, 254, 268 Distribution of, 254. Notobdella, 372 Notothenia, 590 nototheniae, Notobdella, 372 Nototheniidae, 590 novae-zealandiae, Allorchestes, 605, 645 novae-zealandiae, Anthus, 553 novae-zealandiae, Cancer, 604, 608 novae-zealandiae, Carolobatea, 605, 620 novae-zealandiae, Cyanorhamphus, 555 novae-zealandiae, Geonemertes, 676 novae-zealandiae, Hyale, 605, 643 novae-zealandiae, Leptamphopus, 605, 621 novae-zealandiae, Limosa, 556 novae-zealandiae, Livoneca, 606, 651 novae-zealandiae, Nesierax, 555 novae-zealandiae, Prosthemadera, 554. novae-zealandiae, Sphaerium, 46 novae-zealandiae, Spirillina, 352 novae-zealandiae, Syrphus, 125 novae-zealandiae, Tanais, 606, 669 Nubecularia, 313 Nubecularvinae, 313 nummosus, Rubriws, 157, 167 nummularifolium, nerterioides, var. Lpilobium, 407 Nummulinidae, 363 O. obconica, Spirillina, 352 Obeliinae, 307 obliqua, Nodosaria, 341 obliquata, Siphonaria, 33 oblonga, Cassidulina, 332 oblonga, Miliolina, 320 oblonga, Polymorphina, 349 Observations, magnetic, 782 obtusatum, Asplenium, 440 obtusifrons, Harpinia, 605, 619 Odontria, 79, 104 odora, Veronica, 424. Odostomia, 24. Oedicerotidae, 620 Oestrelata, 563 officenale, Taraxacum, 418 Ogmorhinus, 549 Older basie¢ series, Auckland Islands, 726 Olearia, 413, 543 Olearia Lyallii,— Forest, 208, 218 Forest, interior aspect, 208 Leaves, 194 Prostrate stems of seedlings, 194 oleraceum, Lepidium, 400 840 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Indezx. Volume II begins on page 389. Oligochaeta,— Bibliography of, 289 Dispersal of, 254 Distribution of, 254. Report on, by W. B. Benham, 251 oligostegia, Nodosaria, 340 Omalidae, 97 Omalium, 97 Omalogyra, 19 Omalogyridae, 19 Omphalia, 528 Onagraceae, 406 Onchidella, 41 Onchididae, 41 Omscidae, 606, 668 Oniscoidea, 606, 660 Oniscus, 606, 668 Omithochiton, 4. Onosandrus, 59 Onuphis, 245 Oopterus, 79, 86 opalus, Cantharidus, 10 ophione, Lamellaria, 22 Ophiuroidea, 303 Opiliones, 157, 178 orbicularis, Cristellaria, 346 orbignyana, Lagena, 337 orbignyana, var. clathrata, Lagena, 338 Orbulina, 351 Orchestia, 605, 632 Orchidaceae, 427 Oreobolus, 434 Origin of flora of subantarctic islands of New Zealand. 467 ornata, Callophytlis, 507 ornata, Henricia, 298 Ornithobdella, 374 orophylloides, Xanthorhoe, 62, 65, 68 Orthoptera,— Of Campbell Island, 66 Of Snares, 59 orygmaea, Sticta, 531 Ossifraga, 564 Otarvidae, 542 Otiorhynchidae, 108 ovalis, Myro, 157, 174 ovalis, Synteratus, 58, 85 ovina, Festuca, 481 oxyptera, Xanthorhoe, 62, 67 Pachyderris, 79, 121 Pachymenia, 524. pachyrhynchus, Catarractes, 577 pacifica, Cladophora, 495 pacifica, Paradexamine, 605, 632 Pacificana, 157, 176 Pactolotypus, 79, 119 pagaia, Leucania, 59, 67 Palaemon, 604, 614 Palaeontology, Campbell Island, 700 Palisade parenchyma, feeble development of, 205 Pallavicinia, 533 pallida, Gaimardia, 433 pallida, Lophocolea, 533 pallitarsis, Onosandrus, 59 Panazx, 410 Pannaria, 530 Pannariaceae, 530 Pantellarite, 694. Paphia, 47 Paradexamine, 605, 632 Paralumnophora, 126, 136 Paramoera, 605, 625 Parapherusa, 605, 630 parasitica, Conchothyra, Campbell Island, 700 Parassellidae, 606, 648 Paridotea, 606, 660 Paris, Mount, basalt, 696 parisiensis, Discorbina, 356 parkert, Ischnochiton, 1 Parmeliaceae, 531 parmifera, Scoparia, 62, 66, 72 Parorchestia, 603, 605, 636 Parrakeet,— Antipodes Island, 554 Macquarie Island, 555 Red-tronted, 555 Parson-bird, 554 Passeres, 552 parva, Parorchestia, 605, 640 parviflora, Coprosma, 410 patachonica, Aptenodytes, 576 - Patellidae, 6 patelliformis, Discorbina, 355 Patellina, 354. Paterson, Lomaria, 438 Peat, 191, 745 American, analysis of, 748 Classification of, 746 Covering, Campbell Island, 684, 690 European, nitrogen in, 753 Falkland Islands, 751 Formation of, favoured by climate, 191 Formed by Astelia pumila, 751 Formed in stagnant water, 746 Hall on nitrogen in, 753 High nitrogen content, 753 Trish, analysis of, 747 Miller on nitrogen in, 753 Non-nitrogenous matter in, 753 Russian, analysis of, 749 Peat-ash, composition of, 747 Peat-forming bryophytes, 215 Peat-forming plants, 746 Pecten, Campbell Island, 701 pectinatum, Ptilothamnion, 517 pedestris, Schoenophilus, 125, 132 Pelagodroma, 562 Pelecanoides, 566 Pelecanoididae, 566 Pelecypoda, 701 Pelodrilus, 252, 263 Remarks on species of, 257 Peltidiidae, 671 Penantipodes Islands, xi Pendleton, Captain, xii pendula, Breutelia, 536 Peneroplidinae, 325 Penguin, — Big-crested, 578 Grand, 580 King, 576 Royal, 580 Tufted, 576 Victoria, 577 Penguins, effect on vegetation, 232 Se Indez.| SUBANTARCTIC ISLANDS OF NEW ZEALAND. Volume II begins on page 389. penicillata, Deschampsia, 475 penna-marina, Lomaria, 439 Perennials, 195 perforata, Lawrencia, 513 perforata, Porphyra, 503 Perionychopsis, Diporochaeta, 253, 286 Perisporiaceae, 529 Perofskite in porphyry, 691 perplexus, Neojanacus, 21 Perseverance Harbour,— Description, 682 Dykes, 696 Ice erosion, 685 Persistent dead leaves, 204 Petrel,— Dove, 565 Giant, 564 White-chinned, 564 White-headed, 563 Petrie, D., on Gramina, 472 Petroica, 552 pezizoideum, Lopadium, 530 Phaeophyceae, 496 Phalacrocoracidae, 580 Phalacrocorax, 580 Phaloniadae, 74 Phanerogamic plants of Kerguelen, &¢., 457 Phanerogams, ferns, and Lycopodiaceae,— Distribution of, 448 Enumeration of, 397 Phenacohelicidae, 34 Phenacohelix, 35 Phenacohelix sp. 36 Phialella, 307 Philobrya, 43 Philobryidae, 43 phlogophora, Flammulina, 36 Phocidae, 548 Phoebetria, 575 Phonolite, Campbell Island, 694 Photinula, 11 Phoxocephalidae, 618 Phoxocephalus, 604, 618 Phreodrilidae, 252, 256 Phreodrilus, 252, 256 Distribution of species of, 257 Phycodromidae, 126, 139 Physeter, 551 Physeteridae, 551 Physiognomy,— Climatic, 185 Edaphic, 186 Olearia Lyallii forest, 208 Rata forest, 214 Physiological dryness, 191 piciventris, Acalles, 120 Pigs, effect on vegetation, 235 pileopsis, Acmaea, 5 pinguis, Ranunculus, 397 pinnata, Metzgeria, 532 pinnata, Microcladia, 523 pinnatifida, Laurencia, 513 pinnatifida, Schistochila, 534 Pinnoctopus, 49 Pipit,— Antipodes Island, 553 Auckland Island, 553 piscator, Amaurobioides, 156, 164 Pisces, 585 pisum, Biloculina, 315 Pitchstone, Auckland Island, 724 plagiatana, Pyrgotis, 63, 73 Plagiochaeta, 253, 254, 275 Plagiochila, 533 plagiostoma, Thelepus, 247 Planarians, land, report on, by Dendy, 672 planatus, Halicarcinus, 604, 609 planifolius, Juncus, 431 Planispirina, 323 Plantaginaceae, 426 Plantago, 426 Plant formation, Pleurophyllum Hookeri, 220 Plant formations, 207 Antipodes Islands, 227 Auckland Islands, 211 Campbell Island, 482 bs Snares, 208 Uniformity of, in different islands, 207 Plants,— Introduced, 231, 492 Leading physiognomic, 184 Peat-forming, 746 Of Auckland Islands, 752 Of Chonos Archipelago, 751 planulatus, Allodiscus, 34. Platyclinia, 511 Platyptilia, 73 Plaxiphora, 2 Pleonosporium, 518 Plewronectidae, 590 Pleurophyllum, 413, 599 Pleurophyllum criniferum, description of, 188 Pleurophyllum formation, 223 Pleurophyllum Hookeri,— Description of, 188 Formation, 220 Of Campbell Island, 488 Pleurophyllum meadow,— Disappointment Island, 225 Its composition, 221 Pleurophyllum speciosum,— Description of, 186 Leaf-anatomy, 207 plicaticollis, Oopterus, 87 plicatula, Marginella (Glabella), 29 Plocamium, 511 Plover, Auckland Island, 556 plumosa, Cotula, 416 plunketi, Plagiochaeta, 253, 275 Plutella maculipennis (Cruciferarum), 64 Plutellus, 253, 278 Distribution of, 280 In Auckland Island, 255 Plutonic rocks, Auckland Island, 718 Poa, 476 Poa foliosa meadow, 209 Poa litorosa,— Description of, 186 Formation, 221 Meadow, 209 Variation according to environment, 201 Poa ramosissima, its station, 212 Poduridae, 377 polaris, Stilbocarpa, 409 Polinices, 22 Pollination, 200 polyanthos, vax. sanguinolentum, Hymenophyllum, 435 polybranchia, Polydora, 246 842 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Indea. Volume II begins on page 389. Polychaeta,— Affinities of, 236 Dispersal of, 237 Of New Zealand and South America, 236 Report on, by W. B. Benham, 236 Polydora, 246 polygona, Miliolina, 321 Polygonaceae, 426 Polymorphina, 346 Polymorphininae, 346 polypleura, Liotia, 12 Polypodidae, 48 Polypodium, 442 Polypodium diversifolium, heterophylly in, 203 Polypus, 48 Polysvphonia, 513 Polystomella, 364 Polystomellinae, 363 Polytocus, 127, 144 Pontogeneia, 605, 624. Pontogenevidae, 622 porcatus, Balanus, 607, 670 porcellana, Rissoa (Setia), 18 Porcellanopagurus, 604, 610 Porphyra, 503 Porphyrites, Auckland Islands, 727, 734 Porphyry, Campbell Island, 690 Portulacaceae, 402 posticalis, Pseudhelops, 107 Prasophyllum, 427 Pratia, 419 prenanthoides (?), Brechtites, 417 Primulaceae, 420 Prince Hdward Island, 455 Priocella, 563 Prion, 565 Prionorhynchus, 604, 608 Prior, G. T.,— Basalts and dolerites from Auckland Islands, 734 Mineralogical Magazine, 734 : Pristancylus, 95 Procellarudae, 562 Proterodesma, 74 Prosthemadera, 554 prostratum, Apium, 407 protensa, Mangilia, 31 Protocardia, 47 Protyparcha, 71 pruninus, Cantharidus, 11 prunminus, var. minor, Cantharidus, 11 prunimus, var. perobtusus, Cantharidus, 11 psammitis (?), Scoparia, 63 Pseudhelops, 79, 106 Pseudosphaeroma, 606, 653 Psitiaci, 554 Psittacidae, 554 Pteris, 438 Pterophoridae, 73 Ptilothamnion, 517 Ptychomnium, 536 pubescens, Iais, 606, 649 Publications, recent, dealing with the Subantarctic, 185 Puffinidae, 563 Puffinus, 563 pulchella, Jassa, 605, 647 pulchella, Protocardia (Nemocardium), 47 pulchellum, Distichophyllum, 537 Pulvinulina, 360 pumila, Lagenophora, 412 pumilio, var. pectinatus, Oreobolus, 434. punctata, Allophylopsts, 59, 127, 143 punctatum, Polypodium, 442 punctatus, Oniscus, 606, 668 punctifer, Galaxias, 586 punctiventris, Pachyderris, 61, 121 punctulatum, Calliostoma, 12 Puncturella, 9 puniceus, Callochiton, 2 purpurascens, Tayloria, 535 pusilla, Modiolarca, 44 pycnoclada, Cladonia, 531 pygmaea, Bolivina, 331 Pyramidellidae, 23 Pyraustidae, 71 Pyremdae, 28 Pyrenomycetes, 528 Pyrgotis, 73 pyrifera, Macrocystis, 499 pyrula, Nodosaria, 341 Pzechridae, 156, 157 quadrata, Lagena, 339 Q quadrata, var. carinata, Lagena; 339 quadraticeps, Lycastis, 242 quadricincta, Mangilia, 31 quadricollis, Pseudhelops, 58, 107 quadrimaculata, Calliphora, 64, 125 Quartz, Campbell Island, 698 R. Rabbits, 600 radians affinis, Helcioniscus, 6 Radium content,— Method of testing, 778 Of igneous rocks, 778 Of rocks, conclusions regarding, 780 radula, Gigartina, 506 Rail, Auckland Island, 555 Rain-water caught by trunk of tussock, 204 Rallidae, 555 Ralliformes, 555 ramosa, Hyperammina, 325 ramosissima, Plagiochila, 533 ramosissuma, Poa, 478 ramulosum, Stereocaulon, 532 Ranfurlya, 37 ranfurlyt, Phalacrocorax, 583 Ranunculaceae, 397 Ranunculus, 397 Ranunculus pinguis,— Rhizome, 199 Variation according to environment, 201 Raptores, 555 rarescens, Discorbina, 357 rarum, Hymenophyllum, 435° Rata forest, 213 General remarks, 213 Its composition, 213 Its hygrophytic conditions, 215 Metrosideros lucida, 214. Of Adams Island, 216 Physiognomy, 214 The undergrowth, 215 Rata not found in Campbell Island, 486 Rats, 600 rattus, Mus, 600 = Ce iin 7 TET ye" + OE chen gear aaincs te ten it em twenty TR amr NEP 2 ee ete ee 3 ee es 7 Index.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. Volume II begins on page 389. Raynal, T. E., on wreck of “ Grafton,” xxiv Recent elevation, 689, 705 redimiculum, Helcioniscus, 7 redolens, Hierochloe, 472 refulgens, Trucatulina, 357 regia, Diomedea, 572 Relation between surface soil and underlying rock, 761 relaxum, Campylium, 537 ** Reliance,” xii Remarks, general, on rata forest, 213 reniformis, Azorella, 407 reniformis, Cristellaria, 344 Reophax, 327 repanda, Pulvinulina, 360 repanda, var. concamerata, Pulvinulina, 360 repens, Coprosma, 412 repens, Samolus, 420 Reproduction of plant covering after fire, 231 Reptantia, 604, 607 reticulata, Truncatulina, 359 retiferum, Bittiwm, 19 retipora, Cladonia, 532 retroflexa, Cystophora, 502 Reversion shoots, Coprosma ciliata, 204 Rhabdammininae, 325 Rhizammina, 325 Rhizoclonium, 495 Rhizodrilus, 252, 258 Rhizogonium, 536 Rhizome— Of Ranunculus pinguis, 199 Of Stilbocarpa, 199 Rhododrilus, 253, 277 Rhodoglossum, 508 . Rhodomelaceae, 512 Rhodophyceae, 503 Rhodophyllis, 508 Rhodymenia, 510 Rhodymeniaceae, 510 Rhombosolea, 590 Rhyolite, Disappointment Island, 738 Rhyparosomidae, 115 Rhyphidae, 124, 131 Rhyphus, 124, 131 Rich humus soil, 759 Riebeckite, occurrence of, 694 Right whale, southern, 549 rigida, Ulwa, 494 riparia, Uneinia, 434 Rissoa, 15 Rissoidae, 15 Rissoina, 18 riwularis, Corysanthes, 428 Robin, Snares, 552 robusta, Bartramia, 536 robusta, Bolivina, 331 robustus, Baeopterus, 127, 142 Rochefortia, 45 Rock,— Analysis of carbonaceous, 766 Campbell Island basalt, analysis, 767 Rocks— Antipodes Island, analysis, 763 Inland, of Antipodes Island, 230 Subalpine, 226 Unweathered, analysis of, 765 Weathered, analysis of, 765 Rocks and cliffs, coastal, of Antipodes, 227 Rocks and sands, 764 roemeri, Nodosaria, 340 Roots— Adventitious, in Metrosideros lucida, 192 Of herbaceous plants, 199 Relation to depths of soil, 754 rosacea, Discorbina, 355 Rosaceae, 403 rosea, Mitrella, 28 rosea, Rissoa (Cingula), 17 rosea, Scissurella, 7 roseoradiata, Acmaea, 4 Rosette form, 196 Ross, Port,— Basalts, 734 Description, 710 Ross, Sir James, xxiii, 390, 391, 392 Rossii, Bulbinella, 430 rossti, Oirolana, 606, 651 rosstt, Thinornis, 556 Rostkovia, 430 rosulata, Abrotanella, 417 Rotalia, 362 Rotaludae, 352 Rotaluinae, 354. rotula, Liotia, 12 rotulata, OCristellaria, 345 rotundifolia, Corysanthes, 428 Royal Company Islands, xx Rubiaceae, 410 : rubiginosus, Acanthochites (Loboplax), 3 rubra, Festuca, 481 rubrioides, Amaurobius, 156, 159 Rubrius, 157, 167 rubromarginatus, Oalathus, 96 rubrum, Ceramium, 520 rufa, Coelopa, 126 rufipes, Namostygnus, 99 rufoapicata, Rissoa, 16 rugata, Odostomia (Pyrgulina), 24 rugulosus, Cubaris, 606, 668 Rumex, 426 Rumex neglectus, its increase on dunes, 211 S. St. Col Peak trachyte, 692 St. Paul Island, flora of, 462 sabulosa, Odostomia (Menestho), 24 sabulosella, Scoparia, 63, 72 saccatus, Tylimanthus, 533 Sagartia, 385 sagittula, Spiroplecta, 329 Salt-laden winds, effect of, 483 Salticidae, 157, 176 saloimi, Diomedea, 575 Samolus, 420 Sand, titanium in, Campbell Island, 766 Sandstones, Campbell Island, 686 Sandwich Group, 455 sanguinea, Alcira, 29 Sanguineus caelatus, Cantharidus, 11 Sanguisorbae, Acaena, 403 Sanguisorbae, var. antarctica, Acaena, 403 sarissa, Cerithiopsis, 19 sarsi, Biloculina, 314 Samicava, 48 Saxicavidae, 48 844 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Index. Volume II begins on page 389. scapha, Nonionina, 364 Scaphandridae, 32 scaphodes, Protyparcha, 62, 71 scariosum, Lycopodium, 444 Schenck, H., on floras of subantarctic islands, 458 scheuchzerioides, Juncus, 431 Schimper, A. F. W., on flora of subantarctic islands, 458 Schismope, 7 Schistochila, 534 Schists in conglomerate, Carnley Harbour, 726 Schizaea, 443 Schizonewra, 512 schlegeli, Catarractes, 580 schlichti, Lagena, 337 schloenbachi, Cristellaria, 342 Schoenophilus, 125, 132 schlumbergeri, Sigmoilina, 323 schreibersii, Pulvinulina, 361 Scirpus, 433 Scirpus frondosus : Its absence on dunes, 211 Scissurella, 7 Scissurellidae, 7 sclatert, Oatarractes, 578 Scolopacidae, 556 Scoparia, 71 scoparium, Dracophyllum, 420 scopulinus, Larus, 561 scoriadea, Antennaria, 529 scorpiurus, Reophax, 327 Scott, J. H., on Macquarie Island, xxi, xxix, 184, 394 Scrophulariaceae, 424 Serub, Antipodes Island, 229 Serub, mountain, 216 sculptipennis, Heterexis, 61, 113 scutellata, Lachnea, 528 scylla, Badwmna, 156, 160 Scyphacidae, 606, 664 Scyphomedusae, 309 Scyphomedusae and Hydromedusae: Report by Benham, 306 Scyphoniscus, 606, 664 Sea-bear, 542 Sea-elephant, 548 Sea-hawk, 556 Sea-leopard, 549 Sealing gangs, xx, xxi Sealing islands, xii Sealing regulations, xxix Sea-lion, 542 Seals, effect on vegetation, 232 Seals, reserves for, xxix Seasonal changes, 204 Seed-dissemination, 200 Seedling of Suttonia divaricata, 195 Seedlings of Olearia Lyallii, 194 Seedlings, prostrate stems of, 194 Seila, 20 Selago, Azorella, 407 Selidosemidae, 70 semmmulum, Miliolina, 319 Semi-prostrate habit of trees, 192 Cause of, 194 semireticulata, Monilea (Minolia), 12 Semostomeae, 309 Senecio, 417 Senecio Stewartiae, leaves, 194 Septidae, 22 septiformis, Acmaea, 5 Sequence of volcanic rocks,— Adams Island, 731 Campbell Island, 698 North Arm, 733 Serolidae, 606, 652 Serolis, 606, 652 Serpulids, 236 serrata, Biloculina, 314 serrata, Lhrenbergina, 332 serraticauda, Atyloides, 605, 627 serrulata, Orchestia, 605, 632 setarius, Stilbodiscus, 118 setifolia, Deyeuxia, 474 setosum, Dicranoloma, 535 Shag,— Auckland Island, 581 Bounty Island, 583 Campbell Island, 580 Shearwater, Forster’s, 563 Sheep, 599 Effect on vegetation of Campbell Island, 490 Shipwrecks, xxiv Shrubs, 194 Divaricating form, 194 The Dracophyllum form, 195 sibbaldu, Balaenoptera, 550 siebert, Mniodendron, 538 | Sigmoilina, 323 Silphidae, 101 Silver-eye, 553 sumilata, Hydriomena, 65 simplex, Hemipodus, 246 simplex, Panax, 410 Simuluidae, 124. Simulium, 124 Sinclair, Apophloea, 525 Siphonaria, 33 Siphonarudae, 33 Skey, H. F., on magnetic observations, 782 Skua gull, 556 Smith Harbour, 714, 716 Smith, J. Crosby, on soil of Campbell Island, 484 smithu, Empis, 125 Smithii, Hemitelia, 437 Snares, 207 Coastal formations, 209 Coleoptera of, 58 Diptera of, 59 Discovery of, xi Forest, 208 General appearance of vegetation, 207 Geology of, 703 Granite, 704 Insects of, 58 Lepidoptera of, 59 Meadow formations, 209 Orthoptera of, 59 Plant formations, 208 Position, size, &c., xiv, 389, 704 Sealers marooned on, xxi Tussock meadow, 209 Snipe, 556 Soil,— Antipodes Island, 763 Ash-content, 753 Calcareous humus, 762 Campbell Island, 762. Coarse sandy humus, 761 ahs oer 2 thitese 2 he hi GAS ARR, gti mbit etiam pst ee = Ze Te ee PP RR oT RO eT Sa DENT Sen gage One yon eer Index. ] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 845 Volume II begins on page 389. Soil—continued. Stereum, 529, 530 ; Hill humus, 757 Sterna, 559 Humus, its amelioration, 750 Sternidae, 559 Humus, of Florida, 749 Stewartiae, Senecio, 418 Lowland humus, 758 Stichaster, 302 Of Auckland Islands : Cockayne’s remarks, 752 Stichasteridae, 302 Relation to depths of roots, 754 stichidiosum, Ceramium, 521 Rich humus, 759 stichidiosum, var. scopulorum, Ceramium, 523 Samples, notes on, 757 stichidiosum, var. Smithvi, Ceramium, 522 Soil-analysis, table of, 772: Sticta, 531 Soils, methods of analysis, 756 Stictaceae, 531 soldanui, Rotalia, 362 Stilbocarpa, 409 Sonchus, 418 Stilbocarpa bollonsii, 209 South African flora : Connection with New Zealand, 461 Stilbocarpa polaris, xxi, 409 South Georgia, 455 Air-chambers, 206 ** Southern Cross ” expedition, xxx Description of, 189 spathulata, Abrotanella, 417 Rhizome, 199 Species of limited distribution in Auckland Islands, 210 Stilbocarpa robusta, 209 speciosum, Pleurophyllum, 413 Stilbodiscus, 79, 117 spectabile, Calliostoma, 12 Stinkpot, 564 Speight, R., on geology of Auckland Islands, &c., 705 Stiphidion, 156, 157 Sphacelariaceae, 496 Stomatopoda, 604, 615 sphaera, Planispirina, 324 Storm-petrel, 562 Sphaertidae, 46 White-faced, 562 Sphaerium, 46 Stratigraphy and petrology, Auckland Island, 718 Sphaeroidina, 351 Stratiomyidae, 124 Sphaeromidae, 606, 652 strebeli, Huthria, 26 Sphaerophoraceae, 530 Streblocladia, 516 Sphaerophorus, 530 striata, Mnestia, 32 Spheniscidae, 576 striata, Thais (Polytropalicus), 27 sphenoideum, Flabellum, Campbell Island, 702 striatula, Hmarginula, 9 Spheridiidae, 98 striatus, Pactolotypus, 119 Spiders and Opiliones, 155 stricta, Atropis, 480 Distribution of, 155 strictum, Ditrichum, 535 spinicosta, Polytocus, 64, 127, 145 strigilis, Helcioniscus, 7 spinigera, Spirillina, 353 strigosus, Lentinus, 528 spinosa, Lysiosquilla, 604, 615 strombifolia, Plagiochila, 538 Spirillina, 352 Stuckey, F. G. A., on Actiniaria, 384 Spirillininae, 352 g stutchburyi, Chione, 46 Spionidae, 246 Stylidieae, 419 Spiroloculina, 318 styliformis, Cerithiopsis, 20 Spiroplecta, 329 Stylohalina, 539 spirospermia, Geonemertes, 674 Styphelia empetrifolia, life form, 195 Splachnaceae, 535 Stypocaulon, 496 : Sponges, two, from Campbell Island, by H. B. Kirk, 539 Subalpine rocks, 226 spretus, Cadulus, 42 Subalpine tussock meadows of Campbell Island, 489 squamata, Amphiura, 303 Subantaretic,— squamosa, Lagena, 335: Climate, 190 Stagnant water, 746 Distribution of Crustacea, 602 Stangertt, Madotheca, 534 Islands of New Zealand, xiv Starling, 584 Distribution of Notiodrilus and Pelodrilus, 254 Station,— Islands : Magnetic observations, 782 Of Azorella Selago, 230 Islands : Plants not in New Zealand, 466 Of Marsippospermum gracile, 221 Islands, scientific investigation of, xiv Of Poa ramosissima, 212 Islands, vegetation of, 456 Of Veronica elliptica, 210 Plants, mesophytic structure in, 206 stebbingi, Tmetonyx, 604, 618 Plants, xerophytic structure in, 206 Steganopodes, 580 Vegetation a reflection cf the climate, 192 steindachneri, Anthus, 553 subantarctica, Cerithiopsis, 20 Stellaria, 401 subantarctica, Mitrella, 28 Stellerids, 295 subantarctica, Phenacohelix (?), 36 stelliferus, Helcioniscus, 6 subfusca, Rissoa (Cingula), 17 Stems, of hebaceous plants, 199 subfusca, subsp. micronema, Rissoa, 17 stenopetala, Drosera, 404 subglobosa, Cassidulina, 332 Stephani, F., on Hepaticae, 532 subquadrangulata, var. enderbyensis, Geoplana, 672 Stercorariidae, 556 subradicula, Nodosaria, 340 stereocauloides, Sphaerophorus, 530 subrotunda, Miliolina, 319 Stereocaulon, 532 subrugosa, Munida, 604, 612 Stereodon, 538 subscaposus, Ranunculus, 397 - 846 SUBANTARCTIC ISLANDS OF NEW ZEALAND. (Index. ar a ee ey nee Volume II begins on page 889. subspicatum, Trisetum, 475 subsulcata, Chione, 47 sub-tater, Circulus, 13 subteres, Bulimina, 330 subulata, Agrostis, 473 subulata, Astelia, 429 subulatus, Oolobanthus, 402 succincta, Aclis, 23 Succulent leaves, 199 Suess: ‘* Face of the Earth,’’ 680 sulcata, Lagena, 333 Summer-green herbs, 195 sumptuosa, Liochria, 60, 103 superba, Plaxiphora, 3 superba, Todea, 444 superciliosa, Anas, 583 Surface covering, Auckland Island, 714 Surface features,— Auckland Island, 712 Campbell Island, 683 Surface soil and rock, relation between, 761 Suter, H., on Mollusca, 1 sutert, Rissoa (Onoba), 16 suteri, Stichaster, 302 var. laevigatus, 302 Suttonia divaricata,— Description of, 190 Seedling of, 195 Suttonia formation, 216 Swamp, Antipodes Island, 229 Swamps, Auckland Island, 225 Swine, 599: Syllidae, 237 Syllis, 237 Sylvvidae, 552 Syngnathidae, 588 Syngnathus, 588 Synteratus, 79, 85 syntona, Cacoecia, 64, 73 Syrphidae, 125, 133 Syrphus, 125 Systematic botany, report on, by Cheeeseman, 389 rc. Table showing analysis of soils, 772 Tabular view of distribution of phanerogams, ferns and Lycopodiaceae, 448 Talitridae, 632 Tanaidacea, 606, 669 Tanaidae, 669 Tanais, 606, 669 tannensis, T’mesipteris, 445 tapirina, Rhombosolea, 590 Taraxacum, 418 tarsalis, Oopterus, 89 tasmanica, Cocculina, 14. taumakiensis, Odostomia, 24. Tayloria, 535 Telethusidae, 246 tener, Sphaerophorus, 530 Tennantiana, Poa, 476 tenuimargo, Truncatulina, 358 tenuis, Parorchestia, 605, 642 tenuistriata, Daphnella, 32 Terebellidae, 247 Tern,— Swallow-tailed, 559 White-fronted, 560 ternaria, Carex, 434. Terraces,— Auckland Island, 714 Campbell Island, 682, 683 Terrestrial Crustacea, distribution of, 602 “Terror,” 391 tetricus, Catadryobius, 61, 110 Textularia, 328, 702 Teaxtulariidae, 328 Textulariinae, 328 Thais, 2'7 Thaisidae, 27 Thelephoraceae, 530 Thelepus, 247 Thelymitra, 427 Therasia, 35 Thermia, 34. Thinornis, 556 Thomosis, 79, 100 Thomsoni, Microlaena, 472 thomson, Trichoniscus, 606, 661 Thoracica, 607, 670 Thrush, 584 Thuringia, antiseptic action of humus in, 755 Tillaea, 404 Timarete, 247 Timeliidae, 552 Tineidae, "74. Tinoporinae, 362 Tit, yellow-breasted, 552 Titaniterous ironsand, 765 Titanium in Campbell Island sands, 766 Titanium-phosphate, 763 Its insolubility, 763 Tmesipteris, 445 Tmetonyx, 604, 618 Todea, 444 tomentella, Trichocolea, 534 Tomentose leaves, 199 tomentosum, Codiwm, 496 Tortricidae, 73 : totolirata, Daphnella, 32 trailli, Admete, 28 transitans, Alcira, 28 Transition between meadow and forest, 219 Transit of Venus Expedition, — French, xxix German, xxix treadwelli, Bulima, 24. Tree-growth, effect of wind on, 190 Trees, 192 Absence of, in Campbell Island, 483 Cause of absence of, 483, 484 Semi-prostrate habit, 192 Triacanthella, 378 tricarinata, Miliolina, 320 tricarinella, Cristellaria, 343 Trichia, 529 Trichocera, 124. Trichocolea, 534. Trichoniscidae, 606, 660 Trichopticus, 126, 138 Trichotropidae, 22 Trichotropsis, 22 trifida, Carex, 434 Triforidae, 20 trigonula, Miliolina, 320 triloba, Corysanthes, 429 triloba, Globigerina, 350 0 =< eng heel Faore~ A — : » i Indez.] SUBANTARCTIC ISLANDS OF NEW ZEALAND. 847 Volume II begins on page 889. Trioenonychidae, 157, 178 Trioenonyx, 157, 178 triphooki, Pecten, Campbell Island, 701 Triphora, 20 Tripterygion, 597 tripunctatus, Oopterus, 60, 87 triscelis, Scoparia, 62, 71 Trisetum, 475 tristami, Gallinago, 556 Tristan d’Acunha, flora of, 462 tristis, Puffinus, 563 Trochidae, 9 Trochus, 9 Trophon, 26 Truncatulina, 357 truncatulinoides, Pulvinulina, 361 Trunk,— Of Dracophyllum longifoliwm, 194 Of Nothopanax simplex, 194 Trunk-building in tussock form, 198 Tryphosa, 604, 617 tuberculatus, Pelodrilus, 252, 263 tuberculatus, Pseudhelops, 60, 106 tuberculosa, Gigartina, 505 tubicauda, Cymodocella, 606, 657 tubicola, Onwphis, 245 Tubificidae, 252, 258 Tubinares, 562 Tufted form, 198 Tui, 554 tumidellus, Morychus, 58, 102 Tunbridgense, Hymenophyllum, 437 Turbinidae, 14 Turbo, 14. turbo, Discorbina, 354 Turbonilla, 23 Turritella, 21 Turritellidae, 21 Turritidae, 30 Tussock form, 198 Trunk-building, 198 Tussock meadow,— Antipodes Island, 227 Snares, 209 Tussock meadows, effect of sheep on, 490 Tussock, trunk catching rain-water, 204 Tylimanthus, 533 typica, Aora, 605, 645 Ulmaridae, 309 Ulva, 493 Ulwaceae, 493 umbellifera, Omphalia, 528 Umbelliferae, 407 umbilicatula, Nonionina, 363 Uneinia, 434 Undergrowth of rata forest, 215 Undergrowth, serubby, of Danthonia meadow, 219 undulatus (Rhyphus), 124, 131 undulatus subantarcticus, Onithochiton, 4 ungeriana, Truncatulina, 359 ungulata, Paridotea, 606, 660 unicolor, Cyanorhamphus, 554 uniflora, Thelymitra, 427 Uniformity of plant covering in Campbell Island, 487 “Union,” sealing trip of, xii United States Exploring Expedition, xxii, 390, 393 universa, Orbulina, 351 urinatriz, Pelecanoides, 566 Urtica, 426 Urticaceae, 426 Urvilliana, Marginaria, 502 Usnea, 531 utricularis, Adenocystis, 498 Uvigerina, 349 ‘ Vv. Vaginulina, 342 Valleys,— Auckland Island, 712, 714 Campbell Island, 684 Valvifera, 606, 658 Vancouver discovers the Snares, xi variabilis, Oristellaria, 343 variabilis, Truncatulina, 358 varians, Alcira, 29 Variation according to environment, 200 Cassinia Vauvilliersu, 201 Celmisia vernicosa, 202 Coprosma foetidissima, 200 Coprosma parviflora, 201 Dracophyllum longrfolium, 200 Hymenophyllum multifidum, 201 In the cushion form, 201 Ranunculus pinguis, 201 Poa litorosa, 201 variegata, Callophyllis, 507 variegata, Lessonia, 499 variegatus, Bovichtus, 595 variolosum, Chaetangiwm, 505 varvum, Lycopodium, 444 varium, Tripterygion, 597 Vauvillersu, Cassinia, 415 Vegetation,— Effect of albatroses on, 234 Animals on, 232 Birds on, 232 Penguins on, 232 Pigs on, 235 Seals on, 232 Of Auckland Islands: General remarks, 210 Of coastal cliffs, 212 Of Macquarie Island, 230 Of New Zealand (Drude), 192 Of Snares, its general appearance, 207 Subantarctic, a reflection of the climate, 192 Vegctative parts of herbaceous plants, 198 vellereum, Stereum, 529 venator, Omalium, 98 Venericardia, 45, 701 Veneridae, 46 venusta, Miliolina, 321 verecunda, Rissoa (Setia), 18 vernicosa, Celmisia, 414 Veronica, 424 Veronica benthamt, variation in colour, 203 Veronica elliptica, station of, 210 Veronicas, whipcord, adventitious roots, 194 verrucosa, Drillia, 30 verrucosus, Araneus, 156, 162 verrucosus, Athoracophorus (Amphiconophora), 40 verrucosus, var. fasciatus, Athoracophorus (Amphicono- phora), 41 : verrucosus, var. nigricans, Athoracophorus (Amphicono- phora), 40 Vertebrata, 542 848 SUBANTARCTIC ISLANDS OF NEW ZEALAND. [Index. Volume II begins on page 3889. verticillata, Cladonia, 532 | | Whale,— verticillata, Cladophora, 494 Blue, 550 vesicularis, Gypsina, 362 Sperm, 551 vespertilio, Biloculina, 315 Whale-bird, 565 vestitum, Aspidium, 441 : Whale-fishery, revival of, xxiii vestitum, Ceramvum, 521 White-eye, 553 vestitus, Catadryobius, 58, 109 Wildemann on botany of Magellan Lands, 460 vexans, Simulium, 64, 124 Wilkes, Commodore, xxii, 390, 393 viaticus Achorutes, 377 Willit, Desmarestia, 497 villosum, Hymenophyllum, 435 Wilsonis, Bpymenia, 510 villosus, Trichopticus, 126 Wind, 190 virgata, Corallina, 525 Kffect on tree-growth, 190 virgatum, Ceramium, 521 Transporting seeds, 459 virginea huttoni, Haliotis, 8 : Winds,— Vitrinellidae, 13 Causing absence of trees in Campbell Island, 483 vittata, Huthria, 25 Of Campbell Island, 789, 790 vittata, Sterna, 559 Winter aspect of Plewrophyllum meadow, 204 vittatus, Catarrhactes, 577 Winter visit to subantarctic islands of New Zealand, 185 vittatus, Prion, 565 Wrecks at the islands, xxiv, xxvii, xxx viviparu, Spirillina, 352 wuellerstorfi, Truncatulina, 358 Volcanic tuff, Campbell Island, 686 vortex, Ornistellaria, 345 Voyage au Pole sud, 393 Xx. vulgaris, Miliolina, 320 Xanthorhoe, 67 ; vulgaris, Sturnus, 584 Xerophily (Dachnowski), 191 Xerophily in bog-plants, 191 Xerophytic plants in Campbell Island, 483 W. Xerophytic structure in subantaretic plants, 206 Waite, E. R.,— Xiphophora, 500 On Aves, 551 On Mammalia, 542 On Pisces, 585 On Vertebrata, 542 Zaluscodes, 124, 130 Wallace on antarctic element in New Zealand flora, 468 Zaus, 607, 671 Water-storage in cushion plants, 197 zealandica, Turbonilla, 23 Weeds, Acuena, 234. zealandiae, Pecten, Campbell Island, 701 Western clifis,— zebratus, Galaxias, 586 Auckland Island, 707 “ Zelee,” 391 Campbell Island, 682 Zircon in porphyry, 691 Western entrance, Carnley Harbour, 708 Zonal distribution in Auckland Islands, 210 “« West-wind Trift,’? 237, 254 Zosterops, 553 1D INO Oss W/O, UIs Oarded Joun Macnay, Government Printer, Wetiineton, N.Z. é . 3 ? ea ee ee eee ae ees ae ere" —_ ~ HOIMNIIUD 00 40 NYIOUI) EXPLANATION. a as ae a Extreme limit: of 0 range of pack-ice: ANTARCTIC AND 3 : = ; Approximate mean position of pack-ice: === = —~ SUB-ANTARCTIC REGIONS, compiled for the Permanent magnetic observatories: @ Recent temporary magnetic observatories: (eo) Reference to Sea-colouring. PHILOSOPHICAL INSTITUTE OF CANTERBURY. NEW ZEALAND. _ Soa-lovel 100 fathoms. MARCH, 1909, 600 ” 1,000 Drawn by E. N. D'Oyly. 1,600» 2,000 2,600 $,000 . { Soull sootl's (07th poo, 02 (By, osu” Va Mowbray BY. c Adare: aa jf Sea! ° fs ¥ & | SMITHSONIAN INSTITUTION WA 9088 00559 2290 EEO Aig n