p ■m^ v; ,1 i"?7 PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCI ETY OF LONDON. VOL. 168. (EXTRA VOLUME.) LONDON: SOLD BY UAUIUSON AND SONS, ST. MARTIN'S LANE, W.C, AND AM. 1500KSE1.LERS. (.''' Q 41 Ls >!■ Its AN ACCOUNT PETROLOGICAL, BOTANICAL, AND ZOOLOGICAL COLLECTIONS MADE IN KERGUELEN'S LAND AND RODRIGUEZ DURING THE TRANSIT OF VENUS EXPEDITIONS, CAERIED OUT BiT ORDER OF HER MAJESTY'S GOVERNMENT IN THE YEARS 1874-75. LONDON: PRINTED BY UKORGE EDWARD EYRE AND WILLIAM SPOTTISWOODE, PRINTERS TO THE QUEEN'S MOST EXCELLENT MAJESTY. MDCCCLXXIX. PREFACE. When, in the year 187i, Her Majesty's Government detcrniined to despatch several expeditions to observe the Transit of Venus, the Council of the Royal Society resolved to request the Treasury to attach naturalists to those destined for Kerguelen's Land and Rodriguez, two of the least explored and most in- accessible oceanic islands of the southern hemisphere ; and a Committee, consisting of Sir- J. D. Hooker, Professor Huxley, and Mr. P. L. Sclater, was appointed to prepare the application, which was laid before Her Majesty's Government in the following terms : — " It is an unexplained fact in the physical history of our globe, that all known oceanic archipelagos distant from the great continents, with the sole exceptions of the Seychelles and of a solitary islet of the Mascarene group (which islet is Rodriguez), are of volcanic origin. According to the meagre accounts hitherto published, Rodriguez consists of granite overlaid with limestone and other recent rocks, in the caves of which have been found the remains of recently extinct bii'ds of a very singular structure. These facts, taken together with what is known of the Natural History of the volcanic islets of Mauritius and Bom'bon to the west of Rodriguez and of the granitic archipelago of the Seychelles to the north of it, render an investigation of its natural products a matter of exceptional scientific interest, which, if properly carried out, cannot fail to be productive of most important results. " As regards Kerguelen's Land, this large island (100 by 50 miles) was last visited in 18i0 by the Antarctic Expedition under Sir James Ross, in midwinter only, "when it was found to contain a scanty Plora of flowering plants, some of which belong to entirely new types, and an extraordinary profusion of marine animals and plants of the greatest interest, many of them being representatives of North-temperate and Arctic forms of life. " H.M.S. ' Challenger ' will no doubt visit Kerguelen's Land, and collect largely ; but it is evident that many years would be required to obtain even a fair represen- tation of its marine products ; and though we arc not prepared to say that the a 3 [ vi -J scientific objects to ho obtained by a naturalist's visit to Kergucleu's Land arc oi equal importance to those ^ylucll Eodrigucz will yield, avc cannot but regard it as in every respect most desii\ible that, the rare opportunity of sending a collector to Kerguelcn's Land should not be lost." Her Majesty's Government acceded to the request preferred by 1 lie President and Council ; and subsequently, on their* recommendation, the Treasury sanctioned the appointment of four naturalists, three to Rodriguez and one to Kerguelcn's Land. Those selected for the -work in Rodriguez were Dr. I. B. Balfour, who Avas chai'gcd with the duties of botanist and geologist; Mr. George Gulliver, who was directed to investigate the Fauna generally ; and Mr. H. H. Slater, whose special duties consisted in the exploration of caves and in collecting the remains of extinct animals. The naturalist attached to the Kerguelcn's Land expedition was tho Rev. A. E. Eaton, who was well qi^alificd to bring to bear his experience of the Arctic fauna and flora upon those existing in the southern hemisphere under parallel physical conditions. The collections and observations made by these naturalists fulfilled the expec- tations of the Council. A Committee appointed to consider the best means of rendering the collections serviceable to science, recommended that they should be entrusted to competent persons for examination and description, and that their reports should be published as a separate volume of the Philosophical Transactions. This recommendation was adopted by the Council, who requested Sir J. D. Hooker and Dr. Giinther to undertake the editing of the work. With regard to the specimens, the Council directed that complete sets shou.ld be reserved for the National Collections, and that the remainder should be distri- buted to the Museum of the Royal College of Surgeons in London, the Government Museums of Natural History in Edinburgh and Dublin, the University Museums of Oxford and Cambridge, and other Institutions. By Order of the President and Council. Burlington House, March 1879. G. G. STOKES, ^ Secretaries, T. H. HUXLEY, J I^oijal Society. TABLE OF CONTENTS. THE COLLECTIONS FROM KERGUELEX ISLAND. IxTRODTJCTORT NoTES. By the Rev. A. E. Eatox. I. The Bhysical Features of Kerguelen Island page 1 II. Recent visits of Naturalists to Kerguelen Island 1, BOTAKY. Observations on the Botany of Kerguelen Island. By J. D. Hooker, P.B.S. 5 Flowering Blanfs, Ferns, lycopodiacece, and Characece. By J. D. Hooker, P.E.S. {Pis. I, II.) 17 Musci. By W. Mitten, A.L.S. {PI. Ill, Figs. 1-5) 2i Bepaticce. By TV. Mitten, A.L.S. {PI. Ill, Figs. 6-11) 40 Lichens. By the Eev. J. M. Crombie, F.L.S 46 Marine Alga;. By G. Dickie, A.M., M.B. {PI. V., Fig 3) 53 Freshwater Algm. By Hr. P. E. Reinsch. {Pis. IV. and V., Figs. 1, 2) 65 Fungi. By the 'Rey.'M. 3. B^TiKiihEr, M.A., F.L.S. 93 Zoology. Seals and Cetaceans. By W. H. Flower, F.B.S. 95 Birds. By^.B.SBJLRVE, F.L.S. {Pis. VI.-VIII) 101 Eggs of Birds. ^// H. Saunders. F.Z^.5' 163 Fishes. By Dr. A. Gunther, F.B.S. 166 Mollusca. By E. A. Smith, F.Z.S. {PI. IX.) 167 Polyzoa. By G. Busk, F.B.S. {PI. X.) 193 Crustacea. By E. J. Miers, F.L.S. {PI. XL) 200 Entomostraca. By G. S. Brady, 3I.D., F.L.S. {PI. XII.) 215 Arachnida. By the Rev. O. P. Cambridge. {PI. XIII.) 219 [ viii ] Lisecid. Obscrrtflioiis on the Insects of Kei'gnelcn Island. Jh/ llic 1{ev. A. E. Eaton, M -I page 228 Cok'opteni. ii// C. U. AV.vtkuhousk. {P/. A'IF.) 230 Lcphloptero. By the Ricv. A. E. Eaton, JI.A 235 Bipfeni. Bi/ G. 11. Veuuall. (P/. XIV.) 238 Xeufoptei'ii. 5// //((' Rev. A. E. Eaton, J/.^y 218 CoUenibolii. B;/ Sir J. Lubbock, F.Il.S. {PL JUL) 249 Jlallophaffd. Bii C. Giebel. {PI. XIV.) 250 Marine Annelida. By W. MacIntosii, F.R.S. {PI. XV.) 258 Terrestrial Annelida. By E. Ray Lankestek, F.E.S 261- Echinodermata. By E. A. Smith, F.L.S. {PL XVI.-XVII.) .... 270 Actinozoa. By the Rev. A. E. Eaton, M.A 281 Eydroida. By Prof. Allman, 3I.D., P.L.S. {PL XVIII.) .... 282 Spongiida. By H. J. Carter, F.B.S 286 THE COLLECTIONS FROM RODRIGUEZ. Introdtjctort Notes. I, The Physical Features of Rodriguez, ^y Is. B. Balfour, *S'c.D. . . 289 II. Reports of Proceedings of the Naturalists 298 Petrology. By N. S. Maskelyne, F.R.S. . . 296 Botany. Introductory Remarks. By Is. B. Balfour 306 Flowering Plants and Ferns. By Is. B. Balfour. {Pis. XIX.-XXXVII.) 326 Musci. By W. Mitten, A.L.S. {Pis. XXXVII, XXX VIII., A. B.) . 388 HepaticcB. By W. Mitten, A.L.S. {Pis. XXXVIIL, C. D.-XL.) . . 396 Lichens. By the Rev. J. M. Crombie, F.L.S 102 Fungi. By the Ti^Y. M. J. Behkeley, M.A. , F.L.S 413 Algce. By G. Dickie, M.B., F.L.S. 415 Zoology. ILrtinct Fauna : Observations on the Bone Cares of Rodriguez. By H. H. Slater, B.A. 420 Birds. By Dr. A. Guxther, F.R.S., and E. Newton, M.A., F.L.S. {Pis. XLI.-XLIII.) 423 On the Osteology of the Solitaire {Pezophaps solitaria). By E. Newton, M.A., F.L.S., and J. W. Clark, 3I.A. {Pis. XLIV.-L.) .... 438 Reptiles. By Dr. A. Gunther, F.R.S. 452 [ i>^ ] Recent Fauna : MammaUo. By G. E. Dobson, 31. A page 457 Birds. By E. B. Suakpe, F.L.S 459 Beptiles. By Dr. A. Guntuer, F.B.S 470 Fishes. By Dr. A. Guntuer, F.B.S. . 470 Mollusca. By E. A. Smith, F.Z.S. (PI. LI.) 473 Crustacea. By E. J. Miees, F.B.S 485 Myriojjoda and Arachnida. By A. G. Butler, F.L.S. {Bl. LII. ) . . 407 Coleoptera. By C. O. Waterhouse. {PL LIU.) 510 Hymenoptcra, Biptera, and Neuroptera. By E. SiiiTii 534 Lepidoptera. By A. G. Butler, F.L.S. 541 OrUioptera and Semiptera. By A. G. Butler, F.L.S. {PI. LIV.) . 545 Annelida. By Prof. Ed. Grube 554 Turbellaria. By G. GvJaJAYETi,B.A. {PI. LV.) 557 FcUnodermata. By E. A. Smith, F.Z.S. {PI. LI.) 564 Corals. By Dr. E. Bruggemann 569 THE COLLECTIONS FROM KERGUELEN ISLAND. INTRODUCTORY NOTES, By the Rev. A. E. Eaton, M.A., Naturalist to the Expedition. I. The Physical Features of Kergiielen Island. Kerguelen Island is little else than a succession of hills and mountains formed almost exclusively of volcanic rock. In its greatest diameters it measures about 80 miles from X.W. to S.E., and 70 from N.E. to S.W., but no part of the interior is farther than 10 or 12 miles from the sea, for the coast on all sides is exceedingly intricate and abounds in large inlets and nari'ow fiords, which run far inland at frequent intervals between ranges of precipitous hills. A district of considerable extent in the midst of the island is occupied by snowfields, whence glaciers descend east and west towards the sea. This district is bounded on the S.E. by a series of snow-clad mountains and lofty hills extending in a curve northwards from Mount Ross almost across the island. The area intermediate between this range, Royal Sound, and the Mount Crozier hills, and likewise the islands in Swains' Bav and Royal Sound, contain series of interrvipted ridges, the majority ol' which are crowned and terraced horizontally with basalt more or less amygdaloidal, and rarely exceed 600 feet in altitude. Their south-eastern and eastern slopes are generally very favourable for vegetation, being sheltered from the prevailing winds ; but scarcely any of them attain to. heights sufficient to be suitable for the growtli of plants restricted to elevated sites in this island. The Mount Crozier hills, and those on the opposite side of the soimd, are upwards of 2,000 or 3,000 feet high. Their summits are covei-ed with snow xuitil late in the season. Most of them are massive and rather simple in contovu-, but near Mount Crozier a few of the peaks are singu- larly picturesque, bristling with pinnacles, needles, and castellated towers of rock, A ■2 THE COLLECTIONS FROM KEUGUELEN ISLAND. some of which arc visihU^ from the southward, l)ul iu)t many. A similar depai'tm-e from tlio iirt'vailing typo of Ivovguelen Ishiiul scenery is noticeal)h' among the hills near Sprightly ])ay, and the aiguilles on the sontli side of Mount Ross. Au active volcano is reported to exist in the neighbourhood of Bonfire Beach, and at one or two jilaces on the same side of the island it is said that tlun-e ai-e hot springs, ri'sorted to by the sea elephants for the purpose of recreation at certain seasons of the year. A cold mineral spring oozes forth at the liead of a small patt-h of boggy ground very near the Avestern terminus of Swains' Haulover. It lies within a stone's throw of the sea, in a line with the nearest end of one of the nameless islands in the bay and the apex of the semi-pyramidal extremity of a high preci])itous hill, which constitutes a conspicuous landmark on the opposite shore. The Avater is free fi'om smell, but has a pronounced mineral flavour, secnringly of alum. On the steep slopes bordering lloyal Sound, near Swains' Haulover to the S.E., two or tlu-ee small patches of yelloAV clay are visible. The clay appeared to include nothing but small fragments of volcanic rock, similar in composition to the masses ill sitii aroimd it. A little farther on, along the same shore, opposite the south- eastern jioint of Seal Island, one of the hills is intersected by a nearly vertical wall of trap conspicuous from the interA'cning channel. The coal beds found by Dr. Hooker at Christmas Harbom" and Cumberland Bay, and a stratum at the first-named place, containing silicified trunks of trees, are almost the only foi'mations in the island that are not distinctly of igneous origin. Some limestone, however, is stated to occur at Foundry Branch, and after our departure from the country a cast of a fossil conchifer was shoAvn to me by Mr. J. Stone, Assistant Surgeon to H.M.S. " Supply," AA'hich AA'as given to him by one of the sealers, who had picked it up " somewhere near Thumb Peak." The whole island, exclusive of the snoAvfields already mentioned, abounds in freshwater lakes and pools on the hUls and lower ground. Counting large and small together, there were about two dozen of them Avithin a two and a half mile radius of the chief English observatory ; two of these were upAvards of a mile in length, but the majority of them were much smaller. Beyond this limit, hoAvever, some lakes of considerable size existed. They appeared to be uninhabited by fish or by Neiiroptera, but a small Entomostracon {Centropages) AA'^as exceedingly plentiful in many of them. Between the eastern base of the Mount Crozier hills and the coast a broad tract of low-lying land slopes very gently towards the sea. It is saturated with the di'ainage of the adjacent heights, and Avith the excessive precipitation upon it of rain and snow, resulting from its situation immediately to leeward of them. Some portions of it are said to be unsafe to persons attempting to walk over it, either through being undermined by subterranean streamlets, or on account of the exist- ence of dangerous bogs. Streams descending from the steep declivities of the hills wear by degrees narrow trenches in the stiff clayey soil of the low ground, Avhose INTRODUCTORY NOTES.— REV. A. K. EATON. 3 sides, in course of time collapsing by their own weight, meet and groAv together in an arch aboA'e, being kept by roots of plants (especially Azorollo) from crumbling to pieces. In the tunnel thus formed the work of excavation proceeds ; the hidden channel is enlarged until it acquires the dimensions of a chasm, thinly roofed over by a layer of living plants, continuous with the surface of the firm adjacent groiuid, but easily broken through, into whose dark cavernous entrance the stream in mid career tumbles abruptly to a depth of many feet, sooner or later to reissue at a lower level. Years pass by ; the chasm becomes wider, until at length it is dis- closed by the subsidence of the yielding vault into the deep cavity beneath. Dr. Kidder states that some of the pits excavated in this manner by streams in the district in question are upwards of 30 feet in depth ; the tunnels seen by me were of much smaller dimensions. The sides of such trenches and the entrances to the tunnels frequently offer favom-able sites for the more delicate species of llepaticcB. Walking over the island is at certain localities very unsafe, in consequence of the frequent occm*rence of mud-holes, the surface of which is precisely similar in appearance (or veiy nearly so) to many sm-rounding patches of firm bare soil. Holes of this description are produced by Tinderground springs deep enough down to permit the siu-face to remain tolerably dry, while they keep the earth below in the consistency of liquid mortar, the water that issues fi'om the soiu'ce escaping by percolation through the soil. They are sometimes dangerously deep. The Antarctic Expedition, in the course of an overland journey in the northern part of the island, found their average depth on the western coast to be up to a man's waist. In places where the ground is soft and boggy (as is frequently the case on gentle slopes and in the valleys) it is well to take notice of the kind of plants growing upon it. Patches of incoherent Sepaticce and tremulous sheets of moss are best avoided. Stunted Azorella is somewhat uncertain, but is generally firm. Sj)aces overgrown with Accena, as a rule, may also be traversed with safety. But Avherever the surface waves freely, whatever may be growing there, it is advisable to be cautious. So serious are the obstacles to travelling on foot in Kerguelen Island presented by fiord, lake, bog, and ten-aced hill, that some parts of it are absolutely inaccessible by land from districts immediately adjoining them, while others can be reached only by following the devious windings of a complicated maze. Measm-ements of the map afford no very trustworthy bases for estimates of the time needed for a jom-ney of a given length in a certain direction. The climate of Kerguelen is tempestuous, chilly, and wet. Days perfectly calm are of extremely rare occurrence, gales, or at least strong breezes, being almost con- stant. The wind is usually westerly, and cold ; but now and then, in the south of the island, it comes on to blow from the E. and S.E., and this change is attended with a rise of tempera tm'C. In a treeless country, such as " Desolation," the violence of a gale is recognised not by ravages committed by it, but by transient evidences A 2 4 THE COLLECTIONS KUOM KERGUELEN ISLAND. of its I'oivo cxliihitod at the tiiac> of its jjroii'ivss. WattTlaUs (U-sa-iuluig Iruiu (lie dirt's air inteivi'ptod in mid air, and driven backwards in continuous clouds of spray to ihr hills, lVi)n\ whence they leap down, h>a\ iiiu' tlie stream hcil empty below; aud similar clouds swept up from lakelets on the mountain side mil rajjidly away to leeward, like vapour from a cauldron. Ihit perhaps the strangest sights may he ohservetl at such times from the shelter of a clilV, for there the air resounds with the gale, whicli is not felt, while eddies launched from the sununil plunge wiili tremendous force upon the sea, scoiu' the surface into foam, aud hurtle the water to and fro with fitful violence. Suddeu squalls and " willy-waughs " from the hills, such as are usually met with olV mountainous coasts, prevail in many parts of the island, and cause boat navigation to be attended with considerable risk. But the uneoueentrated force of the usual wind is the greatest obstacle to boating. In Observatory Bay, when the wind was off shore,. it was often impossible for three or four days together for anyone to land from the " Volage " and " Sui^ply," thougli their distance from the landing place was only 300 or 400 yards to leeward. Tlu> range of temperature throughout the year does not appear to be excessive, the highest readings of the thermometer in summer being under 70° P., and the lowest in winter seldom being less than 02". At Christmas Harbour it did not descend below 27' during the stay of the " Erebus." Before the English expedition arrived at the island the Americans, early in September, found the temperature one night to be as low as 18°. In the warmer months the readings are not often much hi "-her than 55^ or 56°, or mucli lower than 42°, on the eastern side of the island. The western coast is, however, much more bleak than the other, as is cvideiiced bv the belt of herbage adjacent to the sea attaining a lower altitude there than it does on the eastern side of the island. On exposed parts of the S.W. shore the hill sides are conspicuously green to a height of hardly more than 150 to 200 feet (as it seemed to the eye) ; but in Eoyal Sound, and on the lee side of the island, the upper limit of verdure is between 500 and 700 feet above sea-level. On the high hills and moimtains fog is very prevalent, but districts of only slight elevation are comparatively free from it, and enjoy more fn^quent 'sunshine than other portions of the country. "While the English were in Royal Sound the glimmer of distant lightning was seen one night in the direction of Mount Ross ; but this was a very exceptional occurrence. Such displays of the aurora as were observed were not remarkable for their brilliancy. II. Mecent Visits of NdlaruUsls to Kcrguelen Island. Kerguelen Island has been visited within recent times by five scientific expedi- tions,— the Antarctic under Sir James Ross, the Challenger under Sir George Nares, and three Transit of Venus' expeditions. Christmas Harbom- was selected for head-quarters by the Antarctic Expedition, which remained there from the 15th of May until the 20th of July 1840. It is a INTRODUCTORY NOTES. — REV. A. E. EATON. 5 deep ialet ou the cast coast at the uortli ond of tlic mainland^ bounded on oacli side by steep terraced hills more than 1,000 feet high, and terminated by a sandy beach at the foot of a gentle slope of moderate altitude. At the entrance it measures a mile across, but half Avay u]) it is abruptly narrowed to a quarter of its fomier breadth. Lying lengthwise in the direction of the prevailing winds it is open to every gale that sweeps over the neck of low land at its head ; and consequently it Avould seem to be not the best site in the island tliat could be chosen for the prose- cution of researches in natural history. In fact a collector stationed here Avould have good cause to utter Caliban's remonstrance : — " here you sty me " In this hard rock, whiles you do keep from me " The rest of the island." . . For it apparently is quite cut off by precipices from the country immediately to the south of it, — an obvious disadvantage. Notwithstanding these drawbacks, how- ever, many interesting specimens can be obtained here. A small tarn, not far from the sandy beach, contains a peculiar species of Bmmnctilns (discovered by Mr. Moseley). At the foot of a huge mass of dark volcanic rock that is conspicuous on the south side of the harbour, between the siimmit of the hill and the movith of the bay, is fossil wood in abundance. In a small loay on the same side, close to the Arch Rock at the entrance to the harbour, and 30 feet above the sea, and again in a little cave in shale near the centre of the small bay formed by C. Fran midst of the small graveyard at Betsy Cove, whieli is described by Sir C. Wyville Thomson, op. cif. p. 815. No complete account of the natural history specimens collected by this expedition has been published as yet. The cove is favourably situated for a naturalist's station. To the south and west are the Mount Crozier hills, affording much variety of altitude and exposure. On the east and south-east a broad tract of low marshy land is extended. Probably the situation is not well adapted for marine species that require very sheltered water ; but others which haunt the open coast are found there in plenty. Dredging Avas conducted from the " Gazelle " in the neighbouring sea ; and it may be anticipated that the results obtained by this means will prove to be the most important part of the collection. Considerable attention was bestowed upon the vertebrata, careful etchings being prepared of many species. Large series of specimens of birds in different stages of development, and the skins and skeletons of three species of seals, are other specialities of this collection. It is almost to be regretted that this district should have been worked over twice at the same season in consecutive years, when so much of the island is terra incognita. The American Transit of Venus Expedition occupied a position on a slope near Molloy Point in Royal Sound, close to the commencement of the rocky beach which extends from thence to the Prince of Wales' Eoreland. Speaking roughly, the country adjoining this position is similar to the district in the neighbourhood of Betsy Cove, differing chiefly in its distance from the open sea and its situation on the opposite side of the Mount Crozier hills. The expedition remained at Molloy Poiut from the beginning of September until the commencement of the second week in January. Dr. Kidder, the surgeon at the astronomical station, studied the zoology and botany of the locality, and obtained many interesting observations of the habits of birds, together with a large collection of skins and eggs. The results obtained by him are recorded in Nos. 2 and 3 of the Bulletin of the United States * This publicatdou contains a detailed account of the past history of the island, and of its leading physical features, together with a summary of the proceedings of the expedition while there, and notices of interest- ing Natural History incidents. INTRODUCTORY NOTES.— REV. A. E. EATON. 7 National Museum, TVasliington, 1876. They include the i'ollowing species: — 2 mammals (one a seal) ; 20 birds ; 3 or 4 fishes ; upwards of 12 insects, besides 3 Arachnida ; 7 Crustacea ; 3 or more Annelida ; 11- Mollusca ; several Timicata and Polyzoa, and 4 Echiuodermata ; 17 Phanerogams ; 6 ferns and Lycopodia ; 28 Musci ; 18 Lichens, and 22 marine Alg£B, total 69 or more species of animals, and 91 of plants. The English Ti-ansit of Venus Expedition sailed from Simon's Bay on the 18th of September in H.M.S.S. " Volage " and " Suj)ply." After a stormy passage the vessels joined company at the rendezvous at Three Island Harbour in Royal Soimd on the 11th of October. Leaving this in the afternoon of the following day, they came to anchor in Observatory Bay in the morning of the 13tli, whence they took their final departure on the 27th of February 1875. Collections were made at the following places while the expedition was at the island : — Cat Island, in the after- noon of October 11th ; Observatory Bay and the adjoining district, to a distance of five miles inland, dm-ing most of October, November, December, and February ; isthmus at the head of Carpenter's Cove, January 4th-9th ; Thumb Peak, Decem- ber 7th noon till 10th morning ; vicinage of second astronomical station and hills near Swain's Haulovcr, October 27th, afternoon December 22nd-24th ; shores and islands of Swain's Bay, January 15-30th. The character of the localities just mentioned has been sufficiently described above in the general account of the island. The " isthmus at the head of Car- penter's Cove " is a high pass which divides the Mount Crozier hills from those to the westward of them. The summit and upper portion of this pass is referred to as a locality, in some of the botanical reports, as a "hill N.W. of Mount Crozier." A large bog at the foot of the northern declivity of the pass is described as a " bog near Vulcan Cove," and a waterfall descending the hills a little to the east of it is similarly specified as a " waterfall near Vulcan Cove " ; but their real position is doixbtful, and may be nearer the promontory enclosing an inlet adjacent to the entrance of Foimdry Branch than to the cove in question. The collections represent the faima and flora of the inland and sheltered portions of the island from altitudes of 500 or 600 feet above the sea down to the depth of about 10 fathoms along the coast, and are as a rule rather deficient in duplicate specimens. The groups wliich are most poorly represented are those which at the time of collecting were known to have been the special subjects of investigation by other natiu-alists in the island, e.g. mammals, l)h'ds, and the Phanerogamic plants. In such groups, with the exception of certain particidar species, it was considered that the interests of science wouJd be best consulted by restricting the series oi' examples to minimum proportions, in order that more time might be devoted to the acquisition and preservation of specimens in less popular branches of uatxu'al history. The number of species that came under observation Avcre as follows : — 3 (besides 1 undetermined) native mammals and 2 or 3 introduced ; 22 (with 1 or 2 8 THE COLLKCTIONS FROM KKKGUKLEN ISLAND. (loiil)! Till) birds and oij:gs of 1(» spivios ; t (wilh 1 doulil I'lil) species of lishos ; 21 (and 5 undetermined) native insects, besides 1 introdueed; ,') (and 2 undetermined) terrestrial -:\a'aebniil:i ; iipwartls ol' 12 Crustacea ; 8 Aiiiiclid,-! ; 25M()lhisca; ;} 'I'uni- cata, and 2G Polyzoa ; K5 Eciiiuodermata ; 2 Actinozoa ; 7 livib'oida ; S sijohi'ts ; 20 native (I)esides 1 intn)dnc(<(l) si)ei-ies of llowei-ing; jilanls ; (i ferns and Lycopotb'a ; 1 Cliaraceie ; t;} Musci ; 15 irepaticjc ; ."» Funi;i; r>l or r)2 species, 1)(>sides 9 named forms of Licbens ; 51 marine and 81 freshwater Alga>, in all about 170 species of animals, and 277 of plants. Preliminary notices and descriptions of ncAV si)ceies in the collections were published in the Annals ami Mai^azine of Natural History for 1875 and 1S7(), the Entomoloji'ists Monthly jMagazine for August 1875 and Aua:ust 1870, the Journal of Botany for February 1870, and the Journal of the Linnean Society (Botany) for JiUy 1870. In such a country as Kerguelen Island no one without assistance could execute a fairly complete survey of the botany and zoology of an area lying within the radius of a day's walk from his bead quarters in less than a month or six Aveeks ; and then it woidd hold good only for that period of tb<> year at Avhicb it was made. Much, therefore, has to be alloAved for before the absence of a species from a collection can be attributed without question to its non-existence in the district in which that collection was formed ; while on the other hand, should the physical conditions presented by the district be unfavourable to its occurrence, its absence docs not indicate deficiency in point of completeness in the collection. BOTANY. Observations ox the Botany of Kekguelen Island. By J. D. Hooker, P. U.S. The history of the botany of Kerguelen Island (also called Ker^uelen's Land, and Desolation Island), previous to the visit of the Rev. Mr. Eaton, the last and most com- plete explorer of its flora, is a very brief one. It commences with the visit of Capt. Cook during his third voyage, in the narrative of Avhicli the vegetation of the island is thus described by Mr. Anderson, the surgeon of the " Resolution : " " Perhaps no " place hitherto discovered in either hemisphere, under the same parallel of latitude, " affords so scanty a field for the uatm-alist as this barren spot. The verdure which " appears, when at a httle distance from the shore, would flatter one with the expec- " tation of meeting Avith some herbage ; but in this we were much deceived. For " on landing we discovered that this lively colour was occasioned only by one small " plant, not much imlike some sorts of Saxifrage, which grows in large spreading " tufts, to a considerable way up the hills." Mr. Anderson proceeds then to give some particulars of this plant [Azorella Selago, Hk. f.), of the cabbage {Prhiglea atitiscorhutlca, Br.), of two small plants found in boggy places, which were eaten as salad, one " almost like garden cress and very fiery " (probably Rajinncuhis crassipes, Hk. f.), the other very mild and " having not only male and female, but what bota- " nists call androgynous plants" (? Callitriche). He adds to these a coarse grass {Foa Cookii, Hk. f.), and a smaller sort which is rarer (probably Descliampsia antarctica, Hk.) ; a sort of goose-grass (? Cotula 2}iif»iosa, Hk. f.), and another small plant much like it (this I do not recognise). " In short," he says, " the whole " catalogue of plants does not exceed 16 or 18, including some sorts of moss and a " beautiful Lichen" [Neiiropogon Taylori, Hk. f.) "which grows higher upon the " rocks than the rest of the vegetable productions. Nor is there the least appear- " ance of a shrub in the whole country." The date of Cook's visit was the summer of 1776, and the specimens obtained by Mr. Anderson were deposited in Sir Joseph Banks' Herbarium, which subsequently became the property of the nation, and is preserA^ed in the British Museum. Not having been poisoned, all the Kerguelen Island plants were, when I examined them in 1843, much injm-ed by insects, and many were entirely destroyed. From 177G till 1810, when the Antarctic Expedition imder Capt. (afterwards Admiral Sir James) Ross, anchored in Christmas Harbour, Kerguelen Island is not known to have been visited by any ship of war, or by the Discovery or Sm"veying ships of any nation, though it had become the frequent resort of English and B 10 BOTANY OK KKRGUELEN ISLAND, American scalers. Durinp; the stay o\' thr ;il)ovt'-iiamcil expedition :ill tlic plants enumerated bv Anderson as found l»v him in mid-sununer wci'C refound in mid- winter, togetluM- u itli ninny move, amounting- to nearly 150, ol" whieh 18 were flowerini; plants; the other large classes being- mosses and ITepatica) 35, Lichens 25, and Alga> 51. These have all been described in the botany of the voyage (Flora Antarctica, Tart II., 1S47). The next visit of naturalists to Kerguclen's Land was that of the " Challenger " Expedition in January and February 1874, when Mr. Moscley collected mostdiligently, both in Christmas Harbour and on the east coast GO to 70 niiles south-cast of it. He found 23 flowering plants in all, including three European weeds, all anuvxals and doubtless inijiorted by sealing parties {Cerasfiiim tricialc, JPoa praicnsis and omnia), and tlu-ee species not in the collections of the Antarctic Expedition (two Ho II II II cull and an Unciiiia). He also procured flowering specimens of the two endemic genera Friiigleo and Lyollia, and made large accessions to the cryptogamic flora, especially from the southern localities A'isited. J\Ir. Moscley had also the good fortune to land upon Marion Island, 1,650 miles to the Avest of Kerguelen Island; and on Yong Island (of the Heard group), about 120 miles to the south-east of it, neither of which had been previously visited by any naturalists, and in both of which he found some of the most peculiar of the Kerguelen plants. Mr. Eaton arrived at Kerguelen Island with the Transit of Venus Expedition eai'ly in October 1871, and left towards the end of February 1875, during which time he collected diligently, chiefly at Eoyal Sound, Swains' Bay, and Observatory Bay. He obtained nearly all the flowering plants of previous explorers, and added very largely in the Cryptogams, especially to the Alga3. Nearly contemporaneous with Mr. Eaton's visit was that of the American Transit Expedition, on which Dr. Kidder was the natm-aUst. He arrived in September 1874 and left in January of the following year, having explored some of the same localities as Mr. Eaton. His collections, amounting to about 90 species, are described in the bulletin of the U. S. National Musevim, No. 3, issued in 1876 by the Government Printing Office of Washington. The floweiing plants and ferns are revised by Prof. A. Gray; the mosses are described by Thos. P. James; the Lichens by Prof . E. Tuckerman, and the Algae by Dr. W. G. Farlow. Except amongst the Lichens, there are very few novelties. Dr. Kidder adds a list of seven plants from the Crozets, all identical with Kerguelen Island species.* The botanical results of the German Transit Expedition to Kerguelen Island are not yet published. The three small archipelagos of Kerguelen Island (including the Heard Islands), Marion and Prince Edward's Islands, and the Crozets, are individually and collec- tively the most barren tracts on the Globe, whether in their own latitude or in * He ako mentions " a small vine with blue flowers growing amongst scoriaj," of which no specimens were collected. This is probably some endemic plant unknown to botanists. BOTANY OF KERGUELEN ISLAND.— DR. HOOKER. 11 any higher one, except such as lie within the Antarctic Circle itself ; for no land even within the N. Polar area presents so impoverished a vegetation. The chief interest attached to the flora of these archipelagos lies in the indication it atfords of their being, in all probability, the remains of a much larger land area, -which, thougli peopled with plants mainly from the southern extreme of S. America, 4,000 miles to the westward, possessed an endemic flora of its own, wliich included forest trees of considerable dimensions. Before, however, proceeding to discuss the relationships of their floras, I shall describe that of the largest and the only one that is at all well known. As pointed out in the "Elora Antarctica," the prevalent features of the vegetation of this island as then known were Fuegian ; one species of flowering plant alone, of those that are not peculiar to the island, being characteristic of any other flora, namely, the Cotida plumosa, wliich is found elsewhere only in the Auckland and Campbell Islands, south of New Zealand. More recent collections have con- fii'med and even strengthened thig Fuegian affinity, for of the three additional flowering plants procui'ed by subsequent explorers, one is Fuegian {Ranunculus trullifoUus), another {R. Moseleyi) is closely allied to a Fuegian species, and the third one, TJncinia compacta, is a native of the mountains of New Zealand and Tasmania, and this is so nearly allied to a Fuegian species that it may prove to be a form of a plant common to all high sovithern latitudes. Not only has a further knowledge of the Kerguelen Island flora strengthened its known affinities with the Fuegian, but recent discoveries in the latter flora have done so too ; some of the Kerguelen's grasses especially proving to be more closely allied to Fuegian species than was suspected. The discovery of the flowers of the endemic Kerguelen genus LyalUa is another instance of this affinity. In the Flora Antarctica, judging from the fruit alone, the flowers being unknown, this remarkable plant was provisionally placed in Tortulacece, its resemblance in habit and foliage to the andine genus Pycnophyllum being indicated. Complete specimens collected by Moseley prove its close relationship to the latter genus, in juxta-position with which it had indeed been placed in the Genera Plantarum, where both liad lieen referred con-ectly to CaryophyUece. The elements of the Phsenogamic flora of Kerguelen Island may be thus classified : — 1 Endemic genus, which has no near ally — Pringlea antiscorbutica. 1 Endemic genus allied to an Andean one — LyalUa kerguelensis. 6 Endemic species allied to American congeners — Raniaicnlus crassipes and Iloseleyi, Colobantlms kerguelensis, Accena offinis, Poa Cook'd, Festiica kergtielensis. 5 species common to Fuegia biit not found elsewhere : Ranunculus trullffoli/is, Azorella Selago, Galium anta7'ctictim, Festuca erecla, Deschampsia aiif- arctica. B 2 12 BOTANY OF KERGUKLEN ISLAND. (i species common to America, and also to jS'cw Zealand and the islands sonlli of it. Tillu'o oioschalo, MoiiCui foiifaiia,* Ctillilric/ic obtusaiKjnla* Linimclla aqttallca* J/i/iciis oc/u'tizi'i-ionles, Ayrostis Mayclhotica. (!Most of these arc aquatii" or marsh plants, and those marked witli an asterisk arc also European, and \ery widely dispersed.) 2 species found elsewhere but not in Fueg-ia, Cotula plumosa, common to Lord Auckland's group and CamphcU's Tshind south of New Zealand, and Vuciuia compacta, a native of the mountains of Tasmania and New Zealand. This American affinity of the Iverguelen Island llora thus clearly established by its flowering plants is very strongly manifested by its Cryptogams, amongst which, however, the only evidence of migration from Soutli Africa occurs. Tbis is the case of Polypodiitm vidgare, a widely distributed fern in the north temperate zone, but known in the southern only from the Cape Colony, Marion, and Kerguelen Islands ; what is further curious respecting it is, that the Kerguelen Island individuals arc referable to a variety with pellucid veins, hitherto known only from tbe Sandwicli Islands. As to the local grouping of the Kerguelen Island plants, that of the Phyonogams is not altogether in harmony with the Cryptogams, the former seeming to be by far the most ubiquitously dis^iersed of the two groujjs. All the plants hitherto collected have been from two areas, one, Christmas Ilarbour, in the extreme north, extending about five miles either way ; tlic other, considerably larger, occupies tbe south-east coast, and following it extends for about •iO miles. The distance between these areas is about 60 miles in a N.W. and S.E. direction. Of tbe Pha^nogamic plants, 19 were found in the nortbern area, nearly every one of Avhicb was also found in the south-eastern one, where but two addi- tional species were collected ; whereas of the 150 Cryptogams found in the northern area, a large proportion were not found in the south-eastern, Avhere, hov/ever, nearly four times the number of species were obtained. Again, whilst but one fern w^as found in the north, four occur in the south-east. Of 35 Ilusci and Hepaticce col- lected at Christmas Harbour by the Antarctic Expedition, hardly half Avere found at Swain's Bay, Betsy Cove, or Pvoyal Sound, which localities yielded about 80 additioiial species. Nearly 50 marine Alga; were collected at Christmas Ilarbour, of Avhich 18 did not occm* in the south-eastern coasts, Avhere upAvards of 30 additional species were obtained. In the case of the Lichens, the discrepancy is still more marked, but this is possibly more apparent than real, and is to be attributed in part to the difficulty of defining the species and recognizing them from descriptions ; and m part to the difficulties caused by the irreconcilable views of Lichenologists as to the limits of the species of this order. "^'hatever other causes there may l)e for this anomalous distribution, one, no doubt, is the nature of the Christmas Ilarbour area. This is almost occu])icd by transverse valleys that run east and west completely across the north tip of the BOTANY OF KERGUELEN ISLAND.— DR. HOOKER. 13 island, from sea to sea, are bouiulctl h\ hills 1,200 feet liigli, aud are perennially swept by terrific blasts from the westward. There arc, hence, no shelter on land for the terrestrial flora, and no qmet bays for the projjer development of a varied marine vegetation ; facts which may very Avell account for the paucity of Cryptogams in Christmas Harbour, but not for the presence there of nearly all the flowering plants of the island. Turning again to the soutli-eastern area, its more slieltored valleys and laud-locked harbours favour not only a greater development of Cryptogams, but also a far greater luxuriance of the Phainogams than obtains in Clu'istmas Harbour ; Avbich last fact renders the absence of additional species of Phainogams to the south-eastward all the more remarkable. The question remains, granting that the great majority of the Phccnogams of Kcrguelen Island are derived from South America, how was theu* transport efl'ceted ? Though this question cannot be satisfactorily answered by a reference to the facilities for distant transport possessed by the fruiting organs of the Kerguclen Island plants, it is only proper to refer to these organs in some detail. Obviously, regarding the whole flora, the plants with the most minute seeds or spores and the water-plants are the most widely distributed. Under these categories come — 1. The Fungi, of which all but 2 of the 8 species found are widely dispersed over the globe. 2. The marine Algae, of which only 8 out of the 7-i are peculiar to the island. 3. The fresh water Algae, of which 28 out of 80 are regarded as endemic. 4. The aquatic and marsh Phajnogams, 8 in all, of which 6 are widely dispersed. Of the Phaenogams, whether aquatic, marsh, or terrestial, none have appHauces for wide dispersion except the hooked style of the Rammculus, the reversed barbs of the AccBua (a most powerful aid), and the hooked organ attached to the fruit of TJncinia, also a very adequate aid. Kone of the others have any aid to dispersion, though they have small seeds or fruits. Tui'uing to the natural agents of dispersion, winds are no doubt the most powerful, and sufiicient to account for the transport of the Cryptogamic spores ; these, almost throughout the year, blow from Euegia to Kerguelen Island, and in the opposite du-ection only for very short periods, but appear quite insixfficicnt to transport seeds over 4,000 miles. Oceanic cm'reuts have, doubtless, brought the marine Algae ; but the transport of the seeds of the freshwater plants, of the grasses, and of the two plants with hooked and barbed appendages to the fruit, is not apparent in the case of a country that has no land bu-ds but an endemic one (the Chionis), and of which the water bmls come to land only or chiefly at the breeding season, and this after long periods of oceanic life in a most tempestuous ocean. Even supposing that the sea birds which habitually l)reed in Kerguelen Island did visit Fuegia between the periods of incubation, it is difficult to imagine that any seeds that had adhered to their beaks, feet, or bodies on leaving the latter country Avould not have 1)ecn removed by the buffets of wmds and waves over upwards of 4,000 miles of ocean. The supposition tliat more land formerly existed along the parallels between 14 BOTANY OF KKRGUELEN ISLAND. Fucg-ia and KiTguolcu Ishuul, j)ossil)ly in the lurm ol' iskuuls, iriuaius as the I'orloru hope of the hotanieal gooccrapher. By siu-h stepping stones the land hirds, so mimevous m the Falkland Islands (whieh lie in the direction of such hypothetical inlands), and of whieh the vegetation is identical with tliat ol' coldiT South Amevica, might, favoured hy the prevalent westerly gales, have passed from thence to Ker- guolen Island, having adhering to them fruits and seeds. The ahsenee of such hirds from the present Avi-fauna of Kergueleu Island olfers no ohstaclc to such a speculation, as such immigrants would on arrival speedily he destroyed hy the pre- datoiy guU and petrels of the island. Various phenomena, of very different relative value and nature, hut common to the tliree archipelagos, Kergueleu, the Crozets, and Marion, favour the supposition of these all having hoen peopled with land plants from South America hy means of intermediate tracts of land that have now disappeared ; in other words, tliat these islands constitute the wrecks of either an ancient continent or an archipelago which formerly extended further westwards, and that their present vegetation con- sists of the waifs and strays of a mainly Fuegian flora, together with a few survivals of an endemic one. The extreme southern point of South America, from lat. 52-54° and long. 70° "W. comprising Fuegia, is deflected to the eastward. Following its general direction, the Falkland Islands group is the first land met with (in long. 60° W.) ; its vegetation is comparatively rich and exclusively Fuegian ; it has, no douht, heen brought mainly by the land and freshwater bii'ds which abound there, and are identical with Fuegian ones. South Georgia is the next land met with to the eastward, in long. 35° W. and 54° S. ; of its vegetation nothing is known except for the scanty obser- vations recorded in Cook's voyage, wliich indicate its botanical identity with the Fuegia. Of Bouvet Island, the assumed position of which is long. 5° E. and 54° S., nothing is known ; it was searched for in vain by the Antarctic Expedition in 1843. Marion Island is 37° E. and 46° S., and the Crozets, in 48° E. and 47° S., are respectively about 1,650 and 1,200 miles west of Kergueleu Island, and there is no land intermediate between them. Now, from such specimens as have been obtained of the vegetation of the fii'st of these islands by Mr. Moseley,* it appears to be abnost identical with that of Kergueleu Island ; that is, to be Fuegian with the addition of some of the peculiar Kergueleu Island types,f and the same remark applies to the Crozets, J facts from which Mr. Moseley has drawn identically the same conclusions as those to which I had aiTived thirty-five years previously from a consideration of the Ker- gueleu Island flora alone. He says, speaking of Marion Island (Linn. Journ. XV., * Journ. Linn. Soc. XIV., 387, and XV., 484. t ^Marion Island contains several Fuegian species not hitherto found in Kergueleu Island, namely, Ranunculus biternatus, Hymenophyllum tunbridgense, and probably a llicrochloc (the scented gi-ass mentioned by Moseley), together with a Cape fern Aspidium mohrioides and an Asjjlenium. § See Kidder in BuU. U.S. Nat. Mus. No. 3, p. 3L BOTANY OF KERGUELEN ISLAND. — DR. HOOKER. 15 185), " the occiu-rence of JPringlea on the islaud, as also on the Crozets and Kerguelcn " Island, point to an ancient land connection between these islands, Avhich the " antiquity and extent of denudation of the lavas would appear to bear out. It is " difficidt to see how such seeds as those of Pringlea could have been transported " from one island to another by birds ; and the seeds seem to be remarkably " perishable; besides the distinctness of the genus points to a former Avide extension " of land on which its progenitors became developed. The existence of fossil tree " trunks in the Crozets and Kerguelcn Island points to similar conditions. In the Flora Antarctica, I say, p. 220, referring to the time required for the formation of the innumerable superimposed beds of volcanic rocks, as observed by me in Kerguelen's Land, and for the growths and destructions of successive forest vegetations that once clothed the island, and are now imbedded in strata at great depths, that this time is sufficient " for the destruction of a large body of land " to the northward of it, of which St. Paul's and Amsterdam Island may be the " only remains; or for the subsidence of a chain of mountains running east and " west, of which Prince Edward's Island, Marion, and the Crozets, are the exposed " peaks." And, at p. 210, when discussing the strvictural peculiarities of the Tr'uujlea, I say, " However loth we may be to concede to any of our vegetable pro- " ductions an antiquity greater than another, or to this island (Kerguelcn) a posi- " tion to other lands wholly different from that it now presents, the most casual " inspection of the land where this plant now grows wUl force one of the two " following conclusions upon the mind, either that it w^as created after the extinc- " tion of the now buried and for ever lost vegetation, or that it spread over the " island from another and neighbouring region, where it was imdisturbed dm-ing " the devastation of tliis, but of whose existence no indication remains."* It remains to indicate the faint traces of relationsliip which the Kerguelcn Island vegetation presents with those of a few other spots of land in a lower latitude, and that might be supposed to share some of its peculiarities. Of these the nearest are Amsterdam and St. Paul's Islands, the names of which are often transposed in our best maps (even in the Admiralty Sou.th Polar Chart of 1839). They lie about 800 * These ideas, suggesting the hypothesis that the existing distribution of plants is dependent on foi-mer geographical relations of land and sea, suggested themselves to me during my visit to Kerguelen Island in 1840. The first attempt to apply similar views in extenso to the conditions of a botanically well-kuown country was in the late Professor Edward Forbes' paper " on the distribution of endemic plants, more espe- " cially those of the British Islands, considered with regard to geological changes." " Brit. Assoc. Rejiorts '• for 1843." It had, however, been previously enunciated by Lyell, who thus accounted for the identity of the Sicilian animals and plants with those of the surrounding Mediterranean shores. He supposes these to have "migrated fi-om pre-existing lands, just as the plants and animals of the " Phlajgrean fields have colonised Monte Nuovo since that mountain was'thrown up in the 16th century," and further on ho says, "we are brought therefore to admit the curious result, that the flora and fauna of " the Val di Noto, and some other mountain regions of Sicily, are of higher antiquity than the country " itself, having not only flourished before the lands were raised from the deep, but even before they were " deposited beneath the waters." Principles of Geology, Ed. v. iii., p. 444, &c. IG BOTANY 01- KKRGUELEN ISLAND. miles to tlio X.E. of Korguolon Island, in TS'^ E. lonir. ; tho nortlionimost, Amsterdam Island, is nearly on the aStli and St. Paul's (ui tlH> IVMh i.avallel i.l' latitude, so they both are very little south of the latitude of the Caiu- ol' (l.iod lldjie. I have hrou2:ht together, in a pa])or ])ublished in llu^ .lounial of llic Linna-an Societv (vol. xiv. p. 17 f\ all (he littlr ti:a( was then i^iiowii of ihr llora of these islands, wliieh, likr Kcruucli-ii. arc Aolcanic. Their seanty vei^etation is on the Avhole more temperate than antarctie, and approximates to that of S. Afriea in eontaining such genera as Fhylica, Spartina, ^m\ Dauthouia. Their fern flora is very interesting ; one fern only is common to Kergueleu {Loniaria alpina), one {Neplirodium antarcticum) is peculiar, though allied to a ^Mauritian species, and two others {Blcchnum australe and Asplenium fiircafi(i)i) are natives of the Cape and other countries ; hut what is most singular is, that neither \\\o Poliipod'non vulgarc wov Aspidlum mohrioides have been found in either island, though the fornier is common to the Cape, ]Marion Island, and Kcr- guelen's Land, and the latter to the two first of these localities. Tristan d'Acimlia, in 12" W. long, and 37° S. lat., and the adjacent islets called Nightingale and Inaccessibh', all nearly in the latitude of Amsterdam Island and the Cape of Good Hope, are the only other islaiids whose vegetation demands a passing notice bere.* Their flora is essentially Fuegian, with an admixture of Capo genera, but with none of those cbaracteristics of Kerguelen Island. Of Cajjc types, it contains a JPelargonium and an abundance of both the Phi/Uca and Spartina of Amsterdam Island, together Avith species of Oxalis and Uydrocotyle. The Fuegian and Falkland Island plants of Tristan d'Acunba and its islets, which have not liitberto been found in the islands south, and east of them, are however more numerous than are the Cape genera even, and include Cardamine hirsiita, Nertera depressa, Empe- ti'iim nigrum, var. rubrum, Lagenophora Commersoniana, and Apinm ausircde ; and it contains besides the strictly American genus Chevreidia. Two laud bii-ds, both peculiar, are common in the Tristan group, and they possess a water hen, wbich has a representative in Africa and S. America. I am not aware whether land liirds are found in Amsterdam Island ; if so, they may help to account for the wonderful fact of the Tristan d'Acunba Fhylica and Spartina being found also in it, though separated by 3,000 miles of ocean. In conclusion, I have to state that no trace of the mountain flora of S. Africa has been found in any of the southern groups of islands. * For the Jatest account of this group sec Mc^cley in Journ. Liiiii. .Soc. XIV., 377. FLOWERING PLANTS.— DR. HOOKER. 17 Enumeration op the Plants hitherto collected in Kerguelen Island by THE " Antarctic," " Challenger," and " British Transit of Venus " Expeditions. I. — Floioering Plants, Ferns, Lycopodiaceae, and Churacecp. By J. D. Hooker, P.R.S. 1. Ranunculus crassipes, liook.f. Fi. Antarci. 221, t. 81. Cliristmas Harbour, Observatory and Swain's Bay,Boyal Sonud (a form witli petioles 5-7 iucbes long). I have nothing to add to what I have said of this species in the Antarctic Elora, beyond that I can hardly doubt its being a derivative form of the Euegian R. hiter- iiatiis, Sm., with wliicli it agrees in habit and its thick- walled beaked carpels, biit differs chiefly in its robustness and simple leaves. B. hiternatm has been found by Moseley in Marion Island, where it presents every character of the American plant. 2. Ranunculus trullifolius, Hook.f. Fl. Antarci. 220, t. 82 A. In streamlets and lakes, Boyal Sound, Swain's Bay, Betsy Cove; Moseley, Eaton, Kidder. (Fuegia and the Ealklands). Glaberrimus, caulibus prostratis radicantibus. Folia longe crasse petiolata, obovato- oblonga trulliformia v. fere orbicularia, apice obtuse 3-5-dentata v. lobata, carno- sula, nervis obscuris ; aimculis petiolaribus membranaco-dilatatis. Flares ad nodes solitarii, brcvissime pedicellati. Sepala 3, orbicularia, concava, membranacea. Fe- tala 3, sepalis sequilonga, obovato-oblonga v. spathulata, 3-nervia, nervo medio medium versus fossa nectarifera instructo. Stamina pauca. Carpella numerosa ; matura cuneiformia, compressa, dorso incrassata, stylo gracili subulato. I described this species in the Flora Antarctica from very imperfect specimens gathered by myself in the Falklands in mid-winter, along with the very similar R. hydrophilus, Gaud., and from a careful examination of the remains of the only flower found, which resembled in petals, sepals, and stamens those of its neighbour, I supposed it to be closely allied to it. Good specimens gathered by Cunningham in the Straits of Magalhacns, and by Eaton in Kerguelen, prove that it belongs to another section of the genus, differing from R. hydropJiilns in the itsually trimerous perianth and the long style of the flattened ripe carpels. R. truWfolins is, in fact, referable to St. Hilaire's genus Casalia (now reduced to Ranunculus) , and its nearest ally is B. bonariensis, Poiret (R. Kunthii Trian. and Flanch.), wliich differs by its ovate crenate leaves, long-peduncled flowers, and absence of style in the ripe carpels. R. hydruphilns, again, is probably a form of R. adscendens, St. Ilil. {R. hinnUis, C IS BOTANY oi' KKRGUELEN ISLAND. Collio, in llook. Bot. Boccli. A'dv. p. I, i. ii.\ whk-h lins similar ininuto sul)glol)OSC ripe t-arpols Avitliout :i stylo. 2i. »ion(Vi(/ioii,V\uUi^\n of Chili, and J?, hcmignoslns Stcutl. of I'itu, .uc probably forms of iu tniUiJoVnts, Avhicli, as our ligurc shows, is a very variable plant in foliage and structure. The Jiaiiiiiicitlus, sp. 3, not in flower, of Kidder (Bull. U. S. Nat. Mus. 3, 21), of which Gray says it can hardly be a form of trullifuliKs, no doubt is this, il", as I apprehend, the term caudate as applied to the leaves is a misprint for cordale. Plate I., Pigs. 1-5. — Plants in diifercnt states; of natural size ; G, 7, reduced leaves and stipules ; 8, sepal ; 9, petal ; 10 and 11, stamen ; 12, immature, and 13, mature carpels : — all eularged. 3. Ranunculus Moseleyi, iTooA-./. ; pusillus, glabcrrimns, acaulis, foliis radicalibus, pctiolo in laniinam obovatam v. oblongam integerrimam dilatato, lloribus solitariis peduneulatis minutis 3-1-nieris, petalis lineari-obovatis obtusis eglandulosis, staminibus 1-7, carpcdlis 10-12 maturis oblique subglobosis in stylum brevem gracilem abrupte attenuatis. — Ranuncidus au nov. sp. ; Oliver, in Journ. Linn. Soc. XIY., 3S9. In the lake at Christmas Harbour, Moseley. A very diminutive species, resembling in size and babit B. Umoselloides, Muell, of Australia, but differing in the carpels, &c. In the latter respect it more nearly approaches B. crassipes, from which it differs in all other respects. Its allies are, 110 doubt, to be found amongst the S. American water-loving species. Plate IL, Fig. 1—1 and 2, plants of natural size ; 3, leaf ; 4, flower ; 5, sepal ; G, petal ; 7, stamen ; 8, immature ; and 9, mature carpel : — all enlarged. 4. Pringlea antiscorbutica, Br. MSS. ; Fl. Antarct. 238, t. 90, 9] ; Kidder in Bull. U. S. Nat. Mus., JVo. 321 ; Oliver in Journ. Linn. Soc. XIV., 389 ; Dyer in Broc. Linn. Soc. 1871, xxxiv. ; Mook.f. 1. c. Thi'oughout the island. — (Marion, Crozets, and Heard Islands). Sepula lineari-oblonga, obtusa, membranacea, pilosa. Bctala 0 in exemplaribus perpliu'imis a nobis scrutatis, in jiaucis 1-4, unguiculata, apice roseo-tincta, inconspicua, caduca. Stamina G, suboequalia, filamcntis elongatis complanatis, 4 lougiorilms per paria scpalis anticis posticisquc opposita ; anthera; magnic, lincari- oblongo}, virosccntes ; pollen sphcricum. Disci glandulai 0 v. valdc inconspicuie. Ovarium oblongum, lursutum, 2-loculare, carpellis latcralibus ; stylus brevis, validus, glaber, stigmatc capitato obscure 2-lobo dense villoso. In the Proceedings of the Linnaean Society 1874, p. xxxiv, I have indicated the evidence of Bringlea being a wind-fertilized member of a natural order most or all the species of which are insect-fertilized. These indications are the usual absence of petals and disk-glands, the exserted anthers and long-tufted papillse of the stigma, to wliich is to be added the absence of winged insects in Kerguelen Island. In reference to the last statement, it is a curious fact that wingless flies abound in the FLOWERING PLANTS.— DR. HOOKER. 19 island, and on this vory plant. INIoseley, Journ. Liun. Soc. xv., 51, in his notes on Kcrg'uolon botany, mentions an apterous ily as bii^ as a blow- fly, nestlin^? at the base of the leaves of Frhiglea and laying its eggs in the fluid which is caught there ; every cabbage yielding ten or a dozen specimens. He adds that he did not observe whether it climbs to the inflorescence in sunny weather. Mr. A. W. Bennett, Proc. Linn. Soc. 1871, xxxix., has described the pollen oi Prbiglea as diflFcring from that of nearly all otiier Crucifcrs in being much smaller and perfectly spherical, instead of ellipsoid with three furrows. This he considers to be a striking confirmation of my suggestion that the plant is wind-fertilized, and which is further confirmed by the total absence of haifs on the style. Moseley fovmd one plant with 28 flower-stalks, three of the one season growth, the others appearing to belong to eight preceding seasons. It is a remarkable fact that all attempts to grow this plant in England, Scotland, and Ireland have failed ; the young plants, after attaining a height of a few inches and a good crown of leaves, have invariably succimibed to the combined eirccts of summer's heat, and the attacks of the common parasite fungus, Ci/stopus candichis, which infests the Cctpsella Bursa-ixistoris. Some few, out of many hundreds, sown at different seasons and under very varied conditions, sm'vived one winter, but perished in the following summer. Plate II., Pig. 3. — 1, 2, 3, apetalous flowers ; d, monopetalous, and 5, tripetal- ous flowers ; 6, petal ; 7, ovary ; 8, the same laid open ; 9, ovule : — all enlarged. 5. Colobanthus kerguelensis, Sook.f. Fl. Anfarcf. 249, t. 92. Christmas Ilarbour, Swain's Bay, &c. (Heard Island, Ilosele//.) (Stellar ia media L.) Introduced by sealers. (Cerastmm triviale. Link.) Introduced by sealers. 6. Lyallia kerguelensis, nook. f. Fl. Anlaret. 518, t. 122; Kidder in Bull. U.S. Xal. Mils., No. 3., 22. Oliver in Journ. Linn. Soc. XIV. 390. Dyer in Broc. .Linn. Soc. 1874, xxxiv. Clu'istmas Harboiu- and Boyal Soimd. The flowers have been described from Kidder's specimens by Asa Gray, and from Moseley's by Oliver and Dyer, the descriptions agreeing well. The stamens, n liich appear to be almost constantly tlu-ee and hypogynous, are stated by Oliver to be variable in position. Kidder retains it in Fortulacea', but Bentham and I had long previously placed it in CaryophjllecB in the Genera Plautarum and next to Py- cnophyllmn, a position which the discovery of the flowers confirms. It has many of the characters of Colohanthus, especially the androccium. Plate II., Fig. 2. — 1, plant, of natural size ; 2, leaves; 3, flower and bract; 4, flower laid open ; 5, stamen : — all enlarged. c 2 20 BOTANY or KERGUELEN ISLAND. 7. Montia fontaua, L. Comiuiiu ill wcl placos. (Marion Islaud, Mosclcy, ami w idi'Iy dislribiilril in liie N. and S. (oniperato iv2:ious). 8. Acsena affillis, Uook.f. Fl. An la re/ . 2GS, i. 'JC. 15. Common (lu-ouu,hnut tho island. (Marion and tlio Crozct Islands). Called Korp:nolimpera(e reii'ions). From a drawing of the ripe fruit whieli I maih; Avhcn in Kcrg-uelen in IS lO, T have no hesitation in referring tliis to Subspecies obtusangula, as Hegelmeycr did from his examination of my dried specimens. The IVuit lobes are nearly semi-cireul.ir, and each pair is united by about two thirds of their faces. The free portions are obtusely triefonous at the back. Two forms arc common in Kerguelen, as elsewhere in the south temperate zone, one aquatic with long stem and proportionally large spathulate leaves, the other smaller, terrestrial, subcrect, with obovatc or oblong leaves ; this flowers the most abundantly. 10. Tillaea moschata, I>. C. Bulliarda moschata, i)' Urv. Abundant in inoist places ncai- the sea. (Marion Island, Iloseleij, Crozcts, Kidder; Avidcly spread in high southern latitudes). 11. Azoreila SelagO, nool-.f. Fl. Antarct. 281., t. 99. Very abundant throughout the island. (Marion and Heard Islands, Moseley ; Crozets, Kidder ; Eucgia ; Mac Quarrie Island.) Kidder remarks that tlic flowers are greenish yellow, not pale pink as I foiuid them to l)c in winter. Also, that the leaves have not the bristles on the faces of the lobes as figm'cd in the Flora Antarctica, I find them on specimens from all localities. ]\[os(dey observes, in reference to this plant at Marion Island, that the mounds ii, forms evidently retain and store up a considerable amount of sun's heat, and that this fact probably explains its peculiar mode and form of growth, and that of many otherwise widely different Antarctic plants. He found that a thermometer plunged into the heart of a hummock rose to 50", when the temperature of the air was 45°. 12. Galmm antarcticum, Soohf. Fl. Antarct. 303 bis. Common, but not found at Christmas Harbour. (Crozets, Kidder ; Fuegia and Falkland Islands.) Kidder remarks that the flowers are distinctly pedicelled, and as often 4- as 3- merous, and even 5-merous ones occur. Eaton's specimens confirm this. FLOWEPaNG PLANTS,— DK. HOOKER, 21 13. Cotula (Lkptixella) plumosa, nook.f. Fl. Anlarcl. 26 and 308, t. 20. On cliirs, especially near the sea, often forming immense luxuriant blue-green patches where the soil is enriched by the dung of birds and seals. (Crozets, Kidder ; Lord Auckland, Campbell's, and Mao Quamc Islands.) Reputed by the whalers to be a prompt and effectual emetic. Through a typo- graphical omission of the word not at p. 308 of the Antarctic Flora, this plant is stated to be found on the continent of America. The genus Leptinclla is reduced to a Cotida in the Genera Plantarum. This plant, like the Friiiglca, proved so impatient of heat in this country, that of innumerable seedlings raised at Kcw to several inches high aU perished. 14. Limosella aquatica, L. Common in the freshwater lagoon at Christmas Harbom-, (Fuegia and all temperate regions.) A very small form, with the leaf -blade hardly broader than the petiole. Stamens included. Orary globose ; style rather long. 15. Juncus SCheuzerioides, Gaud.; Sook.f. Flor. Antarct., 79, 358. Common in spongy places. (Euegia, the Falkland, Lord Auckland, and Camp- bell's Islands.) 16. Uncinia COmpacta, JBr. ; Boott in Sook. f. Fl. Tasman, ii. 103, t. 153 B. Royal Sound and Observatory Bay, Moseley, Eaton. (Moujitains of Tasmania and New Zealand.) 17. Deschampsia antarctica, Eook. ic. Fl. 1. 150 (Aii-a) ; Hook. f. FL Antarct. 377, /. 133. Common and ascending to considerable altitudes. (Fuegia, Falkland Islands, South Shetlands.) A true Deschampsia, as that genus is now defined, by its 4-toothcd flowering glume and free caryopsis, 3Iunro. 18. Agrostis magellanica, Lamk.; Eook.f. Fl. Antarct. 373. A. antarc- tica, ihid. 373, t. 132. A multicaulis, ihid. 95. Common throughout the island. (Marion and Ileard Islands, Moseley ; Chili, Fuegia, Falkland, and Campbell's Islands.) Since the publication of this plant as A. antarctica, I have examined a specimen of Lamarck's A. magellanica named by Nees in Arnott's Herbarium, and find it to be identical. Further, JNIunro informs me that it is fairly described by Trinius in his " Agrostideae," and by Kunth in his supplemental volume (p. 175) from a Lamarckian specimen ; he adds that the Kerguelen specimens agree with these descriptions, except in the flowering glume being larger and much longer than the ovary. This glume is sometimes obtuse or rounded, at others deeply divided. The beard on the callus, which is very indistinct on the Kerguelen's plant, is conspicuous on some Fucian ones. 22 BOTANY OF KKRGUELEN ISLAND. ]'.). Poa Cookii, Hook./.; Fl. Aiiho'cl. ;5S2, 1. i;]<) (Fostuca). Forma 1. ; t'uliis i-ulnunu supm-antibiis, inuucula eloiip;a(a inioiTupta. Forma 2.; foliis culmiim superantibus v. cequaniibiis afiuniiKitis ])iini;cntibus, panifula dcnsa sub-cylindracca. Forma 8. ; foliis culmum luquantibus subacutis v. obtusis, panicula miuorc laxiorc, spiculis paiu-ifloris coloratis. Abundant and ascondinc: to a considoral)!*' lK'ii;lil :— l-'orma 1. C'liristnias Ilai'- bour ; Forma 3. Eoyal Sound, ou a biyli liill, Enloii. (Clarion and Uoard islands, Moseley). This fine grass should, ini([Ucstionably, be referred to Po« (as now defined by tlie compressed flowering ghuue, &c.), along with its n(>ar congener Daciylis cccsintosa * of Fucs::ia and the Falklands, from which it diiTers, amongst other characters, in never formino; tussocks. It is scarcely specifically distinct from T'.foHosa, Hook. f. Hand- book of N. Z. Flora 338 (Festuca foliosa, Fl. Antarct. i. 99, t. 55; Fl. Nov. ZeuUl. i. 308) ; and this, again, from the Fuegian Poa lanigero, Nees (Festuca fuegiana, Fl. Antarct. 380). The flowering glimics are often ohscvu'cly, or not at all toothed. The spikelets arc 8-5-flowered and \-l in. long (not eight lines as misprinted for three lines in the Antarctic Flora). A. Gray remarks of Kidder's specimens that they seem to be male only. Poa jn'afensis, L. Introduced by sealers. Poa anmta, L. Introduced by sealers. 20. Festuca erecta, D' TJrv. Common and ascending to a considerable elevation. (Fuegia and the Falkland Islands.) Often forming tussocks ; panicles green or purplish, 21. Festuca kerguelensis, nook. f. Triodia kerguelensis, Fl. Antarct. 370, /. 138 {Poa). Common and ascending to 2,000 feet. Spikelets sometimes 1-flowered. A very variable grass in stature, evidently allied to F. erecta, and more nearly still to F. scojyaria (Fl. Antarct. 98 ; Fl. Nov. Zeald. i. 308), of which possibly it is a dwarf form, as suggested in the Handbook of the New Zealand Flora, p. 341. The naked base of the flowering glume, however, will always distinguish all the specimens I have examined. Filices. 1. Cystopteris fragilis, Pern//. Crevices of rocks near the hill-tops. Royal Sound, Kidder, Eaton. (Fuegia, Falklands, and N. and S. temperate regions generally.) * The name Poa ceespitosa being occupied by Forster, tlioiigb it is Joubtful to what .sjjccies it applies, I propose that oijlabellata for the Tussock grass, which is the Festuca flal/dlata, Lanik. FILICES, LYCOPODIACEiE, CIIARACEiE,— DR. HOOKER. 23 2. Lomaria alpina, Sproiy. Common, often forminjj large beds, but not I'oimd at Christmas Harbour. (Marion Island, Moseleij ; Crozets, Kidder ; all the colder S. temperate regions.) 3. Polypodium (Grammitis) australe, Mctt. Crevices oi' rocks, Observatory Eay, Kidder, Eaton. (Marion Island, Moseley ; Fuegia, and all tli(> colder S. temperate regions.) 1'. Polypodium VUlgare, L. i-ar. Eatoni, Baker, \Gm^ pellueidis. Crevices of rocks by running streams. Observatory Bay, Kidder, Eaton. (Clarion Island, Moseley ; S. Africa ; Sandwich Islands, and N. temperate liemisphere.) This pellucid-iierved variety only occurs elsewhere in the Sandwich Islands. Lycopodiaccce. 5. Lycopodum clavatum, L., tar. magcllanicum ; Hook. f. Fl. Antarct., 113. L. magcllanicum, Sicart:. Not uncommon throughout the island, but not met with at Clu'istmas Harbour. (Yar. magellaniciim, Marion Island, Moseley ; Pucgia, and all the colder S. temperate regions. The typical L. clavatum inhabits aU northern cold damp climates. 6. Lycopodium SelagO, L. var. Saururus, Rook. f. Fl. Antarct. 394. L. Saururus, Lamk. Xot uncommon throughout the island. (Var. Saururus, Marion Island, Moseley ; Tristan d'Acunha, St. Helena, Bom-bon, Peru. The typical form inhabits all damp cold climates.) Characcce. 7. Nitella antarctica, Braua. N. Hookeri, Beinsch in Journ. Linn. Soc. XV. 219. Chara flexilis, Linn ; Fl. Antarct. 395. In the Lake at Christmas Harbour ; and in that next but one to the Observatory, in Observatory Bay, Eaton. 21 BOTAMY OF KEKGUELEN ISLAMD. • II. — Mitsci. By William Mitten, A.L.S. The first investigation of the mosses of Kcrguelcn was madi' hy Dr. J. D. Hooker clui-ing the voyage ol' the " Erebus " and " Terror " in tlio winter of 1810. From the eolleetions made by him there were described J5I s])ecies and varieties, wkieh were arranged as 25 sjiecies in 11 genera. Of the Avliole number six species were considered to be new and undescribed, and tlie remainder to have been found in other regions. The most remarkable species contained in this collection are the Scfiisddiiim marginatum, Weissia stricta, and. W. torl'ifolia. Dm-iug the visit of the Challenger, there were collected by Mr. Moscley, in the summer of 1874, 28 s]ieeies, of which number 20 were additional to those discovered by Dr. Hooker. Snflicient materials were obtained to establish the presence of eight more genera, all previously knoAvn to occur in austral lands, four of the species appearing to be new. Twenty-eight species Avere obtained by Dr. Kidder of the American Ti-ansit Expedition, of whicb number 12 were additions to the Flora, two being described as new. Following the above come the collections made by the Rev. A. E. Eaton, pending the observations of the transit of Venus, w^hich include 38 species, of which 17 were additional to the Flora of Kerguelcn Island, three being undescribed, and by this collection three genera were also added ; thus raising the whole number of the species of mosses inhabiting the Island to 74. This, con- sidering how much has been added by each collector to those which were previously known, is probably a low estimate of the entire moss flora. No genera peculiar to Kerguelcn arc observable in the collections, unless a species here referred to Blindia and the ScMstkUum marginatum (here placed in Streptopogon) should be so considered. The remaining genera are universal in boreal as well as austral regions, with the exception of the three species of Dicranum, all which belong to extra-European sections of that genus. Twenty-three of the Kerguelcn mosses are considered identical with species found in the north of Europe and America, of these Bryum aljnnum and Brachythecium salebrosum had not before been identified in the southern hemisphere. A few distinct and well-marked species have been gathered in Kerguelcn Island ■which are also found at great elevations on the Andes of Quito and of New Grenada. Of these Mielechhoferia campylocarpa and Psilopilum trichodon are conspicuous instances; they probably inhabit the whole Andine chain. Bartramia appressa, Brachythecium paradoxum, and Torlula Bri^icei^s are found also in New Zealand and Tasmania ; but with the exception of Bicrannm, IcerguelenHC there is no species which points to any connexion with the mosses of South Africa. MUSCT.— AV. MITTKN. 25 1. Ditrichium australe, ^mi. 1. c, (Cynontodium). Lophiodon strictus, Hook.f. et JFils. Fl. Antarcl. 130, t. LIX., Pig. 1. In dense fulvous tufts, with old capsules, Moseley. (Lord Auckland's and Campbell's Islands.) In all the specimens referred to this species the dry young foliage is fulvous, the older broTvn or black ; the terminal leaves are frequently longitudinally twisted, otherwise their direction is tlie same as when wet ; the lower portion of the leaf is in outline of an elliptic oblong figure, from which the nerve is continued in a straight line, and is rather suddenly carried out so as to appear without a margin of leaf; a transverse section shows it to be concave above and convex beneath ; the apex is abrupt, rounded, and nearly flat, so as to appear as if dilated, and, as stated in the Flora Antarctica, the species is distinguished from most of its allies by this par- ticular. The substance of the base of the leaf is composed of elongated cells which, although shorter towards the top of the dilated portion, are not dense, so that the entire expansion is of a pellucid fulvous colovu-, the nerve being everywhere smooth, with a few small teeth at its apex.* 2. Ditrichium Hookeri, C. Muller Syn. I., 450 (Leptotrichum) . Royal Sound, with old capsules and young setae, Eaton. 3. Ditrichium COnicum, Mont, in Ann. Sc. Nat. Ser. 3, iv. 100. (Aschi- stodon.) Kear Vulcan Cave, barren, Eaton. The imbrication of the leaves at the apices of the stems, when dry, so as to form an erect or curved point, renders this species not difficult to recognise in a barren state. 1. Asiothecium vaginatum, Book. Muse. Exot. t. 141 (Dicranum). * In the Journal of the Linnean Society, Sept. 1859, there was confused with the Leptotrichum australe therein mentioned, the following ajjparently distinct species, — D. punctulatum, Mitt. ; dioicum ? dense csespi- tosum, dichotome raniosum, foHa inter se remotiuscula a hasi erecta amplcxante oblonga eellulis inferioribus elongatis superioribus abbreviatis rotundatis obscuriusculis veluti punctatis, subito in subulam patcntem inferne canaliculatam apice angustam planiusculani denticulatam miuutissime scabridam subliEvem eellulis punctulatis areolatam prodiicta, perichoetii aha basi latiora et longiora parte subulato pateutiora, theea in pedunculo breviusculo rubro parva ovali-cylindracea erecta leptoderma fulvo-fusca. Flos maseulus in rumi.s termiiiahs, ovatus, e basibus foliorum dilatatis apice retusis vaginantibus involucratus. Distiehiuin capilla- ceuui, Fl. N. Zealand, II., 73. Hab. — New Zealand, Dr. Lyall. Great Barrier Island, Hutton and Kirk. Fagiis Forests, Ifopkins, Dr. Haast. In size colour and general appearance very similar to D. australe, having also the same, but narrower flattened apices to its leaves ; in the recurvation of the subulate portion from the top of the erect base it resembles D. capillaceum, and for this species Dr. Lyall's barren specimens were mistaken, nlthoun-h the leaves are not distichous, but so disposed that each fiftli leaf occupies tiio same vertical position on tlie stem as the first counted from ; the outline of the dilated base is not oval-elliptic as in D. australe, but oblonc obtuse. The fruit in an old state is present on Dr. Haast's specimens ; accompanying these fertile stems were many conspicuous male flowers, which do not appear to arise from the lower parts of fertile stems, but seem to be really distinct male plants. D 26 HOl'ANY OF KERGUELEN ISLAND. Swain's Bay, Eaton. Small l)arroii stems, but not cliircrcut from spccimcus from the Bogotian autl Quitcuiau -V'ndos. 1. Blindia gracillima, Mitt. Dioica, laxc cacspitosa. Caulis elongatus, gracillimus, iulernc uiulus, supcrno foliis rcmotiusculis laxo obtcctus. Folia augiistc lauceoloto-subulata, pagiua I'olii c ccllulis angustis clongatis parietibus pollu- cidis iis(iu(.' ail f iiorvi apico vix dcnticulati loiigitudinis anguste continual a; ccllulis alaribus in auriculam parvam dispositis rubris ; Iblia pcricluctialia (U'ccta, basi obovata, convoluta, sensim subulato-attonuata, ncrvo longiiis cxcurrcutc. Tbeca in seta brevi flcxuosa arcuata pcudula, subrcsupiuata, globosa; operculo oblique rostrato ; pcristomii dcntibus rubris latis tcncris intcgris vcl rarius pcrtusis intus lajvibus extus pai'ietibus transversalibus proniincntibus appcndiculatis ; annulo nullo ; calyptra parva, viridis, nigrcsecns. B. curviscta, Mitt, hi Linn. Soc. Joiirn. XV., 193. lloyal Souuil, in lakes, Avitli young and nearly ripe fruit, Eaton. Stems 2-1 inches long, formmg loose tufts, the upper portions red, the lower black, denuded of leaves, and forming a loose entangled mass. Leaves at the apices of the stems fulvous and sliining, the lower all blackened ; in then* direction the ujiper leaves are but little changed when wet or dry ; they arc 1^—2 lines long ; the areolation consists of elongate cells separated by pellucid walls ; at the angles of the base of the leaf the alary cells are distinct and red. The nerve becomes indistin- guishable at four-fifths of the whole length of the leaf, and is thence continued, and ends without forming a pungent point ; leaves of the perichaitium longer, and their dilated bases about twice the width of the cauline leaves. Seta 2-24- lines, straight in its lower half, thence to the capsule twisted and variously curved. Capsule erect when di'y, Avhcn wet with a swan's-neek-likc curve, and so bent as to become hori- zontal ; when matiu'c spherical without any neck where it is affixed to the seta ; colour reddish brown ; substance thin but firm. Operculum always obliquely beaked, at length of the same coloiu' as the capsule. Peristome perfectly formed ; teeth red, broad at the base, thence with an even outline narrowed to their points, with the exception of a rare perforation there is no trace of their being composed of a double row of cells ; at the base of the teeth the transverse divisions are close together, but above this they arc much wider, and on turning the tooth on edge it is seen that each dissepiment of the articulations is prominent on the outer side, but not on the inner. Spores small, round. Calyptra coriaceous, brownish-green, deeply cleft, with a spreading base. Tab. III., Fig. 1, plant of natural size; 2, cauline leaf; 3, perichGetium with capside ; L, portion of peristome ; all magnified. 2. Blindia microcarpa, Mitt, in Joum. Linn. Soc, XV., 65. Monoica, pulvinatim ctespitosa. Caulis dichotomus, fastigiatim ramosus. Folia patentia, stricta, plus minus falcata curvatave, dimidio inferiore lanceolate supcriore carinato MUSCI.-— W. mXTEN. 27 angustc atteni;ato, integcrrima, ncrro angusto pcrcursa, colliilis elongatis alaribus in am-iculam parvam fuscam dispositis aroolata ; pcviclisetialia brevia, pavva, ovata, convoluta, in acumen sulnilatuni proclucta. Tlicca in pcdunculo gracili foliis caulinis dimidio breviore crecta, parva, ovalis ; opercnlo suljulato obliquo dcmum ore dUatato cyatbiformi fusca ; peristomii dentibus tencris ; calyptra parva, dimidiata, Flos masciilus foliis jiropriis pericbtctialibus similibus inelusus. Kcrguclcn Island, Moscleij. Tliis is tlic species mentioned in tbe Flora Antarctica, p. 128, nndcr TFeissia contecta, as being present in the Hookerian Herbarium, its balntat unknown. In compact, but not coberent tufts. Stems fastigiately branched, about an inch high. Foliage shining, but little altered in dii'cction wet or dry. The minute capsule is scarcely conspicuous amongst the leaves. Leaves at the tops of the stems yellowish green, below brown, erect or slightly falcate, about 2i lines long, com- posed of elongate cells with pellucid walls ; ncrvQ pale broAvu and with the pagina gradually attenuated into a very narrow flat entire point ; alary cells at the angles of the base distinct, brown, forming sub-quadrate masses. Leaves of the pericliEe- tium ^4 as long as those of the stem, and quite concealed amongst them. Seta about 1 line long, straight, pale brown. Capsule as it reaches maturity appearing to pass from oval to nearly globular; after the fall of the operculum by the dilatation of its mouth it becomes cyathiform, with no distinct neck. Operculum with a very oblique sul)ulate beak which is longer than the capsule. Peristome- teeth very thin, broad at base, narrowed upwards into entire points; transverse articulations remote. Calyptra small, coriaceous, brownish, scarcely reaching the base of the operculum. Male inflorescence in a small l)ud below the base of the l^erichaetium. Tab. III., Fig. ii. : 1, plant of natiu-al size ; 2, cauline leaf ; 3, perichsetium with comal leaf, capsule, and male flower ; 4, old capsule ; 5, portion of peristome ; all magnified. 3. Blindia contecta, Eook. f. Sf Wlls. Flor. Antarct. 404 t. 58, f. 3. (Weissia). Christmas Harbour, on rocks, barren, Hooker. In this species the pericliJBtium is composed of enlarged leaves as in Stylostegium, Schimp., but the capsule which is immersed has a peristome.* * These three species aftbrd some considerations respecting their mode of ft-uctification. The genus Blindia, Bruch et Schimp., at tirst included only the European B. acuta, witli the " pcriehatium vagiuans distinctum," the perichaetial leaves being as large as the cauline and dilated below. To this was added by C. Miiller (in the Synopsis) B. crrspiiicia, wliich liad been made into the genus Stylostcf/iiim in the Bryologia Europca, ditiiiring from Blindia in its gyniuostomous capsule immersed in enlarged but not vaginaut perichaetial leaves, in these pai'ticulars analogous to some species of Grimmia of the section Schis- tidium, in which B. Cfrspificia had itself sometimes been placed. The distinction between Blindia .and Stylostegium is reduced by the presence of a peristome in B. contecta (which may bo said to be a Sti/loste- fjium with a peristome) by the immersed capsule in Stylostegium, and the exserted one in Blindia. In B. D 2 28 150TANY ('V KKRGUELEN ISLAND. 1. Dicranum (Isocakvus, Ji?//.) tortifolium, Hook. f. cl Wils. Fl. Ant- arct. 101, /. l.")!*,./: T) (Woissia). Hal)., Christmas Harbour, on gravelly banks, Iloo/.rr. Under the shoot, of a walcrrall noar Vulcan Cove, with old eapsules and young setiv, Ualo/i. In eonii)act tufts 1-li inch high. Foliage very green above, below becoming brown. Old capsules black and shining ; young calyptras orange brown. 2. Dicranum (Tsocakits) strictum, llouh-. f. ci Wils. Fl. Auhivcl., tot, /. ir.L',./: 1 (Wcissia). Christmas Harbour, ou rocks' near the sea, Tlookcr. This has been described as dioicous, hut I In- male flower is terminal on a branch arising some distance below tlu> perichtctiuni. The peristome has rather broad tliii\ teeth ; in the solitary example which could be; examined, the teeth ap- peared to be partly adherent in pairs, the median line is obsolete. This species is closely related to 7). for/ifoliio)/. 3. Dicranum (Ukmicampvlus, Hi//.) robustum, Jfook. f. e/ iFl/s. Fi. Antorcf. '100, /. 152,/. 8, var. lucidum; I), pungcns, var. lucidum, Hook. f. ci Wils. I. c. Hah. Christmas Harbour, Hooker, Moseleij. Known only in a barren state. 3. Dicranum (IIkmicampylus) kerguelense, C. Mailer, Syn. i. 370. D. Boryanum, Schimegr ; Hook.f. et Wils. Fl. Anlarcl. 106. D. dichotomum, ISeauv. Frodr. 51 (Cecalyphum) . Christmas Harbour, Hooker. On an elevated moor, Eoyal Sound, Falon. microcarpa the pericha;tium is formed of leaves reduced in size like those which usually include the anthe- ridia, and thus another modification of the porichjetium is produced, all other particulars being as in Blindia proper. Thus, by the diiference in tiic leaves of the pericluetium, the species are separable into several groups : — Stvlostegiuni, B. & S. ; — theca in perichajtio e foliis caulinis ampliatis immersa. Blindia, B. & S. ; — theca e perichsetio c foliis basi vaginantibus caulinorum magnitudinis exserta. Ilomogastrium ; — theca e perichsetio microphyllo exserta. The difl'erences in the leaves of the perichaHium between Stylostegium and Blindia are analogous to those which exist between the GrimmicB of the sections Schistidium and Grimmia ; between Hedwigidium 8nd Brannin; lietween some Bnrtrumi Rhacomitrium crispulum). 30 BOTANY OF KERGUELEN ISLAND. Kcrguolon Island, Jlosc/ei/, Hill N.W. of jVlount C'rozior, bavron, Ealou. Vory closely rolated to G. Si/mp/ii/odoii and G. cnicrsa, C. ]\lullov, and also to D. cri^piiht^, ITook. f. et AVils. ; all arc possibly forms of one species. 0. Griminia (DiurTODON) crispulus, Ilook.f. et JfHs. Flor. Anlarci. 124', ct 102, /. 57, /. n. Chi'istniMs TTavbouv, in gravelly beds of vivub^ts, Hooker. 7. Griinmia (ruiAcoMrriuuM) lanuginosa, Jyill. ; Brid. i. 215. llab. — Korgueleu Island, Moseley ; Ivoyal Sound and ]\ear Vulcan Coa'c, barren, Eatoi. AM tlie specimens arc less robust tban tliose collected by Dr. Hooker in Ilermito Island ; from tbe Avbitening of the tips of the leaves they are very hoary. Many of the specimens brought from southern regions which appear to dilTcr in only slight particulars from northern states have been described as distinct; of these, lihacoiHiti'ium Jirmnni De Notaris, from Chili, is a fulvous brown moss, a. Geronticnm, C. MiiUcr (Hedwig'ia, 1870), is possibly the same. B. senile, Schimp. (Lecliler, 1089, from Magellan), Avith leaf points crisped and hoary. It. incanum, C. MiiUcr (Hedwigia, 1870), from Cape of Good Hope, is, if specimens from the top of Table Mountain belong to it, scarcely in any particular different from Arctic examples. 8. Grimmia (Rhacomitrium) 'pYOtensum, A. Brmm ; Hook. f. el IFUs. Flor. Anfarct. 402. Clu'istmas Island, barren, Hooker. 9. [G. FRONDOSA, James in Bull. U. S. Nat. Mus. 3, 25, is another Kerguelen Island species, found by Kiddev.] 1. Orthotrichnm crassifolium. Hook. f. el Wils. Fl. Antarct., p. 125, lab. Ivii.y. S. Clu-istmas Harbour, common. Hooker, Moseley ; Hoyal Sound, Eaton. The specimens from Kerguelen have the points of the perichastial leaves reach- ing to three-fom-ths of the length of the capsule, which is thus only emergent, and in this respect they agree with some of the specimens gathered in Hermite Island by Dr. Hooker. No importance can be attached to this particular character, as in Dr. Hooker's specimens fi-om Lord Auckland's Islands, emergent and cxserted capsules may be seen on the same stems. The capsules are either smooth or with a few folds regularly placed on one side, the remainder being smooth, and are more m-colate than any of the specimens collected by Dr. Hooker. The iitflorescence consists, as usual in the genus, of a male flower near the base of the pcrichoetiuni in all the specimens. 2. Orthotrichum atratum, Mitt, in Linn. Soc. Joum., XV., p. 66. Monoi- cum. Caulis humilis, caespitosus. Folia patentia, sicca incm-va, laxe contorta, lan- ceolata, apice lata obtusiuscule acuta, nervo sub summo apice evanescentc, cellulis MUSCI. — W. MITTEN. 31 fere ubique parvis rotuiidatis obscuris ; periclisetialia majora. Thcca in peclunculo longitudinc pcricliictii subsequali ovalis, loovis, sicca infra os contracta, infcme collo crasso; operculo convcxo, rostro angusto; pcristomii dentibus 16, vcl plus minus cohoerentibus 8. Calyptra uigro-fusca, calva, ad medium usque thec» descendens, nitida. Kerguelen Island, Moseley. Stems not more than lialf an inch higli. Leaves a line long ; a few of the youngest greenish, the rest all blaclv, coriaceous. Capsule pale straw-coloured, somewhat fleshy, smooth when deoperculate, very slightly contracted just l^elow the mouth at the base, when chy shortly plicate. In aU its parts larger than O. crassifolinm, with leaves twice as wide, and without the horny appearance ; it is, however, more nearly allied to that species than to any other, and approaches in some respects the O. anomalum, HedAV., which ascends far towai'ds the Polar regions. 3. Orthotrichum rupestre, Schlelch. ; Brid. i. 279. Royal Sound, with fruit nearly matm-e, Eaton. The specimen is in good state, and appears to agree in all respects with the Em-opean, except that no internal peristome has been found ; it does not correspond so well with either of the very closely allied species, O. Skirmii or O. aqmlatnm, which agree in being destitute of cilia. 1. Zygodon Brownii, Schwaegr. t. 317 h. Kerguelen Island, Moseley. The minute scrap rather establishes the fact that a species of the o-enus in- habits Kerguelen Island than provides materials for identifying with certainty that to which it is here referred. Tortula (Syxtrichia) Princeps, DeNotaHs; Barbula MuUeri, jBr«c/i eif Schiuip., Bryol. Burop. t. 28. T. Euegiana, IHtt., Joiirn. of Linn. Soc, Sept. 1859. IIiisc. Aiistr. Amer. 174. Barbula, S. magellaniea, C. Miiller in Bot. Zeit. 1862 349; B. antarctica, Sampe ; Tortula antarctica, T. cuspidata, ct T. rubella. Kooh.f. et Wils. Fl. Tasmanica, pi. clxxii.,/. 8, 9, 10. Pvoyal Soimd, with abundant mature capsules; Observatory Bay, with older fruit, Baton. The first examination of the Kerguelen specimens yielded no male inflorescence, they were therefore considered to be T. fuegiana, with which in size, colour, and appearance they appeared to be identical, this being supposed to be a dioicous species, as no male flowers were observed in Lechler's Magellan specimens No. 1088, from Cabo Negro. The same specimens were again described by C. Miiller as dioicous, imder the name of Barbula S. magellaniea. In seeking for the male flowers amongst Mr. Eaton's abundant specimens, it was, after the examination of many stems, ascertained that although no antheridia were present in the fertile flowers, a small proportion of the stems had a male flower without archegouiaj cither 32 BOTANY OK KERGUELEN ISLAND. tonuinal on a short branch, or hitoral IViun {\w t;ro\vth ol' iiinovaiions. Finally it was tlisc'ovm-oil that there niii^ht bo present on the same stem, lldwcrs conlaiiiiiin' anthcriilia aecompanicd by others eontainius; archegonia, and nlidxr Imlli llicse another Howcr in uliich l)oth orj;'ans Aveiv intcniiixcd. TIhk, willi s]i(>cinu'ns in small qnantity to examine, the inlloreseeneo miglit In- (l(>scril)e(l as nioiioic-ous dioi- eons or synoieous, as mii;-ht ehanee to happen to tlie iavestij^ator. Tlu> Enropean T. Priuceps was at iirst correctly described l)y Do Notaris as ])oly- gamous in the Bryologia Europea, where it is figured as Barhida trailer}. It is there described as hermaphrodite, with a remark in a subs(>qucii( noie that it occa- sionally produced tlowers containing archegonia only. In Schimper's Synopsis and in the IBryologia Britannica it is simply stated to be synoicons. An examination f)!" De Notaris's original s])ecimen shoAvs synoieous t'cn'tilo tlowers with innovations ol' the stem terminated by llowcrs Avith archegonia alone; in this particular coinciding with British specimens. The distribution of this species appears to be very wide, and it would seem to be the preponderating ii" not the only species of the genus in southern regions. From N.W.America it extends to Mexico, Chili, and the Straits of Magellan ; in AlVica it is found at the Cape of Good Hope, and may be identical with the JJarbiila mollis, Schimp., of the Abyssinian ^lountains ; it occurs in N.W. India ; it inhabits also New Zealand, Tasmania, and Australia, from whence several species have been de- scribed as dioicous, viz., Barbula Latvobeana, C. Miiller (Bot. Zeit. 1861', 358), 7i. Preissiania (ejusd. Synops. I. Q\-2),B. pandHr(rfoUa (ejusd. et Hampe, Linnrca 1S.")3, 493). Xo specimen, however, amongst those sent by Baron F. von Mueller to th(> Kew Herbarium has been examined without finding its inllorescence monoicous or synoieous. There is also Tortula S. pusilla, J. Angstr. from Magellan, described as dioicous? and Barbula Lechlerl, C. MuUer (Bot. Zeit. 1859, 229), as monoicous. All these species or supposed species may be well distinguishable, but if tlie certainty of the condition of their inflorescence is removed from their descriptions, the remainder becomes applicable to T. Pnnceps, in Avhich the outline of the leaves even on the same stems is, as in the European T. ruralis, subject to a great amoimt of variation. 2. Tortula (Barbula) serrulata, Eook. et Grev. in Breiost. Edinb. Journ. i. 291, t. 12. Kerguelen Island ; a few small barren stems with other mosses, Moseley. 3. Tortula (Bakbula) erubescens. Mitt, in Uooh. f. Handbook of Noxo Zeald. Flora, ii. 421 (Didymodon). Kerguelen Island ; a few fragments, Moseley. Very closely related to the T. rtibella so widely distributed in northern regions, differing chiefly in the longer operculum and larger size of the whole plant. 1. StreptopOgOn australis. Mitt, in Linn. Soc. Joum. XV. G6. Humilis. Folia inferiora patentia, spathidato-ligulata, obtusiuscule acuta, nervo in apice de- MUSCI.— W. MITTEN. 33 sincnte, marginc apicem versus denticulata ; superiora duplo latiora, a basi erectiore scnsim recurva, patcutia, apice cum nervo in acumen longitudiue variabilc sensim cducto, marginc supcrnc serrulata. Royal Sound'; a single stem, Eaton. Two small stems amongst other mosses without precise locality, Moseley. The small quantity found of this moss would he insufficient to give any idea of what might be supposed to be the usual appearance of the species were it not evidently a close congener to a veiy ambiguous moss found on thatch in the south of Britain, and which has been known first as a supposed gemmiferous variety of Leplodont'mm jlexifolhnn (Sm.), and since as Didi/modon gemma scens. Mitt. MSS. From this the Kcrguelen species dilfcrs in the form of its lower leaves. In the British moss all the leaves are acuminate and tipped with a globular mass of indi- vidually obovatc green gemm83 of a loose cellular substance, and gemnioe of the same form are present on the points of some of the upper leaves of S. australis. Both species appear to be small, the British one is seldom more than half an inch in height ; the entire plant, excepting a few rootlets, and the rarely present archcgonia, which are red, is of a yellowish green. In the dry state it affords nothing to attract observation, but when wet, every leaf being terminated by its mass of gemmae, it is unlike any other European moss, excepting the more robust Ovthotrichum phyllanthum (Brid.). It comes nearer to some species of Strepto- pogon ; the areolation of the leaves of Calijmperes or of Syrrhopodon are widely different. The genus Streptopogon founded on S. erytlirodontus (Tayl.), with the additional species discovered in the Quitenian Andes by Dr. Sj)ruce, and those from the Bogotian Andes by Lindig and Weir, contains a number of species all seeming to have a tufted Orthotrichoid habit. They differ among themselves considerably, some of the Andean species having the leaf with a callous margin which is wanting in others, and the capsule immersed or shortly exsertcd from perichaitial leaves which are not very different from the cauline. ^S*. mnioides, Schw. t. 310 {Barhula), however, has the perichajtium leaves much elongated, and different from those of the stem, simulating those of Uolomltrium, and on this account should stand apart from the other species, thus — Streptopogox, Wils. Theca in perichtetio e foliis caulinis subsimilibus im- mersa, emergens, vel brevitcr exserta. Calyptra breviter midtitida. Calyptopogon, Mitt. Theca in pericha^tio e foliis clongatis a cauUnis diffor- mibus exserta. Calyptra profimde plurifida. The first group contains all the species of which the fruit is knoAvn, and which correspond to the typical S. erytlirodontus, together with probably some others wliich are known only in a barren state, including the two ambiguous species aS*. australis and S. gemmascens. The second group consists of S. mnioides alone. 2. Streptopogon r marginatus ;— Schistidium marginatum, Hook. f. and Wils. Flor. AntarcL, 399, /. 151. I', vi. i; 3 J. DOT ANT OF KERGUELEN ISLAND. Christmas narbmir, forming l:u'i;v pati-lios on wet rot'lvS, Ilookcv. This, whii'h appoai-s dostituto of peristome, is in other respects more nearly related to Streptopogon than to any other i>cniis, and il" ineluded in it Avould oeeupy a position analogous to that of Stylostcgium acspUicinnt and S. conleclum before mentioned nnder Blimliit. 1. ElltOSthodon laxus, llouk. /. ct lllls. Fl. AnlanU., '.VM, t. 153, T. 5. (Physeomitrium)- Christmas Uarbour, Hooker. Royal Soimd, barren, and Swain's Bay, Avilli ncai'ly matiu'o eapsules, Eaton. Traecs of an internal peristome are present witliin the extcM-nal teeth. 1. Bartramia (riin.oN-oTis) appressa, llouk. f. ci iFih. Fi. New Zeaid. ii. 8i), t. 8G, f. :•. Royal Sound, barren ; Observatory Bay, with a few nearly ripened capsules ; and hill N. W. of ]\Iount Crozier, a tall barren slender state, Eaton. 2. Bartramia (Puilonotis) australis, Illtt. in Rook. Uandh. New Zealtl. Flor., 418. Swain's Bav and Boval Sound, all baiTcn, Eaton. The few small stems growing among other mosses appear to belong to this species. 3. Bartramia (Breutelia) pendula. Hook. Muse. Exot. t. 21. Ivcrguclon Island, Moseley. lloyal Sound; hill N.W. of Mt. Crozier; near Vulcan Cove, with abundant immatm-e fruit, Eaton. 1. Bartramia (Eubahtkamia) patens, Brid. Sp. 3lusc. iii. 82. Iverguclen Island, 2Ioseley. Royal Sound, with old fruit ; and hill N.W. of Mt. Crozier, Eaton. 5. Bartramia (Eubarteamia) robusta, Hook.f. et Wils.lFl. Aoitaret. t. 51). Kerguelen Island, Moseley. Royal Sound, with old capsules and young setae rising, veiy fine tall specimens, and Swain's Bay, Eaton. (Heard Island, Moseley.) [B. FLAViCAKS, Mitt., is enumerated by James as amongst the U. S. collections, collected at the rear of the American Transit House.] 1. Bryum (Webera) nutans, Schreb. ; Hedw. Muse. Frond, i. t. 1. Near Vulcan Cove ; hill K. W. of Mt. Crozier, a small state with unripe fruit growing amongst Fsilopilum trichodon, Eaton. 2. Brynm (Webera) elongatum, Dicks. Sw^ain's Bay, a single stem with ripe capsule, Eaton. 3. Bryum (Webera) crudum, Uedio. Muse. Frond, i. t. 88 (Mnium). Kerguelen Island, Moseley. Swain's Bay, with fruit just mature, Eaton. 4. Bryum (Webera) albicans, Wahlenb. Christmas Harbom% Hooker, Moseley. Near Vulcan Cove, Eaton. Specimens all barren. 5. Bryum (EccREMOTHEciiiM)pendulum, Homsch. Roval Sound ; and Cat Island, Royal Sound, Eaton. MUSCI. — W. MITTEN. 35 The inflorescence, "whicli is usually syuoicous in capsuliferous flowers, is some- times acoompaniod by unisexual flowers upon the same stem. 6. Bryum (Ecckejiotuecium) Eatoni, 3[itl. iu Joum. Linn. Soc, XV., p. 195. Synoicum. Caulis humilis, graeilis, iunovationibus infra comalibus paucis ramosus. Folia erecto-patentia, inferiora minora, superiora elliptico-lanceolata, nervo in acumen tenue lasvc vel denticulis paucis asperum exciu'rente, margine limbo tenvii e seriebus cellularum elongatarum 4-5 composite anguste reflexo intcgerrima, cellulis angustis limitibus tcncris areolata ; folia comalia longiora, basi subauriculato-dilatata, angulis rotundutis laxis areolatis. Seta clongata, recta, apic(> anguste; ciirvata. Theca pendula, sporangio ovato collo suba^quilongo ; operculo depresse conico acuminulato ; peristomio parvo, dentibus pallidis subsubulatis, apice punctulatis, processibus apice punctulatis ciliisquc in unum angustissimum conflatis in meni- brana usque ad dentium longitudinis ?j exserta impositis, annulo triplici cu'cumdato. Swain's Bay and Pi-oyal Sound, mth frmt ripened, Eaton. The very narrow leaves retain the same position in both the wet and dry state, they are narrower than observed in any form of B. 'pendulum. Tab. III. f. iv. ; 1, natui-al size ; 2, cauline leaf; 3, leaf from perichsetium ; 4, capsule ; 5, portion of peristome ; all magnified. 7. Bryum (Eccrejiothecium) bimum, Schreh. ; Bryol. Europ. t. 21. Christmas Harbour, Hooker. Near Swain's Bay ; and Royal Sound, with ripe fruit. Eaton. The specimens vary in size, the stems in some being nearly three inches high, the lower leaves are all blackened. 8. Bryum (Eccremothecium) alpinum, Lhm. Royal Soimd, with shining red foliage ; and Swain's Bay, all barren, Eaton. The red-leaved specimens are exactly similar to those states of this species which are fovmd in sub-alpine regions in Europe ; those states which are found in the plains have never the histrous appearance which adorns this handsome moss. The small specimen from Swain's Bay was mistaken for B. IcBvigatum, Hook. f. et "Wils. (also a Kerguelen species), to which in colour it has a great resemblance, and the similarity was increased by the points of the upper leaves being broad and obtuse ; the lower leaves are, however, of the usual form. 9. Bryum (EccREMOTnEciuM) argenteum, Linn. On sea clift's near Observatory, barren, Eaton. A small silvery state with the leaf points not produced. 10. Bryum (Eccremothecium) kerguelense, Mitt, in Joum. Linn. Soc. XV. 67. Monoicum, crospitosum. Caulis brevis, ramosus. Folia erecto-patentia, imbri- cata, inferiora rameaque ovali-lanceolata, acuta, carinato-concava, nervo rubro per- cursa, mari^ine integerrimo, cellulis angustioribus in seriebus duabus limbum sub- indistinctum formantibus ; reliquis suboblongis ; comalia longiora latioraque ; peri- chaitialia interna minora. Theca in pedunculo breviuscido rubro superiK; flexuoso E 2 36 liOTANY OF KKKGUKLEN ISLAND. I'urvato Iiorizontalis, (cmii-mombranacca, nilida; sporaui^-io ovnii ciilln recto aMpii- lougo sousim aiigustato ; oir satis parvo coairtato; ()[)orcul() I'oiivoxo ajjicc bievis- simo aouto ; piM-istomii dcntibu^^ ])alli(lis iiitiM-iii Iran'inciitis (-xtcrno iis(|ui' ad incdiimi adba>iriitil)iis. Kcrgudon Island, Moseleij. Stems imdudiny tlic numerous brancbes about '■^ linos bigb, and willi ihr foliage about balf a line wide. Leaves apprcsscd -wben dry, a few at tb(> apices ol: tbe brancbes green, tbe lower all dark brown. Seta 3 lines long. Ca])sule about la- lines long, oebraeeous, almost sbining. Tbe male llowers are terminal on brancbes arising below tbe perieluetium. Tbis small sjiecies appears to be nearly allied to B. demissum, Hook., but its capsule is symmetrical, and tbe peristome is dilFerent. Tab. III. fig. iii.; 1, jjlants nat. size; 2, entire plant; 3, eaulincleal'; t, [)eri- cluvtial and i-omal leaves ; 5, portion of peristome; all viagnijicd. 11. Bryum Isevigatum, var. ii., Hook. f. and IFils. Flor. Aiilarcf., U^), t. 151, f. 3. Cbristmas Harbour, barren. Hooker. 12. Bryum Wahlenbergii, Schwcvg. Cbristmas Karbour, common. Hooker. [B. "Warneum, Bland. ; gayanum, Mont. ; torqtjescens, Br. and Scb. ; and TALLKSCEXS, SclnvDDg., are all enumerated by James as Found by Kidder (Bull. LJ. S. Kat. Mus. 3, 2G.] 1. Mielichhoferia campylocarpa, Hook, el Am. in Hook. Icon, n., t. 13G (AVeissia). Kerguelen Island, Iloseley. Near Swain's Bay, with unripe fruit, Haion. First described from tbe Andes, wbere it was gatbered by Jameson; it was afterwards found in Mexico, and may be one of tbosc species extending tbrougbout the Andine chain. M. basillaris, Bruch et Scbimp., from tbe Abyssinian mountains, with entirely the same stature and appearance, differs in some particulars of the peristomial teeth, and in tbe nerve of the leaf vanishing below tbe point. Plagiothecium antarcticum, Mitt, in Joum. Linn. Soc. xv. 71. Monoi- cum, ctcspitosum, ramis aseendentibus. Folia compressa, subfalcata, nitida; caulina ovata, acuminata, intcgerrima, enervia ; ramea ovato-lanceolata, tcnuiter acuminata, subenervia ; omnia basi subcordata, cellulis angustis elongatis areolata ; pericbaitialia convoluta, late ovata, breviter acuminata. Theca in pedu.nculo elongate rubro ovalis, insequaKs, suberecta inclinatavc ; operculo breviter conico ; peristomio in- temo ciliis in unum coalitis inter processus carinatos dcntium longitiidinis imjoositis in membranam usque ad dentium dimidiam longitudinem exsertam- insidcntibus. B-oyal Sound, with matm'e and old fruit, Ha ton. Marion Island, Iloseley. Stems forming extensive soft patches, with shining foliage about balf a line wide. Seta about half an meh long, when dry twisted. Capsule obovate, the neck col- lapsing plicate, and so curved that the whole capsule is inclined ; mouth large ; MUSCl.— W. MITTEN. 37 palo peristome prominent. The male llower, as is frequent in this genus, forms one of a cluster of small l)ud-like Howers at the base of the pcrichx'tium. Closely resemWes the Em'opean P. nitididum, TFahL, scarcely differing exeoi)t in the base of its leaves. This is the species wliich is mentioned in Hooker's Hand- book of the New Zealand Flora, ii. 170, as Hypnum pulchellum Hicks ? from the Canterbury Alps. Tab. III. rig. V. ; 1, 2)laut uat. size ; 2, eauliue leaf; 3, pericluetium witli male flower at base ; 1, capsule ; 5, portion of peristome — all maguijied. [P. HoNiANUM, Sm., is enumerated by James as having been collected by Kidder in the IT. S. Transit Expedition.] 1. Acrocladium politum, Hook. f. el Wils. Fl. Antarct. ii., t. 151., f. 2 (Hypnum). Mltl. 1. c. Hab. — Christmas Harbour, slender, tufted state. Hooker, lloyal Sound, small and barren, Eaton. This moss resembles some species oi\ Flagiothecium, but seems to differ in habit, its branches with conduplicate bifarious leaves having so close a resemblance to those of Phyllogonium elegans, Hook. f. et Wils., that it is frequently mistaken for tbat plant. In the review of the genus Orthorhynchnm, Ilcich. by C. MLiller (Linnoea Band, 3G, p. 28), one of the species to bo referred to this genus, the O. Hampeamim, C. Muller, sent from Australia Felix by Baron F. von Mueller, must, from the description, be Acrocladium politum, of which specimens have been seen from Von Mueller, not however exactly corresponding in locality. 1. Stereodon cupressiformis, Linn. (Hypnum). Base of sea cHflf, lloyal Sound, barren, Eaton. The small specimens obtained exhibit this] variable species in that form which in Eiu'ope is found on the roofs of buildings or on the ground ; they are very unlike S. chrysogasier, C. Muller, so common in New Zealand. 1. Amblystegium uncinatum, Hedic. Christmas Harbom-, Hooker. Near Vulcan Cove, a tall robust form with nearly mature fruit ; and lloyal Sound, a similar state, but Imrrcn, Eaton. 2. Amblystegium fluitans, Dill. West side of Swain's Bay, barren, Eaton. A large state, with all but the terminal leaves of a brown colour. 3. Amblystegium riparium, Linn. In the lake at Christmas Harbour, Hooker. Specimen in a very imperfect state. Also found l)y the U. S. Transit Expedition growing with Ranitnculus crassipes. i. Amblystegium kerguelense, Mitt. Dioicum? Caulis decumbens, ramis confertis ascendentibus pinnatim ramosis. Folia caulina laxe imbricata, stricta vel curvata, ovato-lanceolata, subulato-acuminata, integerrima, nervo basi lato sensim angustato et in acumen cvanido percursa ; ccllulis parvis oblongis limiti- busquc pellucidis ad angidos paucis rcctaugidis latioribus areolata ; folia ramea 38 BOTANY OF KERGUELEN ISLAND. crccto-patoiuia, aupistiora, Jicrvo (.-rassioro. iiypmiui lilii-iiuiin var. ot 11. scrj)cns, var. Flor. Antarct., p. Ill) ot llS. Christmas 1 [arbour, Hooker. Near Swain's Bay, baiTon, Eaton. The single pat eh of this moss gathered by Mr. Eaton exhibits the species as very closely resembling A. fiUcinum, Linn., when it has not assumed a jiinnate form ; it is larger than A. serpens. The foliage is fulvous, ncilhcr wet nor dry is it altered in appearani'(\ 5. Amblystegiuin decussatum, Mook.f. et JFils. FL Neio Zeald. ii. t. uo, f. 2. (Uypnum.) l\oyal Sound, a slender straggling state, with irregular branches and an upright Umn, iimo\vj;st Bri/uin 2)eudHlum; near Swain's Bay, an upright state more robust and more branched ; near Vulcan Cove, a still larger state, Avith stems three inches liigh; all barren, ^rt/o«. All the specimens referred to this species have but little external resemblance! to the complete state found fertile in New Zealand, but they agree very closely in the areolation of thcii- leaves, and it is probable they arc only slender forms similar to those produced by A.fiUcinum. 1. Sciaromium conspissatum, nook. f. cl Wils. Fl. Antarct. 419, 1. 155, f. 3. (llypnum). Christmas Harbour ; Hooker, Moseley. A short barren state. All the Kerguclen specimens are smaller than those from the Falkland Islands. 1. Brachythecium subpilosum, Hook. f. et mis. Fl. Antarct. 418, t. 151, f. 1. (llypnum). Kerguclen Island, Moseley. More robust than the original [specimens from Cape Ilorn, and in this I'cspect nearer to the Hyimum rutahiilum, var. 5, Fl. Antarct., from the Falkland Islands, which has since been named H. subplicatnm, Hampe. If, however, the species may be supposed to vary as much in aspect as the European i?. rutabulum, these slightly larger forms may be fairly considered within the limits of probable variation. Intermediate between the Hermite Island specimens and those from Kerguclen arc some barren mosses from Otago, New Zealand, and some others collected in the Australian Alps by Von Mueller, to which it is probable the de- scription of Dr. Hampe's Hypnimi austro-alpinum may apply, as he says that the seta is tliick and rough, and the capsule short, which are the most prominent characters appertaining to B. sitbj^ilostmi. 2. Brachythecium salebrosum, Hoffm. (Hypnum). Hypnum rutabu- lum, var. 1, Hook.f. et TPlls. Fl. Antarct. 117. Chi-istmas Harbom-, Hooker. Hill N.W. of Mount Crozier, a fine silky state in large tufts, with stems 2-3 inches long ; Swain's Bay, in boggy ground on the west side, a smaller state, all barren, Eaton. This species is distinguished from B. rutahilinn by the form of the leaves on the principal stems, which are not so dilated at their base, the outline being more nearly MUSCI.— W, MITTEN, 39 ovate and not deltoid. Speeimens collected by Dr. Lyall in the Ai-ctic regions at Beechy Island, correspond very nearly A^atli the Kerguclcn moss. 3. Brachythecium paradoxum, llook.f.'et Wils. Fl. Antard. 449, t. 155, f. 2. (Hypniun). Royal Sound, and Swain's Bay, with matm-c fruit, Eaton. This species, which is found also in Ncav Zealand and Fucgia, varies in size ; the Kcrguelen specimens are smaller than those from New Zealand ; its affinity is with the European B. vcluUmim (Linn.), which is sometimes seen with falcate leaves, and then presents an appearance very differcp.t from its more usual state. 1. Psilopilum trichodon, nook, et Wils. m Rook. Lond. Journ. Bot. vi. 289. (Polytrichum). Hill N.W. of Mount Crozier, with narrow capsules, Eaton. Originally described from the Andes of New Grenada, where it was foimd near the snow by Purdie ; it was afterwards gathered by Jameson in a similar situation in the Andes of Quito. Pogonatum alpinum. Bill. Swain's Bay, with unripe fruit, Eaton. This species occru's also in Australia, and has been described as P. ^;sf which wcm-c dilTcrenl from those obtained by Dr. Hooker, and (5 genera were also added to the ilora. Fourteen s])ecies were found by ^Ir. Eaton ; of these S s])eeies and 2 genera were additional to those previously known, bringing the whoh^ number of the Ilcpaticce up to 25. The HepaticfC of Kerguelen are allied }nost nearly to those of the Auckland and Campbell's Islands, and of Fuegia. ^oterocJada porplnjvo7'hiza, Leioscyphus i^nllens, and Teinnoma quadripctrt'da, are found also in Euegia. Jioif/ernuuinia colorata, and Sip)ipliyo(iyna podophylla, are found at the Cape oi' Good Hope. The former is, however, veiy A\idely distri- buted in austral regions. As with the mosses, it is remarkable how many addi- tions were made to the flora by the small number of specimens obtained by each collector. 1. Plagiochila heterodonta, Hook. f. et Tayl. Fl. Antarct. 428, t. 157, f. 2. Christmas Harbour, on moist rocks, Hooker. Royal Sound, barren, Ealon. The specimens closely resemble those gathered by Dr. Hooker; it appears to be always a small species. 2. Plagiochila minutula, Uook.f. ct Tayl. Flor. Antarct. 427, 1. 157, f. 1. Christmas Harbour, on rocks and the groimd, Jlooker. 1. Leioscyphus turgescens, Ilook.f. ct Toyl. Ft. Antarct. 150, t. 64, f. 2. Hab. Eoyal Sound, amongst DitricJmm Hookeri, Futon. (Lord Auckland's group). 2. Leioscyphus ^^ollens. Mitt, in Joum. Zinn. Sac. xv., 68. Caulis procum- beus ascendensque, parce ramosus. Folia sursum secunda, conniventia, imbricata, orbiculata, caviuscula, integerrima, ccllulis rotuudis parietibus crassiusculis areolata. Aniphigastria crecto-patcntia, laneeolata, profunde bifida, laciniis elongatis subulatis. Folia involucralia majora, conformia ; amphigastrio parvo quadrifido laciniis dentatis integerrimisve. Perianthium obovatum, ore truncate integerrimo. Pboyal Sound, associated with F. turgescens, barren, Eaton. Stems from an inch to an inch and a half long, seldom branched, with the leaves i line wide. Leaves pale olive-green, becoming in age brown, rather firm, not IIEPATiaE. — W. MITTEN. 41 collapsing when dry, composed of rounded cells -vvhich at first contain small round granules that disappear in the older leaves. Stipules -} line long, the one imme- diately under the perianth is small and easily overlooked. Perianth compressed. No capsuliferous stems have been seen. It appears that in this species, and in some others of the same genus, the compressed truncate perianth is the result of the small size of the involucral stipule, which in the coalescence of the leaves of which the perianth is theoretically formed, is too small to affect its form, the reverse of which is so evident in the perianth of ZojiJiocolea. Tab. III., Fig. vi., plant nat. size; 2. leaf detached ; 3. stipule from the stem; 4. perianth as seen laterally with involucral leaves ; 5. stipule next the perianth ; all magnijied. 1. Lophocolea pallidovirens, SooJc. f. et Tayl. Fl. Antarct. 439, t. 159, t. 9. Kerguelen Island, Iloseley. Near Vulcan Cove, barren, Baton. (Fuegia). 2. Lophocolea Novae Zealandise, Lehm. et Lindenb. (Jimgermannia). Hoyal Sound, fragments amongst Ditrichitm Sookeri. Hill N.W. of Mount Crozier, with young perianths, Eaton. (New Zealand and Lord Auckland's group). 3. Lophocolea humifusa, Uook.f. et. Tayl. Fl. Antarct. 436, t. 159, f. v. Christmas Harbour, Hooker ; near Observatory Bay, barren, Eaton. The specimens are pale yellowish green, and seem not different from L. bidentata, with which it agrees in perianth. 1. Teinnoma quadripartita, Jlook. Mmc. Exot. 117 (Jungermannia). Kerguelen Island, a few small fragments, Moseley. Gathered also amongst Dicrana at Christmas Harbour by Hooker. 1. Jungermannia cylindriformis. Mitt, in Joum. Linn. Soc. xv. Exilis. Caulis procumbens, ascendens, subsimplex, vix radicans. Folia subalterna, antice incurva, oblongo-ovalia, obtusa, sinu parvo obtuso obtuse bidentata, dentibus sa?pe conniventer incurvis ; involucralia minora, acute bidentata vel caulinis conformia. Perianthium elongatum, cylindraceum, obtusum, apice plicatum. Royal Soimd, in very small quantity Arith perianths amongst Ditrichum Sookeri, and hill N.W. of Mount Crozier, with Scapania clandestina, Eaton. Stems about 2 lines long. Leaves ]■ line long, brownish green. Perianth 1 line long, of the same colour as the leaves. This minute plant is nearly related to J. inflata, Huds., having the same cylindrical perianth, and involucral leaves not much different from those of the stem, which arc the characters of the genus Gymnocolea, Dumort, which comprises besides the European J. iujlata, and the J. turbinata, Iladdi. Tab. III., Fig. vii. ; 1, plant nat. size ; 2 and 3, perianth and involucral leaves, dorsal and lateral view ; 4, caulinc leaf, expanded ; all magnified. F 42 BOTANY OF KI:KGUELEN ISLAND. 2. Juiigermannia leucorhiza, l^Utt. in Joum. Lhm. Soc. xv. (58. Caulis procuinbt'Uis, railii-fllis pallidis. i'olia laxo inscrta, quadrala subrotuiulave, sinii acuto obtusovo bilobalM, intordum lobo altero miiioiv ; lubis aciitis obLusisvc, iu- curvis; ccUulis rotundatis et Dvali-hoxagonis aroolata. Korguoleii Island, in very small quautity amongst mosses, barren, Moselei/. Stems loss tliau 1 inch long, Avitli the leaves I lino wide. Leaves green, tinged with brown.' lueomploto spccimen.s of a speeies not bci'ore U(;Liecd in llu' Anlarclie regions, but which appears to be near to the European J. venlricosa, Dieks, and to some states of J. barhafa. 0. Jungerinaunia colorata, Lchm. ci Limlenh. Christmas Uarbour, abundant on the hills, Uooker, Moselen (with perianths). 1 . SoleUOStoma humilis, ITooh. f. et Tayl. Fl. Aularcl. ii. 431, t. 158, 1'. (!. ^^.lungcrmannia) ; J. inundata, Flor. Nov. Zealand. 128, t. 93, f. 3. Ilab. Christinas Uai'liour, barren, Jlooker. A few fragments Avith one perianth, Moseley. Both S. hmniVis and J. inundata were originally described as stipulate species, no amphigastria liave, however, been since found on the specimens. It is probable that the figure of the supposed stipule of ,/. hninilifi, may have been drawn from a frairnient of Lcioscyi)hi(s tnrgescenn. Scapania, Lind (ex parte). Perianthium terminale, la3ve, a tergo ventreque compressum, ante cajisulae emissionem apicc dccurvum, hcrbaceo-membranaceum, ore tnmcato. luvolucri folia 2, libera, eaulinis conformia. — Plantse terricolse. Rami erecti ascendcntesve, simplices vel furcati. Polia fere ubique sequalia, bifaria, equi- tantia, profunda biloba, laciniis suba^qualibus apicibus rotvmdatis vel plus minus bifidis, textura e cellulis parvis. Amphigastria nulla. This description is that of the Synopsis Hepaticarum, with slight modification, it applies to S. densifolia, oertebralis, and chloroleaca, all so intimately related that the possibility of theii- being forms of one species may be conjectured. These differ from the chiefly Eiu-opean species which were included in the original idea of Sco2)ania, and which arc now by right of priority assigned to MaHinelUa, Gray, in having leaves not keeled m the space between the equal lobes, a peculiarity which gives the plants a different aspect. The perianth known from a single example on S. vertebralis, is like that found in Martinellia, but is narrowed upwards, trvmcate, the mouth bent over and denticulate. 1. Scapania densifolia. Hook. 3Iusc. Exot. 36 (Jungcrmannia). Kcrguelcn Island, Moseley. The specimens agree with those gathered by Mcnzies, and are of the same brown colom'. The distinction between S. densifolia and its congeners may be thus stated: — • S. densifoliu, Hook., lobis foliorum apice iategris rarius emarginatis. — ^S*. vertebralis, Tayl., lobis apice exsectis.— aS*. chloroleuca. Hook. f. et Tayl., lobis apice bifidis. HEPATIC^.— W. .MITTEN. 13 2. Scapania clandestina, Ilont. Bot. Criipt. Astrolabe, t. IG, f. i. Balan- tiopsis iucrassata, Milt. In Jour,i. Soc. Linn. XT. 197. Hill N.W. of Mt. Crozier, in very small quantity Avith J. cyUndriformis, Eaton. The stems of this small plant are about ?, IlcIi high, and with the leaves ?, line wide. Leaves firm, with small round cells ; lohes unequal and differing in their du-ection, the dorsal patent, the ventral nearly twice as large and divergent. In the Kpvii;uelen specimens the space between the lobes is keeled and curved, and both the lobes are denticulate, except the superior edge of the ventral lolic which is only denticulate towards the apex, and like that of the dorsal lobe is terminated by two larger teeth (hence bidentate, with a small rounded sinus). In this particular they nearly resemble the leaves of Balantiopsls diplophylla and B. evinacea, Tayl. (Scapania), but dili'or in their dense areolation. No authentic specimen has been s(vn of S. clandestina, Mont., but the figure quoted agrees except in the arcuation of the carina. A single stem picked from a tuft of Aneuva from New Zealand has the lobes more nearly equal, the carina straight, very much longer, and all the marginal teeth morfe spinif orm ; it is probable as suspected in the Synopsis Hepati- carum, that the plant in a complete state would be different from the imperfect specimens yet seen. This species departs from ^S*. densifol'm and its allies in the leaves being carinate, and thus corresponds to jSIurtinelllu ; it has, however, the apices of its leaves bedentate, which give it a different look from any of the species referred to that genus. 1. Cesia atrocapilla, Book./, et Tayl Fl. Antarct. 423. Foul haven, on clay banks, Hooker ; in small blackish patches closely inter- woven, Moseley. From the examination of some branches of the specimens collected by Dr. Hooker it appears that fertile shoots would have theii- upper leaves nearly or quite entire and nearly orbicular in foi"m. 1. Lembidiuni ventrosum, Mitt, m Jotmi. Soc. Linn, xv., 69. Caulis humilis, late compacteque cajspitosus, ascendens vel erect us, arcuatus, crassus, ra- mosus, innovationibus flagelliformibus ex amphigastriorum angulis emittens. Folia inferiora remota, superiora majora, insertione fere verticalia, patentia, a])iccm caulis versus imbricata, rotundata, profunde coucava, apice rotundata sinuve subindistincto subretusa, cellulis parvis parietibus angustis areolata. Amphigastria parva, cauli appressa, subtriangulari-ovata, apice subemarginata. Pcrianthium in ramo supcrne valde incrassato, foliis amphigastrioque involucralibus convolutis ovatis apice brcvi- ter bitri-denticulatis, trigonum, ovatum, apice plicatmn, ore laciniis conniventibus denticulatis obtusum. Hill N.W. of Mt. Crozier, in dense tu[ts on the earth, with capsules just rising, Eaton. In extensive brownish olive-green patches. Stems about 4 lines high, with the leaves scarcely \ line wide, closely congested anil cohering with very slender ' F 2 •li. 150TANY OK KEKGUKLKN ISI.ANl). hyaline rootlets. Perianth largo for the size of the plant, arising JVoni the apex ol" a tliickeuoil braneh ; apex obtuse before the egress of tiie rather large spherieal eapsuU', but al'terwanls sub-truueatc. Spores minute, round, smooth, Inowii, accompanied by rusilVnni mouiliate fibres. L. nutans. Hook. f. cl Tayl. Fl. Nod. Zcald. KJO, t. 65, f. S, is a larger species, and appeared by itself dilVerent from any genus that has been described, whereas L. rentrositm resembles the Jitiiprrmauiiia Franclsci of iiooker's Jirit. Jungermanniie, t. iO, a species which also produces tliiekly ileshy stolons, is irregular in the emar- "ination of its leaves, has the same kind of stipules as well as perianth, and is therefore a species of Lcmbidtam . How this genus or group of species may Im; distinguished from the Cephalozia of Dumortier must remain for examination. Tab. III., rig. viii. ; 1, plant of nat. size ; 2, portion of stem witli leaves and stipules ; 3, perianth and involucral leaves on lateral branch ; — all maynificd. 1. Herpocladium fissum, Mitt, in Joiu-ii. Linn. Soc. xv., GO. Caulis pcr- pusillus, lirnuis, crassiusculus. Folia alterna, patentia, ovata, obtusa, apice incurva, sinu parvo acuto breviter acute bidentata, concava, basi utroque caulem ad medium diamctrum tegentia, margine dorsali interdum llcxura siuuata rarius vmidentata, cellulis deusis obseuris areolata. Amphigastria foliis similia, patula divaricatave, apice obtusa, Integra. Kerguelen Island, Moselci/. Stems 3—1 lines long, wdth the leaves -]- line wide. The entire plant almost black. Tab. III., Fig. ix. ; 1, plant of nat. size ; 2, portion of stem, with leaves and stipule from the dorsal side ; 3, lateral view of leaf and spreading stipule ; i, leaf detached and expanded ; — all magnified. 1. Tylimanthus viridis, Jlilf. in Joum. Linn. Soc. XV., 197. Ilumilis. Caulis ercctus ascendensvc, apice decurvus, subsimplex. Folia distichc expansa, oblongo-quadi-ata, apice oblique sinu lato subtruncato-bilobata ; lobis obtusis, dor- sali majore apieem versus interdum subdentatis, cellulis parvis rotundatis limitibus pellucidis areolata. Royal Sound, and hill N.W. of Mount Crozier, all barren, Eaton. Stems h inch high, green, with the leaves scarcely 2 lines wide. Leaves green, frequently convex from the recurvation of the margin. This nearly resembles T. ienellus, Tayl. {Gymnanthe) from Tasmania, but it seems to be a smaller species. Tab. III., Fig. x. ; 1, plant of nat, size ; 2, portion of stem with leaves enlarged. 1. Marsupidium excisum, 3Iitt. in Joum. Linn. Soc. XV., 09. Caulis pri- marius repens, exLnde ascendens, pallidus. Folia inferiora minora, deinde superiore minora, omnia oblongo-quadrata, concava, sinu obtuso bidentata, integerrima sub- crenatave, lobis latis acutis incurvis, cellulis protuberantibus papulosa. Eoval ^o\)Ln^i\fii]i Acrocladiimi j)olituni and Fogonatum alpinum, barren, Eaton. lIEi'ATIC.E.— W. MITTEN. 45 Primary stems of the same colour as the leaves, fleshy, obscure, creeping ; from those arise erect or ascending simple or branched shoots, which are arcuate, their points attenuated and decurved. The leaves where largest are about l line long, and when flattened of the same mdth, of a pale obscm-e olive-grecu ; bases not decurrent ; insertion variable but generally oblique ; margins entii-e, or obtusely sub- crenate ; areolation of hexagonal or rounded cells with thin walls, enclosing a few green granules and projecting on lioth surfaces, but most on the external, as hyaluie papuhv. PapuUe of the same kind are also present on the yoimgei* stems, but less prominent. No kind of inflorescence has been seen on this species, and its location here is conjectm-ed from its having the same habit as If. KiiightU, from New Zealand. Tab. III., Fig. xi. ; 1, plant of nat. size ; '1, part of stem with leaves ; 3, cells from middle of leaf; both marjn'ified. 1. Fossombronia australis, Mitt, in Joum. Soc. Linn. XV., 73. Caulis coespitosus, prostratus vcl ascondeus, arcuatus, radicellis pui'piu'eis. Folia sub- quach-ata^ angulata, margine tlexuosa, antice incm-va. Perianthium turbinatum, margine flexuosum, angulatum, scmina rotunda limbo hyalino Isevia. Kerguelcn Island, Moseleij. Eoyal Sound, and near Vulcan Cove, with young capsules, 'Eaton. (Heard Island, Moselei/). Some of the specimens are very large, with arcuate stems more than an inch long, producing many purple rootlets. The leaves are 1 line long by about li wide, green, with pellucid cells. 2. Fossombronia pusilla, Linn. Christmas Harbour, amongst moss. Hooker. 1. NoterOClada porphyrorhiza, Nees ; N. confluens, Fl. Anlarct. 41G, t. IGl, f. 7. Christmas Harbour, on moist banks, Hooker. 1. Symphogyna pcdophylla, Tkunlj. (Jungermannia) ; Gottsche, Lindenh. et Nees Si/n. ZTepat. 481. Near Vulcan Cove; Eoyal Sound; hill X.W. of Mount Crozier; all Avithout fructification, JSaton. 1. Aneura multifida, Linn. (Jungermannia). West side of Swain's Bay, on boggy ground, and near Vulcan Cove, all barren. Baton. 2. Aneura pinguis, Linn. (Jungermannia). West side of Swain's Bay, small and barren, LJaton. 1. Marchantia polymorpha, Linn. Christmas Harbour ; ILooker, Jloselcy ; Iloyal Sound and Swain's Bay, Ealon. All the specimens produce scyphi, but arc otherwise barren. 4G BOTANY OF KERGUELEN ISLAND. IV. — Lichens. By the Rev. J. ]\l. CitoMBiE, l^.L.S. The first record that wo can lind ol" the Lichcu-tlora of lliis remote island, is con- taiued in ii preliminary account of the Antarctic Lichens collected by Dr. .T. D. Hooker * dnrinu: the voyage of the "Erebus " and "Terror,'" Avliicli was published by him and Dr. Thomas Taylor iu the " London Journal oj' 13otany " (1811), vol. iii. pp. G34-G58. The Kerguelen Island lichens there enumerated amoixnt in nmnber to 17 species, named by Dr. Taylor; but at least one lialJ' of the names attributed to them ai'c misapplied, and therefore must be excluded, owing chiefly to the determinations having been attempted in the absence of such microscopical analysis of the specimens as is now found to be essential for their discrimination. The number was subsequently raised to 27 sjiccies and varieties, AAdicn the list Avas revised by the Rev. Churchill Babington for publication in Dr. Hooker's " Plora Antarctica " (1817), vol. ii. pp. 519-512. A considerable proportion of the names iu this later list must however be rejected for the same reason as those erased from the previous one. Unfortunately authentic examples of sevei*al of Dr. Taylor's critical species are wanting in the Kew Herbariiun ; f and his collection (now in the Herbarium of the Boston Society of Natural History), according to Professor Edw. Tuckcrman, contains very little that is illustrative of his Kerguelen Island determinations. I have lately published a further revision of the Kerguelen Island Lichens collected by Dr. Hooker, in the " Journal of Botany " for April 1877, wherein the number of the species is reckoned to be 18 or 19 besides 2 named forms. ]\Ir. jMoseley of the Challenger Expedition gathered in this island upwards of 13 species and 1 named form. ( Vide Crombie in Jom-n. Linn. Soc. Bot. 1877.) Dr. Kidder of the American Transit of Venus Expedition collected in tlie vicinage of Molloy Point 13 or 14 species and 1 named form. These with others from the Taylor collection are specified by Prof. Ed. Tuckcrman in Bulletin U.S. Nat. 3Ius. Xo. 3 (187G), and are noticed by me in the " Journal of Botany " for April 1877. The collection made by Mr. Eaton between the end of October 1874 and the end of Eebruary 1875, in the district immediately to the westward of Dr. Kidder's * One (or more) species of Licheus was obtaiiied in Kerguelen Island in 1776 by Mr. Anderson, the Surgeon and Naturalist who accompanied Captain Cook. — A. E. E. I Dr. Taylor died shortly after the publicition of his first rough determination of the Antarctic Lichens, and it was impossible to recover from the heap of his unananged materials, which were in a confused state, all of the specimens which should have been returned. 1 strongly suspect, from the state of his notes sent to me from time to time, that he did not attend sufficiently to localities, and tiiat some of the specimens in the Herbarium labelled as from Kerguelen Island did not come from that island. — J. D. H. LICHENS.— .J. M. CEOIMBIE. 47 station, compvisos 50 or 51 spccios and 9 named forms. Of thcsn about 30 Avcrc described as neAV species referred to known genera, in ibe Joui'ual of Botany for November 1875 and January 1876, and again witb fuller diagnoses with wbicli Dr. Nylander (Avbo most kindly assisted mo in tbeii- determination) favoured me in tlie Jom-nal of the Linnajan Society (Botany) for July 1870. Though several of the new species bear a superficial resemblance to some of our northern lichens, yet on analysis they are found to be quite distinct, and for the most part are peculiar to Kerguelen Island. The results obtained by the German I'ransit of Venus and Surveying Expedition at Betsy Cove, are not yet published. The total number of species obtained from the island is 61, and 10 varieties. Traces of a few other species exist in the various collections, consisting either of sterile thalli or undeveloped apothecia which are necessarily indeterminable. 1. Lichina antarctica, Cromb. in Joum. Bot. v. 21 (1876) ; et m Journ. Linn. Soe. xv. 181. Observatory Bay, on dry rocks near the sea, Eaton. 1. Amphidium molybdoplilacum, JSfyl. in Joum. Bol. iv. 333 (1875) (errone molybdophoeum) ; Cromb. In Journ. Linn. Soc. xv. 181, et in Joum. Bat. vi. 103, 106. Lecanora melanaspis, Bab. in Flor. Antarct. 536 {cxcl. Syn. L. dichroa) Pannaria glaucella, Tuckerm. in Bull. U. S. Nat. Mus. ui. 28. On earth and stones, Christmas Harbour, Hooker ; on stones in wet places, Swain's Harbour, Eaton ; Molloy Point, Kidder. 1. Stereocaulon cymosum, Cromb. in Joum. Linn. Soc. xv. 182, et in Journ. Bot. vi. 103. S. coralliuum, ILook.f. et Taijl. Flor. Antarct. 528. On rocks, altitude 6-1200 feet, Christmas Harbour, Hooker, Moseley ; top of a hUl on west side of Carpenter's Cove, barren. Smith Dorien {Eaton). 1. Cladonia fimbriata, Hojfm. ; Cromb. in Joum. Linn. Soc. xv. 182. C. pyxidata, Linn. ; Tuckerm. In Bull. Torr. Bot. Club, October 1875, et Bull. JJ. S. Nat. Mus. 3, 29. Dry slopes, Swain's and Observatory Bays, Eaton ; Molloy Point, Kidder. Var. costata, Flk. Observatory Bay, Molloy. 2. Cladonia COrnuta, Lhm. ; Cromb. in Journ. Linn. Soc. 1877. Kerguelen Island, Moseley. 3. Cladonia acuminata, Ach. ; Cromb. in Joum. Linn. Soc. xv. 182. C. phyllopbora, Tayl. Christmas Harboui-, Hooker ; observatory Bay, common, but sparingly fertile, Eaton. 1. Neuopogon melaxanthus, Ach. (Usnea) ; Cromb. in Joum. Linn. Soc. XV. 182; et in Journ. Bot. vi. 103, 106 (1877). Usnea sulphurea, 3Iul(. ; Tuckerm. hi Bull. Torr. Bot. Club, 1875, et In Bull. JJ. S. Nat. Mus. 3, 27. Eamalina scopulorum e, Uook.f. et Tayl. Flor. Antarct. 522. 48 nOTANY OF ICERGUELEN ISLAND. Eocks and boulders, Christmas llarbour, Aiiderso)!, TTonkr)' ; on the upper slopes and tops of hills, Ealou. Vak. souediifku, Cnmh. I. c. Connnoii at all altitudes, Fjtdon. Vau. CTi.i vr\, Ni/I. Crombic, I. c. Observatory Bay, Eaton. 2. N. Taylori, llool: f. ; Flor. Anforct. 521, t. 1J)5, Fig. i. (Usnca) ; Cromh. ill Joiirn. Linn. Soc. xv. 1S3, rt in Jonru. Bot. vi. 103 (1877). Ivocks ascomling- to 1,200 feet ; Christmas Harbour, llool; e r, Moscley ; Swain's Bay and Carpenter's Cove (hut not near Observatory Bay), JS'a^OM ; Molloy Point, Kiildcr. 1. Parmelia Stygioides, 'Nyl. ; Cromh. in Joum. Bot. iv. 333 (1875), et in Joiini. Linu. Soc. xv. 183. Dry rocks and stony slopes, Swain's Bay, Eaton. 1. Peltigera rufescens, rav. spvu-ia, DC. ; Cromb. in Jonrn. Linn. Soc. xv. 183 ; Joum. Bot. vi. 103. Pcltidea venosa, ILook.f. et Tayl. Flor. Antarct. 525. On wet moss, &c., Christmas Harbour, Hooker ; Swain's Bay, Eaton. 2. Peltigera polydactyla, forma hymenina, Ach.; Cromb. in Jonrn. Linn. Soc. xv. 183. P. polydactyla, Ftor. Antarct. 521'. P. horizontalis, Ach. ; Flor. Antarct. I.e. 525. Amonn^st moss, &e., Christmas Uarbour, Hooker ; Observatory Bay, Eaton. 1. Pannaria dichroa, Jlook. f. et Tayl. in Jjond. Joum. Bot. iii. G13; Cromh. in Joum. Bot. vi. 100 (1877), et in Jonrn. Linn. Soc. xvi. P. Taylori, Tuckerm. in Bull. Torr. Bot. Clith, October 1875, et in Bull. V. S. Nat. litis. 3, 28. P. placodicopsis, Nyl. in Joum. Bot. iv. 33 1< (1875) ; Cromb. in Joum. Linn. Soc. xv. 183. Lecanora melanaspis, Ach.; Ilook.f. et Tayl. Flor. Antarct. 536. On rocks, Christmas Harbour, Hooker, Moseley ; 01)servatory Bay, sparingly, Eaton ; Molloy Point, Kidder. 2. Pannaria Obscurior, Nyl. ; Cromb. in Joum. Bot. iv. 331 (1875), et in Joum. Linn. Soc. xv. 183. On decayed n^oss, Observatory Bay, Eaton. 1. Psoroma hirsutulum, Nyl.; Cromb. in Joum. Bot. iv. 333 (1875), e/f in Joum. Linn. Soc. xv. 181. On moss and dead stems of Accena, very local. Observatory Bay, Eaton. 1. Lecanora (Placqpsis) gelida, i?»«. ; Flor. Antarct. n. ?)o^; Tuckerm. in Bull. Torr. Bot. Club, Oct. 1875, et in Bull. 11. S. Nat. Mus. 3, 29 ; Cromb. in Joum. Linn. Soc. xv. 184, et in Joum. Bot. vi. 104, lOG (1877). On stones, Christmas Harbour, Hooker ; Observatory Bay, Eaton ; Molloy Point, Kidder. Var. viTELLiNA, Bab. in Flor. Antarct. 1. c. ; Cromb. 1. c. Cbristro.as Harbour, Hooker, Moseley. Var. LATEKiTiA, Nyl. ; Cromb. 1. c. Placodium bicolor, Tuckerm. 1. c. Christ- mas Harbour, Hooker, Iloseley. Swain's Bay and Royal Sound, Eaton. LICHENS.— J. .M. CROMBIE. 49 2. Lecanora (PLAcorsis) macropthalma, Hook. f. el Taiji. in Lond. Jmirn.Bol. iii. GGU (Urcoolaria) ; Tackenn. hi Bull. Torr. Bot Clnh, Oct. 1875, et ill Bull. U. S. Nal. Mas. 3, 29; Cromb. in Journ. Linn. Soc. xv. 185 and xvi., et in Journ. Bol. vi. 104. On stones in moist places, Hooker, Eaton, &c. Lecanora (Placodium) elegans, Ach.; Cromb. in Joum. Linn. Soc. xr. 181, el in Joum. Bot. vi. lOi- (1877) ; Tackertn. in Bull. U. S. Nat. Mns. 3, 28. L. murorum, var. 3, Flor. An'tarct. 535. On rocks and stones, Christmas llarbonr, Hooker; Observatory and Swain's Bavs, Eaton. Var. LUCENS, Nyl. ; Cromb. 1. c. On dead stems of Aca'na and Pringlea, Ob- servatory Bay, Eaton. •4. Lecanora SUbunicolor, Nyl. ; Cromb. in Joum. Bot. v. 19 (1876), el in Joum. Linn. Soc. xv. 184. On rocks, Royal Sound, very sparingly, Eaton. 5. Lecanora vitellinella, Nyl. ; Cromb. in Joum. Bot. iv. 334 (1875), et vi. 104 (1877), et in Joum. Linn. Soc. xv. 184. L. candclaria, Ach.; Flor Antarct. 537. Maritime rocks, Christmas Harbour, Hooker ; Observatory and Swain's Bays, Eaton. 6. Lecanora cyphelliformis, Cromb. in Joum. Linn. Soc. xvi. Christmas Harbour, Moseley. 7. Lecanora diphyella, Nyl. ; Cromb. in Joum. Bot. V. 21, et in Joum. Linn. Soc. xv. 184. On rocks at low elevations, Observatory Bay, Eaton. 8. Lecanora atro-Caesia, Nyl. ; Cromb. in Joum. Bot. iv. 334 (1875), et vi. 104 (1877), et in Journ. Linn. Soc. xv. 185. L. confluens, Hook. f. et Tayl. in Lond. Journ. Bot. iii. G36. L. albo-coerulescans, Ach. ? Bab. in Flor. Antarct. 538. Rocks at Christmas Harbour, Hooker ; Observatory Bay and stony slopes at Volage Bay, plentiful, Eaton. 9. Lecanora brocella, Nyl.; Cromb. in Journ. Bot. V. 21 (1870), et vi. 104 (1877), et in Journ. Linn. Soc. xv. 185. On dead moss, &c., Christmas Harbour, Hooker ; Observatory Bay, Eaton. 10. Lecanora umbrina, Ach. ; Cromb. in Joum. Linn. Soc. xv. 185. On dead plants, Observatory Bay, Eaton. 11. '.Lecanora kergueliensis, Nijl.; Cromb. in Joum. Bot. vi. lOG (1877). TJrceolina kergueliensis, Tuckerm. in Bull. JJ. S. Nat. Mus. 3, 29. Bocks at Molloy Point, Kidder. 12. Lecanora sublutescens, Nyl.; Cromb. in Joum. Bot.v. 21 (187G), et in Journ. Linn. Soc. xv. ISO. G 50 BOTANY OF KERGUIOLEN ISLAND. On a shailod sea olill" near Observatory Bay, c(>U)uring the roek, Ealon. [L. ciTUiNA, Acli., L. liuVTUKOCAur.v, Fr., and L, JLlAuiiNi, Ach., enumerated iu the Flora Antarcliea, 53G, from very impcrrcct materials, arc too doubtful to be eiuunerated.] I. Pertusaria perrimosa, Nyl. ; Cromb. iu Journ. Linn. Soc. xv. 186 ; Joiini. JUui. vi. li)i. r. communis, D C ; Flor. Autarci. 5i0. PLocauora tartarea, Ach.; F/or. Aularct. 530. On roelly have hcon added io 1 lie 53 in his collection. Jiulgin;;:; Troni the nuniher ol" species apparently indicjenous to vm- sheltered situations ■which jjo to form the 18 that are not vepr(^sentc>d in his collec- tion, this conjecture may have been not Far IVoni the mark. But the advantages ol' situation atforded in Swain's ]?ay lor the growth of various Algtr absent from the almost waveless shores ol" Observation Bay would have avaihnl nothing, had it not been for the liberality and kindness of Cai)lain Faii-iax, ]\.N., in command of II. M.S. " A^'olagc." Having invited Mr. Eaton to be his guest for thi'co weeks, he conveyed him in his gig to almost every part of the bay that was accessible by boat in Kerguelen Island weather, and surrendered his cabins without reservation to the reception of buckets and specimens of all descriptions, excluding only seals and cetacca accommodated elsewhere. The local distribution of the species round the coast may be ascertained roughly from a comparison of the constituents of the collections above mcntion(>d. Of the 71 species, 11 axe common to all of the collections, and 8 common to three out of the four, making together 22 species, which may be regarded as plants generally distributed round the island ; 11 are common to Mr. Eaton's collection and one of the other three, and 1 species to Dr. Hooker's and Dr. Kidder's, — together making 15 local plants, mostly of frequent occurrence ; 5 are common to Dr. Hooker's collection and Mr. Moscley's (gathered in Christmas Havboiu"), and 29 arc in one of the collections only, making 31 scarce or rare species. Of the 29, there are in Dr. Hooker's collection 7 species, in Mr. Moseley's 1, in Dr. Kidder's 1, and in Mr. Eaton's 17. As to their general geograpliical range, 20, or rather more than a quarter of them, are found in various parts of the .shores of Europe, and some are cosmopoli- tan. The follo-ndng 8, so far as is known, are peculiar to the island : — Desmareb'lia chordalis, Sphacelaria corymhosa and S. affinis, Melobesia kergnelena, Nitophyllimi fusco-rubmm, Epymenia variolosa, Ptilota Eatoni, and CallUhamnion simile. The following are the numbers of the species after their respective families : — IHicacese, 2. SporoclmacejB, ■!. Laminariaceaj, 2. Dictyotaceae, 2. Chordariacege, 3. Ectocarpacese, 3. Rhodomelacese, 4. Laurenciacese, 2. CoraUinacese, 3. Sphajrococcoideae, 8. Gelidiacese, 1. Bhodymeniacese, 4. Cryptonemiaceae, 11. Ceramiaceaj, 7. SiphonacesB, 3. ITlvaceae, 5. Coufervacese, 7. Of these 16 belong to the Olive, 40 to the Red, and 15 to the Green Series. MARINE ALG^.— G. DICKIE. 55 There ai'e also included iu the present paper, for convcnicncej 4 fresliTrater species : — Bostrychia caga, Vaiichcrki Dillioi/nii, TJlva cristata, and Prasiola Jluvtaiilis. 1. D'Urvillea utilis, Bory ; Flov. Antarct. 454; Dickie in, Journ. Linn. Soc. xy. 43, 198 ; Farloio iu Bull. U. S. Nat. 3Ius. 3, 30. On exposed rocks at and below half-tide level, not in very sheltered situations ; abundant. — In the Southern ocean, between lat. 45° and 55° S., reaching to lat. G5° S. in the meridian of New Zealand (Hooker). 2. D'Urvillea Harveyi, Mook.f. Flor. Antarct. 456, t. clxY., clxvi. ; Dickie in Journ. Linn. Soc. xv. 41, 198; Farloio in Bull. U. S. Nat. Mus. 3, 30. In positions still more open than D. utilis. (Cape Horn and the Falklands.) 1. Desmarestia Rossii, ITook. f. and Sarv., Flor. Antarct. 467, t. clxxii., clxxiii. ; Dickie in Journ. Linn. Soc. xv. 44, 198. Swain's Bay on rocks in 3 fathoms, at the end of an island about 2 miles within the entrance of the bay, exposed to a slight swell from the open sea ; local and not common, Baton. (Fuegia, Falkland Islands, Heard Island, Moseley.) 2. Desmarestia chordalis, Mook.f. and Sarv., Flor. Antarct. 467; Dickie in Journ. Linn. Soc. xv. 44, 198. Swain's Bay, in 3 fathoms, with the preceding ; very local, Baton. Christmas Harbovu', Hooker, lloseley. (Kergueleu Island only.) A very graceful species. The fronds, upwards of 4 feet in length, are arranged in a manner similar to those of a fern, and cause the plant, as seen in situ from a boat, to bear a general resemblance in contour to such species as Aspidium filix- mas. 3. Desmarestia aculeata, Lyngh. Var. media, Grev.; Hook. f. and JSarv., Flor. Antarct. 466; Dickie in Journ. Linn. Soc. xv. 44, 198. Between tide-marks, Swain's Bay, Eaton. Cockbiu'n Island {Hooker) ; and in North temperate and Arctic seas.) 4. Desmarestia viridis, La/nix. ; D. viridis (and var. /3. distans). Hook., Flor. Antarct. ii. 466 ; Dickie in Journ. Linn. Soc. xv. 44, 198 ; Farlow in Bull. U. S. Nat. Mus. 3, 30. Christmas Harbour, Hooker, Moseley. In 2 fathoms, Royal Sound and Swain's Bay ; common. Baton. (Marion Island ; the Falklands ; Cape Horn ; American coast from New York northwards ; Unalaschka ; Hunde Island ; W. coast of Em-ope.) 1. Macrocystis pyrifera, Ag. ; Flor. Antarct. 461, t. clxix., clxx. ; Dickie in Journ. Linn. Soc. xv. 44, 198 ; Farloio in Bull. U. S. Nat. 3Ius. 3, 30. Abundant along rocky portions of the coast. (Antarctic Sea, from lat. 40° to 60° S. ; New Zealand; Indian Ocean; Marion Island ; Chili; California.) 1. Lessonia fuscescens, Bory. ; Flor. Antarct. 457, t. clxvii., clxviii. A., and clxxi. D. ; Dickie in Journ. Linn. Soc. xv. 44. 56 BOTANY ol KKHOrKLICN ISLAND. In exposed situations; Cln-istmas llarlunu-, rarr, / 1 oo/ccr uml .l/osclei/. ((Miili, Fucgia, I'alkland Islands. Cnckl)urn Island, Auckland and Campbell Islands.) 1. AspcrOCOCCUS siuuOSUS, Itofh; F/or. An/ai'cf. ii. 4G8 ; JXtckic i)i Joiiru. Linn. Soc. xv. IDS. Crevices of rocks bcl ween tide-marks, Observatory Bay (two vcrv small speci- mens), £}iaia in very shallow water along the shore in Observatory Bay, Faloii. (Atlantic coasts of France and Britain ; Baffin's Bay.) 1. EctOCarpuS geminatus, Ilook.f. and Ilarv. Flor. Antarct. 4(59; Dickie in Journ. Linn. Soc. xv. il, 199. Plentiful on Desmarestia at Christmas Harbour, Moseley. Very slender soli- tary young plants on Annelid tubes, at 5 fathoms ; stronger and more bushy, with trichosporangia only (but these abundant), on Myiilus at 1 fathom, and in tide pools, Observatory Bay ; frequent in Swain's Bay among Cladophora Jlexuosu in pools and shallow Avatcr, Eaton. (Falklands and Cape Uorn.) 1. Sphacellaria COrymbosa, Dickie in Joum. of Dot. v. 50 (187(5), et in Journ. Linn. Soc. xv. 199. 'r S. funicularis, Mont. ; Flor. Antarct. 469 ; Farlow in Bull. U. S. Nat. Mus., 3, 30 ; fronde estuposa, fills eagspitosis, ramis infei-ne panels dichotomis superne subpinnatim decomjiositis, ramulis alternis corymbosis. On shells of Jlytilus and on Annelid tubes; Swain's and Observatory Bays, Fat on. The specimens arc 2 to 3 inches long, caespitose, but without fruit. Dr. Hooker's plant obtained at the Falklands, and Dr. Kidder's from the vicinage of Molloy Point in Boyal Sound, are probably the same as the preceding. 2. Sphacellaria afRnis, Dickie in Journ. of Bot. V. 50 (1870), et in Journ. MARINE ALG.E.— G. DICKIE. 57 Linn. Soc. xr. 199 ; filis dense oipspitosis crcctis parce clicliotomis, articulis diametro subipqnalibus vol paulo longioribiis, trichosporangiis solitariis obovatis brcvitcr pediecllatis. On shells of Mytilus in rather open situations ; Swain's Bay, Eaton. The specimens are about 4 inch in height, and are similar in habit to the British S. radicans. 1. Rhodomela Hookeriana, Ag. ; Bhodomela Gaimardi nook. f. mid Haw. Flor. Antaixt. Ibl, t. clxxxiv. {iion Gaud.) ; Farloio in Bull.TJ. S. Nat. Mus. 187G, 3, 30 ; Dickie in Journ. Linn. Soc. xv. 199. Swain's and Observatory Bays, frequent, Eaton ; near Molloy Point, one speci- men, Kidder. (Falklands and Tuegia.) 1. Polysiphonia abscissa, Hook. f. and Harv. Flor. Antarct. 480, t. clxxxiii. '1 ; Dickie in Journ. Linn. Soc. xv. 199. Forma microcarpa ; P. microcarpa, Hook. f. and Harv. 179, t. clxxxii. 3 ; Harv. Ner. Ansf. 42. On roots of Macrocystis, and on tubes of Annelides, in 1 to 5 fathoms, Obser- vatory Bay, ^rt/o». (New Zealand; Tasmania; Fuegia). There are two forms of this species, one of them smaller and more rigid than the other. The P. microcarpa of the Flora Antarctica represents one of them, P. abscissa the other. 2. Polysiphonia anisogona, Uook. f. and Harv. Flor. Antarct. 478, t. clxxxii. 2 ; Dickie in Journ. Linn. Soc. xv. 44. Kerguelen Island, Moseley. (Falklands and Fuegia, Hooker.) 1. Dasya Berkeleyi, Mont. ; Dickie in Journ. Linn. Soc. xv. 44, et 199 (var. ^ Dacisli) ; Farloio in Bull. U. S. Nat. 3Ius. 3, 30. Polysiphonia punicea. Hook. f. and Haro. Flor. Antarct. i. 182; (Heterosiphonia) Berkeleyi? var. (6 Dacisii, idem, 480. Swain's Bay, on the seaward sides of islands, Eaton ; Royal Sound, Kidder. (Auckland Islands, Marion Island, Falklands, Fuegia, Chiloe.) Mr. Eaton's specimens belong to the var. 0 Davisii, having a habit and colour different from those of the typical plant. There are examples in different stages ; but in all the ramvili are heterosiphonous. 1. Bostrychia vaga, Hook. f. and Harv. Flor. Antarct. 484, pi. clxxxvi. i. (Stictosiphonia). Christmas Harbour, on rocks and stones above high water mark, and in damp places a considerable distance from the sea, abundant, Hooker. 1. Delisea pulchra, 3Iont. ; Flor. Antarct. 484 ; Dickie in Journ. Linn. Soc. XV. 45. Clu-istmas Harbour, Hooker, Moseley. (Heard Island, S. Tasmania, W. and E. Australia.) 1. Ptilonia magellanica, 3Iont. ; Dickie in Journ. Linn, Soc. xv. 45, 200 ; H 58 liOIANY OK KEHGUELEN ISLAND. Farloio in Jiu/l. U. S. Nat. Jffiis. 1S7(), o, .'50. I'locainium ? ina2:ollanicum, JIoolc. f. and Jlorv. Flor. Antord. 171. Thamuoplioi-a magi'llauk-a, Jlont. Christmas Harbour, Ilookcr, JIosclci/. In tideways and on parts ol" islands open to a sli!::h( suell from llie ouli-r sea, not in sheltered waters; Swain's Bay, Eaton ; Royal Sound, Kidder. (Falklands and Euegia.) 1 . Melobesia aiitarctica, Ag. ; Dickie in Joum. Linn. Soc. xv. 200. M. verrucata, Lamx ; Dickie I. c. 15. M. vcrrucata, var. antarctiea, Hook, and Rarv. Flor. Anturct. 1S2. On Ballia, Sf'C, Swain's Bay, Eaton ; Christmas Uarbour, Hooker, (i^'uegia, Ealklands, Tasmania, New Zealand, Auckland, Antarctic Seas.) -. Melobesia lichenoides. Ell. and Sol. ; Dickie in Joum. Linn. Soc. xv. 200. Swain's Bay, common, Eaton. The only example preserved Avas grappled in about 2 fathoms in a tideway between two islands, inerusting two sponges (Microciona atrosangninea, Bk., and Ualichondria incrnstans, Jtn. ; both British siiccies.) It is normal in habit, texture, and in the character of the keraniidia, and is very luxm-iant, measuiing about 3 inches square. (St. Paul's Island, Norfolk Island, the Mediterranean, France, Britain, Baffin's Bay.) 3. Melobesia kerguelena, Dickie in Joum. Dot. V. 51, 1870, et in Jotmi. Linn. Soc. xv. 200; simplex, 2\ poll, diam., dui'a, crassa, tantum in medio subtus adhierens, subconvexa, eircumscriptione orbiculari, margine Isevi parce undulata, kcramidiis conspicuis numerosis pleiiimque in seriebus concentrice disjiositis. Swain's Bay in 2-3 fathoms, with the preceding, Eaton. ]\Ir. Eaton has an impression tliat this grows upon Ballia or Ptilota, but I should rather be disposed to suspect that it was attached to rocks. The description was taken from an almost complete sf)ecimen ; there are fragments of others whose contoLU' is less regular, probably through interference of external objects. All of them are in coloiu- of a very pale buff or dull yellowish hue, varied with pale red tints. 1. Delesseria Lyallii, Hook. f. and Harv. Flor. Antarct. 471, t. clxxvi; Dickie in Joum. Linn. Soc. xv. 45, 200 ; Farloio in Enll. U. S. Nat. Miis. 1870, 3, 30. Christmas Harboui', Hooker (but not seen attached), Moselei/. Observatory Bay (iU grown), and Swain's Bay (well grown), abundant, Eaton. (Marion Island, Falklands.) Dr. Hooker obtaining only wrecked specimens at Christmas Harbour, was led to suppose that this species was a resident of the exposed coast. It is however common in the very sheltered waters of Observatory Bay, though it certainly does not tluivc there ; and it is abundant and luxuriant round the islands iu Swain's Bay, in 3-5 fathoms. MARINE ALGJE.—G. DICKIE. 59 2. Delesseria Davisii, Hook. f. and Harv., Flor. Antaret, 470, t. clxxv. ; Dickie in Jonrn. Linn. Soc. xv. 45, 200. Swain's Bay ; normal on Mi/tilus in sheltered places at or just below low-water mark; Taryin;? in more open situations, Eaton. (Falklands and Fuegia, Hooker.) 3. Delesseria quercifolia, Bonj. ; Flor. Antaret. 471 ; Dickie in Journ. Linn. Soc. xv. 200. Swain's Bay, frequent, Eaton. (Falklands and Fuegia, Hooker.) 4. Delesseria Crassinervia, Mont. ; Flor. Antaret. 471 ; Dickie in Journ. Linn. Soc. xv. 200. Swain's Bay, near the siu'facc of the water ; Observatory Bay, only one ill-grown cxamjilc, Eaton, (Fuegia, Falkland, Auckland and Campbell Islands.) 1. Nitophyllum fusco-rubrum. Hook. f. and Harv., Flor. Antaret. 472 ; Dickie in Journ. Linn. Soc. xv. 45, 200 ; Farlow in Bull. U. S. Nat. 3Ius. 1870, 3, 30. Christmas Harbour, Hooker. Abundant in open water in Swain's Bay, not found in sheltered places, Eaton. There is a variety in the collection with coc- cidia. Mouth of Royal Sound, Kidder. 2. ? Nitophyllum multinerve, Hook. f. and Harv., Flor. Antaret. 473 ; Dickie in Journ. Linn. Soc. xv. 45. Clmstmas Harbom* (one specimen referred to tliis species with doubt, Moseley). (Falklands and Fuegia, Hooker.) 3. Nitophyllum lividum, Hook. f. and Harv., Flor. Antaret. 4<72, t. clxxix. ; Farloio in Bull. U. S. Nat. 3Ius. 1876, 3, 30 ; Dickie in Journ. Linn. Soc. XV. 201. In very sheltered water (one example only) at 6 to 10 fothoms, Swain's Bay, Eaton. Royal Sound, Kidder. (Falklands, Hooker.) 4. Nitophyllum laciniatum, Hook. f. and Harv., in London Journ. of Bot. iv. 256 ; Dickie in Journ. Linn. Soc. xv. 201. N. Bonnemaisoni car. laciniatum, Hook.f. and Harv. I. c. 474. Swain's Bay, frequent, in 3 to 5 fathoms, Eaton. (Falklands and Fuegia, Hooker.) 1. Chsetangium variolosum, Ilont. ; Dickie in Journ. Linn. Soc. xv. 45, 201. Notogenia variolosa, Mont. ; Flor. Antaret. 487. Clu"istmas Harbour, very abundant. Hooker, Moseley. 01)servatory and Swam's Bays, abundant between tide marks. (Fuegia, Falklands, Auckland Islands.) 1. Plocamium Hookeri, Harv. in Flor. Antaret. 474 ; Dickie in Journ. Linn. Soc. xv. 15, 201. Christmas Harbour, Hooker, Moseley. Swain's Bay, local, at 2 to 3 fathoms, in situations open to a slight swell from the outer sea, Eaton. (Heard Islands, Moseley.) H 2 60 nOTANY OF KRROrELEN ISLAND. 1. Rhodophyllis capcnsis, 7v7;/-. ,- Dickie in Jouni. Linn. Soc. xv. 201. Swain's and Oltsrrvatory Hays, spariuiJijly, ou tubes ol" Anuclides, in 3-5 rathoms, Eaton. (Table ami Simon's liays.) The few specimens collected by Mr. Eaton are duai T and very narrow . 'They all liav(> thi> structure of tbc genus, and must, I think, bo rcterred (o llie above species. 1. Rliodymcnia palmata, i/»». ; Flor. Anfarct. ^*lh; Farlow in Bull. U. S. Xa/. JlKfi. 187(>, o, 30; J)irki(' in Joiivn. Linn. Soc. xv. 201. Swain's 13ay and lloyal Sound, common in tide pools and shallow water ; veiy luxuriant specimens. Also a dwarf form of olivaceous complexion, growing between tidi> marks, dry at low Avatcr, in Swain's Bay, Eaton. (i'\ilkland Islands; Fuegia ; Unalaschka ; Greenland ; Newfoundland ; Scandinavian, British, and Preneh coasts.) 2. Rhodymenia COrallina, Grer.; Elor. Antarct. 475; Farloio in Bull, r. S. yal. Miis. lbH>, 3, 30. On roots of Macroci/sfis, Christmas Harbour, rare, Rookcr ; lioyal Soimd, Kidder. 1. Phyllophora cuneifolia, Ilook. f. and JIarv., Flor. Antarct. d'SG ; Dickie in Journ. Linn. Soc. xv. 201. P. Broditri, Turn. ? Flor. Antarct. I. c. ; Dickie I. c. Christmas Harbour, rare, Hooker. Swain's Bay, rare, in very sheltered water, at 5-10 fathoms, Eaton. (Falkland Islands.) Professor Agardh (loc. cit.) considers with the authors of this species that it is probably a form of P. Brodia;i. 1. Ahnfeltia plicata, Ends.; Dickie in Journ. Linn. Soc. xv. 10, 201. Gigartina plicata, Eook.f. and Earv., Flor. Antarct. 487. Local, between tide marks near Observatory Bay, Eaton ; Christmas Harbour, abundant, Eooker. (Falkland Islands ; temperate and colder seas in the northern hemisphere.) 1. Callophyllis variegata, Bonj. ; Dickie in Joum. Linn. Soc. xv. 40, 201 ; Far low in Bull. U. S. Nat. Mus. 187G, 3, 31. Rhodymenia variegata {in jjai't), Eook.f. and Earv. Flor. Antarct. 475. Christmas Harbour, Eooker. Swain's and Observatory Bays, in sheltered situ- ations, Eaton. Royal Sound, Kidder. (Auckland Islnnds ; New Guinea ; Falk- lands ; Fuegia ; Chili ; Peru ; California.) Mr. Eaton's collection comprises different forms of this very variable species : — from Observatory Bay, on Mytilus in sheltered Avater, a variety with small marginal kalidia; from Swain's Bay var. 3 atro-sanguinea, also a narrow variety (?), torn at the apex and proliferous ; and in addition var. y on roots of Ilacrocystis in very sheltered water. :\rARINE ALGiE.— G. DICKIE. 61 2. Callophyllis dichotoma, Kutz. Ehodomenia dicliotoma, Hook. f. and Harv., Flor. Antarct. 18(5, t. Ixxii. 1. Swains Bay, one specimen only, Eaton. (Marion Island, Moseley ; Campbell Island, Ilooker.) The specimen obtained at Kcrguelcn Island by Mr. Eaton has the structure and kalidia of Callop/n/Uis. The last are not marginal, and therefore it is not a form of C. variegata. [This species was not mentioned in Dr. Dickie's MS., nor in his list in the Linnfcan Society's Journal ; but the name and remarks were noted by him on the sheet containing tlic specimen in the collection, A. E. E,] 3. Callophyllis tenera, J. Ag. ; Dickie in Journ. Linn. Soc. xv. 202. Local in very sheltered Avater, Swain's Bay, Eaton. (South Shctlands.) 1. Kallymenia dentata, S^ihr. (Halymenia), Dickie in Journ. Linn. Soc. XV. 46 [vars. a. and y), 202. Swain's Bay ; and (at 1 fathom, of inferior growth) Observatory Bay, Eaton. ( Cape of Good Hope.) 1. Gigartina Radula, Esp.; Dickie in Journ. Linn. Soc. xv. 46, 202; Farloiv in Bull. U. S. Xal. JIus. 1876, 3, 30. Iridsea Radula, Hook. f. and Harv., Flor. Antarct. 485. Christmas Harbour, Hooker. Swain's and Observatory Bays, abundant on rocks from low- water mark to 1 fathom or more, Eaton. (Cape of Good Hope ; New Zealand ; Auckland and Campbell Islands ; California.) The collection includes various forms of this species. 1. Iridaea capensis, J. Ag. ; Dickie in Journ. Linn. Soc. xv. 46. Kerguelen Island, Moseley. (Cape of Good Hope.) 2. Iridsea laminarioides, Dory. ; Dickie in Journ. Linn. Soc. xv. 46. Kerguelen Island, Moseley. (Auckland Islands, and the S.W. shores of Chili.) Mr. Moseley's collection contains several specimens which belong, I think, to this species. 1. Epymenia variolosa, Kiltz. ; Dickie in Journ. Linn. Soc. xv. 45. Rho- dymenia variolosa, ILook. f. and ILarv., Flor. Antarct. 476, clxxx. ; Dickie I. c. ; Farloio in Bull. U. S. Nat. Mus. 3, 30. Christmas Harbom', Hooker. Swain's Bay, local, Eaton. Royal Sound, Kidder. 1. Halymenia latissima, Hook.f. and Harv., Flor. Antarct. 189, t. Ixxiii., 1, 2 ; Dickie in Journ. Linn. Soc. xv. 202. Swain's and Observatory Bays ; common on rocks at low-water mark, and on Mylilus at 1 fathom, Ealon. (Auckland and Campbell Islands, Hooker.) 1. Ceramium rubmm, Ag. ; Flor. Antarct. 488 ; Dickie in Journ. Linn. Soc. XV. 40, 202. C. rubrum var. secundatum, Lyngb. ; Farloio in Bull. TJ. S. Nat. Mus. 3, 31. 62 nOTANY OF KERGUELEN ISLAND. Chvistmns ITavbour, very abundani, Ilookcr. Conmion in Swain's Eay, Ixoyal Sound, and near Viilcau Cove, Eaton. (General in llio colder seas oL" both hemi- spheres.) 2. Ceramium diaphanum, J. Ag. -, Flor. Autarcf. 488. Christmas Harbour, abundant, Hooker. (Cape oF Good Hope and Atlantie coasts of Europe.) 1. Ptilota Eatoni, lyickio in Jovru. of Bol. V. 51, 1870; ci in Jovrn. Linn. Soc. XV. 202.; raehide iilirornii S-O-poUieari, pinnis oppositis in:e(pialibiis, unA, ma] ore alterml minore, pinnulis pinnaruni longiorum apices versus pecrtinatis, reliquiis bases harum versiis, eaiterisque oniiiibns snbulatis ex serio articulorum niagnorum subsimpliei structis, spboorosporis ad apices pinnularuni subsolitariis, favellis terniinalibus, ramis involucri 4-5 peetinatis eonniventibus. Swain's ]?ay, in 2 to 5 fathoms, on the side and end of a promontory and of an island about two miles from the entrance of the bay, exposed to tlie tide and a slight swell from the outer sea ; very local. Usually parasitic upon BaUia, sometimes attached to iliy^/7»s; Eaton. This species resembles P. Harveyi in the character of the involucre, but differs from it in general habit, and in the structure of the larger and smaller pinnules. It is also dissimilar in colour, being dull j)urplc. Plate V., Fig. iii. : — 1, portion of frond of nut. size ; 2, portion of stem Avith young branch ; 3, apex of fully grown branch ; 1, ditto with sphtcrospores ; 5, sphac- rospores; all much enlarged. 1. Ballia callitricha, Ag. ; Dickie in Jonrn. Linn. Soc. xv. 4G, 202 ; Farloio in Biill. U. S. Nat. Mus. 1876, 3, 31. Ballia Brunonis var. [i Ilombroniana, L[ook. f. and Harv., Flor. Antarct. 488. On Mf/tihis, roots of Ilacrocijstis, and Annelid tubes, from tide pools down to G fathoms ; very common in Christmas Harbour, Swain's Bay, and Eoyal Sound. (Falklands ; Marion Island ; Australia ; Tasmania ; New Zealand ; Auckland Islands.) 1. Callithamnion S\mi\Q, Hook. f. and Harv.^ Flor. Antarct. 4Sd ; Dickie in Jov.rn. Linn. Soc. xv. 202. Chi-istmas Harbour, rare. Hooker. On Miftilus, Annelid tubes, and roots of Macrocijstis, in 1 to 5 fathoms, in Observatory and Swain's Bays ; frequent, Eaton. 2. Callithamnion Ptilota, Hook. f. and Harv., Flor. Antarct. 489, t. clxxxix. 1 ; Farloio. in Dull. U. S. Nat. Mus. 3, 31. Eoyal Sound, Kidder. (Crozets, Hooker.) 8. Callithamnion Rothii, Lyngh. ; Dickie in Journ. Linn. Soc. xv. 203. In tide pools and at the extreme verge of low water, on Mytilus, in Swain's and Observatory Bays, local, Eaton. (Atlantic shores from Greenland to Africa j N.E. shores of the United States.) MARINE ALGJE..—G. DICKIE. 63 I can see no essential difference between Mr. Eaton's specimens and the plant from the northern hemisphere. Tliey agree in habit, and in the arrangement of the tetraspores. The articulations are a little longer than those of British examples. 1. Codium adhaerens, Af/. ; Fwlow in Bull. Z7. S. Nat. Mus. 1876, 3, 31 ; Dickie ill Joiwii. Liiiii. Sac. xv. 203. On rocks in about 2 fathoms in Observatory Bay ; frequent, Eatou (Em'ope ; Cape of Good Hope ; Maviritius ; Ceylon ; Australia ; Friendly and Loo-choo Islands.) 2. Codium tomentosum, Stackh. ; Flor. Antarct. 491 ; Dickie iit, Journ. Linn. Soc. xv., 16. Christmas Harbom*, Hooker. (Tongabu ; Banda Islands ; and the colder seas of both hemispheres). 1. Bryopsis plumosa, Qh-ev. ; Flor. Antarct. 492 ; Dickie in Journ. Linn, Soc. XV. 203. Dwarf or very young specimens on Annelid tubes in 5 fathoms, Observatory Bay, scarce, Eaton. (Greenland ; widely distributed throughovit both the tem- perate zones, and even in some of the warmer seas). 1. Vaucheria Dillwynii, Ag. ; Flor. Antarct. ii. 192. On the ground amongst Pengmn rookeries, Christmas Harbour, Hooker. 1- Ulva latissima, Linn. ; Flor. Antarct. 199 ; Dickie in Journ. Linn. Soc. XV. 17, 203 ; Farloio in Dull. U. S. Nat. Mus., 1876, 3, 31. Christmas Harbom-, very common, Sooker. Royal Soimd; Swain's Bay, Eaton. (Widely distributed in both hemispheres.) 2. Ulva(?) cristata, Rook.f. and Hare; Flor. Antarct. 199. In moist clefts of rocks overhanging Christmas Harbom", growing with Tnjp- othallus \_Palmo(licti/0)i, Ktz.] anastomosans. Hooker. 1. Porphyra laciniata, Ag. ; Flor. Antarct. 500 ; Dickie in Journ. Linn. Soc. XV. 46, 203. Christmas Harbour, very abundant. Hooker. Common in shallow water, Obser- vatory Bay. (Temperate and colder seas of both hemispheres.) 2. Porphyra vulgaris, Ag. ; Flor. Antarct. ii. 500 ; Dickie in Journ. Linn. Soc. XV. 16. Christmas Harbour, very abundant, Hooker. (Widely distributed in the northern and southern hemispheres.) 1. Enteromorpha COmpressa, Link.; Flor. Antarct. 500; Dickie in Journ. Linn. Soc. xv. 16, 203. Very abundant on rocks and in tide-pools ; Christmas Harbom-, Hooker ; Hoyal Soimd and Swain's Bay, Eaton. (Cosmopolitan). 2. Enteromorpha intestinalis. Link. ; Flor. Antarct. 500. Christmas Harbour, Hooker. (Cosmopolitan.) 64| liOTANY OV KKlUa'KLKN ISLAM). 1. Prasiola fluviatilis, Sommcrfeldt, Sujiplcm. Fl. Lap2). 1S2(\ (/csir Dickie in Arctic Muinial, 1S7(!.) i'. Saulori, Mcneghinl MS., Kiitz. Sp. Alf/. 1S|<), ]). 173 ; Dickie ill Joio'ii. Liini. Soc. xv. 203. Oa wet rofks in the bed of a streamlet descendinijf from ti pateli of snow, iilioiH •100 feet above the sea, on the jiyrainidal hill on the W. sich- of Swain's .I5ay. (European .Ups to 9,300 ft. ; Spitsbergen ; streams of W. Greenland.) 1. Cladophora rupestris, I/nm.; Flor. .hihircl. 105; Dickie ill Joitrii. .Linn. Soc. xv. 17. Christmas L[ari)our, on rucks, Jlouhcj'. (tJeiieral between tlie aretie circle and tlie ^^Fediterranean ; only at Kcrguelen Island in the southern hemisphere). -. Cladophora arcta, Ktz. ; Flor. Antarct. 195 ; Dickie in Jotirn. Linn. Soc. XV. 17, -03. C. llookcriana, Kh. Sp. Al(j., p. 118. Very abundant on Jli/tiliis at low-water mark, and in shallow' water. Observatory '^aj, Uafon. (Falkland Islands; Pueyia ; Cierman and N. Atlantic Oceans ; (Jlrcen- laad.) 3. Cladophora simpliciuscula, Hook. f. and llarc. Flor. Ant. \\){\, t. cxeii. 1, 1-3 ; Dickie in Journ. Linn. Soc. xv. 203. One vei'y small specimen, probably very young, too imperfect to be idenlil'icd with absolute certainty, but which I think may be referred to this species, was obtained on an Annelid tube in Observatory Bay at 5 fatlioms, Eaton, (l^'alklands and Puenia, Hooker.) 4. Cladophora flexuosa, Griff ; Dickie in Journ. Linn. Soc. xv. 203. In tide-pools and at 5 fathoms in Observatory and Swain's Bay^s, specimens iVoni the deeper water are poor, Eaton. (Shores of Europe ; Massachusetts Bay.) 1. Rhiozoclonium riparium, Both.; Cladophora riparia, iZbo/.. / and Hare. Flor. Antarct. 195. Christmas Harbour, Hooker. (Cumberland Sound ; British coasts, &c.) 2. RhizOClonium ambiguum, Ktz.; Conferva amhi(jua. Hook. f. and Hai'v. Flor. Antarct. 191, t. cxci. 1. Christmas Hai'bour, in the sea, Hooker. 1. Chaetomorpha Linum, Ktz. ; Conferva Linum, Moth. ; Flor. Antarct. 193. Christmas llarbour, on rocks near high-water mark, Hooker. FRESHWATER ALG.l::.— 1'. F. REINSCH. 65 VI. — Fresh-water Algcc collected by the Bee. A. E. Eaton. AlgcB aquce dulcis Insulcc Kerguelensis, auctorc Paulo Friderico Reinsch. (Cum notulis cle distributione geograpliica a G. Dickie adjectis.) Tanto amj)liorcs iiotitia; do plantis simplicissima structura ac simplicissimis orgaiiis in tcrris diversissimis crc'scentibus, quo evidentius Itxctum agnoscitur notan- duni : has plantas iisdem logibus non subjcctas ex quibus depcndet plantaruui in systomatc suporiorum diffusio in orbi. Specierum plantarum microseopicarum diffusio universa determinatus rationibus peculiaribiis : diffusione aeris meatus in superficie terrae effecta, mobilitate levissima cellularum propagativarum earumque vi vitali diu permanente in statu ipso siccato ; neque minus efi6.citur diffusio rationibus vita) multo simplicioribus accommodatis ad diversissima terra? csela. Materia bujus enumerationis algarum Insula) sclecta a CI. A. Eaton in cxpe- ditione transitus Veneris in bieme 1874-5 continet numeros quatuordecim. Speci- minum inquirendi causa ab Herbario Regio Kewensi mibi traditorum sunt : ampul- lula) tres cum algis aquae dulcis in spiritu viui asservatis, decem folia cum algis siccatis et capsula parvida cum algis siccatis. Omnia qua; ampullula major con- tinuerat : Specimina compluria muscorum aquatilium densissime algis variis [Schi- zoslpho7i Spec, nova et Nostoc Spec, var.) vcstita cum singulis speciminibus Nitellce antarticce et Vaiicha^iarum ca3spitulis parvulis intermixta : mibi dedit mate- riam pro maximam partem bujus enumerationis. In bac ampullula inventa) erant 81 species algarum aqua) dulcis ad i5 genera spectantes ; barum algarum sunt novaj species 28, nova genera 3. Omnes in bac enumeratione recejitte species sunt conjuncta) in proeparatorum collectione Integra, nunc in Herbario Ilogio Kewensi dcposita. Insulse Kerguelensis Specierum algarum aquee dulcis bucusque cognitarum niunerus totus est 106, numerus generum G7. Ab bis sunt Diatomopbycea; - - - 21 species, 13 genera. Pbycocbromopbycea) - - 33 species, 18 genera. Cbloropbyllopbyceae - - - 50 species, 31 genera. Melanopbycea) et Rhodopbycea) - 2 species, 2 genera. Omnes Pamilia) Algarum aqua; dulcis, Ulothrichaceis, Chroolepideis, Confer- vaceis, Sphceropleaceis exceptis, inveniuntur in Insula Kerguelensi. A Cladophoris, Chcetojihoris, Rhizocloniis nulla species est observata. In ordinem systematis ad- ducta riora Algarum aqua; dulcis Insula; hucusque cognita, est ba^c. I 66 BOTANY OF KERGUELEN ISLAND. DLVTOMOPlIYCEiE. 13 genera, 21 species. (2 Spec, novcc, 3 Spoe. inecrt.) PnYCOCnKOJIOPUYCE.Ii, Chroococcaeeto. — 5 s^-enera, G species (al) liis 1 novoo, 1 incevia). Oseillariaeetc. — 3 genera, 3 s])eeies (ab his 2 noA^tc, 1 forma noAa). Nostoehacea]!. — 3 genera, lU speeies (ab his 7 iioviu, 1 forma nova). KiATilaiiaccuD. — 3 genera, 5 species (ab bis 2 novae, 2 i'ormic novic). ScytonemaceoB. — 3 genera, i species (ab his 1 nova, 2 forma; novso). Sii'osipboniaceoe. — 1 genus, 5 species (ab his 4 nova;, 1 forma nova). ClILOllOPlIYLLOrilYCEiE. Pabnellaccce. — 7 genera, 9 speeies (2 novoe format). ProtoeoecacefT. — t genera, 5 species (1 genus novum, 1 species nova). Volvocine?e. — 1 genus, 1 speeies. (Si)ec. nova ?) Desmidieae. — i genera, 5 species (1 nova, 3 formai nova)). Zygnemese. — it genera, 7 species (1 nova, 1 forma nova). Vaucberiacea?. — 3 genera, 6 species (2 novre). Ulvacea). — 1 genus, 1 species. Oedogoniacca;. — 2 genera, 5 species (2 species incertse). Chsetopboraceic. — 7 genera, 10 species. a. Cbsetophorcse. — 4 genera, 6 species (1 genus novum, 5 species novae, 1 forma nova). b. Gongrosii'cte. — 3 genera, 4 species (1 nova). Ehodophyce^. 1 genus, 1 species nova. Melanophyce^. 1 genus novum, 1 species nova. Diatomophyce;e.* 1. Stauroneis goeppertiana, Bleisch ; Itahenhorst Alg. Europ. Nr. 1^2; specimiiia kergueleueusia accuratissimc consentiunt in magnitudine ac forma cellularum cum speciniiiiibus Silesiacis in Collect. Algar. E-abenhorst. communi- catis. Areola transversalis in speciminibus Silesiacis plurimis paulo est angus- * Materiam enumerationis Diatomacearum in ampullula majore reperi in singulis terrosis corpusculis Juris radiculis Nitella antarcticm partim adhaerentibus partim in liquore fluitantibus. FRESHWATER ALGiE.— P. ¥. REINSCII. 67 tior,— Longit. 0,0224 mm. y/' Engl.) Latit. 0,0058 mm. (,, J-^'" Engl.)— (Distkib. Silesia. — Considered by some authorities to be probably a form of S. cUlatata, which is widely distributed in Eiu-opo, G. Dickie.) 2. Stauroneis anceps, Ehrenherg ; forma linearis. Maximo consentiunt speeimina ciun specimiuihus Em'opieis a Erlangen et a Ealaisc leg. Jircbisson, — (Distkib. Europa tota, California, Cayeime.) 3. Stauroneis Phcenicenteron, Ehrenherg. — Longit. 0,0952 mm. {^.}.J" Engl.) — (DiSTRiB. Europa frequens, Ameriea, Persia.) 1. Aclinanthes exilis, KUtzing. Longit. 0,0221 mm. (,,'^"' Engl.) Latit. 0,0028 mm. (y] -'" Engl.) In quoque latere in medio ccllulas (a fronte visas) nodu- lus singulus ; in phirimis speeiminibus Europa^is tantum in intcriore latere. In maguitudine ac forma maximc consentiunt cum speeiminibus e Jura Eranconia, c Himgaria, et e Falaise (Gallia). In Vaucheria; cellulis nidulans. — (Distrib. in Europa vulgaris.) 1. Larirella diaphana, Bleisch. Longit. 0,1008 mm. (^y Engl.) Latit. 0,01-18 mm. (J^ " Engl.) In speeiminibus singulis. — (Distrib. Silesia; an forma S. spleudiclcB in Europa vulgaris, G. Dickie.) 1. CampylodisCUS, species nova ; Beinsch. in. Journ. Linn. Soc. xv. 205 ; magnus, elliptico-ovalis, utroque polo rotimdato-obtuso, costis marginalibus radi- alibus crassis usque ad tertiam partem latitudinis superficieis se pertinentibus in quoque latere 22is — 21is, areolas 21as — 22as rectangulares includentibus, area media Icevi ; frustulse a latere visffi simpliciter spiraliter curvatse areolis 21is — 22is rcctangularibus instructsB. — Diam. longit. 0,132 mm. dV/" Engl.) — Diam. transvers. 0,0066 mm. (Jy Engl.)— Costte in 0,02 mm. (y^V" Engl.) tres. In speeiminibus singulis inter Sckizosiphontis kerguelensis ca^spites. A Cmnpylodiscis frustulis oblongis Campylod. Liirirclla, Ehrenberg, (Abb. Ber- lin. Acad. 18-15, p. 362), mihi tantum ex diagnosi nota, proxima species videtm-. 1. Gomphonema Brebissonii, KUtzing, Spec. Alg. p. 66; Ealfs Brit. Infus. p. 887. Gomph. acuminatum var. Smith. Brit. Died. — Longit. 0,0478 mm. {^^" Engl.) Latit. (in summo) 0,0112 mm. (yl/" Engl.) — Cum speeiminibus o locis divcrsis Germaniae et Austriai et e Ealaise Gallia maxime consentiunt. In Vaiicherice sessiUs et sericece fills. — (Distrib. in Europa vulgaris, an forma G. dichotomi ? G. Dickie.) 1. Amphiprora Spec, nova, Beinsch in Journ. Linn. Soc. xv. 205 ; parva, rectangularis, subtilissime striata, medio parum constricta, utroque polo late truncato-rotundato, lineis interniediis duabus in medio cellula; oequaliter ex- trorsum curvatis aream mediam cruciformem laevem cii-cumcingentibus, nodulo singulo et in quoque latere cellulte in decussi bnearum incluso et in summo utrius- quc linea;. Longit. 0,0333 mm. {,>,,'" Engl.) Latit. 0,0081 mm. {.jls" Engl.) Aniph. Pockorngaua Grunow : dimensionibus duplo majoribus ceUulis ovato- I 2 68 BOTANY (il- KKT^GUELEN ISLAND. oMoap'is voliindatis, nndulo et>n(rali, ^litipJi. minor Vvwv^ovx : (liinonsi(>nil)ii(s nmjdri- bus cidlulis i'lli]>ti('o-(>l)lonuis jiolis rolundatis, striis radialis diirenint. 1. Navicula elliptica, Kilfzhir/, var. Cocconkides; llaheuhorst, jilfjcujl. Eitrop. 1.. ]>. ISO, dimcnsioni^ll-^ iluiilo ininoriliiis. I ioni;il. 0,0201 — 0,022l' mm, (,,',,, — ^V" l^iiJi'l-'* l^^tit- 0,0i;i— 0,0108 mm. (,,',; — , W Engl.) In opero novissimo do Diatomat'ois, Atlas dor niatomaoconkundo, Heft II. tab. YII., lip:. ")"), Xariciihi est dolincata (e Ciorinania) quiv maximo oonsontit in forma, magnitudiuo, ac struct lu-a ecllubp cum plautula KerguoI(>usis. — (DisTiiiu. in Europa froquons, Java, Nova Zolandia.) 2. Navicula dicephala, Wtrenberg. Longit. 0,0218 jnm. («V" Engl.) Spccimina ad formam pertinent sunimis capituliformibus distinct ins disjunctis. — (DiSTiiiiJ. Europa). 3. Navicula minutissima, Gnmoir. ]']. minimis cellularis linearl-oblongis, nodulo medio et linoa loiigitudiiiali distineta, indistinetc transversalitcv striatis. — Longit. 0,0112 mm. (, ,i,,"' Engl.) Latit. 0,0028 mm. {^g-" Engl.) Hasc Navicula in speciminibiis numerosis in massa ex Diatomaceis exstituta ; non sine dubio Navicula kerguelensis ad bane speciem est posito. — (Distuib. Europa oriental is.) ■4. NaviculSB Spec. Cellnlis lanceolatis apieibus capituliformibus porrcctis, nodulo centrali ct liuea media iudistincta, marginibus distincte striatis striis ad mediam non pertinentibus. Longit. 0,0278 mm. (yV" Engl.) Latit. 0,0050 mm. (^b"' Engl.) 1. Amphora gracilis, Uhrenherg. Longit. 0,0357 mm. (J/" Engl.) Latit. 0,0190 mm. {^h) Engl.) CoUuUr ovato-ovales a])ieibus truneatis, nodulis circa tertiam partem diamctri transversalis a niargine distantibus, areola media subtiliter longitudinalitcr striate. Specimina a Ealaise (Gallia) et c Germania consentiunt in magnitudine ac forma cellularum. Amphorcv gracilis in Scbmidt Atlas d. Diato- maceenkunde, vii. fasc. tab. 20, fig. 101, cellukc, qute ad Amphoram angustam Ehrenberg pertinent, graeiliorcs et apieibus attenuatis. — (Distrib. Europa, Mexico, ct in Kurdistania fossilis.) 1. Pinnularia viridula. Smith Brit. Diatom. 57, tab. xviii. fig. 179; Hahenh. Eur. Aty. i. p. 21 i. Eorma apieibus subito attenuatis, striis transversa- libus lineam mediam attingcntibus distinctissimis. Longit. 0,0357mm. (g'g'" Engl.) Latit. 0,0123mm. {j\-^" Engl.) Plantulse Falaise (A. de Brebisson, leg.) et plantula^ Erlangcnsis in magnitudine consentiunt, sed differunt polis sensim attenuatis striis subtilioribus. (Distuib. Europa, America.) 2. Pinnularia Viridis, Ehrenberg. Longit. 0,0018mm. {.^.,"' Engl.) Latit. OjOlomm. (rr.f" Engl.) Specimina paulo minora speciminibus formsc ajmd Erlau- gam communis. (Distkib. Europa, America, Persia.) 3. Pinnularise species ; cellulis ovato-ellipticis, polis rotundatis, nodulo cen- FRESHWATER ALGiE.— P. F. REINSCir. 69 trail firmo, strii'^ transvcrsalibus distinctis lincatn niotliam attingcntibus, Longit. 0,0168— 0,01 ncmm. (yi.T — rlr." Engl-) Latit. 0,0081mm. (vrJV" Engl.) •1. Pinnularise species; cellulis late ovato-cllipticis, polis siihito angiistatis apicibus rotundatis, nodulo centrali firmo striis transvcrsalibus distinctis lineam mediam attingontibus. Longit. 0,0221.mm. (,,V"' Engl.) Latit. 0,0112mm. (jl-.j'" Engl.) 1. Synedra Vaucherige, KUtzing. ; forma apicibus obtusis. Longit. 0,01i8mm. (,V" Engl.) Latit. 0,0028— 0,00o6mm. {.V^—^]^" Engl.) Individua brcvitcr stipitata basi radiatim conjuncta in Schizosiphonte kergnc- lensi, et in Vancherice collulis. (Distrib. in Europa frequens.) 1. Eunotia pectinalis, JDllhoyn. Longit. 0,10Gmm. (^"' Engl.). Latit. 0,0393mm. (5V" Engl.) (Distrib. in Europa vvdgaris.) 1. Denticula thermalis, Kutzing., var. minor. Longit. 0,0168mm. (yi^'" Engl.) Latit. 0,0056mm. (yfs'' Engl.) Cellultc in quoquo latere 9is nodulis in- structa?. (Distrib. Aquis calidis Gallia?, Ilungariaj, ct Italitio.) 1. Cymbella gastroid.es, Ehrenberg. Longit. 0,0121mm. {r]~" Engl.) Latit. 0,013mm. {^~" Engl.) Specimiua minora speciminibus e locis variis Germaniaa. (Distrib. Per totani Europam.) Phycochromophtce^. 1. ChroOCOCCUS macrocOCCUS, Babenh., Alg. Fl. Eur. i. p. 33. Proto- coccus macrococcus, KiUz., Tab. Fhyc. i. tab. 2. Eorma cytiodermatc tcnuiore, cytioplasmate grossius granuloso. Formam typicam tantummodo observavi in familia singula tricellulari inter algas uniccllulares Hormoslphonti adhserentes. Diam. cellular, indivis. 0,047Smm. (4V'" Engl.) Eormam in famUiis singulis bi- ct tricellularibus in massis minoribus algarum variarum foliis muscorum adhserentibus observavi. Hasc forma pertinet ad formam Chr. macrococci=C/iroococcns aureus, Kiitz.,Tab. Phyc. ii. tab. 2, Chrooc. inacrococc. Pialicnb., var. 3. ; c^i^^ioplasmatis cellularum colorem nunc pallide-flavum in statu vivcnte cellulip fuisso aurco-lutcum non dubito. (Distrib. Europa tota.) 1. Microcystis Olivacea, Kntz., Tab. FJnjc. i. tab. 9. Diam. cellular. 0,0011mm. (3-^y'" Engl.) Diam. famil. 0,066— 0,0896mm. (3V— 2^3'" Engl.) Observavi tantum familias singulas inter alias algas uniccllulares muscis adluc- rentes. In colore quoquc obscm'c olivaceo cum specim. Europseis conscntit. (Distrib. Germania.) 2. Microcystis parasitica, Kutz., Tab. Thyc. i. tab. 9, fig. i. In pbyseumatum structura, magnitudinc ac colore cellularum cum speciminibus Europseis et cum icone Klitzingiana maximc conscntiens. Pbyscumata minora ct majoi-a, partim cobserentia in Nitelke antarcticce cellulis affixa. 70 BOTANY OF ICERGUELEN ISLAND. Diam. collular. 0,003mni. (- ', ,'" Eng'l.) DImui. pliyscMimalum 0,0278— 0,11 2mm. (.\ — ,'„ ' Enu'l.) (Disi'iMu. Europa.) 1. Glocethece iuvoluta, Ilcinsc/i. in Jonm. Linn. Soc. xv. 20G; thallo non limitato intor al£::as minores dispoVso ; collulis oblongo-cycliiidricis utroqm' poUi rotumlatis, diamctro transvcrsali duplo longioribus, singulis nut geminis, togumcntis biuis crassissimis distincto plurilamcllosis circumvclatis, cytioplasraate pallidc- an-uirinoo subtilitcr granuloso, plcvum(|no granulo singulo majorc insti-urto. Loiigit. I'ollular. (c. imlum. oxter.) 0,0278 — 0,03^3mm. (.'. — ,.'.,'" Engl.) Longii. oi>llular. (c. iudum. intor.) 0,0218— 0,O278nu-a. (s'.f— 7'.,'" l^ngl.) Longit. eolliil. (sine indum.) 0,0112— 0,013mm. (, Jy,— ily'" Engl.) Inter algas minores musois aqnatiois insidcntes. Hoec Glocethece aliqua simi- litudine conscntit cum G. monococca, Eabonli, El. Eur. i. p. Vil^Gloascajysa ■iiionococca, Kiitz., Tab. Phyc. i. tab. 23, itidcm ropi'rta pl(>vum([uo in statu uni- ct biccUulari; quic species uova attamen est divcrsa indumcutistructura valdo, distincte lameUosa et cellulis lougioribus a Glooeth. monococca, cujus intcgiimcntum semper est structura bomogcnea. 2. Glooecapsa magna, Kiitz., Tab. Thjc. i. tab. 22, fig. 7. CelliilcV! siuguloe et geminje spba^rica; colore intensive ajrugineo-viridi donsissimc positie, pliysoumata spbterica plerumque cobscrentia indumento colorato velata formantes. Cellular, diam. 0,028 — 0,0011mm. (jj j — ^\~" Engl.) Diam. pliy.scu- matis 0,033— 0,()5mm. {^t^i-i" Engl.) Inter Scijtonema castanemn, Kiitz., iu massis pamilis Ilormosiphonti coriaceo adhterentibus " prope Viilcan Cove." Non est mibi certissimc, banc plantulam perti- nere ad Gl. magnam propter speciminum observatorum minimum numerum. (Dis- TRIB. Europa, Grcenlandia.) 1. Anacystis marginata, McnegMni. Eamiiia) singiilse quarum diameter Ojl7inm. (J'" Engl.), inter Algarum massas minores natantes. (Distrib. Europa.) 1. Leptothrix hyalina, Beinsch in Joum. Linn. Soc. XV. 206 ; aggrcgata, csespitulos disperses et radicantes muscis aqtxaticis affixas formans, tricbomatibus byalinis, vaginis distinctissimis crassis liyalinis, supcrnc saepissime vaeuis ct in smnmo apcrtis, cellulis teiiuissimis diametro aequalibus, cytioplasmatc piinctulato. Diam. tricbomat. 0,0028 — 0,0011mm. (y^^^" — avW" Engl.) Csespitulorum altitude, 0,084— 0,112mm. (^— jV'" Engl.) In foliis muscorum. Leptothrix radians, Kiitz., Tab. Phyc. ii. tab. 59, proxima species distinguitm" vasjinis multo augustioribus cellulis crassioribus. 1. Lyngbya major, Kiitz., Tab. Phyc. i. tab. dO,Jig. 8; var. kerguelenensis ; tricbomatibus inter alias algas dispersis subrectis, cellulis intensive ajrugineis sub till tor distincte granulatis, diametro 8plo — lOplo brevioribus, vaginis amplis byalinis (interdum fuscescentibus) distincte 8-12 — lamellosis, cellulis interstialibus FRESHWATER ALGM.—P. F. REINSCH. 71 iiullis. Diam. tricliomat. (c. vagin.), 0,0361— 0,0M8mm. (^^ — rf"'EngL) Vaginar. crassitudo, 0,0081— 0,0112uim. (.^i.—j-V^'" Engl.) Diam. cellular. 0,0196— 0,0221mm. (xj o — oV Engl.) Tricliomatum longitudo 8— 15mm. Inter alias algas natantes et affixas in dispersis tricliomatibus. (Distrib, I/, majoris in Europa oricntali.) Hujus formse cellulte celliilis iuterstitialibus uon interruptae cylindrum con- tinmmi formaut, diamotro trichou^atum apicem versus non decrescente, ultima cellula late rotuudata, vaginic in tricliomatum summis utplurimum sunt apertse ct cellulis vacucG. In fere omnibus tricliomatibus a me visis Microthmmiu novi elegautis plantulas observavi, qua) Lyiighi/ie sunt affixte radiculis eontortis ssepe eircum circa triclioma jirocurrentibus. Speciminum formae typicoe ex mari Adriatico tricbomata paulo sunt crassiora, sajpissime occurrunt celluliB interstitiales colore rubro-lutescente distincta) ceteris cellulis tricbomatis. 1. Limnactis minutula, Kiltz., Tab. Phyc. a. tab. 63, fig. 1 ; var. tricbo- matibus rectis sensini attenuatis margine crenulatis, cellulis distincte separatis diametro triplo-quadruplo brevioribus, cytioplasmate dense grossius granuloso, cellulis summis diametro usque quadi-uplo longioribus, byalinis distinctis, vaginis byalinis, cellulis perdm-autibus spbnericis cellularum diametro sequalibus. Diam. tricbo- matum, 0,0056— 0,0075mm. ( .i^— ^f^'" Engl.) Ill ScMzosiphontis kerguelensis tricliomatibus in ctcspituHs parvulis u.sque 0,28mm. {]'" Engl.) latis. Distrib. Gallia, Germania, Dania, Suecia, Britannia. 1. Dasyactis Kunzeana, KUtz. Diam. tricbomat. 0,0056— 0,006Smm. ((»TANY l)F KKRGUELKN ISLAND. s minilms l']urop.Tis in thallo C/Kc/opfiora/'inuci Aonfoc/iidia nidiihmtihus. Tn sing'ulis trichomatibus intor Tohipollirichin Jlacckla-, Kiitz., (':i>s])i(ul()s nidulantihiis. In singulis trifliomatibus ccllula' inleriorcs In-cvioros et iiulistim-tiiis ilisjuuctio. (Dis- TiiiB. Gcrmania.) 0. MastigOthrix miuuta, llclnsch in Jonnt. Linn. Soc. XV. 207 ; Iriclio- matihus ilistiiicto articulatis apicihus rcctis, articiilis inlerioi-ibus dimidio latiiudiiic birvioribus (ot paulo maL;-is), s])oris ]iov(hirantil)ns obovalibiis iisqiu^ siibspba-ritMs diaiurtro dimiilio trichomatis latitiulinis a'quantc. Latit. tricliomaluiii (in basi) 0,0081— 0,()0i)7mm. (.jj.^— .5?j,v"' Engl.) Diam. spono pordui'., 0,()011nim. (,,l/" Engl.) Inter algas miuores {Leptothrix, Coleocluete) in foliis musooruin aquatilium insidentos. A Jlastiff. tentginea, Kiitz., dimonsionibus dn])lo magis minoriljus distincta species. In trichomatibus singulis vaginiu inlima pars paulo incrassala et lamellosa, scd ccllula pevdurans non inclusa a lamcllis. 1. Hydrocoleum Eatoni, Bcinsch in Joum. lAnn. Soc. XV. 207; fasciciiHs liberis inter alias algas dispersis usque ad ISmins. longis in sununis sensiin aitenu- atis, tricbomatibus olivaceo-vividibus (a latere visis), Sis — 12is cousociatis et l(>viter contortis subtilissimc distinctc articulatis, cellulis distinctis omnibus liomogeneis, diametro qiuntuplo breviorilms, cytioplasmate dense punctulato, vaginis aclirois membranaccis duris subtiliter lamcUosis, tricliomatum fasciculi latitudiuis dimidio crassis. Diam. fasciculi (in medio parte) 0,056 — 0,08Gmm. (^ig — ^V" Engl.) ; (in suuimis) 0,0221— 0,0333mm. (.'^ — gV'" Engl.) Diam. tricliomatum 0,0011— 0,005Gmm. {sh — sW Engl.) Vagin. crassitudo 0,0028mm. (,!/" Engl.) Inter muscos aquaticas et aliis algis {Vcmcheria, Schizosiphon) immixtum. Hoc Hydrocoleum consentit cum U. hclcctico, Niigeli, in f'asciculorum dispo- sitione, scd difTert dimensionibus i'asciculorum quintuplo magis majoribus, triclio- matum di verso colore et cellulis brevioribus. Tab. IV. Eig. i. — 1, fasciculi media j)ars (-',-) ; — 2, fasciculi summa pars {^^-^). NostOC hyd.rocoleoid.es, Bcinsch ill Joiiru. Linn. Soc. xv. 208 ; subtil- issimmu, pliyseumatc in modo Hydrocolcorum teretiformi prolongato perider- mate distincto hyalino cuicto ex trichomatibus et rectis et paulo contortis (5is — lOis) fasciculatim conjunctis formato, trichomatibus pallidc serugineis parallelis leviter contortis vaginulis hyalinis velatis, cellulis vegetativis post divisionem diametro paulo longioribus, cytioplasmate punctulato, cellulis perdurantibns ceteris paulo majoribus sj)ha3ricis in trichomatibus sparsis. Diam. trichomat. 0,0022 — 0,0028mm. {^I^^^—tIi'" Engl.) Diam. fasciculi (in media parte) 0,0112— 0,221mm. (t-L-,V-"' Engl.) Inter Tolijpothrix NcegeUi, Kutz., et in massa parvula Diatomacearum foliis muscorum aquatilium et Nitcllcv adhairente. Hsec plantula paradoxa secundum structuram et physeumatis et trichomatum FRESHWATER ALG.E. — P. F. REINSCH. 73 JVosioclticI/s generis bonam spceiem se ostendit. Trichomata integumento communi distinctissime clausa vix sunt disccrnonda a tricliomatibus specierum singularum. Species imica NostochkUs generis hucusque cognita physeumate filamentosa, a ceteris speciebus physeumate sive piano sive sphoerico sat distincta. Tab. IV. Fig. iv. — 1, physeumatis pars media (-y-) ;— 2, pbyscumatis summa pars, trichoma singulum usque in apiceni cxcurrens (— f-). 2. NostOC polysaccum, Seimck in Jonrn. Linn. Soc. xv. 208 ; physeumate coriaceo irregularitcr sph;erico et subreniformi colore subaureo-fasco magnitudine seminis sinapcos ad Pisi sativi, intus loculamentoso ac dissepimentis coloratis lamel- losis et radialiter ct transversaliter positis percurso, i^cridermatc firnio coriaceo fuscescente, tricliomatibus centralibus paulo flexuosis, cellulis sphsericis colore pallide olivaceo, cytiodermate distincte dupKciter striato, cellulis perdurantibus sphnericis ceteris cellulis paulo latioribus. Diam. cellular. 0,0041mm. (o^y"' Engl.) Diam. cellular, perdurant. 0,005Gnuii. (-j'^'" Engl.) Diam. physeumatis, 2,5 — 3mm. Forma (an status peculiaris evolutionis ?). Physeumate ex trichomatibus brevi- oribus vaginis amplis hyalinis homogenis (in modo Sormosiphontis) inclusis lax- issime cohan*entibus exstitiito. Tab. IV. Fig. i. — 1, Physeumatis sectionis transversalis pars usque ad periphe- riam physeumatis se pertiuens, vesiculaj trichomata includentes, radialiter dispositaj, parietes vesicularum subcoloratcB (Y )• 2, Format physeumatis pars peripherioe, sectio transversalis ; physeuma ex vesiculis numerosissimis, trichomatibus singulis inclusis formatum, trichomata breviora in modo SonnosipJiontis indumento crasso subhyalino inclusa (^,'). 3, Specimina plantulae (in spiritu vini asservatae) magnitudine naturali. 3. NostOC polysporum, Eeinsch. in Jonrn. Linn. Soc. xv. 208; physeumate spba;rico magnitudine pisi niinoris, indumento crasso hyalino distinctissime pluri- lamelloso velato, trichomatibus laxius positis subcontortis pallide serugiucis, cellulis spha,>ricis arctissime conjunctis, post divisionem transverse ellipticis, cellulis perdu- rantibus numerosissimis spha?ricis ceteris cellulis duplo majoribus cytiodermate crasso. Diam. cellular. 0,0028mm. {jl~f"' Engl.) Diam. cellular, perdurantium 0,0041 - 0,0058mm. (, f ,— ^^" Engl.) . Diam. physeumatis, 3—4 mm. Inter alias algas fluitans (in paucis speciminibus observatum). A persimilibus : N. gymnosphcsricum et N. ccsruleum, Kiitzing, Tab. Phycol. ii. tab. 3, fig. 3, 4, differt indumento pluiilamelloso, cellulis perdurantibus numerosi- oribus. 4. NostOC species, e minoribus, physeumate irregularitcr polyedrico, textura cartilaginea, colore rubro-fusco, tricliomatibus contortis, cellulis subsphicricis arc- tissime adjaccntibus, cellulis perdurantibus spluei'icis ceteris cellulis paululo majo- ribus cytiodermate crasso disfcincto. Diam. cellular. 0,003 — 0,0041mm. (jJ^ — -^"' Engl.) Diam. physeumatis 1,8mm. K 74 BOTANY OF KKRGUELEN ISLAND. inter Zi/f/noiiaiii. In :i iVrshwaler pool, Swain's Bay. In toxtuva, forma irroi!;ulari phvseumatis minus in tricliomatum forma N. ediiU Berkeley ])ersimilo. SpccinKMi unicum obscrvatum specicm accuratius constituendani milii non permit tit. 5. NostOC paludosum, Kid.:., Tab. Phyc. ii. lab. H, fif/. 2. Spocimina siumila observata iusidentia plant ulis Bulbuc/ia'fcis foliis muscorum insiclentibus et in triehomatiun erassitudine ct in cellularum forma raaxime conscntivmt cum speci- minibus Gerraanicis et eum iconc Kiitziugiana. Diam. cellular, trichomat. 0,0011 — 0,0018mm. Wcnr— TWff'" En-l.) Ab omnibus Kostochidis spcciebus pagnitis species cellulis minimis. J)istrib. In Europa vultraris. G. NostOC leptonema, Meinsch in Joui'ii. Linn. Soc. xv. 209 ; pbyseuma- tibus usque somiui sinapeos lequalibus spbairicis paulo clasticis arctissimc con- junctis colucrentibus, indumento exteriore subcrasso hyalino liomogcnco, triclio- matibus prolongatis nuiltipliciter contortis laxius (in majoribus) et densius (in minoribus) intricatis, cellulis oblongis polls attcnuatis laxe se adtingcutibus ; cellulis perduvantibus splurricis usque subovalibus sparsis ceteris cellulis duple paulo magis majorilnis. Diam. cellular. 0,0015— 0,0021.mm. {-^rf^—gW" Engl.) Diam. cellular, perdur. 0,00oGmm. (ttts'" Engl.) Diam. pliyscumatis 0,2— 1,5mm. In muscorum caulibus et foliis pliyseumatibus cohajrentibus, partim corpora uvseforniia formans. A Nostochidihns physcumate spbierico Nostoc aureum, Kiitz., Tab. Phyc. ii. tab. 1, fi"-. i (planta marina) proximum in magnitudine et tcxtura physeumatis ac erassitudine tricliomatum ; lioc Nostoc diffcrt trichomatibus brevissimis valde cur- vatis cellulis perdiu-antibus minoribus. Inveniuntm' interdum muscorum foliis iusidentia corpora ex pareneliymatice conjunctis physeumatibus varia magnitudine formata. Forma: Crystallophorum. Physeumate corporibus crystallisatis subsphgericis inclusis ex crvstallis (Ferri oxydati ?) radialiter dispositis formatis. Diam. corpor. crystallisat. 0,0224— 0,05mm. (^L__y" Engl.) 1. Anabaina COnfervoides, Bemsch in Joum. Linn. Soc. XV. 209; e subtilioribus stratum tenue formans, trichomatibus prolongatis rectissimis parallelis in muco commvmi nidulantibus, cellulis distinctissimis rectangularibus usque sub- quadi-aticis, spatiis iuterloculatis angustioribus distinctis sejunctis, diametro trans- versali paulo longioribus (usque duplo), cytioplasmate subtiliter granuloso colore pallide a;rugineo ; cellulis pei-dm-antibus ellipticis ceteris cellulis paulo latioribus et longioribusr Diam. ceUular. 0,0022— 0,0028mm. ( [iitulos t-ourcrtoi; r;i(li;ilit(M' dispositos iisqiio (i mins altos in miiscis aquatiois aflixos fonnans, trieliomatibus rndianlihiis o I)asi ropcliio (liclioionio- ramosissiniis in sumiuis fastig-iatis, pseudornmulis iillimis t'orymlmsis i'asciculiitis apii'iluis imnlalini suiiaii^'iistatis, vag'iiiis psoudiir;uniili)niiii ullinioi'uii! I'liscis inlc- geiTimis irlullanim diaiuotro suba'qualiter crassis, vagiiiis triidioinatum inrerioris partis crassiorilms donso sulitilitiT laniollosis, collnlis omnibus uupialibns distinctis, dianietro subioqualibus apiconi pscudoraniuloruni versus non dccresceutibus, cytio- plasiuatc colore pallide olivaeeo-viridi granulis majoribus distinctis dense repleto, cellulis perdurantibus singulis aut compluribus basilaribus subsplucricis diamelro cellularuni ivqualibus. Diani. tricdionial. (in diversis locis raensuratus) 0,01G8 — 0,0333 mm. (, ig — ,5'..'" Engl.) Diani. pscudoramulonuu uKimorum 0,013 — 0,0108 "nn. (,),,-, i,>Eng-l.) liab. in muscis aquatieis eaules densissime })elicuUL' i'ormiter inducens. Haec species elegant issima in ciespitiilis muscis in caliculo vitreo inclusis co- piosissima est reperta, in ])rimis specimiuis majoris eaules densissime erant obtecti. Cellularum funiculus singulis locis baud raro et sim})]ieitcr et dupliciter contortus, qute partes tricboraatum paulo sunt incrassatie ; basin tricbomatum versus cellu- larum funiculi ])lerumque sunt contorti ; tricbomatvun iniimjc partis vaginae pluvi- lamcllosic et tricbomatum intima pars cuneiformiter angustata in lilum siugulum producta. 1. Tolypothrix flaccida, 7v7;/c. Tab. Fhi/c. ii. tab. 32, f(/. 2. Torma cellulis diametro transversali aequalibus et i^aulo longioribus. Diam. tricbom. 0,0056—0,0081 mm. (^|s— ^"' Engl.) In cpespitulis parvulis in foliis muscorum aquaticorum insidcns. (Distrib. 2'. Jlaccidw, Britannia, Gallia, Germauia, Ilelvetia.) U.ycc formae sunt peculiares ut in forma typica, cellulie perdurantes complures postpositse, celluke compku'es funiculi tricbomatum saepissime interstitiis byalinis sunt disjunctic et tricbomatum summa pars cellulis vacua. 2. Tolypotlirix Nsegelii, Kiitz. Haec Tohjpuihrix a forma typica est distincta trieliomatibus paulo tenuioribus, pseudoramulis crebrioribus, quiB sunt brcvissimae in singulis tricboniatiluis ; summa pars cellulse perdurantis singular in pseudoramulorum basi nomiunquani est trun- cata. Inter ScMzosiphontis csespites et affixi et fluctuantes caespituli. (Distrib. T. li'ceyelii, Helvetia.) 1. Schizothrix hyalina, KiUz. Spec. Alg. Tab. Fhyc. ii. tab. i^O,/iff. 1. Var. EAiiosissiiiA, Reinsch in Jov.rn. Linn. Soc. xv. 211 ; tricbomatibus Schizo- siphonti insidentibus subtilissimis, funiculis et submoniliformibus et subcylindraceis pallide serugineis, vaginulis amplis byalinis cinctis ; pseudoramulis numerosis erectis flagelliforme attenuatis. Diam. tricbom. (cum vaginulis) 0,0022 — 0,0056 mm. FRESHWATER ALGJE.—V. V. REINSCK. 77 (oio— Bf^"' Engl.) Diam. tricliom. intern. 0,0011 mm. (o'so"' Engl.) Altitvido plantulae 0,8 mm. (i'" Engl.) In Schizoslphontis kerguelensis trichomatibus in csespitulis dispcvsis. (Distrib. S. hyalince, Montibus Europte.) Ilanc formam peculiarom, verisimile spccicm propviam, tantummodo in paucis scd bonis speciminibus ol)scvvavi, qii;e erant apta ad constitucndum genus. Est similitudo maxima cum Schizothr. hi/aliaa iu tricbomatum et A'aginarum crassi- tudine et cellularum funiculi forma, quamquam incrementi modus et loci natalis est diversissimus. 1. Sirosiphon vermicular is, Beinsch iu Joum. Linn. Soc. XV. 211 ; e minimis, ca^spitulis parvulis tricbomatil)us subrectis summis attenuatis procum- bentibus intertextis, plus minusve ramosis, ramulis alternantibus apiccm versus sensim attenuatis ramulis simimis diametri tricliomatis primarii dimidio tenuioribus, tricliomatiuxL cellulis uniseriatis arctissime conjuuctis, cytiodermate sulicrasso flrmo fuscescente, cytioplasmate subtilissime granuloso, ramulorum cellulis apicem ram.u- lorum versus angulosis confervoideis, vaginis (tricliomat. primarior.) tenuioribus (vix cellular, diametri octavam partem) simpliciter striatis cellulas arctissime iuclu- dentibus ; cellulis interstitialibus nullis. Diam. tricliom. priniar. 0,0112 mm. (yirj'" Engl.) Diam. ramulorum 0,0056 mm. {^l-^ Engl.) In csespitulis singulis inter alios Sirosiphontes Hormosiphonti coriaceo prope Vulcan Cove adhaerentes. Ab omnibus Sirosipliontibiis bucvisque cognitis species minutissima. Sirosiphon in ramvxlorum cellulis diversis a cellulis tricliomatis primarii. Sirosiphonti sylvestri, Itzigsobn. proximus sed sat distinctus trichomatibus tenuiorilius cellulis cytio- dermate tenuiore indistincte articulatis. 2. Sirosiphon pulvinatus, KiUz. ,- var. cellulis tricbomatis primarii cytio- dermate crassissimo colorato absque ordine biseriatis cellularum ramulorum uniseri- atis aut absque ordine biseriatis. Diam. cellular. 0,0056 — 0,0068 mm. {^i-g — 3-^ Engl.) Diam. cellular, c. vagina 0,013 mm. (yJ-y'" Engl.) Tricliomat. crassit. 0,0221 — 0,0306 mm. {^\ — ^^"' Engl.) Tricbomata perpauca dispersa. Forsan pro- pria species. (Distrib. S. pnlvinati, Europa, Anieric. boreal.) Var. ; trichomatibus irregularitcr ramosis, ramulis apice obtusis numcrosis sub- contortis, cellulis omnibus sequalibus subovalibus, cytiodermate tenuiore hyaline decolorato, iiTegulariter liiseriatis. Dimensionibus iisdem ju'aiced. 3. Sirosiphon species nova, Beinsch in Joum. Linn. Soc. xv. 212; e minoribus, trichomatibus singulis inter alias algas dispersis, irregularitcr j)innato- ramosis, raniis Inlateralibus trichomati primario iioqualiter formatis et aequaliter crassis, summis non attenuatis, cellulis subspbtericis spatiis hyalinis disjunctis, cytiodermate tenui homogcnco subhyallno, cytioplasmate subhomogeueo pallide- aerugineo, vaginis crassis hyalinis subhomogeneis decoloratis, cellulis intcrstiti- 7S BOTANY l)F KKRGU1-:LKN ISLAND. alibus? Diani. c-ollulnr. 0,001.1—0,0056 mm. (,,l^ — ^1^"' Engl.) Triehomat. crassit. 0,0112—0,013 mm. (,io— r.vr"' l^'^gl-) In trichomatibus singulis inter alias Sirosip/ioiUcs ct intor Scytoiiemam casta- nenm inter Hormosiphoiiteni coriaccnm (near Vulcan Cove). S. velntiniis ct ^S". hormoides Kiitzing trichomatibus crassioribus fasoiculato- ramosis et dicbotomc ramosis distiucti. S. paiiniformis, Kiitz., distinguitur ramis elougatis tricbomatc primario tenuioribus et ccllulis interstitialibus. k Sirosiplion kerguelensis, llehisch. in Journ. Linn. Soc. XV. 212 ; trichomatibus ramosissimis, tricbomate pi'imario procumbcntc ramis irrcgulariter i*amosis ramulis ultimis apiccm versus ooqualiter latis, cellulis tricbomatis piimaiii ac ramulorum ovalibus ustjuo irrcgulariter spbajricis in seriem simpliccm dispositis, intcrvallis byalinis nsque cellularum lougitudini a;quantibus disjunctis, articulis tubuliformibus angustissimis (lacnnis tubulil'ormibus in muco vagina)) conjunctis, cytioplasmate subbomogeneo dilute a^rugineo, cytiodermatc subtili dccolorato (cellu- larum trichom. primarii crassiore fuscescentc), cellulis summis ramulorum coha3- I'cntibus lyngbyaceis, vaginis crassis byalinis dccoloratis subbomogoneis (vaginis tricbom. primarii sublamellosis aureis). Diam. tricbom. primar. 0,0278 — 0,0333mm. (-.1, — g'.^ Engl.) Diam. ramulorum 0,021-8 mm. (^V" Engl.) Diam. cellular. 0,013 mm. (yiy'" Engl.) In tricbomatibus singulis inter alias Siroslphontes. Cum prsecedente. Hie Sirosiplion primo pro forraam propriam Sirosiphontis ocellati babitus, cm est pereimilis m tricbomatis ramificatione et crassitudinc, sed propter propriam de ceteris Sirosiphontibus discedentem structuram tricbomatis propriam speciem se offert. Tab. IV. Eig. ^^. — 1, tricbomatis pars summa ifS^ ; — 2, tricbomatis pars maxime aucta, X20. In singulis speciminibus observavi Sirosiphontem sequentem quern bujus Si7'0- siphontis varietatem puto. Triebomata ramosa ramis subintegris adscendentibns, cellulis ovalibus usque subspbrericis, intervallis byalinis disjunctis. Articuli tu.buli- formes cellulas singulas conjungentes plurimum desunt. 5. Sirosiphon Oliver!, -BeM2scA in Jonm. Linn. Soc. XV. 213; csespitulis parviibs, tricbomatibus adscendentibns prolongatis subramosis, ramulis singiilis (et raro ramulis compluribns bre\aoribus approximatis) et leviter contortis, e serie simplice cellularum formatis, cellulis ovalibus diametro dimidio brevioribus (et panic magis et minus), cytiodermate firmo crasso extrorsnm fuscescente, cytioplas- mate subbomogeneo obsciu'e-semgineo, vagina membranacea simplici subtenui, cel- lulis interstitialibus nullis. Diam. tricbomatum (cum vaginis) 0,0196 — 0,0224 mm. In csespitnlis parvulis inter Sormosiphontem coriacewn cum csespitulis Scytone- matis castanei intermixtis ; cum preeced. FRESHWATER ALG^.— P. F. REINSCH. 79 5. velittino, Kiitz., ct S. Jioi'Mo/de, Kiitz., spceiebus })roximis in cellularum forma ac dispositione cliffert triclioiuatil)us sul)iutcgris, vagiuis tenuioribus. Tab. IV. Fig. ii. — 1, trichomatis summa pars (ff--) ; — 2, tricliomatis pars maxinie aucta, vagina dupliciter striata, cellularum cytioderma dupliciter striatum, cellula singula longitudinaliter divisa (\t~). 6. Sirosiphon secundatus, KUtziug, Tab. Fhycol. ii. tab. S1,fiff. 1; forma tricliomato primario partim incrassato, ramis prolongatis apice incrassato ; ccllulis parvulis tricliomatis primarii numerosis absque ordine dispositis, ccllulis ramorum uni- aut irrcgidariter l)iseriatis cytiodermatibus crassis confluentibus. Diam. tri- chomatis primarii 0,0333—0,0393 mm. (^Jg — ^"' Engl.) Diam. ramorum 0,022i— 0,0278 mm. {^\ — ^"' Engl.) Diam. cellular. 0,0056 mm. (^"' Engl.) In specimine singulo obscrvato, inter alias Sirosiphontes. Cum prsecedente. (DisTRiB. Europa.) Chlorophyllophtce^ . 1. Gloeocystis vesiculosa, Nmjeli. Cellulae indivisae usque ad 0,011 mm. (.j-^"' Engl.) diam. ; familise bicellulares 0,0058 mm. {■^^" Engl.) diam. Inter algas unicellulares adhserentes foliis muscorum. (Distrib. Gei*mania, Helvetia.) 1. Palmella mucosa, KiXtz. Tab. Phyc. i., tab. 16, fig. 7 ; cellul. diam. 0,0056—0,0112 mm. (^fg — rl?'" Engl.) Inter alias algas unicellulares. (Distrib. Europa.) Distinguitur a forma communi ccllulis paulo minoribus et integumentis crassio- ribus distinctius limitatis. 1. PleurOCOCCUS vestitus, Reinsch, Algenfl. Frank., p. 56, tab. iii., fig. 4). Var. minor; ccllulis spliajricis singulis aut binis aut qnaternis et compluribus sphserice conjunctas familias foimantibus, cytioplasmate dense subtiliter granuloso, cytiodermate crasso (interdum colorato) verruculis acutis dispersis instructo. Diam. ceUular. 0,0112—0,013 mm. (^-i-^ — ^"' Engl.) Inter alias algas imicellulares. (Distrib. P. vestiti, Germania). 2. PleurOCOCCUS anguloSUS, Coi^da. Protoc. palustris Kiitz. Tab. Fhyc. i. tab. 9 ; forma, ccllulis splitericis in familias m in ores in modo Merenchymatis cohse- rentes collocatis. Cellular, diam. 0,00 il — 0,0056 mm. (g^y — ^4^'" Engl.) Diam. famiHar. 0,0224^0,0278 mm. {^—^"' Engl.) Cum praecedente. (Distrib. P. angulosi, Europa.) Scenedesmus acutus, Meyen. In singulis spcciminibus observatum inter Zygnemce ca^spites ; in a fresliAvater pool on the W. of Swain's Bay. (Distrib. Em-opa.) 1. Botryococcus Braunii, Kutz. Maxim e consentit cum spcciminibus Europseis e diversis locis. Inter caespites ScMzosiphontis et in massis parvulis algarum unicellularium muscis aquaticis 80 BOTANY OF KKRGUELEN ISLAND. aclhfcrentibus. Invciiiuutur familia; et viridcs et fuscesrentes. (Distrib. Gci'Diaiiia, Helvetia.) 1. Oocystis Nl'egelii, -//. Ih-anu. Longit. ei-llnl,ir. 0,0278 — 0,0;5()(; mm. (t's— uV" l^oi?'-^ '"'tit- fi'lliil:^-. 0,0108mm. (-piV" Engl.) ^[agnitiuliue ai- lorma eelliilaium ac evtio])lasiiiatis lextura Hiaxiiu(> eonsenfiuiit speeim. eum speeiminibiis Gernianieis. Indiuueiiluiu ramiliarmn l)i- ant ([uadii- eellulariuiu distinete duplieiter striatum. In singulis familiis in luassa jiarvula Phycochromophycearum, uuicellulariiun Jlormosiphoiiti adliiurentc ; marshy ground near A'ulean Cove. (Distrib. Ger- niania) . 1. Dictyosphaerium Ehrenbergii, ]\'(cgrli ; celluHs ])anlo majoribus. Diam. eellular. 0,0081- 0,0112 mm. (.i^ — ^l,,'" Engl.) Inter algas rarias muscoriim foliis insidentes. (Distrib. Europa meridionalis.) 1. Pediastrum ellipticum, lia/fs Brit. Dcsmid.; var. .EQUILOBUM; coeno- bio elliptieo continuo, cellulis disei regularitcr 5-6-gonis, mcmbrana byaliua acbroa laevissima, cellulis peripliericis leviter obtusangulc-cmarginatis, lobulis a^qualibus cellula> dimidio brevioribus apicc truncatulis. Longit. maxima coenobii 0,278 — 0,336mm. (i— I'" Engl.) Diam. cellular. 0,0278—0,032 mm. (,',-.-^i,/" Engl.) In speciminibus duo1)us inter Uormosipliontis pbyseumata observatum. (Dis- trib. P. ellipfici, Britannia.) Asterosphcerium,* genus novum Frotococcacearum. Coeuobium spliaeri- cum, intus excavatum, libcre natans, e cellvilis angulosis parencbymaticc arctissime coujunctis (sicut in Fed last r is), cxtrorsum pyi'iformc ampliatis et subito augus- tatis formatum. 1. Asterosphserium elegans, Beinsch in Jonm. Linn. Soc. XV. 213. Coenobium spbairicuni e cellulis 64 aut 128 formatum. Diam. coenobii ex cellulis 128 format! 0,1 M mm. ( iV" Engl.) Inter algas minores libere natans (in panels speciminibus observatum.) Hoc genus proxime se continuatur generibus Protococcacearum coenobio ex cellulis parencliymatice coujunctis formate {Ilydrodiction, Pediastrum, Coslastrum, Staiirorjcnia). Coenobii dispositio fit in quoque liemispbsera secundum seriem : 1, 6, 11, 16, 21, 26 (seriem aritbmeticam prima? ordiuis cum numero differen- tially 5). Quo dispositionis modo hoc genus disccdit a Pediastris, generi joroximo. Pediastrorum plurimum specierum dispositio coenobii fit, in speciminil)us regula- ritcr formatis, secundum seriem : 1, 5, 10, 16 (seriem arithmeticam secuudge ordinis cum serie diflferentiali prima : i, 5, 6, ct numero difterentiali=l). In Astei'osjilucrii coenobiis legem dispositionis ccllularum, ut fere fit, ad expli- canda coenobia pervenii-e in omnibus casibns, certissime jnito ; sicut per analogiam in Protococcaceis coenobio pluricellulari spheerico ( Coelastrum et Sorastrum), quorum coenobia abnormiter disposita rarissime observari possunt. •' a^Trif Stella, o-rpaTpx globus. FRESITWATEK ALGiE.— P. F. REINSCH. 81 Omnium spccimiuum o1)scrvatorum cellulse erant vacuae, vcluti sfcpe obscrvamus in Fcdiastris majorilius. Tab. IV., Fig. viii. — 1, specimen integrum ex ccUulis 128 cxstitutum (— ,-) ; — 2, cocnol)ii mavginis pars magis aucto {--'\'X). 1. Gloeocystis botryoid.es, Ncegeli, Gatt. einzell. Alg. Cellular, diam. 0,0022—0,001; mm. (^^r-— t^/" Engl.) Thallus gelatinosus, ccllulis singulis ct quaternatis, tegumcntis crassis hyalinis distinctis. In massis parvulis cum aliis algis Fhycochromaceis inter Uormosiphontem coria- ceum var., ]n-opc Vulcan Cove. (Distuib. Em-opa orientalis.) 2. GloeOCOCCUS species. Diam. cellular. 0,0011 — 0,0050 mm. {.-,h ^ a] s'" Engl.) CcUuUe subspbtcricic in familiis 1- efc 8-cellularibus consociatie, cytioplas- mate colore intensive viridi, locello byalino decolorato singulo instructo. In familiis singulis dispersis inter alias algas Sormosij^honti coriaceo, var. adbserentes. 1. Polyedrium tetraetricum, Ncegeli. Cellular angulis acutiusculis (vix acvileolatis), marginibus lateralibus subrectis. Diam. cellular. 0,0221mm. (^y Engl.) In speciminibus singulis inter alias algas unicellulares Hormosiphonti coriaceo var. adbserentes, prope Vulcan Cove. (Distrib. Europa australis.) 2. Polyedrium minimi, ^^. Braun, Alg. unicellul., p. 91, forma; cellular regulariter tctragonai (quadraticae) marginibus lateralibus omnibus sequalibus (vix leviter repandis), angulis obtuso rotundatis. La tit. 0,0066 — 0,0075 mm. (3^ — 3^00'" Engl.) In speciminibus singulis in massa parvula algarum Zi/gogonio loruloso var. adhserente. (Distrib. P. minimi Europa orientalis.) JPolyedntim Fynacidium, Heinscb, Algenflora von Franken. 1866, p. 80, tab. iii. a. — d., complurcs formas comprebendit. Specimen fig. d. dclineatum est P. mini- mum Al. Braun, " latcribiis alternis profundius emarginatis ; " specimen fig. a. delincatiun cum speciminibus Kerguelensibus exacte consentit ; siDecimen fig. b. formam reprresentat marginibus jcqualiter emarginatis. 1. Chlam.yd0C0CCi species, Beinsch in Journ. Linn. Soc. xv. 211 ; cellulis globosis vel ellipticis magnitudine paulo diversis, cytioplasmate et sul)- homogeneo et granuloso (granulis amylaceis dense rej)leto), in statu progressiore corpusculis spbicricis majorilnxs colore intensive luteo-purpurasccnte binis-quin- ternis instructo (cellulis f:lialil)us, Zoogonidiis), cytiodermate byalino crassissimo plurilamelloso (interdum unilateraliter incrassato). Cellular, diam. (ante divis.) 0,0278—0,0393 mm. (^V — -h'" Engl.) Diam. post divisionem 0,0196—0,0221 mm. (rU-i*V"'Engl.) Hab. in foliis musci aquatici. Hujus plantulfe vera natura initio mibi erat aliquid dubia. In ccllularum pluri- mum magnitudine cellidas filius non procrcantium, cytioplasmatis colore, cytio- L 82 JWTANY or KERGUELEN ISLAND. dcniKitis stnictun valdo conscntions c-uai Chroococco atirro, niliilomiims invciii- \mtnr cclluhi' sini;-ula' cylioplasniaiis vnldc divcrsa struciura a C/ironroccis. 8cd post (>l)s(>rvatis collulis minorilms (Znogonidiis) simMhibio in colm'vontia organica i-iun i-ellulis majovilms Chroococcoldcis, hujiis plantula; positio in systemato est constituta. Qiuo i-cllula> I'orina lato pyiiformi, polo subito angustato, cytiodermati; tonui. cvtioplasniato Iiomogcnoo colore iutensivc purpunn) crant inventa) in con- sovtio cellularuni niajonun cytioplasmate vacuarum. In singulis (H^llidis sunt inclusa^ compUiros t'(dlula> iilialcs splian-icjp colore luloo-i)uv]mrco, aliis cellulis sunt corpuscula bina (intcrduni singulum, c.cllula super.). Isonnullaruni cellulanan cvtioplasma deusissime est repleia (•orpusculis amylaceis. Chlamiidococci species dua" cojynita? dilVerunt cytiodermate multo tenuiore noti lanieUoso. Cellulie filialos (huv ( "t") ; I'ollula singula, cytiodermate unilateraliter incrassato, cytioplasmate coUulis lilialil)us {f/oiikUis) a. compluriluis (^'"P) ; cellula singula cytiodermate tenuiore, cytioplasmate corpuseulis amylaceis dcnsissime replete (■",'?). Cellula minor pyriformis (Zoof/oi/id/mi/). 1. Cosmarium pseudo-nitidulum, Nordstedt, By dr. till Kaenned. am st/dl. Novges nesutld. Lund. 1872, torn, ix., p. 46, tab. i., fig. 4) ; var. scmiccUularum semicircularium cytioderma in apicc intus nodulo singulo incrassatum. Longit. 0,033ram. (^V" Engl.) Lat. (),02i8 mm. (v,V"' l^nsl-) In speciminibus singulis in massis minoribus algarum variarum in muscis aquaticis adbierentibus. 2. Cosmarium crenatum, Brrh. var. kerguclcnsc ; cellula in ambitu late ovali, diamctro longitudinali diametro transversali paulo longiore {I), semi- cellulis subsemicii-cidaribus basi arctissime se adtingentibus incisura non disjunctis, margine undato exciso, gibberulis truncatulis 14is — 15is instructo, superficie verru- culis in seriebus radialibus dispositis verruculosa, areola media lacvi, semicellulis c vertice risis ambitu ellipticis (in laterum medio Icviter tumidis), isthmi latitude I- diametri transversal is. Diam. transv. 0,033 mm. (^L'" Engl.) ; diam. longit. 0,039 mm. (^"' Engl.) Isthmi latitude 0,0007 mm. {.^^"' Engl.) In siiecimine singulo observatum inter Vaucherice et Schizothrichis csespites. (DiSTRLB. Em-opa, America borealis, Greenlandia.) Formis singulis Cosm. jmlcJierrimi Nordstedt (Symb. ad Elor. Brasil. Desmid. p. 175. tab. iii., fi"". 24) simillimum in semicellularum ambitu et forma (C pttl- cherrim. |3. horeale, Nordst. Desmid. Spetsberg. et Becren Eiland. p. 32, tab. vi. fio-. 14), sed differt semicellulis e vertice visis in medio utrinque non productis, a fronte visis in medio Isevibus. 1. Staurastrum kerguelense, Beinsch in Joum. Linn. Soc. xv. 214; semicellulis a latere late trapezicis angulis longe productis, margine terminali sub- recto a vertice visis regulariter trigonis, marginibus lateralibus rectis angulis in cornulum rectum margine regulariter crenulatiun longe productis, cornulis summis bispinosis, cytiodermate Isevi seriebus tribus verrucularum marginibus semicellulte FRESHWATER ALGiE.— P. F. REINSCII. 83 parallclis ct in cornulis excurrentibiis ornato, istlimi latitudiuc quinta pars cellulae latitudinis. Latit. ccllula^ 0,1038 mm. (.,'„'" Engl.) ; isthmi latitudo 0,0110 mm. (Tl^"'Engl.) Observavi tantum spccimiiia dua in mas.sa algarum muscis aquaticis ad- ha;rente. S. gracili, E,alfs. simile semicellularum forma, sed diffcrt dimcnsionibus duplo magis majoribus, cornulis multo gracilioribus. E Familia pidclicrrima Dcsm'aViaccarum spocierum niimerosissima, sunt reperta tantum Cosmavia dua liac Stuurastrum J?almogla;cc species ct Euaslrum hinale var. 1. Euastrum binale, Turpin, var. gibbosum; semicellulis in sciagrapliia trapezicis, margine terminali recto in m.edio levissime cmarginato, augulis obtusis non productis, marginibus lateralibus gibberulis birds sequalibus rotuudatis, super- ficie semicellulte in quoque latere gibberulis birds fequalibus iustructa, semi- cellulis a latere apice truncatis. Longit. 0,0306 mm. (^V" Engl.) Latit. 0,0221mm. (yL"' Engl.) Latit. margin, tcrmiu. 0,013 mm. (jib" Engl.) Istbmi latitudo 0,00-11 mm. (jJ-,'" Engl.) Li singulis speciminibus inter algas unicellulares Hormosiphonti adbaerentes. (DisTRiB. Europa, America borealis.) A ceteris formis Eiiastri h'malis Iuec forma distinguitur supcrficie gibbosa semicellularum. Euastrtim hinale var. dissimile, Noi-dstedt (Desmid. Ai-ctoa?, Konigl. Wetensk. Akad. Forbandl. Stockholm 1875, Nr. G, p. 31, tab. viii. fig. 31), persimile in semicellularum sciagrapbia, differt lobulis basalibus leviter repandis, augulis paulo productis, superficie non gibbosa. 1. PalmOglCBae species ; cellulis cUipticis polls angustatis, diametro trans- rersali dimidio diamctri longitudiualis breviore, cytiodermate subcrasso, cytio- plasmate granulis singulis majoribus instructo in massa gelatinosa irregulariter expausa nidulantibus. Longit. cellular. 0,0068 — 0,0081 mm. {^\^ — 2I3'" Engl.) Latit. cellular. 0,0041 mm. (.^"' Engl.) In massis parvulis Hormsiphonti adhasrentibus. Granulis amylaceis cytioplasmatis ad Falmoglaas spectans, a P. maerococca et micrococca distinguitur cellulis minoribus et polls angustatis. 1. Vaucheria sessilis, Vancher. ■ Maxime consentit cum speciminibus Europaiis. Oosporie matura; membrana trilamellosa. Filum unicum fructiferum observari potuerat in massa ex algis diversis compo- sita. (DiSTRiB. Europa, America borealis.) 2. Vaucheria sericea, Lyngbye. Filum singulum florescens observatum. Oogonium ad fecundationcm aptum, anthcridia biua horizontaliter flcxa nondum aperta. In filo singulo oljservato oosporas maturas evolventc oosportu in oogonio laxc inclusie. (Disthib. Europa, America borealis.) 3. Vaucheria pachyderma, Symn. Vaucu. Dillwyni, JFeh. et Mohr. exp. L 2 84 BOTANY OF KKKGUELEN ISLAND. Fila comitluria oospovis maturis obscrvata. Oos])oranun iniMuljvana iilurilanu-Uosa (luplo crassior mombraua /'. scsnilii'. luca^spitulo parvulo ex lllis intcrtoxtis Vancheric lila portincrc ad /'. acriceain, pacbydcrinain an ad spccicm ]iro])riaHi. I. Vaucheria geniinata, l>e Ctoulolle. i'ila coinplmia oogouiis imniaiuris siuc dubio ad /'. (joiiinalinn spcctantia ; tballi ramuliis latci'alis mimitus Mores evolvons paulo loui^'ior ct gracilior ramulo spe'ciminiuu Europa'arum, cornulum jaiu ill positionc propria, suiunio — ad obsrrvaloroiu verso— ininimc latcraliter contorto. J^. hamatic vi\m\\\\\% lateralis diinidio brevior a(! rainiili oogonia proereantcs duplo longiores. Antbcridiuin a basi lairvatuiu in uuo aui'ractii contortum. (Distrib. Europa, America borealis.) Status cvolutiouis partium florum Vauchcyiarum perfectc congrunt cuiu eodem statu evolutionis iloruni Nitelhe aiitarclide. Tempusanul, rcspoudcus statu aualogo vitcC liarum plantularuni in uostris latitiidinibus ver est (menses Aprilis, Mali, usque ad initium mensis Junii). Quarum plantularum plujonomena vitalia normam dare ad dijudieaudas iilla) regionis terra; rationes iu respectu commiitationum temporum quadrijjartitanim anui, verisimilc videtur. 1. Olpidium caudatum, llelnsch iu Juum. Linn. Soc. XV. 215 ; cellulis spbterieis siuc radiculis substrato viveuti insidentibus, in polo processu singulo spini formi cellukc diametro subicquaute postremo aperto iustructis, oytiodcrniate distiucto subcrasso, eytioplasmate dense granuloso. — Diam. cellular. 0,0112 — 0,013 mm. (ri^j—rirT'" Engl.) In Schizosiphontis kerguelensis tricliomatibus. 0. ampullaceum {Chytridiimi ampullaceum, A. Braun, Ber. d. Berlin. Acad. 1855, p. GO; llabeuliorst, El. Eur. Alg. ii. p. 282) est distinctum ab lioc Olpidio dimensionibus duplo minoribus (0,0001 mm. ; .,{ ^"' Engl. diam.). Tab. IV., Eig. vi. — 1, Schizosiphonlis tricbomates pars cum plantula parasitica insidente, ff?^ ; — 2, cellula singula parasitica major ampliflcata, ^\^. 1. Chytridium pyrifornie, Beinsch in Joum. Linn. Soc. XV. 215 ; cellulis zoogonidiis nondum egressis operculose apertis ovato-pyriformibus, basi sensim angustata, iu radiculum in substrato vivente radicantem prolongatis diametro transversali dimidio diametri longitudinalis angustiore, eytioplasmate dense subti- litcr granuloso, cytiodermate distincto dupliciter striate, cellulis zoogonidiis egressis subcylindricis usque subcuneatis, opereulo transversaliter a cellula se sejimgente subbemisphrerico apice rotuudato (non aeuminato), radiculo usque tertiam partem diametri longitudinalis cellula3 a;quante, in medio plus minusve incrassato apiculo prolongate deorsum verso. Diam. transvers. cellula; 0,013 — FRESHWATER ALGiE.— P. F. REINSCII. 85 0,0168 mm. (yVf— t-V^"' Engl.) Diam. longitud. cellultc 0,0258—0,0278 mm. ( J— tVt'" Engl.) In Vauchericc coUulis. A Chytridiis cognitis proximimi in celhilarum forma Chytr. \Olla, A. Braun, (Verjung. p. 198. Ber. Berlin. Acad. 1855, p. 380 ; llabenli. Fl. Eur. ii. p. 277), quod Chi/triclkim distiuguitur cellulis latioribus operculo obtuse umbilicato ; C. acumutatum et C. brevipes, A. Brauu, sunt distincta operculis acuminatis. In omnibus Van die rice cellulis, qua; portavcrunt plautulas parasiticas, sunt observatse prolongationcs utriculiformcs dense jjositoe e VaiicherUc cellula egresses. Parasita invenitur plerumque in iisdcm locis Vaiicherice cellulse infectsc ubi sunt evoluta> liffic prolongationcs utriculiformcs. In his locis abnormiter transmutatis ajiparent parietcs intcrcalares quae separant lumen transmutatum a ccHuIob cetero luminc. Certissime adducta est transmutatio abnormis Vattchericc a plantulis parasiticis. Complurcs casus hucusquc sunt observati, in quibus eft'eruntur transmutationes morjjhologiccB plantarum altiorum per plantidis parasiticis uniccllularibus.* Singula fila Vancherice abnormiter transmutatre observavi, quibus desunt Cliytridii cellula?, sed in cytioplasmate Vaucherm sunt impositse cellulse sphaericai magnitudine raria manifesto alieme Vauchericc cellulse. Utrum aliquam connexionem esse geneticam inter Cliytridii cellulas pyriformes Vauclierice insidentes et cellulas entopbyticas, an non, incertum est. 1. Micro thamnion cladophoroides, Beinsch in Joum. Linn. Soc. xv. 21G ; e maximis, fruticulosum, fills solitariis erectis regulariter ramosis, radiculis singulis contortis in substrato (algis viventibus) insidentibus, ramulis erecto-patentibus attenuatis unilateraliter dispositis (in speciminibus minoribus) aut vcrticillatim dispositis (in speciminibus majoribus), cellulis fill primarii apicem versus paulo in- crassatis diametro Iplo-Gplo longioribus, cellulis ramulorum in basi paulo con- strictis diametro 10plo-20plo longioribus, cytioplasmate omnium cellularum sub- bomogeneo, colore pallide luteo-olivaceo, granulis singulis dispersis instructo. Fili primarii cellularum latit. 0,0056 mm. (^f «'" Engl.) Ramulorum cellularum latit- 0,0028—0,0041 mm. (yi^ — jb"' Engl.) Plantulse altit. 0,556 mm. (|"' Engl.) In Lynr/byce majoris, Kiitz. forma tricliomatibus et in Chlorococci spec, cellulis radiculis brevissimis aflG.xum. Ilajc plantula elegantissima Cladophoris singulis in habitu baud dissimiles, sed sat distincta a Cladophoris cytioplasmate subhomo- geneo ac dimensionibus minimis, ad Microthamnia spectat quibuscum consentit in cytioplasmatis structura. Generis specierum triuni bucusquo cognitarum nulla aliqua sirailitudinc consentit cum plantula Insula; Kerguelensis. 1. Stigeoclonium Hookeri, Beinsch in Joum. Linn. Soc. xv. 216 ; luite viride, parasiticum, ciespitulos cbaitopboneformos basi radicantc formans ; fills a;tate provectiore inferue nudtc et subintegric supernc ramosissimis, basi • Synchi/lrium Turaxucl (Ue Baig. tt Worou. Ijur. il. Natuif. Gescllsch. Freiburg, iii. 2. tub. i. ii., fig. 1-7. 86 BOTANY OF KERGUELEN ISLAND. radiculis anastomosanlihus instmctis, ramis spicatis (|)k^"' Engl.) In muscorum foliis, in Nitellce et Vancherice cellulis, et in Schizosiphontis kergueleims tricliomatibus. Hoc Stigeoclonium ramulis prolongatis tenuissimis iiagellit'ormibus erectis ex ramulis crassioribus ortis aliqua similitudine conscntit cum S. setigero, Kiitz. (Tab. Phyc. iii. tab. 5), quod distinguitur ca;spitulis multo majoribus fluctuantibus (usque tres lineas longis). CjEspituli tantum fila propagativa procreantes baberi possunt pro Speciem pro- priam. In singulis speciminibus jilantulaj inveniuntur et fila sterilescentia et fila propagativa. Ulteriora paulo crassiora sed brevioi-a saepe inveniuntur ex uno ramulo evohrta cum filis sterilcscentibus. Cscspitulos quoque singiilos in Hor- mosiphontis sp. n. pbyseumate crescentes una cum Choreoclomi procumhentis gen. n. caespitulis observavi ; in his plantularum duarum infimse partes adeo sunt inter se coalitffi ut plantulas duas valdc diversas in cobajsione genetica putarc possis. Choreoclonium, genus novum.* Plantula parasitica ex filis ramosis pro- cumbentibus densius aut laxius intricatis substrate dense adpressis interdum parcn- chymatice inter so conjunctis formata; cellula; rectaugulares usque quadraticse. * Kofim expando, K/.ini clunis. FRESHWATER ALGiE.— P. F. REINSCH. 87 Propag-atio ? — Synon. Gemis s. n. in Rcinscli, Contribut., p. 76, tab. iv. (Cblo- ropb.) descriptvim ct dcliucatum genus ad Chtctophoraceas spcctans, Stigeoclonio proxm7im. 1. Choreoclonium procumbens, Meinsch in Journ. Linn. Soc. XV. 217. Colhilar. diam. 0,0028— 0,00 11 nim. (^I, — ^J^'" Engl.) Cellular, longit. 0,0112—0,0224 mm (^]"-,— ,,'./" Engl.) In foliis muscorum et in N'Uellce cellulis. Hanc plantulara primo obscrvavi anno 1872 in Gcrraania in plantis aquaticis {Hottonia, Tltrieidcrria) crescentem, deinde in comphiriljus formis variis locis Gei-- mania?. In contributionibus mcis formas varias in uno goncre conjunctas sine nomine rccepi ; post plantulam Korgulenensem inventam nimirum dubitavc possum in idontitate plantularum e locis duobus remotissimis. Tab. IV., Fig. ix. — 1, folii musci aquatici pars cum plantula singula minore in nervo folii crescenti {'^^) ; — 2, alteri folii pars cum plantula majore obtecta (— 1~). 1. Draparnaldia subtilis, Meinsch in Journ. Linn. Soc. xv. 218; fills ramisque primariis liyalinis, ramis e basi rcpetito dicbotome ramosissimis, ramulis fiu'catis acutis plerumque in pilum byalinum ex cellulis compluribus exstitutum longe productis, cellulis infimis fili primarii diametro sequalibus cytiodermate crasso lamelloso, cytioplasmate subbomageneo subtilissime granulate, cellulis superi- oribus diametro usque duplo longioribus, cellulis raraulorum diametro usque triplo longioribus, cytioj)lasmate dense granulose. Diam. fili primarii 0,0168 — 0,0232 mm. (^4^ — ^"'.) Diam. ramulorum 0,0050— 0,0084 mm. (^J-^_^iy Engl.) Plantulte altitude 1 — 2 mm. In Vaucho'ice ccllidis et in mviscis aquaticis in plantulis dispersis radiculis numerosis radicantibus. Haec plantula elegans tantummodo in speciminibus paucis observata differt a ceteris Draparnaldiis et magnitudine et loco natali. 2. Draparnaldia distans, KUfs., Tab. Fkyc. iii., tab. ll, fig. 2; forma tenuis, cellulis fili primarii duplo-quadruplo diametro transversali longioribus, ramis primariis perpaucis, ramulis sparsis crebrioribus brevioribus cum ramulis longiori- bus in ambitu lanceolatis perpaucis iutermixtis, ramulis ultiuiis plerumque in pilum acbroum cellulare attenuatis, cellulis ramulorum tumidis diametro subicqualibus. Diam. ceUular. fili prim. 0,0278—0,050 mm. (yV— tj^s'" Engl.) In speciminibus exsiccatis cum Zygnemate iutermixtis. " In a fresbwater pool on tbe W. of Swain's Bay." (Distrib. Europa.) 1. Proterderma viride, Kutzing. Familice singulge in foliis musci aqua- tici laxius insidentes, in magnitudine cellularum ac forma (0,0081 mm. ; ^,^-3-'" Engl, diam.) cum speciminibus Franconicis maxime consentiunt. 1. ZygOgOnium torulosi, KMz., Tab. Phi/c., tab. 14, fig. 1; forma crassior. Ccllultc diametri trausversalis dimidio brevioribus (ante divisioncm usque sequalibus) cytiodermate interiore crassissimo plurilamelloso, cytiodermate extcriore 88 BOTANY OF KERGUET-EN IST.AND. siibtoniloso. "Diam. collular. 0,0108—0,019(5 mm. (,1.-— ,1,,"' l^'"Sl-) l"*'^^- filorum {c. iiuluin.) 0,03,".— 0,031)3 mm. (,A.— .,/.,"' Engl.) ill civspitulis inttn- Tlonnosiphon coriacciim, \ar. " In nioisi places near Vul- can Cove." — DiSTKiii. Z. torulosl Eiiropa oiicnlalis. In illis singulis observantur sicut in spociininibiis I'uioprois ccllukc siibspliivriciii lafci-alcs filis adluvrcntos inclumciito crassissinio vclatte. Qiia> cellulho — nullo modo colluhu propagativa' — orinntur in boc Zi/ffof/oiiio ct in Z. (ii/oiikiIo divisiono lougi- tndinali intordum incidcnto coUularum singiilarum iili. — li;v!C ibrma a forma typica in Tab. Plnx'ol. delinoata cidlulis anguslioribus cytiodermato crassioro ct Jilis cras- sioribus distinguitur. Z. tondosum, Kiitz., cum scric specierum : Z. rricetornm, anomaliim, delicatulum, a CI. llabcnborst (PI. Eur. Alg. ii., p. 2-jl) in una specie contrabuutur, sed cbaractcrcs constanter observattc borum Zygogoniorum a speciebus Kiitzingianis discederc milii non pcrmiserunt. -■ ZygOgonium tenuissimum, Eeinsch in Journ. Linn. Soc. XV. 218; filis tenuissiinis eel lulls dian^clro duplo longioribus (ct paulo minus) rcgulariter rcctaugularibus, cytiodcrmatc subcrasso bomogcnco byalino, cytioplasmatc contracto colore lutco-viridi grauulis majoribus instructo. Diam. cellular. 0,0008 — 0,008imm. (xk--b"' Engl.) In singulis filis inter Scijtonemam castanetim dispersis, " near Vulcan Cove." — DilTert a Z. dellcatulo at Z. salino cellulis longioribus, a Z. graeiU et Z. Balpsii cellulis brevioribus, ab omnibus Zygogoniis autem filis multo tenuioribus. 1. Spirogyra longata, Kntz., Tab. Fhyc. v. tab. 20, fig. l ; cellular, diam. 0^039— 0,05mm. {^i—i^" Engl.) Lougitudo ccllularum Splum— 7plum latitu- dinis. In a freshwater pool W. of Swain's Bay (spccim. exsiccat.). Structura fascias spiralis latsc anfractibus lis — 5is maxime consentit cum speciminibus Europa?is. Eila omnia incopulata sunt latiora (usque duplo) filis forma; communis Europea; per totam Em-opam dififusaj. — (Distuib. Europa, America borealis.) 2. Spirogyra Spec. -, Ccllularum diam. 0,0278— 0,0393mm. (V5-A'" Engl.) Longitudo iplura — opium latitudinis. Eila omnia incopulata ad quandam speciem ISpirogyrm spectantia, qutc pertinet ad Spirogyras cytiodermate in utroquc polo ccllukc nee protenso nee replicato ; sed structura fasciarum spiralium adco est transmutata ut numerus fasciarum et forma certe non potest explicari. Eila dispersa inter Draparnaldiam distantem. " In a fresbwater pool on the W. of Swain's Bay" (specim. exsiccat.). 1. Sirogonium sticticum, Kiltz. Cellular, diam. 0,045— 0,050mm. (^1. ^1.,'" Engl.) Eascitc chloropjhyllacecc terna; — quaternae in quaque cellula nucleis ex substantia proteinicis cum jodinis agentia fuscescentibus majoribus (nunc decoloratis) instructse. Qui nuclei sunt majores nucleis speciminum ex Germania. FRESHWATER ALG^.— P. F. REINSCH. 89 In singulis filis iucopulafis inter Zyynemce crespitcs. "In a freshwater pool W. of Swain's Bay." — (DisritiB. Europa borealis et centralis.) 1. Zygnema Vaucheri, A minimi-^, iH-ulis inannatis vix conspicumn, parasitit'iiiii, lilis intogorrimis croctis singulis aul perpaucis aggregatis, artii-ulis inferioril)us sul)C'uneiformil)us apico paulo incrassatis, ccUalis cortiealibus tis — 6is obtcctis supcrioribus niulis roctangularibus, raimilis vcTticilloniiii intcgoiTiniis (rarius singulis raniul is iiisi me- tis) irqualibus, a])icil)us paulo angustatis ex collulis Sis — 7is cxsiitulis apicciii iili vorsus scnsini docivsi'cntibus, int'iMMoriljus arficulovuni longitudinc siibaH^uanfiljiis, siunniis cluplo-triplo longioribus, ccllulis ranuilovum rectangularil)us us(iu(> sub- quadi-atiois, cytiodermatc extiTiorc tcnuissinio vix conspiruo, cytioplasmate sub- liomogeneo colore obscure olivaceo-viridi ; rructiftcatio ? — Diani. articuloruni 0,00il— 0,005Gmm. (., l^ — g-JV" Engl.) Diani. raiiuilonini 0,00dlmm. (7;,'; /" Engl.) Plantul:o altitudo 0,37— 0,45mm. {]—\"' Engl.) In Xitelhe cellulis ot in muscoruni I'oliis, in lilis singulis .sjjarsis cum aliis alo'is ( Tohipothrix, Leptothrix) iutcrmixtis, rarius in caispitulis parvulis. B. tumidum,\\\ Charavulgari crescens (Ecinscb, Contributionos, p. G9, tab. xliv., Ilbodopli. fig. 1-5) a speeiebus bucusquc cognitis proxima species, sed valde diversa dimensionibus omnium partium multo majoribus (3-llinoas longa), verti- ciUorum ramulis- numerosis rcpctito dieliotomc ramosis ; in ramulorum ccllularum forma aliqua'similitudino conscntit. [The fresli-watcr species recorded by lieinsch are 106, to these may be added the lollowin"' mentioned in the Antarctic Hora: — Oscillataria purpurea, Hook. f.& Ilarv., Calothrix olivacea, H. f. & H., Ulva cristata, H. f. & H., Mastodia iessellata, fl. f. & H., TrypothuUns anastomosans, IT. f. & H., Nostoc commune, Vaucher, and N. microscopicum, Carm., making a total of 113 species. This interesting Antarctic island, therefore, so far as explored, appears to be very rich in certain forms of fresh- water Algae. — G. Dickie.] FUNGI— M. J. BERKELEY. 93 VII. — Fungi. By the Rev. M. J. Berkeley, M.A., F.L.S. [The Fungi collected iu Kcrgueleu Island amount to 9 or 10 (the tenth being still an undetermined form).* Dr. Hooker obtained 2 species in the winter (May and June) 1810 ; Mr. Moseley 3 in addition to the same, during summer (December and January) 1873-1; Mr. Eaton, also in summer, 5 determinable species, and 1 that could not be identified {see footnote), besides the species found by Dr. Hooker. Until a few days before Midsummer {i. e. Christmas) no Fungi were seen in the vicinage of the English Observatory Bay. The first to appear was the common mushroom, a single specimen of which was found on an island in the sound by some ofBcers from H.M.S. " Volage." Later in the summer the other four species came up in a few places on the mainland. They were not by any means of frequent occurrence, and probably scarcely any of them would be found at the time of year corresponding with the date of Dr. Hooker's visit to the island. — A. E. Eaton.'] 1. Agaricus (Galera) kerguelensis, Berk, in Jonm. Bot. v. 51 (1876) ; et in Jonrn. Linn. Soc. xv. 22. Caispitosus, fulvus, pileo e breviter campanulato convexo loevi carnuloso, margine tenui striata, stipite oequali apice pulverulento- granulato, lamellis distantibus ventricosis adnatis. Amongst moss in a bog on the eastward portion of the base of a promontory E. of Vulcan Cove, Jauuary 1875, Eaton. Ca?spitose, attached by abundant mycelium. Pileus ^ inch across ; stem nearly 1 inch high, ^ to f line thick ; principal gills about 12 in number, shortly but truly adnate, and not in the least decurrent. It is far more fleshy than any variety of A. hypnorum, to which species no doubt it is closely allied ; and while agreeing with A. embolus in possessing comparatively few gills, it differs from that species in the mode of their attachment. 2. Agaricus (Galera) hypnorum, Batsch.; Berk, in Joum. Linn. Soc. sv. 53. Hab. — On Azorella. January 1871. Moseley. Spores • 0004 inch long. 3. Agaricus (Naucoria) furfuraceus, Pers. ; Berk, in Jonrn. Linn. Soc. XV. 221. * This species is relerred to by Mr. Eaton (in Proc. Roy. Soc. 1875, May. xxiii. 355) as " a peculiar " parasite on Azorella, which grows out of the rosettes " of the leaves " in the form of a clear jelly, which " becomes changed into a firm yellowish substance of iudefinite form." It was common on the sides of hills in tlio neighbourhood of the observatory towards the end of December, and a series of examples was preserved in spirit, but they could not be worked out. 94 BOTANY OF KERGUELEN ISLAND. lu tlio same bog as A. kcrffiic/ciiid,'^-. and at the same lime, Eaton. -i. Agaricus (Xaucoiua) glebarum, Berk, in Flor. Anlarct. 4X7, t. clxii. fig. iii. ; ct ill Joiini. Linn. Soc. xv. 53. On AzorcUa, January 187-1', Kerguelen Island, Ilooler, Eaton. (Marion Island, Mosetey. On tufts of Botaa:, Falkland Islands, Hooker.) Spores •()()();> inch long. 5. Agaricus (Psalliota) campestris, Linn. ; Berk, in Jovrn. Linn. Soc. XV. 221. A. (P.) arvcnsis, Eaton in Proc. Itoij. Soc. xxiii. 355. On an island near Observatory Bay, in Royal Sonnd, IGth December 1874. A solitary specimen, Eaton. (Almost cosmopolitan.) 1. Coprinus atramentarius. Fries ; Eaton in Froc. Boy. Soc. xxiii. 355 ; et in Joiirn. Linn. Soc. xv. 222 (footnote). Two or tbree specimens were found singly close to the margins of two of the lakes among tbo hills near Observatory Bay, in February 1875, Eaton. 2. Coprinus tomentOSUS, Fries ; Berk, in Journ. Linn. Soc. xv. 53. On dung, January 1874., Moseley. 1. Peziza (SAiicoscYPn.i:) kerguelensis, Berk, m Flor. Antarct. 451, t. clxiv. fig. iii. ; Cooke, Ilycol. fig. 13 1 ; et in Journ. Linn. Soc. xv. 53, 222. Christmas Harbour, May and June, on bare boggy ground near the sea, growing amongst ConfercfP, Hooker. Amongst dwarfed Leptinella plumosa, on wet ground close to the shore, growing in rings, rare. One ring on an island in Swain's Bay, January, and another on the mainland near Observatory Bay, February 1875, Eaton. Royal Sound and Betsy Cove, Moseley. (Hermite Island, Cape Horn, alt. 1,000 ft., Hooker.) 1 . Sphseria herbarum, Pers. On dead stems of Prinrjlea, Eaton. ZOOLOGY Seals and Cetaceans. By William Henry Flower, F.R.S. Thkee species of seals and two of cetacea are known to be indigenous to Kerguelen Island. Single examples of two of the seals and one of the cetacea have been submitted to me, comprising — a skeleton of a female " sea leopard " Oymorhimis leptonyx (Blainv.), one of a very young male "sea elephant " Ilacrorhinus leoninus (L.), and one of a full grown female cetacean of the genus Globiceplialus. These are now in the British Museum. PiNNIPEDIA. Ogmorhinus leptonyx (Blainv.). JBhoca leptonyx, Blainville, Joiu-n. Phys. 1820, xci. 288. Stenorhynclms leptonyx, E. Cuvier, Mem. du Mus. 1824, xi, 190, pi. xiii. 1 (not Stenorhynchus, Lamk. 1819, of the Crustacea ; nor Stenorhynchtts, Lat. 1823 of the Insecta) . Ogmorhinus leptonyx, Peters Monatsb. K. Akad. d. Wissensehaft. Berlin, 1875 Juni 10, &c. " Sea Leopard." — The female example killed in Observatory Bay, Boyal Sound, 13th October 1874, judging from the condition of the bones, was not quite full grown. The skeleton does not differ to any appreciable degree from another of corresponding age and sex in the Museum of the Eoyal College of Surgeons, received from Tasmania. It has 7 cervical, 14 thoracic, 6 liunbar, and 3 sacral vertebrae. The tail was left in the skin. Extreme length from muzzle to tip of taU 7 feet 8 inches. Dist. — This species is extensively distributed in the southern hemisphere, having been met mth in South Georgia, the Ealklands, Kerguelen Island, South Australia, Tasmania, New Zealand, Campbell Island, &c. Professor Peters has lately proposed the name Ogmorhinus for the genus to which this seal belongs, the name Stenorhynclms, by which it is generally known, having been preoccupied in zoology. Macrorhinus leoninus, (L.) Fhoca leonina, Lin. Syst. Nat. ed. xii. 1766, i. 55. PA. elephantina, Molina, Saggio 1782, p. 200. 96 SKALS AND CETACEANS.— W. II. KI.OWKK. PIi. proboscidca, rorron, NO v. aiix Tcr. Austr. IMO, ii. :)l. Mao'orhiniis proboscideits, F. Cuv., ]\[oin. dii Mus. 1824, xi. 200, pi. xiii. (nut, Jfao'Oi'Iiiiiiis Lat., Fani. \a(. dii lu'i;iu' Anim. lS2r), a i;'onus of llio Colcopiera). 3[iroiiiiga palagoniva, J. E. Gray, (irifllth's Animal Kingdom, 1827, v. 179. jronnif/a fh'pli((ti/iii(i, Gray, Cat. Osteol. Spot-. Uril. Mus. 1847, p. 33; idinn Cat. Seals and Wlialos, 186G, p. 38, fii?. 13 (skull). Cjjstophora probosckJca, Wyv. Thomsou, iu Good AVords, 1874, November, p. 748 (dial. Expcd.). C. leouina, J. W. Clark, in Nature, 1875, Sept. 2, p. 366 (German Tr. Exped.). JIacrorliiniis leoninus, Kidder, Bulletin U. S. Nat. ;Mus. 1870, iii. 39 (Am. Tr. Exped.). " Sea Lion and Lioness " Anson's A'^oy. 17 IS, p. 122 and figui-e. " Sea Elephant " of authors ; Eaton, Proc. lioy. Soc. 1875, xxiii. 502 ; " Platy- vhine seals," idem, op. c-it. p. 353. Two young males easting their coats were kilh^l with stones on the beach near Thnmb Peak on tlie 8th December 1874; and several of each sex oC corresponding age and size were shot with explosive bullets in Swain's Bay in the following month. These last were A'alueless as specimens. Dist. — The sea elephant has long been known as an inhabitant of Kerguelen Island, and formerly was widely distributed along the coasts of the Antarctic and southern temperate seas. Mr. Eaton in Proc. Boy. Soc. 1875, xxiii. 502, says : — " Some examples are " uniformly reddish brown ; others are pale, blotched and spotted with darker grey. " They usually lie just above the beach, separately, in hollows among the Accena and " Azorella, Avheve tliey are sheltered from the wind. On being approached they " make no attempt to move away (possibly because there are no land animals, in- " digenous to the country, capable of molesting them, to cause them to acquire a " habit of flight) but raise up the fore part of their body, open the mouth wide, and " utter a peculiar slobbering cry." There is much confusion as to the synonymy of this species. It is the FJioca leouina of the Systema Naturoe, ed. xii., founded upon the " sea lion and lioness " of Juan Fernandez, described and figiued in Anson's Voyage, 1748, the P. ele- phantina of Molina, 1782, and the P. proboscldea of Perron, 1815, and of many later authors. Leonina thei'efore is the earliest specific appellation. With regard to the generic name, after JPhoca (wdiich of course is inadmissible, having since the dismemberment of the group been restricted by common consent of naturalists to P. vitidina and its immediate allies), Cystophora Nilsson, 1820, has priority for those who hold that there is not suificient difference between the southern Sea Elephant and the well-known Hooded Seal {^Cystophora cristata, Fab., or borealis, Nils.) of the Arctic seas to separate them generically. But those who hold the contrary opinion (and the remarkable dissimilarity of tlie auditory ossicles SEALS AND CETACEANS.— W. H. FLOWER. 97 inclines mc to this -vaew) should adopt JIacroi'himts F. Cuv., 1821., which was afterwards needlessly superseded hy Jlonoif/a, J. E. Gray. It should he mentioned that Jlacrorhinus F. Cuv., is often (as in Agassiz, Nomenclator Zoologicus) quoted as if it had been first used in the Diet, de Sc. Nat. xxxix., 1826, and priority is accordingly misattributed to Latreille's use of the same name in the Fnm. Nat. du Regno Animal, 1825, for a genus of the Cole- opt era. Otaria gazella, (Peters). Arcfoj)Iioc(' (jazello, Peters, Monatsb. K. Akad. d. Wisseusch. Berlin, 10 Juni 1875. [The Fur Seal of Kerguelen Island does not resort to the sheltered waters visited by the American and English Expeditions. A single example was captm-ed by the German ExpecUtion at Betsy Cove, and according to Wyville Thomson (op. cit. p. 748) the Challenger Expedition seems to have obtained two in Fuller's Harbour. It is due to Dr. Gilnther to state that, prior to my leaving England, he informed me of the occurrence at Kerguelen Island of an undetermined small species of seal besides the Sea Leopard and Sea Elephant, and showed me an unpublished drawing of its head. While the expedition was at Observatory Bay, I wrote to Captain Fuller of the " Roswell King," endeavouring to negotiate for a complete skeleton and skin of the Fur Seal, which he expected to meet with at Swain's Islands,— a small group off the N. E. coast of the main island. He returned in about a month afterwards without having killed or seen one in the course of his trip. The species was doubtless A. gazella, Peters. — A. E. ^.J Cetacea. Balsena australis, Desmoulius. [A pair of Bight Whales were seen occasionally among the islands in Roval Sound, between the English Station at Observatory Bay, and the Prince of Wales' Foreland. I noticed portions of two old skeletons of this species stranded on the eastern shore of a promontory immediately to the eastward of Vulcan Cove, about five miles (an hour and a half's walk) from the head of Carpenter's Cove. They comprised the skulls, the lower jaws, the cervical and some of the other verte- l)rse, the scapuhe, and a few ribs. They were in tolerable condition, but it was impossible to convey even the scapulse over so great an extent of bog and rocky hills to a boat at Carpenter's Cove. — A. E. E.] Globicephalus melas, Traill. / Catodoii sij'nieml, Lacepcde, Hist. Nat. de Cctaces, 1801, p. 216, pi. xiii. Belphimis melas, Traill, Nicholson's Journ. 1809, xxii. 81-83. D. globiceps, Cuv. Ann. du Mus. 1812, xix., pi. i. 2. Globicephalus, Lesson, Compt. de Bullou, 1828, i. Globiocephalm svineval, Gray, Zool. Ereb. & Ter., 1846, p. 32. N 98 ZOOLOGY OF KKRGUELKN ISLAND. G. macro>'/ii/>ic/ii!s, aiu-torum (not Gray, op. fit., p. 33) ; Van Bi'uod. & Gerv., Osteoi^v. dor Cot. pi. lii.; Hector Trans. N. Zi>al. Iiistit. vii., pi. x.Kvi. 3. The " Blacklish," " Ciiaiug Whale," or " Pilot Whale." This was the animal nientionnl in ]\tr. Eaton's Tirat lleport (Proc. Roy. Soc. 1875, xxiii. 501), as found by Mr. iMidshipuiau Porrest, dead in shallow water in Swain's Bay. The example is an adult female, all the epiphyses being united to tlio bodies of the vertebra?. The skeleton was coniplcte ; but about live of the caudal vertebrip and a few of the terminal phalauc^es have been lost. There are 7 cervical vcrtebitc, of which five are coherent, 11 dorsal, and 35 lumbo-caudal. Maxillte visible as a narrow strip along the whole of the outer border of the rostrum. Detention -i,)__-r?v (teeth much worn). Extreme length of body in the flesh? 19 ft. 1 in. Principal measurements of the skull : — Totallength ..... Length of rostrum from the tip to the centre of a line drawn across the antcorbital notches Width of base of rostrum at the antcorbital notches AVidth of rostrum in its middle Greatest width of skull .... Breadth of occipital condyles Mandible, length of - - - - 5, length of tooth line of - . - „ height of, at coronoid process lean detect no difference between tbis skeleton and that of the " Blackfish " from the seas round Tasmania, five examples of which, presented by Mr. W. L. Crowther, F.R.C.S., of Hobart Town, are in the Museum of the Royal College of Siu'geons. Indeed I must go furtber, and add that I can find no tangible osteological grounds for separating either the Kerguelen or the Tasmanian specimens from the well known Caaing Whale, Glohicejihalus melas, Traill, of the N. Atlantic, the Grindval of the Faroese. When only one or two examples of eacb are compared together, it is not difficult to discover distinctions between them, because numerous individual variations occur in different parts of the skeleton ; but when larger series of both Nortbern and Southern specimens are examined, these variations do not appear to resolve themselves into constant characters by which the one can be dis- tinguished from tbe other. The reasons for giving to the specimens from distant seas a distinctive appellation have been mainly derived from the sui:)position that it was impossible that cetacea from widely remote localities could be specifically identical with one another ; but this is an assumption too great to be made without proof, and in the present instance is especially inadmissible, considering that G-lobiceplmli almost, if not quite, identical have been observed by voyagers in all 20 in iches 13-5 55 10-3 55 8-7 55 19-0 55 71 55 20-6 55 6-2 55 6-1 55 SEALS AND CETACEANS.— W. H. FLOWER. 99 intermediate localities. If an assumption such as this could he allowed, it would sti'ike at the root of all our knowledp^o of the geographical distribution of animals. It is, however, possible that future observers with still more ample opportunities may succeed better in discriminating between the two. Meanwhile, until some valid distinctive characters shall be pointed out, I think it will be safest and most in ac- cordance with the true principles of zoological nomenclature to designate them all by the specific name melas, imder which one member of the group Avas first clearly described by Traill. The Catodon sv'meml of Lacepede (1801) is possibly founded on the present species; but the description is so vague and inaccurate that it can hardlt be conceded priority over the excellent account and figm-e given by Ti'aill (1809). Cuvier, unacquainted with Traill's memoir, described and figured the animal independently in 1812 under the name of Delphimis glohiceps. This hybrid word was modified by Lesson in 1828 into the generic term GJohlcephalKs. Dr. Gray in 1816 revived the specific term svineval, and definitely applied it to this species in his subsequent catalogues, altering Lesson's generic designation at the same time to Globiocephahis. All the specimens of Glohice^ylialus from the southern hemisphere are often catalogued in systematic works under the name of G. macrorhynchus. Gray. In the Osteographie des Cetaces, pi. lii., a skull from New Zealand, closely resembling the one from Kerguelen, is figm'ed by Van Beneden and Gervais under this name ; but nothing is apparent in the figure to distinguish it from the skulls of G. melas from the coasts of Brittany and Iceland which are figured with it, excepting characters appertaining to a somewhat inferior age. The letterpress of tliis section of their work, which it is hoped may clear up some of the difiiculties existing at present in the history and synonymy of this group, has not yet appeared, A skull figured by Dr. Hector in the Transactions of the New Zealand Institute (vol. VII., p]. xvi., fig. 3) under the name in question also resembles that of G. melas. And M. Fischer (Journal de Zoologie, 1872, i. 273) has been unable to detect any distinc- tion of specific value between a skeleton of a Globicephalus brought from the Cape of Good Hope, by M. Verreaux, and those of individuals from the northern seas. The present specimen from Kerguelen Island, as well as the five from Tasmania men- tioned at the commencement of these remarks, agree in the form of the cranium and the number of their teeth ( . ) far more closely with skulls of Globicephali from Iceland (as figiu-ed in Van Beneden & Gervais Ostcogr. des Get. pi. li.), the Faroe Islands (Mus. Roy. Col. Sm-geons), the British Coast (British Museum ; figured in Gray Cat. Seals and Whales, 1866, p. 316, fig. 62), and the Mediterranean (as figured in Van Beneden & Gervais Osteogr. des Cet. pi. li.), than they do with the type of Gray's G. macrorhynchus. In all of them a strip of the maxilla is distinctly visible along the whole of the outer border of the rostrum. N 2 100 ZOOLOGY or KKKGUELEN ISLAND. It follows tlunvloiv, jii(Ii::iiii;' IVoiu the ostooloi;ii'al cliaraetors only (i'or tlu> (Inscriptions oftlu> oxtornal app^'aianco of the " blacklish " of the south seas Avhicli we possess at i)n'sent are too vague to furnish good zoological characters) that Globicep/ifi/i almost if not quite identical arc rouml in thr Ntnlli Atlantic and (he ^leditcrranean, as well as in the seas of tlic Cape of (Jood Hope, Jverguelen, Tasmania, and New Zealand. [Several deeayed skulls of GlobiccplKthis were found here and there on the shores of retired inlets iu lloyal Sound, and a pair of these animals often fre- quented a rocky bay adjacent to Observatory Bay, sounding along the edge of the kelp, and coming quite close to the land. — A. E. iv".] Note. — GlohicephaJus macrorhynchus, Gray, Zool. l^reb. and Ter. IStd, p. 83, is a well marked second species of this genus, whose geogra))hical distribution is not yet clearly ascertained. The species was based upon a skull in the jNIuseum of the Koyal College of Surgeons (Osteol. Cat. No. 251'.)) presented by E. D. Bennett, Esq., and thus characterised by Dr. Gray : — '■ Nose of skull short and broad, rounded in front, nearly as broad in the middle " as at the prcorbital notch. Teeth sub-cylindrical !*. Lower jaw rounded in " front." This skull, which is quite adult, l)esid(;s being smaller than that of G. me/as, has the premaxillary bones so wide in their anterior half that they extend from side to side over the whole of the upper surface* of the rostrum, concealing the maxillai completely. A precisely similar skull from (iuadcloupe is figured by Van Beneden and Gervais in the Osteographia des Cetaces, pi. lii. 3, under the name of G. inter- medins, and another, from the Atlantic coast of the United States, by Cope in the Proceedings of the Academy of Natural Sciences of Philadelphia, 1876, ]). 129, as G. brachi/ptencs. This is certainly not the species to which the Kerguelen specimen can be referred. BIRDS.— R. B. SHAUPE. 101 Birds. By B. Bowdler Sharpe, F.L.S., F.Z.S., i^-c, Senior AssMant in the Zoological Department, British Musetim. (Plates VI.-VIII.) I PROPOSE on the present oceasion to give a complete account of the Avifauna of Korguelen Island, founded on the collections in the British Museum (partly made by the Eev. Mr. Eaton, jiartly by the naturalists of the Antarctic expedition) and on the reports of the German and American expeditions, which have been recently published.* The American expedition obtained twenty-one species, the German naturalists recording twenty -three, t With regard to the Antarctic Expedition, Sir Wyville Thomson thus writes : — " This expedition had the extraordinary advantage of having Dr. Hooker attached " to it as one of the assistant-surgeons, and the surgeons to the Erehus and " Terror, Dr. M'Cormick and Mr. Robertson, and the assistant-surgeon of the " Terror, Dr. Lyall, were all zealous naturalists, and co-operated heartily with " Dr. Hooker in his Avork, so that every possible advantage Avas taken of the " sixty-eight days of their stay in Christmas Harbour. Their visit was, however, " in the depth of winter, and although the actual difference between the winter " and summer temperature is not so great as might have been anticipated, the " winter weather is so boisterous and unsettled, that on forty-five of the sixty-eight " days it lilew a gale, and on three days only neither snow nor rain fell." Under these circumstances, therefore, it is not a little creditable to the officers of this expe- dition that they managed to collect a series of nearhj every species obtained by the more recent visitors to the island, while they procu.red several species Avhich none of the latter have met with. Before proceeding to a detailed account of the species, it may not be uninterest- ing to give a list of the birds noticed in Chi-istmas Harbom- by Captain Cook, when he visited the island one hundred years ago. We are able to give a tolerably correct idea of these fi-om the paintings by Ellis preserved in the British Museum. Ellis was a draughtsman employed by Sir Joseph Banks, whom he accompanied on Cook's voyage. All these paintings are w^ell executed, and the species are recognisable ; each one has written on the back the locality, and there can be little * Contributions to the Natural History of Kerguelen Island, made in connection with the American Transit of Venus Expedition, 1874-75; by J. H. Kidder, M.D. 1. Oniithology. Edited by D. Elliott Cones, Washington, 1875. 8v<). pj). i-ix, 1-51. (Bulletin of the Unitcil States National Museum, No. 2.) The same. By J. 11. Kidder, M.D. I'art II. pp. 1-122 (Bull. U. S. Nat. Mus., part 3). Uebersicht der auf der Expedition Sr. Maj. Schiff " Gazelle " gesammelten Vogel. Zusammengestellt von J. Cabanis and A. Kciehenow, J. f. O., 1876, pp. 319-330. •f One species ol" Prion is deducted, as I consider that P. uriel, recorded by them, is nothing but the young of P. Iinliir. (Vide infra.) 102 ZOOLOGY OF KERGUELEN ISLAND. doubt Imt that (Mtlior tli(>sr drawings t hrmsolvos or the actual spccimons from wlikdi tlu\v Avorc taken, CornuHl the types ol' some of Latham's speeies. The following species are figured from the "Island of Desolation," as Cook ealled Kerguolen Island. No. 39. Ossifiti-pj gigmitea. No. 46. Aptenodytos longirostris. No. 43. Prion dosoliitus. No. 54. Stcnia vittiitn. No. 45. EiulyptoH s«ltntor* (fig. mala). No. 59. Chioiiis minor. In the Narrative of the " Wreck of tlic Favorite " f is given a very good list of the birds of Kcrguelcn Island, with iigiu'es of many of the species (pp. 173-199). Who was the author of this list I liav(> not been able to discover, but that assistance was rendered by some one who had access to the collections of the Antarctic Expe- dition I have very little doubt, as nearly every bird procured by the latter is mentioned. Sir J. Hooker, to whose kindness I owe the loan of the work, does not remember the name of the compiler, but he informs me that the late Professor HcnsloAv took a great interest in it. Lastly, Sir J. Uooker has allowed me to incorporate in this paper certain notes extracted from his "Journal," an act of courtesy that I gratefully acknowledge; so that by references and quotations I hope to be able to present to ornithologists a tolerably correct idea of the Avifauna of Kerguelen Island, as at present deter- mined. The notes on the habits of the birds are contributed by the Rev. A. E. Eaton, whose initials are in each case appended. Many of the soft parts of the birds described are taken from the carefid notes made by Dr. Kidder in his " Report," some few from Gould's works. CHIONID.ffi. Chionis minor. Chionis minor, ILartl. Bev. Zool. 1841, p. 5 ; id. op. cit. 1812, pi. 2, figs. 2, 2a, 2 b ; Lafr. t. c. p. 402 ; Gray, List Anseres, §'c. B. 31. p. 52 (1844) ; Gray 8f Mitch. Gen. B. Hi., p. o22, pi. 133 (1845) ; Schl. Uandl. Dierk. p. 4<00,pl. viii,fig. 98 (1857) ; id. de Bier. fig. to p. 232 ; Gray, Sandl. B. Hi., p. 21, No. 10,050 (1871) ; Kidder, Bull. U. S. Nat. Mus. No. it., pp. 1-4 (1875) ; id. Sf Coues, op. cit. Hi., p. 7; Beichen. J. f. O. 1876, p. 89; Cab. Sf Beichen. t. c. p. S27,pl. 1, Cbionarchus minor, Kidder and Coues, Bull. U. S. Nat. Mus. Hi., pp. 85-116. * This figure is not so good as most of them, and apparently represents two immature Penguins, which may be the young of E. saltator. j Narrative of the wreck of the " Favorite " on the Island of Desolation ; detailing the adventures, sufferings, and privations of John Nunn ; an historical account of the Island, and its whale and seal fisheries. Edited by W. B. Clarke, M.D. 8vo. pp. i-xx, 1-236. London, 1850. BIRDS.— 1{. B. SHARPE. 103 Ad. pure all)us, priinarioram rccticumque scapis flavicanti-albis ; rostro nigro ; regionc palpebral! et oculari pallide coccinea ; carunciila lorali nigra ; iridc piu-- pm-asceuti-nigra ; tarso pedibusque sordidc albis, carnco tinctis. Long. tot. 17-5, culmen 1-35, alse 9*4, caudae 5-4, tarsi 1-7. ? mari siiuilis, sed caruncula lorali paullo minore. Long. tot. 16-0, alge 8-9, eaudai 4-9, tarsi 1-6. Juv. similis adiiltis, sed rostro multo minore et reniigibus pallide roseo apicatis. In the American account of their Kerguelen collections, it is stated by Dr. Kidder (Bull. TJ. S. Nat. Mus. iii., p. 89) that he has only succeeded in finding a record of four specimens of this species in European collections. The wi-iter has omitted to notice that in the British Museum "List of Anseres," &c. (p. 52), pub- lished by the late jNIr. G. E,. Gray in 1844, no less than seven specimens are mentioned, and I now give a complete list of the birds at present in the national collection. I would call attention to the fact that the liird I'vom the Crozettes seems to have darker legs than the Kerguelen bird : whether this occm's only in the cbied skin, or is to be found in the living Sheathbill, must be left for future visitors to these islands to determine. a. b. ad. c. juv. st. Kerguelen Island. Lieut. Alexr. Smith. d. ? ad. sk. Kerguelen Island. Antarctic Expedition. f. g. ad. sk. Christmas Harbour, May 30, 1840. Antarctic Expedition. h. ad. sk. Crozette Islands. Captain Armson. Su- J. Hooker's journal contains the following note : — " The young birds have pink tips to their wings. When hard pressed it takes to the water (but this very seldom), and swims slowly." [Where the coast is rocky and sheltered, Sheathbills arc common. In some of the most favovu-able situations they were in ilocks of a dozen, thirty, or more ; but the birds which are breeding live in separate pairs. They frequent the shore to feed between the tide marks upon mussels (whose shells they break by pecking at them), isopod Crustacea, Enteroniorplia, and Tllva. Other kinds of seaweeds do not seem to be relished by them ; one was observed to peck at a piece of Delesseria, but it did not eat it. They are also very assiduous in theii" attendance upon colonies of Shags and Crested Penguins, whose eggs they greedily devour. The sitting birds stretch out their necks and croak at the Sheathbills sauntering past their nests ; but the marauders, keeping just out of reach of their bills, pay little regard to them, and proceed in a business-like manner to eat up the first eggs they may chance to find unguarded. It occasionally happens that while an old Shag is gesticulating violently at a chionis in front of her, his friend pecks from behind at the eggs which iii the excitement of the moment arc not covered comiiletely by her. Wlien she finds out what is taking place, she diives him away with a croak, and true to her sex affects to have won her point in the affray. Eeseating herself upon the 104. ZOOLOGY OF KKUGUELEN ISLAND. nest with ijrcat dii^nity ol' ilcpoi-lniciit, and gvntlv iciilarinu; with lun- bill llic brokiMi **ggs iiiultT luT featliors, slio resigns lierseli' to the task of trying to liatcli theui. The Sheatbbills Avitbdraw. Some time alt(M' IIk y have gone away, I be broken eggs are inspeeted, and if there is only a smal] bole ])eeked in eaeb of lliem, (bey aif kei)r in the nest. Their ai)i)earanee and manner of caressing one auoth(>v led the 1)liie jackets to call Sheatbbills " white |)ig(>ons." In their gait and (light they closely resemble Ptarmigan ; and likt> these they utter their cry when starting on the wing, as well as during (ligb(. After they have attained a fair amount of speed they sail along from time to time Avitb outstri'tcbrd wings. < )n aligbdiig at their destination they often greet one nnotlicv with a gmtlc ebiiekle while (bey nod their beads. Shorth' l)efore they began to build it became evident that they were about to do so. Scarcely a hen could fly anywhere without being attended by two males, 'flie holes that seemed available for nests underwent freqniMit and searching examination by them, and in a short while the nests began to be built. Each male then became extremely jealous of his mate ; and if a neighbour happened to approach too near the nest, he was at once warned oil" by the legitimate owner. Mr. Midshipman Stuart of H.M.S. Volage saw^ some lighting, springing up into the air like partridges. The nest is usually built in a hole between or beliind rocks which constitute the extreme limit of the shore where the land plants meet it. Its position can generally be ascertained approximately at a glance by looking for the most conspicuous boulder on the beach frecpu'nted by the pair of birds ; for Sheatbbills usually alight upon a rock of tliat description just before they enter their hole. Another easy method of finding the nest, is to startle the cock bird whilst he is feeding along the shore alone, for he is apt to fl.y to his mate when alarmed. The holes selected for nests are generally rather spacious within ; and pieces of building materials are frequently dropped by the l)irds outside the entrances. The nest is of very simple construction without a lining ; it consists of a heap of dried seed-stalks of Pringlea antiscorbutiea, or tufts of Festuca erecta, slightly hollowed out on the toj). Occa- sionally old bm'rows of Prion and Halohccna are occupied by sbeathbills : when this is done the nest is constmcted, not in the terminal chaml)er of the Petrel's buiTOW, but a foot or two feet within the entrance of the tunnel, wliich is thus far enlarged for the purpose by the Chionis. One nest was in a burrow of this kind at the base of a large stone standing on a slope amongst Accena, upwards of 10 yards from the shore. The usual number of eggs in a nest is two or one ; three is exceptional. Those in the same nest are not always alike in colour, nor in the style of their marking. The first eggs were foimd on the 23rd of December ; the first brood of nestlings BIRDS.— R. B. SIIARPE. 105 about tlic middle of January. The young are clothed with unicolorous slate grey doAvn. The " sheath " of the bill is immoyablc. In the chick it is concrete with the horny substance of the bill, and is only indicated by a faint line of demarcation. In the course of growth it becomes developed into a distinct lamina, embracing the base of the upper mandible. The bird-louse parasitic on tliis species was not observed by us. The Patagonian ehionis (C. alba), according to Giebcl, is infested with an wnAescrihed Ziothenm. {See Zeitscbr. f. d. gesammt. Naturwiss. Berlin, 1861, Heft. 8, 9, p. 311.)— A. E. E.] ANATIDiE. Querquedula eatoni. (Plate VI.) Querquedula sp., Gray, List Anseres, ^c., Brit. Miis., p. 138 (1844.). Querquedula Kerguelensis, of the "Wreck of the 'Favorite,'" p. 186 (1850, descr. nulla). Querquedtila eatoni, Sharpe, Ibis, 1875, p. 828 ; Coues and Kidder, Bull. U. S., Nat. Mus., i\. p. 4; iid., op. cit. m.. p. 7; Cah. and Beichen. J.f. 0. 1876,^. 329. c?. supra brunneus, plumis plurimis griseo marginatis, rufescenti-fulvo maculatis aut fasciatis ; scapularibvis nigricantioribus ; pileo paullo rufescentiore, plumis nigro medialiter striatis ; facie laterali et gutture albicantibus, minute nigro strio- latis, mento fulvescenti-albo ; corpore reliquo subtus albicante, brunneo marmorato, plumis plerisque pectoralibus versus basin griseo brunneis aut medialiter brunneo striatis ; hypochondriis brunneis, albido terminatis et rufescenti-fulvo transfasciatis ; subcaudalibus rufescenti-fulvis, nigro adumbratis, longioribus nigricantibus fulvo terminatis ; tectricibus alarum superioribus cinerascenti-brunneis, majoribus pallide badio terminatis, fasciam alarem formantibus ; remigibus cinerascenti-brunneis, secundariis extus purpureo aereis albido terminatis, speculum alare sereum vix sub certa luce olivascente nitens exhibentibus ; secundario proximo nigricante vel aspectu externo viridi nitente, mediaUter cinerascente strigato, albo apicato ; secundariis interioribus nigricantibus extus pallide brunnescentibus albo limbatis ; rcctricibus mediis nigricantibus, reliquis brunneis albo marginatis, nounuJlis rufe- scenti-fulvo notatis ; tectricibus subalaribus brunneis, inferioribus intimis et axilla- ribus albis brunneo maculatis, tectricibus majoribus cinerascentibus alaj inferiori concoloribus ; rostro virescenti-plumbeo, culmine nigro ; pedibus oclu'ascentibus vel sordide viridescentibus ; iride iiurpurescenti-nigra. Long. tot. 15*5, alae 8'5, caudae 4-8, tarsi 1-2. The above is a copy of the description published by me in the " Ibis ;" and I then described the female as being similar to the male, but distinguished by the want of o 106 ZOOLOGY OF KRROrELEN ISLAND. the :ilar spoculuiu. Accordini;' to Dr. Coucs' observations, however, the female diffci-s only iu having the green speculum duller and tinged with brown : this Avas seen iu specimens determhied by dissection by Dr. Kidder. It is probably a young female that I diagnosed in the original article as foUows : — ^' mari siniilis sed speeulo alari absente, secuudariis albo tcrmiuatis ; caudft bruniioa, rufeseenti-fulvo fasciatini mannorata. Tliis ]ilain-coloured Teal is allied to Q. (jibherifrons and Q. cvcccoidcs. From the former it is at onee to be distinguislied by the fawu-coloured bar on the wing and the bronzy speculum, the wing-bar being broadly white, and the speculum black in Q. gihherifrons. Q. creccoides resembles Q. eatonl in having the fawn-coloured wing-bar ; but then the speculum is black, and the greater part of the bUl is yellow. Q. eatoni also has the axillaries Avhitisli barred Avith brown, whereas they are quite white in the allied species ; and, moreover, it has remains of rui'ous-bulf bars on most of the feathers of the upper surface, the back being uniform in the other species. a. ? ad. sk. Kerguelen Island. Rev. A. E. Eaton. h. ad. c. d. ad. sk. Kerguelen Island. Antarctic Expedition. [We found this bu'd very tame, and only after the officers from the ships had bagged upwards of 2,000 head within a radius of eight miles of the observatory did the survivors acquire habits of caution, and learn to restrain their curiosity. On the bogs they are conspicuous objects. Until they commence to breed these Teal associate in small " springs," and not like mallards in "bunches." Every day w^hen the tide is out they leave the bogs and bills to congregate upon the rocks and the mud of the estuaries left bare by the ebb. The scene presented at their favourite resorts at these times is an animated one. Hundi-eds of Gulls flocking from their resting places in the neighbouring cliffs are filling the aii- with their clamoiu', and leisvu-ely walking about where the mussels lie thickest upon the mud-bank. The Teal in small parties are busily dabbling along the water's edge and round the stones in tide pools for the abundant isopod Sph(Broma gigas. Here and there a SheathbUl rambles apart over the sea-weed covered rocks. Elegant in form and graceful in their flight the Terns flit lightly to and fro, hovering now and again and plunging into the kelp, or play- fully engaged in clamorous pursuit of their successful companions, presently aKght upon a boulder side by side twittering amicably. Here comes a bustling Cormorant hurrying past, tiu'ning his head from side to side occasionally as he looks about him in his flight, a contrast to the sooty Albatross which is sailing so silently across the sky. On the return of the tide the Teal withdraw from the shore, some of them to the cliffs, others to the liill sides and to marshes bordering streams and lakes, where BIRDS.— R. B. 8HARPE. 107 they strip blossom and seed from the Kcrguelen cabbage {Fringlea), and sift the soft groimd for food. Tlieir nests are built in the large majority of instances in crowded patches of Priiigleo near the sea. Very few were found under solitary plants, and when they Averc the plant was siu-e to be groAving close beside a lump of Azorella. Scarcely ever was a nest placed amidst Acmia only. All of them were thoroughly protected from the weather by the leaves, and well concealed. They were all neatly lined with a thick layer of down. The teal build apart from one another, and not in company, consequently I never found more than thirty nests in one day. Tliere appears to be much ii-regularitv in their time of breeding. The greater number of them had eggs in December, but whilst some had even hatched theirs so early as the day of the Transit {i.e. December 9), others had not begim to lay before the first week in Februaiy. The eggs are few in number. Whether this is due to the effects of inbreeding attendant upon the isolation of the species is open to conjecture. It may be occasioned by the coldness of the climate, as in the case of Ai'ctic ducks. Usually there are only three, but occasionally there are four or five eggs in a nest. This last is the largest number of eggs observed together, and of ducklings in any one brood. The louse of this Teal was not found. The British Q. crecca has a Menopon, a Nirmus and a Lipeiirits. In the South African Museum at Cape Town there are two undetermined speci- mens of Q. -eutoni from the Crozettes. — A. E. -E".] LARIDJE. Larus dominicanus. Gaviota Mayor, Azara, Aptint. ii., p. 338, No. 409. Larus marinus, pt. Vieill. N. Diet. d'Sist. Nat. xxi., p. 507. Larus dominicanus, Lieht. Verz. Doubl. p. 82; Neuioied. Beitr. Natnrg. Brasil. iv., pt. 2, p)- 850 ; Gray, List Anseres, §fC., Brit. Miis. p. 169 ; id. Voy. Ereb. and Terror, Birds, p. 18 ; id. and Mitch. Gen. B. p. 654, ^;Z. 180 ; JReicheub. Sandb. Loiigipennes, pi. -s.xvi., Jig. 883; Schl. 3Ins. P. B. Lari, p. 12; Blasiiis, J.f. O. 1865, p. 378 ; Layard, B. S. Afr. p. 367 ; Gray, Handl. B. iii., p. 112 ; Finsch, J.f. 0. 1872, p. 241 ; Scl. and Salv. P. 3. S. 1871, p. 576; Gurney in Anderss. B. Dam. Ld. p. 357 ; Bxiller, B. N. Zeal. p. 270, pi. 29>,fig.2 ; Scl. and Salv. Nomencl. Av. Neotr.p. 149; Finscli, J.f. O. 1874, p. 203; Slunpe, Voy. Ereb. and Terror, App. p. 32; Cones and Kidder, Bull. TJ. S. Nat. Mm. no. ii., p. 13; iid. op. cit. no. 3; Cab. Sf Beichen. J.f. O. 1876, jx 328. Larus littoreus, Forsler, Descr. Anim. p. 56. Larus antipodus. Gray, List Ansei'es, SfC. p. 169 ; id. Ibis, 1862, p. 245. Dominicanus vociferus, Bruch, J.f. O. 1853, p. 100, 1855, p. 281. o 2 lOS ZOOLOOY OF KERGUELEN ISLAND. Dominicmuis polaj^k-us, BnuL, J. f. O. 1853, p. 100, j,I. 2. fiy. 3, 1855, p. 280; 7.'/). Cnisp. ii.,/). 21 1.; hi. C. 7?. x^lil, p. 770. Doininifaiuis votula, Bmch, 11. cc. pi. 2,Ji>. Cy//.v/>. ii. j). 21 1. Dominicauus autipodus, Bi-nch, II. cc.pl. 2, Jig. 8; lip. Co)/.sp. ii., /;. 214. Larus vcrreauxi, J3p. Bev. Zoul. 1851, />. 7; /'/. T<'w;;/. 1851, p. 211; Cwy, 7/rt«rf/. ^. iu., p. 112. Dominicanus Iritzci, Bnich, J.f. O. 1855, yj. 280; />>. Co//*/;, ii., ^;. 21.1' ; id. C. B. xfii., p. 770. Doiiiinicanus vrnvauxi, Brnch, J.f. O. 1855, p. 281. Dominicauus azariu, Bp. Consp. ii., p. 211. Clupoilarus verreaiixl, Bp. Consp. ii., 7;. 221, i^- 38 ; Gould, B. Ansfr.pl. 21; Ilulfon, Ibis, 1867, p. 185, 1872, p. 248; Fimcli, J.f. 0. 1872, p. 210, 187 i, p. 203. 110 ZOOLOGY OF KERGUELEN ISLAND. Lcstvis antarcticiis, Lesson, Traift; p. GIG; Scl. 1\ Z. S. 18G0, p. 390; Jhholl, Ibis, 18G1, p. 105; Scl. .$• Salv., P. Z. S., 1871, p. 579 {pi.) ; lid. Nomencl. Av. Neolr. p. l-t8 ; Cab. S^- Beichcu., J.f. O. 187G,^j. 328. Stert'oravius autarcticus, Grny, List Anseres, Brit. jffKs. p. 1G7 ; Lp. Consp. ii. p. 207 ; Fclz. Jteis. Novam, Tog. p. 150 ; Giglio/i, Faun. Vcrtebr. Oceano, p. 61 ; Bitllei', B. N. Zeal. p. 207 ; Saunders, P. Z. S. 1870,^3. 321. Mogalcstris antarctica, Gould, P. Z. S. 1869, p. 98. Lostris fuscus, EUman, Zool. 18G1, p. 7172. Buphagus autarcticus, Cones, Fr. Phil. Ac. 1863, p. 127 ; /(/. B. JV. TFest, p. 604. Storcoravius catan-actes {nee L.) ; Schl. Mus. P. B. Lari, p. 45 ; Gould, Ilandb. B. Austr. ii., p. 389 ; Layard, B. S. Ap. p. 366 ; Slunpe, Voy. Ereb. Sf Terror, App. i). 32. Buphagus skua autarcticus, Cones 8f Kidder, Bull. U. S. Nat. Mus. No. 2, p. 9; iid. op. cif.. No. 3, p. 9. Ad. t'uliginoso-hruuucus, pliuuis dorsalibus ct scapulaiil)us albiclo terminaliter lavatis, colli postici plumis et quibusclau\ dorsalibus medialiter fulvo striatis ; alis saturatius brimneis, angustc cineraceo marginatis ; romigibus cinerascenti-brunneis, scapis albidis, primariis versus basin conspicue albis ; dorso postico ct txropygio cinerascenti - brumieis, supracaiidalibus magis cbocolatinis ; cauda cincrascenti- bvunnea, versus apieem nigricante, scapis albis : corporc subtus toto fuliginoso, plumis paucis sordidc ferruginco lavatis, hypochondriis quibusdani obscure albicaute lavatis ; subalaribus satiu*ate brunneis, majoribus et remigibus subtus cinerascenti- brunneis, primariis intus ad basin conspicue albis ; rostro nigro ; pedibus ex vires- cente scliistaceo-nigris ; inde intense clialybeo. Long. tot. 23, culm, 2-8, alac 16-3, caudie 00, tarsi, 2-8. The above description is taken from the only skin brought by Mr. Eaton, and is in nearly uniform brown plumage. Two other skins are in the Museum, from Christmas Ilarbom', where they were prociu-ed by the Antarctic Expedition. These two show more clearly the buff streakings on the neck, and have the feathers of the mantle mottled with yellowish buff or whitish centres. Mr. Howard Saunders, who has examined om* series, has called attention to the remarkable colouring of the skins obtained by the Antarctic Expedition in the Southern Seas among the ice. They are very light coloui'cd, and have quite a fringe of yellow round the neck, while the peculiar pale ashy brown colour of the under parts is very conspicuous. I agree with Mr. Saunders that it would not be wise to sepai*ate these Antarctic bh'ds from the ordinary S. antarcticus -ndthout further information regarding them, for the Kerguelen bu-ds show a certain approach to the golden neck-feathers ; and moreover, I find on referring to the registers that one is said to be from the Auckland Islands. It may have been sent home in the same box as the Auckland col- lection, and registered by mistake with them. It still preserves its original ticket. BIRDS.— 1{. B. SriARPE. Ill " Antarctic Gull, Male," but no locality is attached, and its companion bird, a female, is also without locality. These are the two birds catalogued l)y Mr. Gray in his List of Ansercs (p. 168) as from the Antarctic Seas. I further agree with Mr. Saunders that no one comparing the two species of Skuas would unite aS*. antarcticus with ^S*. catarractes, the bills being so dillerent, as will be seen in the figures in the plate. Besides the undermentioned specimens from Kerguelen Island, the British Museum contains examples from South Africa {S'u- A. Smith), Campbell Island {Lieut. A. Smith), Antarctic Seas, and the Pack Ice, Antarctic Ocean {Antarctic Expedition), New Zealand {Sir G. Grey), Norfolk Island {F. 31. Raijner), and from Lat. 36^ 8' S., Long. 88° 55' E. {Sir G. Grey). a. ad. Royal Sound, Kerguelen Island ; Rev. A. E. Eaton. h. c. ad, Christmas Harboiu", Kerguelen Island ; Antarctic Exj)edition. [Every marsh near Royal Sound used to have its pair of Skuas. Many were destroyed within a radius of four miles from the ships ; and before the expedition sailed from the island it was impossible to walk far without coming across dead bodies of the poor creatures. The cause of this useless slaughter was the menacing aspect of the bii'ds, who swooped with fierce impetuosity dii-ectly towards the face of any one approaching their domain, rising only just in time to clear his head, and uttering short despairing cries. They did not feign to be crippled quite so much as the Skuas in Spitzbergen, but preferred intimidation as a means of averting danger from their nest. When they thought they had succeeded in making the enemy retreat, they celebrated theu' triumph standing face to face upon the ground, with their wings extended vertically so as almost to meet above their back, whilst one or both loudly chaunted a paean, consisting of a dozen notes or so delivered in the tones of a carrion crow. In October they also used to croak now and then dming theu- flight ; and this croak, which was discontinued in the breeding season, was very like the lower croak of a raven. Indeed, it was at fii'st difficult to re- assure oneself that they were not a species of Corvus as they circled in the air far oif , and the Blue Jackets used to call them " Black Crows " for some time, but before long the designation "Molly-hawks" came to be applied to them. This change of name took place at the commencent of petrel digging. If Blue Petrels were tm-ned loose in the daytime, they were almost invariably chased by Skuas, and killed on the wing before they had flown half a mile. Petrels of one sort or another seem to constitute the staple diet of these Skuas. They hunt for them in the evening when it is l)ecoming dusk, flying rapidly along the hillsides close to the ground, like hawks, ready to pounce upon any that they may see emerging from the mouth of their bmTows. Again in the early morning they are upon the wing to waylay Petrels returning late from the sea. Nor are they idle during the rest of the day. I have mentioned their fondness for eggs in my paper in the ' Proceedings.' U-2 ZOOLOGY OF KERGUEI.EN ISLAND. The nosl is built amongst Azorr/la, wlioro tlio ^touikI is dry and slii^hily raised. It consists of a hollow scraped in the soil lined Mith dead tufts of Feshtca crccta. Tlio e£:c:s are two in number, and do nut vary iiiiuli in colonr. The statements in Capt. Hiitton's paper on Birds inhabit ini;- the Southern Ocean [Quart. Journ. Sc. (180(5), vi. 77], that tlu> Iver^uelen Skua breeds on flats among grass two feet liigh, and lays three eggs, and that the young are dark brown spotted with wbite, do not, accord with our observations. They young are dark brown, A\ithont any pale i^pots Avhatever. Tlu> ordinary food-call of the nestlings is rather i)laintiveand tremulous ; they also quack like Mallards. The old Skuas were much puzzled when they saw rabbits come out of Petrels' holes. They hovered for a long time over their heads, and at length used to stand beside the mouths of the burrows waiting for th(> young ones to creep forth, just as if they were watching for Petrels. It is doubtful Avheth(>r they will succeed in ridding the island of these mischievous vermin, although the young birds reared by me readily fed upon rabbits procuicd with tlie sling. Twenty or tliirty adult Skuas used to assemble every afternoon upon a small sheltered lake near Swain's Harbour, where they washed and basked. Under the name of S. catarrhactes, two examples of this Skua from the Crozettes arc exhibited in the South African Museum. Louse not found. Stercororhts serves as host to species of Docophorus, Nirmus, Lipenrus, and Colpocephaliis (Giebel). — A. E. E^ Sterna virgata. Sterna arctica, (pt.) Gvay, List Anseres Brit. Mus. p. 178 (1844). Sterna virgata. Cab. J. f. O. 1875, p. 449 ; Smmders, P. Z. S. 1876, p. 64G ; Cab. 8r Beichen. J.f. O. 1876, i?. 328. Sterna vittata. Cones, in Kidder's Report, Bull. TJ. S. Nat. Ifus. ii. p. 17 (1875, nee. Gin.) ; Cones 8f Kidder, op. cit. m., p. 11 (1876). Ad. supra fumoso-cinereus, tectricibus alarum dovso concoloribus ; remigibus fumoso-cinereis intus albicanti-cinereis, scapis pallidissimc brunneis, primario primo extus nigricante, scapa piu-e alba, secundariis albo terminatis ; uropygio et supra- caudalibus albis vix cinereo lavatis ; rectricibus cinereis, intus albis ; pileo summo uuchaque nigris ; facie laterali pallide cinerea ; striga lata supereiliari a basi narium ducta, alba ; corpora reliquo subtus fumoso-cinereo, subalaribus et subcaudalibus concoloribus, iis imis vix albicantibus ; rostro et pedibus corallinis ; iridc indigotico- uigra. Long. tot. 12, culm. 11, alee 103, cauda; 3'1, rectrice extima 5'4, tarsi 0-65. Juv. si;pra nigricans, plumis ad basin cinereis, subterminaliter nigris, ochrascenti- brvmneo marginatis et transfasciatis ; lU'opygio et supracaudaHbus albicanti-cinereis, ochi'ascenti-brunneo lavatis ; remigibus secundariis et rectricibus eodem modo quam dorsum tevminaliter marmoratis ; regione parotica cinerea postia nigra ; corpora BIRDS.— U. 15. SHARPE. 113 subtus ubrasconti-brunneo, nigricauti-brunuoo transvcrsim marmornto, plumis ad basin albis, subalaribus pure albis. Tbis seems to be quite a distinct species. It is closely allied to Sterna ant- aretica of New Zealand, and in coloration is almost exactly similar. The latter bird, however, according to Dr. Bulhn- (B. N. Zeal. p. 238), has the bill and feet " bright yellow," whereas these j^arts in S. vittata are coral-red. a. iinm. sk. Kerguclen Island. Antarctic Expedition. h. ad., c. juv. Royal Sound, Kerguelen Island. Rev. A. E. Eaton. [Crustacea and the yoimg of Notothenia inhabiting the kelp afford subsistence to the Terns in Kerguelen Island. Skimming along the shore in pairs and threes, and over the belt of Macroeystis, these with their graceful flight and elegant form enliven the prevailing desolation of the coast scenery. In general their eggs are laid on barren stony ground one or two hundred yards from the beach, and the brows of terraces of the seaside hills are more frequently selected for the nesting places than any other situations. There is rarely (if ever) more than one egg in a nest, and this is deposited in a slight hollow amongst the stones,' and generally upon the bare soil. The birds breed irregularly throughout the early summer months. Fresh eggs were taken by the officers of the " Volagc " about the third week in November ; a nestling recently hatched was found on the 8th of December, and on the 23rd of the same month a young Tern fully fledged was shot while on the wing. Its beak and feet were black. Another yoimg one able to fly was kept for some time on board the " Supply." It became quite tame, and ate boiled rice and soaked bread readily. After we had been a few days at sea it died, probably through want of water. The adult examples in the collection were secured in the absence of a gun by a couple of stones thrown at them consecutively from a sling while they were passing overhead. A third specimen was caught by hand during our last ten minutes on the island. It was sitting upon a rock near the pier with the eye on my side of its head blinded, so I crept up quietly and laid hold of it. The two specimens marked as Sterna meridionalis from the Crozettes in the South Afi'ican Museum are identical with the Kerg^^elen Tern. The louse of this species was not obtained. Sterna is infested with species of Nirmus, Docophorus, Colpoccphalus, and Analges, according to Giebel. A. E. E.] Sterna vittata. "Wreathed Tern, Lath. Gen. Syn. iii., j)^- 2, p- 359. Sterna vittata, Gm. S. N. i., p. 609 {ex Lath.) ; Lath. Ind. Orn. ii., p. 807 ; Felz. Heis. Novara, vog. p. 152 ; Saunders, P. Z. S., 1876, ^j. 61^7. Ad. pileo nuchaque nigen'imis ; corpore reliquo supra clai'6 cinereo ; tectricibus alanun dorso concoloril)us ; remigil)us cinereis, intus albis, primariis extus nigri- canti-cinereis, rachidibus albis ; secundariis et scapularibus conspicu^ albo terminatis ; uropygio ct supracaudalibus pure albis ; caud^ maximc:^ f m-catii, purd alb& ; facie p Ill ZOOLOGY OV KERGUELKN ISLANO. laterali cinoivft, fasciA distincta a navium liasi, iufrii oculos usque ad uucliam latcr- alciu duc'tu, c't UKu-ula inl'raooulari, all)is ; covporo relicpio subtus saturate cinoreo, giilA inaifis albicantc ; tibiis, crisso, subalavibus et sul)t'audalibus, pur^ albis. Long. tot. 15-3, culm, li, aliD 10"0, caudse 3-0, rcctric. cxt. 7"0, tarsi 0-7. Juv. siuiilis a\-i juvoni S. rirgatiP, ot codom modo nigro ct ochrasccnti-bruunco varius. The view of Ilevr Von Pclzcln, and Mr. Howard Saunders, who rel'or this Tern to the Sterna viltata of Gmolin, is doublless connect. The latter gentleman ob- sei"ves : — " Gmelin's description, rounded on Latham, i'uirly suits this species, " although I am inclined to doul)t the correctness of the locality assigned, viz., " Christmas Island, especially as there is no mention in Cook's A^oyages of any Tern " being found there, except the Sooty Tern, of wliich there is a full description." Many of the bii-ds described by Latham were taken from specimens or drawings in the collection of Sir J. Banks. A list of the few birds of Kcrguelen Island as noticed on Cook's voyage has been drawn up at the commencement of the present paper, and I believe that No. 54, representing the Tern from tbe " Island of Deso- lution." fonned the type of Sferna vittata, and that the locality Christmas Island is a mistake for Christmas Harbour, which was the place where Cook's expedition landed on Kerguelen Island. The picture represents S. vittata, and not S. virgata, and tliat the species occm's there is proved by the vmder-mentioned specimen collected in the island by the Antarctic Expedition. Of. c? ad., ft. juv. S.Paul's Island, 3diQ..\dih^ {J.Macgillvvray). Capt. Stanley. c. ad. Kerguelen Island. Antarctic Expedition. PROCELLARnD.ffi. Pelecanoides iirinatrix. Tec-tee, Forster, Icon. Ined. No. 88. Diving Petrel, Forster, Voyage i. p. 189 ; Lath. Gen. Syu. iii. x>t- 2, p. 413 ; id. Gen. Hist. X. p. 194. Procellaria urinatrix, Gm. S. N. i. S. 560 {ea^ Lath.). Pelecanoides urinatrix, Lacep. Mem. de I'Inst. 1800, j3. 517 ; Gray, List Anseres Brit. Mus. p. 158 ; id. Gen. B. iii., p. 646 ; id. Voy. Ereh. 8f Terror, Birds, p. 17 ; id. Ibis, 1862, p. 243 ; Goues, Br. Bhilad. Acad. 1866, p. 190; Gray, Mandl. B. iii., p. 102, No. 10825 ; Sharpe, Voy. Ereb. 8f Terror, App. p. 33 ; Buller, B. N. Zeal. p. 313 ; Cones Sf Kidder, Bidl. N S. Nat. Mas. ii., p. 36 ; iid. op. cit. iii., p. 17. Halodroma tu-inatrix, Illiger, Prodr. Syst. Manim. SfC. p. 274; Beichenb. Sandb. Longipennes.pl. is.., Jigs. 762, 763 ; Bp. Consp. ii., p. 206 ; Schl. Mus. B. B. Brocell. p. 37 ; Gould, Randb. B. Austr. ii., p. 483 ; Finsch. J.f. O. 1872, p. 256, 1874, p. 210 ; Cal). 8r Beichen. J. f. 0., 1876, p. 328. Procellaria tridactyla, i^orsifer, Bescr. Aiiim.p.l4<9; Bllmau, Zool. 1861,^. 7473. BIRDS.— R. B. SHARPE. 115 Puffiniiria urinatrix, Gould, B. Austr. vii., pi. 60. Pelccanoides herardi, Quoij et Gaim. Voy. Uranie, ^;. 135, i)l. 37 ; Temm. PL Col. 517 ; Gra//, List Anseres Bril. Mus. p. 158 ; id. Gen. B. iii., p. 046 ; Gould, P. Z. S. 1859, 1). 98 ; Scl. P. Z. S. 1860, p. 390 ; Cones, Fr. Bhilad. Acad. 1866, p. 190 ; (?;•«//, iZ««f?;. B. iii., ^j. 102, JVb. 10827 ; Buller, B. N. Zeal. p. 314. Halodroma berardi, Beichenh. Sandb. Longipennes,pl. \x.,fig. 764; Bp. Consp. ii., p. 206; Schl. Mus. P. P. Procell., p. 38; Pi/WcA., J"./. 0. 1874, ^x 210; Scl. 8f Salv. Nomencl. Av. Neotr. p. 149. P Ad. supra niger, plumis basaliter cineraccis ; scapularibus magis cinereis albo terminatis ; alis dorso concoloribus, primariis intus bininnescentibus, secundariis magis cineraceis angiist^ albo terminatis ; rectricibus uigricantibus, intus ciiieraceis angust6 albido terminatis ; loris et regione oculari uigricantibus ; regione parotic^ obscui'6 cinerea, phimis vix albo terminaliter marginatis ; genis et corpora subtus toto albis, genis, guttm-e imo, colli et pectoris lateribus cineraceis albo terminatis ; bypocbondriis cineraceo lavatis; subalaribus albis, majoribus cineraceo lavatis; remigibus infra cinerascenti-brunneis ; rostro iiigro, ad basin mandibulse coeruleo ; ii'ide cinerascente ; pedibus cceruleis. The specimen described above is one of Mr. Eaton's. Considerable variation exists with regard to the amount of grey mottling on the throat and sides of the body and neck, but I cannot determine whether this arises from difference of sex or age. Mr. Gould has figured the adult bird with a pure white breast, and the young one with the under parts shaded with grey, but neither of them have the banded appearance which is apparent in most of the specimens in the British Mviseum. The young bu-ds in the collection, some of which have remains of down still adhering to them, are quite as pure white on the breast as the adults. The principal differences between Pelecanoides urinatrix and its allies P. berardi and P. (jarnoti, have been stated by authors to consist in size and the colom* of the feet. Thus Professor Schlegel {I. c), who is followed by Dr. Coues in his well- known paper (Pr. Philad. Acad. 1866, p. 190), separates them on characters which may be tabulated as follows : — a. feet blackish ; bill longer and more slender ; size larger . . garnoti. h. feet bluish : bill robust ; size smaller ..... urinatrix. c. feet pale with black membranes ; size small • . . berat^di. The colour of the feet, therefore, plays an important part in the differentiation of the species, so that it will be of interest to note the colouring of these organs, as given in variovis works. Thus for P. urinatrix Dr. Eadder gives the " tarsus and foot lavender blue," while Mr. Gould writes: — "Tarsi and toes beautiful light blue; webs trans- parent bluish white, tinged with brown." Forster gives the "feet blue, the soles and webs black," while Dr. Duller has them as follows: — "Legs and feet cobalt, tinged with green, the wcl)s bluish white." According to the last-named observer, in P. berardi the legs and feet arc " yellowish with dark webs." p 2 116 ZOOUHiY 01' Ki:KCiUi:LKN ISLAM). It is, therefore, no wonder, \\ i(li ilie diserepaney in eoloiir incntioncd in tli(> works of the above-named authors, lliat Dr. Cones, after study in i;- the scries brouglit home by Dr. Kidiler from Kergueleu Laud, sliould observe as foUows : — "As very strongly intimated in my paper, satisfactory diagnoses ol' tlu; three currently reported species nl' tlie genus are wanting. ]S\)r is my faith in ilicir disliiietncss increased on linding that these s|i('ciiuens, w liici: IVoni (lie Ideality iindnnhh-dlv represent the original P. lO'i/Ki/rix, are fully up to the dinu^iisions of tlu; supposed larger P. ganiotl from the west coast of South America. Observed variation in the colour of the feet, which is one point that has been relied upon, lessens the probability of distinctness, especially as the ascribed coloration does not coincide in every ease witli the dimensions. The siz(> and proportions of the examples examined, as carefully measured in the flesh by Dr. Kidder, warrant iiic in adducing the P. garnoti of Lesson as a synonym of P. nrinutrix, to which I still refrain, however, from adding the P. hcrardi of Quoy and Gaimard." (Report ; Bull. N. S. Mus. N. H. /. c.) Although my measurements of skins do not agree exactly with those of Dr. Kidder made from the recently-killed bii-d, I add the dimensions of the series in the British Museum, arranged according to the size of the bill, irrespective of localitv. Total Length. Ciihiicn. Wing. Tail. Tarsus. (1. Auckland Isles {Antarctic E.rp.) - - - 7-4 0-55 4-15 1-6 0-6 b. Kcrgiielen Island {A. E. Eaton) - - - 7-9 0-6 4-7 1-75 10 c. Strait.s of M.igelhm - . . . 7-5 0-6 4-45 1-65 0-95 d. $ Kei-gueleii Island {A. E. Eaton') 7-6 0-65 4 •55 1-75 0-95 c. Christmas Harbour (Antarctic E.rp.) 9-3 0-65 4 95 1-65 10 f. Auckland Islands {Smith) . - - - 7-6 0-65 4-85 1-8 10.-) 9' i> )) ») " 8-0 0-65 4-7 1-7 1-05 h. Straits of Magellan . . . - 7-5 0-65 4-7 1-7 1-05 i. New Zealand {Hector) - - - . 90 0-7 51 1-75 105 k. Kerguolen Island {Antarctic Exp.) 8-0 0-7 4-55 1-6 1-05 •• « >» » " " " 8-0 0-7 4-75 1-65 0-95 m. New Zealand (^Sir George Grey) - - - 7-8 0-7 4-6 1-65 0-95 n. Chili {Brydges) ----- 80 0-8 5-2 1-7 1-25 "• » « - 7-7 0-8 5-4 1'8 1-25 p. Valparaiso (Brett) - - — 0-8.5 5-45 1-7 1-3 Although the general length of the skin in these birds goes for very little, the bill is a reliable character, and from the above table of measurements it will be seen that even in the same localities the size of the bill and feet is not constant. My con- clusions differ from those of Dr. Cones, insomuch that I consider that P. berardi is nothing but the young bird of P. tirmatrix, and that P. garnoti on the contrary must be held to be distinct on account of its very much larger size ; at aU events the examples from Western South America indicate a distinct race. BIRDS.— R. B. SIIARPE. 117 Tlic British Museum possesses specimens from the Auckland Islands {Lieut. A. Smith, Antarctic Expedition), from New Zealand {Br. Hector, Sir G. Grey), and from the Straits of ^lagellan, as avcII as the following from Kerguelen Island. a.l). c?. ?. ad. Observatory Bay, Kerguelen Island. Her. A. E. Eaton. c. ad. Christmas llarhour, July 20, 1810. Antarctic Expedition. (/. ad. Kerguelen Island. Antarctic Exj)cdition. [This bird, as Prof. "Wyvillc Thomson well observes, has a close general likeness to Mergidus cdle. Both of them have a hun-icd flight ; both of them, while flying, dive into the sea without any interruption in the action of their wings, and also emerge from beneath the surface flying, and they both of them swim with the tail rather deep in the water. But tliis resemblance does not extend to other particulars of their habits. The Rotche, when breeding, usually flies and fishes in small flocks of six or a dozen birds, and builds in communities of considerable size, which are excessively noisy. Diving Petrels, on the other hand, are more domestic in their mode of hving, fishing and flying for the most part in pairs or alone, and building sporadically. They had begun to pair when we reached Kerguelen Island. The first e^^ was found on the 31st of October. Then' burrows are about as small in diameter as the holes of Bank Martins {Cotyle riparia) or Kingfishers {Alcedo ispida). They are made in dry banks and slopes wbere the grovind is easily penetrable, and terminate in an enlarged chamber on whose floor the egg is deposited. There is no specially constructed nest. Some of the burrows are branched, but the branches are without terminal enlargements, and do not appear to be put to any use by the birds. Before the egg is laid, both of the parents may be found in the nest- chambei', and may often be heard moaning in the daytime : but when the females begin to sit, their call is seldom heard, excepting at night, when the male in his flight to and from the hole, and his mate on her nest, make a considerable noise. There seems to be a difference of a semi-tone between the moans of the two sexes. The call resembles the syllable "oo " pronounced with the mouth closed while a slurred chromatic ascent is being made from E. to C. in the tenor. This kind of Petrel has much difficulty in taking flight from ground which is comparatively level ; it it only by running against the wind, or by starting from a lump of Azorella, that the birds are able to rise upon the wing if they happen to ahght upon a flat. During my walks on calm nights I used frequently to hear them fluttering along the ground in the dark, and (if I had a lantern) easily caught them by uncovering the light and turning it on to them. They sometimes lay still in my hand without trying to escape ; hut when they flew oflt' from it, they did so in a manner which showed that they were not at all crippled. They flew to light on board U.M.S. " Supply " on dark nights in October, wlicn there Avas snow upon the deck. lis ZOOLOUY OF KERGUELEN ISLAND. There is a pair of P. itri)wtria- from tlie Crozcttos in tlio S. Africau Museum. One of the birds dug out near Observatory Bay was inconveniently crowdcnl Avitb a species of Fitlcv and JSlyniiis setoiiis, sp. n. A. E. E.'\ Daption capensis. White and biaek-spotted Petrel, IJdioariJs, N. II. Birds, pi. 1)0. Le Petrel tachcte vulgairement Damier, Briss. Oi'n. vi., p. 14(5. ProccUaria capensis, Linn. S. N. i., p. 213 ; Uahn Sf Kuster, Vog. Asieu, Lief, xix., taf. G ; Gray, List Anseres, SfC, Brit. 3Ins. p. 164; Schl. Miis. P. B. Proeell. p. 14 ; Pels. Beis. Novara, Fog. p. 146 ; Layard, B. S. Afr. p. 361. Daption capensis, Steph. Gen. Zool. xiii., p. 241 ; Gould, B. Austr. vii., ^;/. 53 ; Bp. Coi/sp. ii., p. 188 ; Beichenb. Ilandb. Longipennes, taf. xii., Jigs. 337, 338 ; Lttwr. N. Am. B. p. 828; Gonld, Ilandb. B. Austr. ii., ^3. 469; Goues, Pr. Philad. Acad. 1866, p. 162 ; Degl. and Gerbe, Orn. Eur. ii., p. 469 ; Giglioli, Faun. Vertehr. nelV Oceano, p. 46 ; Cab. Von der BecJcen Beis. iii., p. 62 ; Finsch 8f Sartl. Vog. Osfafr. p. 816; Finsch, J. f. 0. 1872, p. 256, 1874, |x 208; Buller, B. N. Zeal. p. 299 ; Scl. and Salv. Nomencl. Av. Neotr, p. 149 ; Sharpe, Voy. Erebus and Terror, Birds, App. p. 33 ; Kidder, Bull. U. S. Nat. IIiis. p. 39, note ; Cab. Sf- Beichen. J.f. 0. 1876, ^5. 329. Procellaria pimctata, Ellman, Zool. 1861, p. 7473. Fulmarus capensis, Gray, Sandl. B. iii., p. 107. Calopetes capensis, Sundev. Av. Ileth. Tent. p. 142. Ad. Scliistaceo-niger, albo variegatus, plumis ad basin cineraceis subtermiualitcr albis, dorsi plumis ad apicem triangulariter schistaceo-nigris ; scapu.laribus dorso con- coloribiis et eodem modo notatis ; tectricibus alarum minimis nigricanti-brunneis; re- liquis brunneis ad basin albis et august^ albo extus limbatis, majoribus intimis purd albis, quibusdam bruiiiieo terminatis ; tectricibus primariorum ct rcmigibus brunneis, intiis albis brunneo terminatis ; caud^ alba lat6 brunneo terminata ; facie laterali pileo concolori ; fascia parva suboculari alba ; gula summa brunnea, plumis ad basin Delate albis ; gutture et colli lateribus albis, plumis brunneo terminatis ; corpore reliquo subtiis pur6 albo, axillaribus, tectricibus subalaribus majoribus et subcauda- libus albis brunneo terminatis ; subalaribus omnibus marginem alarum formantibus nigricanti-brunneis ; rostro nigricanti-ln'unneo ; pedibus satm'at^ brunneis. Long. tot. 15'5, culm. 1"35, alee 10*5, caudse 4*0, tarsi 1*7. There seems to be very little variation in the plumage of these birds, but the throat is more uniform dusky iu some individuals, while in others it is mottled with white ; the amount of white mottling on the mantle also varies somewhat, the more uniformly colovired birds being apparently the younger ones. Mr. Eaton did not bring home a specimen of the Cape Petrel from Kerguelen Island, but he noticed it in the vicinity, as did also Dr. Kidder (1. c.) It is included BIRDS.— R. B. SHARPE. 119 in the list of Drs. Cabanis aud lleiclicuow, as having heen obtained at Kci't^uelcn, and John Nunn's narrative speaks of it as occumng there, being called Ijy him and his companions the " Spotted Eaglet," or "Spotted Night-hawk." lie says it was extremely abundant near the island. As will be seen below, the British Museum possesses two specimens captured in the vicinity by the Antarctic Expedition. That the Cape Pigeon breeds on Kcrguelen Island is proved by the following note in Sir J. Hooker's Journal : " It builds in sheltered ledges of cliffs about 50 or 100 " feet above the level of the sea. I found two on a nest, but quite matm-e. Its " note is a short hoarse croak." In the British Museum are specimens from the Cape Seas {Caj)t. Harry, E. M. Langioorthij), Lat. 34° 37' S., Long. 22° 29' {D. Bleioitt), Western Australia {Sir G. Grey), New Zealand {Br. Sector), oif Valparaiso {Capt. W. S. Brett), as well as the following. a. b. ad. Oil Kerguelen Island. Antarctic Expedition. [Cape Pigeons were plentiful on 9th of October near Bligh's Cap, and on the foUo\^dng day there were many of them off the coast in the neighbourhood of Mt. Campbell. A few followed the ships into Observatory Bay, but they did not stay there. I do not know whether they ever breed in the island. In December and the first week in January there were many of them stUl about the entrance of Royal Sound and Swain's Bay ; but when we passed by at the end of the latter month none were visible in either of these places, nor did we see any in March between Kerguelen Island and the cape. It is possible that their nests might be found in the cliffs of the Prince of Wales Eoreland and the islands in the mouth of Swain's Bay, or in those of the more exposed parts of the coast farther west ; but there is no evidence of their breeding there. A. E. E.^ Majaquens .ffiquinoctiaUs, The Great Black Petrel, Edwards, Nat. Hist. B. ii., pi. 89. Le Puffin du Cap de Bonne Esperance, Briss. Orn. vi., ]j. 137. ProceUaria sequinoctialis, Liim. S. N. i., p. 213 ; Bitrm. Th. Bras, iii,, p. 3i5 ; Schl. Mus. P. B. Brocell. p. 19 ; Lmjard, B. S. Afr. p. 360 ; Finscli 8f Eartl. Vog. Ostafr.p. 817; Cab. ^ Beichen. J.f. O. 1876, j;. 329. ProceUaria sequinoxialis, Vieill. N. Diet. d'Mist. Nat. xxv., p. 422. Puffinus sequinoctialis, Steph. Gen. Zool. xiii., j^^- h P- 229; Gray, List An- seres, SfC, Brit. Mus. p. 160 ; id. Gen. B. iii., p. 647 ; Beichenb. Handb. Lon- gipennes, pi. xii., figs. 340, 341 ; Pelz. Beis. Novara, p. 142. ProceUaria nigra, Forsier, Descr. Anim. p. 26. Priofinus sequinoctialis, Jacq. Sf Bucher. Voy. Bole Sud. iii., p. 146. Majaquens sequinoctialis, Beichenb. Handb. Lougipennes, pi. xii., figs. 340, 341 ; Bp. Consp. ii., p. 200 ; id. C B. xlii., p. 768 ; Cones, Br. Bhilad. Acad. 1864, p. 118; Giglioli, Faun. Verlebr. Oceano, p. 35; Scl. Sf Sale. Nomencl. Ac. 120 ZO()LOGY OK KKKC.UELEX ISLAND. Neotr. p. Ill); Cones and Kidder, Bull. I'. S. i\'(iL Jliis. U., j). 2') ; /id. op. oil. iii., ji. 13. Pulmanis a>quinoc'tialis, Gray, Hand/. />'. \\\.,j). lOS, no. 10i)15. 9 Ad. supra l'uligiuoso-ni!»er, pluiuis (lorsalil)us vix Iji-iniiioscciitc iiKir!;inalit(>r lavatis; alls niu:ric'auti-l)ninncis, tectricibus alaniin l)riinnc() lavalis; ivmiyilms nigiricanti-brunncis, iutus i-hocolatinis, S('a))is iirimarionini ])allid(' et conspifuo In'iuiiiois ; uropygio ot siipracaiulalil)us ciucroo-uigricautibus ; caudilk, finerco-nigrA, reetrk'ibus mediauis nigricauti-brimneis ; facie et coUo lateralibus fuliginoso-uigris, fasciii lati\ e geuis medianis per mpntmii durtA, all)A; cor])orc rcliquo subtus fuligi- noso-nigro, plumis omnibus bruunco lavatis; subalaribus l)ninnois, cinorasconte lavatis; rostro virescenti-albo, suturibus iiigris ; iridc iiit:ni. Long. tot. 19'5, cuhncn 2"55, al;r 15'0, cauda^ 5"5, tarsi 21'. The above description is taken from the only skin brougbt back by ]\lr. Eaton. The Tvhite mark is "well developed on the chLii and the fore part of the cheeks, occupying all the latter, excepting a bi'owii patch at the base of tlic lower mandible. The way in which the wliite facial markings vary is shown in the following examples : — 1. ? Pacific Ocean. Sir Joseph Banks (spec, a of Gray's List of Anseros, ISM, p. IGO). The white mark is conlincd to a small spot ou the chin. The locality is doubtless erroneous. 2. c? Simon's Bay, Cape of Good Hope. The white mark occupies the entire chin and upper throat, and encroaches slightly on the centre of the cheeks. 3. Cape of Good Hope. The wliite marking is developed on the chin as in the preceding bird, occupying ou one side the anterior portion of the cheeks, but totally absent on tlie other cheek, which is sooty brown. This is spec, e of Gray's list, con- sidered by him to be a variety or a young bird. Considering that the Australian 31. consplcillatus only differs in the extent of the white markings on the head, it might be supposed that it was not really distinct from M. (vquinocflalis, in which these white markings form such a varying pecu- liarity, the extent of white being, as shown above, sometimes not the same on both sides of the face. The AustraHan bird, however, has a white band across the crown, and a second one extending from the cheeks round the side of the head below the ear-coverts, as well as a white patch on the chin ; and in addition to this the nostrils and sides of the mandibles are yellowish horn-colom', according to Mr. Gould. In M. (rqninoctialis none of this yellow colouring in the bill is seen. Mr. Layard, who has had opportunities of observing both species ali\c, considers them distinct, as does also Dr. Coues, who has devoted great time and attention to the Petrels. The following specimens are in the British Museum : — a. ad. Ptoval Sound, Kerguelen Island. Rev. A. E. Eaton. h. ad. Pacific Ocean ? Sir Joseph Banks. c. ad. st. Cape of Good Hope. M. Verreaux. BIRDS.— R. B. SHARPE. 121 [la Kcrguolou Islaud a hole similar to a deserted rabbit's earth, excavated in Avet ground, with Avater standing (in early summer) an inch or two inches deep within the entrance, especially if it is in a slope near the sea, may be regarded as the burrow most likely to be that of a White-chinned Petrel. If it is occuj)ied by the birds there Avill probably be some green shoots of Accena clipped off from plants near its mouth dropped by them in the water. Dm-ing the season while the birds arc j)amng before their egg is laid they make an extraordinary cackle in the nest-chamber ; the sound of approaching footsteps, or a thump upon the groiuid some distance away from the nest, and even a shout at the mouth of the burrow, will cause them to commence it in the day time. During the night this call is uttered by the female sitting on her nest or in the entrance of the tunnel ; and she can be heard at the distance of a quarter of a mile when there is a calm. Much trouble may be saved in digging out the nest by soimding with the sjoadc along the course of the 1)urrow until the situ- ation of the nest-chamber is ascertained. This is spherical and tolerably large. Being in most instances near the siu'face of the ground, care must be taken in the removal of its roof, or the bird's back may be broken by the spade while she is sitting upon her egg. As soon as the chamber is laid open it is well to catch the hen by her beak and drag her out of the hole while she is still dazzled by the light, giving her no time to use her claws. On being released she usually makes no attempt to fly, unless she is purposely chased down the hill ; but after waddling away a few yards she returns to her burrow (or to where its entrance used to be before it was dug into and choked with clods), and begins at once to dig her way into the tunnel through the obstructions with which it has been blocked up. She takes little notice of bystanders so long as they remain still, passing leisurely by them or even over their feet if they happen to be in her way. The nest is built of mud and pieces of plants arranged in the form of an inverted saucer three or foiu* inches high, slightly hollowed out at the top. A space is left between its base and the sides of the nest- chamber. Some of the bu'ds had no white patch under their chin ; when it was present it varied in extent in different examples. In most instances it formed a small triangular blotch occupying the apical portion of the angle enclosed by the lower mandible ; but in a few cases the white was limited to one or two feathers only. In none of the Kerguelen specimens did the patch extend to the forehead, as it does in the birds from Australia. The S. African Museum contains the following examples of this species : — 1 (young, in doAvn) from the Crozettes ; 2 from Table Bay with the white chin patch very large ; and 1 from Australia with the white on each side of the face below the eyes, and a white baud across the forehead, in addition to the chin patch. — A. E. K] 122 ZOOLOGY OF KEllGUELEN ISLAND. Puffiuus kuhli. Lo rullln, Buff. PI. Eiil. x., pi. !)(;2. Proccllaria puflinus, Tevwi. Man. d'Oni. ii., p. 805. Nectris cinerea, Kiihl, Bcitr. ZooL p. IIS, ^j/. 11, Jig. 12 {uec Gm.) ; Linderm. Vog. OriecJieiil. p. 170. Piiffinus cinercus, Ctir. JRegiie Anhn. 1829 ; Less. Tra'itc, p. 613 ; Savi, Oni. Tosc. iii., p. 38 ; Schl. liev. Cr'tt. j). cxxxii. ; IFcbb Sf Berlh. Orn. Canar. p. 43; BoUe, J. f. O. 1855, j9. 177, 1857, p. 311; Vcruoii Uarconvl, Ann. N. U. (2) XV., p. 438; Loche, Uxpl. Sci. Alger. Ois. ii., p. 174; Degl. et' Gerhe, Orn. Eio'. ii., p. 370 ; Ehces Sf Buckleg, Ibis, 1870, p. 330 ; Gigl. Fann. Vertebr. nelV Oceano, p. 47 ; Boderl. Avif. Sicll. p. 228 ; Frltsch, Vog. Eur. tab. 58, fig. 1 ; Godmau, Ibis, IS12, p. 223 ; Bree, B. Eur. y.,p. 86 (1876). PufiBnus kuhli, Boie, Isis, 1830, ^j. 258 ; Bp. Consp. ii., p. 202 ; Belz. Beis. Novora, Vug. p. 142 ; Groij, Handl. B. iii., p. 102 ; Salvad. TJcc. Ital. p. 298 ; Shelley, B. Egypt, p. 357 ; Seugl. Orn. N. O. Afr. iv., p. 1367 ; Scl. ^ Salv. Nomencl.Ac. Neotr. p. 149; Irby, B. Gibr. p)- 217. Nectris macrorhyncha, Heiigl. Syst. TIebers. p. 68, no. 711. Nectris gama, Hortl. Orn. 3Iadog. p. 84 (nee Bp.). Procellaria cinerea, Schl. Mus. P. B. Procell. p. 24. Ad. supra bruuneus, pileo concolori, plumis clorsalibus ct scapularibus cineras- centi-bruuneo marginatis ; scapularibus longioribus uigricantibus brunneo margi- uatis ; tectricibus alarum nigricauti-brunneis, majoribus pallidioribus brunneis vix cineraceo lavatis et angustissim^ albido limbatis ; remigibus nigricanti-bruuueis, intus chocolatiuis ; supracaudalibus cineraceo-brunneis, ad apieem albicantibus brun- neo vermiculatis ; rectricibus centralibus saturate brunneis, reliquis cinerasccnti- brunneis versus apieem saturatius brunneis ; facie laterali pileo concolori, vix satu- ratiorc ; genis ct colli lateribus brunneis obscurci albido irroratis ; corpore reliquo subtus albo, pectoris lateribus obscure cineraceo lavatis ; subalaribus ct axillaribus albis, margine alari satvu-atc brunneo ; tectricibus majoribus cxternis versus ajiicem briuuieo lavatis ; rostro flavo, versus apieem nigro ; pedibus flavis ; iride nigricante. Long. tot. 20"5, culm. 2-15, alse 13"4, caudte 6*6, tarsi, 2'05. ? mari similis. Jnv. similis adultis sed sordidior, subtus sordidius albus ; rostro nigricante ; pedibus coerulescentibus. On comparing the two specimens from Kerguelen Island, collected by the Antarctic Expedition, with Mediterranean examples of P. kuhli in the British Museum, I am unable to find any real specific distinctions between them, the Northern birds being perhaps a shade larger, and having rather a stouter bill. BIRDS. — R. B. SHAKl'E. " 123 a, h. ad. Korguclcn Island. Antarctic Expedition. It is prol^ablj the foregoing species which is alluded to in Nixnn's " Narrative " under the name of Ft/ffiniis major. In the above-mei\tioncd work occiu's the following sentence : " As there appears to be some doubts whether this species is " an inhabitant of Desolation, some elucidation is required ; it is considered to have " an extremely wide range." The description given agrees best with P. major, which is found in the Cape Seas. Adomastor cinereus is stated by Captain Hutton to go to Kcrguelen Island (Ibis, 1865, p. 286) to breed. No specimens have been obtained by the recent expeditions, but so good an observer as Captain Hutton would not include the species without good reason ; it may, therefore, be looked for. Thalassoica tenuirostris. Procellaria glacialis, Forster, Descr. Anhn.p. 25. {^lec L.) Procellaria tenuii-ostris, Audub. Orn. Bioyr. v., p. 333 ; id. B. N. Amer. vii., p. 210 ; Cass. TJ. S. Expl. Exp. p. 409 ; Lawr. in Baird's B. N. Amer. p. 826 ; Elliot, B. N. Amer. Intr. cum fig. {head). Procellaria glacialoides. Smith, Illiistr. Zool. S. Afr. pi. 51 ; Gray, List Anseres Brit. Mtis. p. 162 ; id. Gen. B. iii., p. 648 ; Gould, B. Austr. vii, pi. 48 ; Beichenb. Sandb. Longipennes, pi. xiii., fig. 789 ; Cass. TJ. S. Expl. Exp., Birds, p. 409 ; Belz. Beis. Novara, p. 146 ; Lmjard, B. S. Afr. p. 361 ; Buller, B. N. Zeal. p. 301; Sharpe, Voy. Ereb. and Terr. Ajjp.p.BS. PrioceUa garnoti, Hombr. 8f Jacq. Voy. Pole Sud, pi. 32, fig. 43. Thalassoica glacialoides, Relchenb. Handb. Longipennes, pi. xiii., fig. 789 {pess.) ; Bp. Consp. ii., p. 192 ; Gould, Mandb. B. Austr. ii., p. 467 ; Cones, Proc. Philad. Acad. 1866,^. 31; GiglioU, Faun. Vertebr. Oceano,p. 47. Thalassoica glacialoides var. polaris et tenuirostis, Bp. Consp. ii., ^;. 192. Thalassoica polaris, Bp. C. B. xlii., pi. 768. Procellaria smithi, Schl. JIus. P. B. Procell. p. 22 ; Flnsch, J. f 0. 1872, p. 255, 1874, p. 174. Ad. supra clare argentescenti-canus, plumis celat6 albicantibus ; pileo antico et facie lateraK albicantibus, regione parotica. vix cano lavata ; colli lateribus clare canis ; tectricibus alarum et scapularibus dorso concoloribus ; remigibus cinerascenti- brunneis, intiis versus basin albicantibus, secundariis extiis cinereo lavatis ; cauda. omninb cana; corpore subtAs toto purissimc albo, ci'isso laterali cano lavato ; sub- alaril)us et axillaribus purissime al])is, margine alari cano ; rostro cai'nescente, versus apicem caraescenti-cornco, ad apicem nigro ; pedibus cinereis, tarso pallid^ coccineo lavato. Long. tot. 16"5, culmen 1"8, alse 12*8, caudae 5"5, tarsi 2*15. The description is taken from a specimen procured at Kerguelen Island by Lieut. Alexander Smith, who accompanied the Antarctic Expedition, and on his retiu"n Q 2 121 ZOOLOGY OK KKRGUELEN ISLAND. pivsontod it to the Eritisli JMuseuin (C'f. (iray, Cat. .Viisc-rcs, p. 1G2). Koiu; ol' llic recent expeditions seem to have obtained it, but John Is'unn's Narrative states that this speeies '' Mas used, when yoiniij,', liy (uii- parly as I'ood, and our supplies wore " obtained bv diirijinu: the vounsi: birds IVniu ihr l)nrro\\s in the sand or (iiss(i(d\- " banks on tlie Uh' or S.E. si(h> ol' the ishmd. 'IMu* bird was known and eaten by us " under tlie name of the AVhite Nit^lit-liawk." Tlic Antarctic lixpedition also pro- cureil a. specimen of this s])ecies at Keri;'ueleu Ishind. Clreat diU'enmcc exists in the size and thickness of the bill and lens. Sjwcimeus art' in tlie national eoUectiou I'roni South Africa (Typo of P. (jldc'ut- loides : Sir A. Swit//) : Antarctic Seas {Jiilarctic Exjjeditiou) ; Louis Thilippe Island ; Lat. 44° S., Long-. llOi, W. {J. 3Iacglllivray) ; Straits ol' Magellan ; and olY Valiiaraiso {Copt. TV. S. Brett) ; as well as the following- from the ]n-escnt locality ; — a. ad. Kergueleu Island ; Lieut. A. Smith. b. ad. „ „ Antarctic Expedition. Sir J. Hooker's Journal contains tlie following important note : — " A smaller " bird than D. capensis, about the size of a fowl, of a dusky brown colour, with a *' white bar across the wings. It is not uncommon." This surely must have been Thatassteca antarctica, figured in the " A'oyage of the Erebus and Terror." (Estrelata brevirostris. Procellaria giisea, Euhl. Beltr. Zoot. 1820, p. 144, ficj. 9 {nee Gm.) ; ScJd. Mns. P. B. Procett. p. 9. Procellaria brevirostris. Less. 3Ian. d'Oni. ii., p. Gil ; Gray, List Anseres, etc., Brit. Mns. p. 163; id. Gen. B. iii., p. 048. CEstrelata grisea, Coi'.es, Pr. Phitad. Acad. 1860, p. 148 {nee Gm.). Fulmarus griseus, Gray, Handl. B. iii., p. 107. (Estrelata kidderi, Coues, Bull. U. S. Kat. Mas. ii., p. 28 ; Kidder, I. c. p. 15. CEstrelata brevirostris, Salvin in Rowley's Orn. 3Iisc. p>. 235. ? Ad. saturate cinereus, plumis clariore cinereo marginatis ad basin multo palli- dioribus ; pileo lateral! eoncolori cinereo, regione oculari saturatiore, frontc lorisquc vix pallidioribus ; corpore rcliquo subtus cinereo, plumis basaliter albieantibus, pectoris lateribus clar& cinereo lavatis ; tectricibus alarum superioribus et subala- ribus cinerascenti-brunneis, cinereo lavatis; tectricibus primariorum et remigibus nigricanti-cinereis, intiis pallidioribus ; infra pallide cinereis ; canda sordide ciuerea ; rostro nigro ; tarso pedibusque fuscis, unguibus uigris. Long. tot. 12-8, culm. 1-1, alffi 10-2, caudal 4-0, tarsi 1-35. This species may almost be said to have been rediscovered by the recent expedi- tions to Kerguelen Island, as it had not been satisfactorily identified by natm'alists for some years, and was apparently unknown to Dr. Coues when he wrote his- well- known papers on tlie Procellariidce . In his account of the American collection of BIRDS.— K. 1!. SHARPE. 125 birds from Korguelen, lie namofl this species after Dr. Kidder, who accompanied the American expedition as naturalist; hut Mr. Salvia shortly after showed that it was the same as the bird described l)y Lesson in 1828. In justice to the late Mr. George Robert Gray it must be noted that he correctly identified the si^ecimens brought by the Antarctic Expedition {Cf. List Anseres, p. 103). This useful little list appears to have been overlooked by Dr. Coues. Afterwards in the "Handlist" (p. 107), Gray made the mistake of referring (E. hrevirostris as a synonym to Fterodroma macroptera ; and the Kerguelen Ijirds he referred with a query to Fulmarns griseus {Kuhl.). In this arrangement he was following Dr. Coues' idcntillcations of 18GG. With regard to the name nni color of Gould, quoted by Gray as a synonym and noticed by Coues, I cannot find that it was ever published. A specimen with this name attached, in Mr. Gould's handwriting, is in the Museum, and is doubtless the authority for Mr. Gray's quotation. a. ad. Tristan d'Acunha. Capt. Carmichael, R.N.* b. c. ad. Kerguelen Island. Lieut. Alex. Smith. d. e.f. ad. Kerguelen Island. Antarctic Expedition. g. a. d. Christmas Harbour, Kergi;elen Island, Eeb. 1840. Antarctic Expedition. ? . ad. Royal Sound, Kerguelen Island. Rev. A. E. Eaton. [This Petrel is less common than (E. lessoni about Observatory Bay. I am unacquainted with its call, the birds not even screaming when dug out, althougli they- bit and scratched the hand. They buiTOw into clayey soil near lakes and upland marshes. The burrow is rather smaller in diameter than that of 31, cequltwc- tialis, but in aU other respects is very similar to it. During the early portion of the breeding season the floor of the tunnel leading to the nest chamber is flooded with water an inch or two inches deep ; and any one who saw it then for the first time in his life, with water trickling in a little streamlet out of its mouth, or stand- ing stagnant within it, would readily suppose it to be an old hole abandoned long ago. Towards the beginning of autumn (Eebruary), however, the ground becoming dry, the water disappears. The nest is composed of damp and decayed A'egetable matter, comprising sprigs of Accena and Azoi'ella, tufts of Festuca erecta, &c. It is two or three inches in height, and slightly concave. The fii'st nest was taken on the 8th of November. The embryo was tolerably advanced in growth. In January, in a branch of a Majaqueus burrow, was a nestling which seemed to be the young of (E. hreviroslris ; the Majaqueus egg was in the chamber of the main burrow, to which there was only one entrance. — A. F. E.] * Printed in Gray's List of Anseres " Tristan de Chusan,'' which we believe to be a misprint for Tristan d'Acunha. 12G ZOOLlKiY OK KERGUELEN ISLAND. CEstrelata Icssoui. ProcoUavia lossoni, Ganiof, Ann. Sci. Nat. vii., /*. 54, pi. 4. ; Less. Traite,p. Gil ; Gray, List Auso'cs Brit. JT/is. p. 103; «/. Gen. JJ. iii., p. (MS; Gould, B. Anstr. vii., pi. -19 ; Beichenb. Uandb. JLonffipenncs, ]}/. xx.,fi.ff. 330 ; Flnsch, J.f. 0., 1872, p. 255, 1871, ix 207; Bn/ler, B. N. Zeal. p. 303, ^j/. 2d, Jiff. 3 ; SImrpe, Toy. Ereh. ami Terror, Birds, App. p. 33; Cab. cj- Beichen. J.f. O. 1870,7;. 320. ProcoUavia leucoccpliala, Forster, Descr. Anim. p. 200 ; Gould, Ann. N. IL. xiii., ji. 303 ; Belz. Beis. Novara, Vug. p. 145. CEstrelata leucoccpliala, Bp. Consp. ii., p. 189 ; Gould, Handb. B. Austr. ii., /). 451. Ivhantistes lessoni, Bp. C. B. xlii.,^}. 708. CEstrelata lessoni, Cass. Br. Bhilad. Acad. ] 862, p. 327 ; Cones, op. cit. 1800, p. 142 ; Giglioli, Faun. T'crtebr. Oceana, p. 40 ; Coues and Kidder, Bull. JJ. S. Nat. Mas. ii., p. 27 ; iid. op. cit. iii., p. 14. 6 Ad. siiin'a clard canus, plumls piu'ius cincrco niarginatis, basalitcr albidis, scapis linealiter nigris ; pileo antico albido, postic^ ciuei-eo, plumis ad basin cinerasccnti- bus : loris cinerascentibus ; rcgione oculari nigricante, et rogionc parotica nigri- eanti-scbistaceo lavato ; facie lateral! reliquii, alba ; colli et pectoris latcribus clare ciuei-eis albido tanquam irroratis ; corpore reliquo snbtus pure aibo, latcribus cor- poris et lij^ocbondriis angustd nigro striolatis ; subalaribus cincrascenti-brunneis, angust^ cinereo terminatis ; axillaribus cinerascentibus, basalitcr et apicaliter albi- dis ; alis supra nigricanti-brunneis, tectricibus mediauis ct majoribus cinereis albo niarginatis ; remigibus sordid^ cinerascenti-nigris, intus cincreo-brunncis ; sca- pularibus ad basin cinereis, ad apicem nigris ; uropygio saturate l)runnco, cinereo lavato; supracaudalil)us canis, albido terminatis; caudi'l alba, plumis cano irro- ratis, rectricibus duabus mediis canis ; rostro nigro ; pedibus carneis, digitis supra nigris ; membrano ad apicem nigro ; ii-ide saturate brunnea. Long. tot. 15'5, culm. I'o, alse 12-4, caudge 5'5, tarsi 1"7. ? Ad. mari similis, seb pileo puriore albo, et scapis plumarum dorsalium et bypocbondriarum minus distinct^ indicatis. Long. tot. 10'5, alge 12*3, cauda; 5'2, tarsi 1-7. The descriptions are taken from a pair of birds collected by Mr. Eaton in Ker- guelen Island. Compared with some otber specimens from Australian seas, they do not show any differences, unless it be that the black on the feet is not so exten- ded. Dr. Kidder describes the latter as follows : — Tarsus and foot flesh-iiink, " black along upper sm"faces of digits, and on the web near the claw." This agrees mth the markings exhibited by the Kerguelen Island birds, but in a South Australian specimen quite the terminal half of the webs are black, and the bu'ds collected by the " Ptattlesuake " also show this peculiarity. In the plumage a BIRDS. — K. r,. SIIARPE. 127 certain amount of variation is shown in the vermiculations of grey on the head, and in the amount of grey on tlic tail, the latter hemg almost entirely white in some with grey mottliugs, while in others it is almost uniform grey, mottled only on the outer web. For a description of the young birds. Dr. Kidder's paper (I.e.) must be consulted. This species, wliich was obtained by all the recent expeditions, does not appear to have been collected in Kerguclen Island during the Antarctic Expedition, nor is it mentioned in Nunn's Narrative. In the National collection are examples from New Zealand {Sir George Grey) ; Southern Seas {Antarctic Expedition) ; South Indian Ocean, Lat. 40f° S., Long. 125^° E., Jan. 14, 1847 {J. Mcic{/illivrai/) ; South Pacific Ocean, Lat. 44° S., Long. 110^° W. {J. 3Iacgillivray) ; Lat. 36° 39 S. ; Long. 10° 3' S. {Sir George Grey), and the following : — a. b. S. ?. ad. Royal Sound, Kerguclen Island. Rev. A. E. Eaton. [In Cajitain Huttou's i)aper before referred to, allusion is made to an undeter- mined species of Petrels to which the euphonious sobriquet Procellaria diaholica has been applied. It was said to be a bird inhabiting Desolation Island, which fiew about by night uttering unearthly shrieks. There are good reasons for supposing the sprite to be Lesson's Petrel. It is difficult to describe the cry of this bird. Eor a long time there was no finding out which of the Petrels gave utterance during its flight to its weird sounds. Whenever its cry Avas heard, I went out with a lantern to endeavovu* to get a sight of the bird, but without success. At last, near Thumb Peak, we dug up some large birds whose ovitcries, when caught by the beak, plainly identified them beyond all question with GE. lessoni. The burrow of this Petrel can be recognised externally by its being about as large as an ordinary rabbit's hole and dry, and by its entrance being generally sparsely bestrewn with green shoots of Accena. It is usually excavated in Azorella, the tunnel is short, the large terminal chamber contains no special nest, and when the hand is cautiously introduced to feel after the e^^, it is promptly and severely bitten by the old bird. It is therefore well to take the precaution of dragging her forth from the interior before an attempt is made to secure the egg. Her removal can be easily effected. While she is stooping forward at the entrance of the nest- chambcr looking out in readiness to bite, a piece of stick is presented to her, which she seizes instantly, and Avhilst it is being shaken to make her hold it fast, her beak is suddenly grasped with the hand, and she is drawn up by it out of the burrow shrieking loudly. Care must meanwhile be taken to prevent her from thrusting her claws into the hand. Nests were foimd from the extreme confines of the sea shore to an altitude of about 300 feet above the mean level. They were common amongst Azorella at the foot of the cliils near Thumb Peak, and on the summit of the lower terraces; also on some of the hills near the Swain's Harl)our Transit Station. There were also some nests near the princij)al station on a slope by a freshwater lake on the landward side of a hill. 128 ZOOLOI;Y of KKRCrKl.KX ISLAM). Tlieiv is a spccimon of Lesson's Petrel in the S. Afrieaii Museum, captured in Lat. :vr UV S., Lon;-. aSf l.'V E.— J. K E.] (Estrelata mollis. Procellaria mollis, Gould, Jini. X. U. V6i\, p. 3(53; id. B. ^liislr. \\\., pi. ')0; Cass. r. S. Jbxpl. Krp. 1858, p. ilO ; Schl. Mks. P. B. BvoccU. p. 11 ; Velz. Bets. Xorara, rOff.p.UG; Fuiscli, J. f. O. ISli, ]). 2ob,lSl2, p. 201 ; Cab. §- lleiclieii. J.f. O. 1876, 7;. 329. Proeellaria inexpcetata, Forstcr, Descr. J)i. p. 201. Cookilaria mollis, Bp. Consp. ii., p. 100. llliautistcs mollis, Bp. C. R. xlii., p. 70S. (Estrelata mollis, Coties, Br. Bhllad. Acad. 18GG, p. 150 ; Gould. Jluudb. B. Aiislr. ii., jt). 453; GigUoU, Faun. Vertehr. Oceano, p. '^2. Ad. eiuerascenti-hrunneus, plumis clariore ciuereo marginatis ; plumis lTontalil)us albido margiuatis ; alls saturatius l)ruimeis, tectricibus nuijoribus vix ciiicrco lavatis ; supracaudalibus cauduquc cineraceis ; loris albis ; regionc oculari et paro- ticA. anticil ciuerco-nigris ; I'aeie latcrali et corpore su1)tus loto ])urt! albis, colli latcribus et hy|)Ocliondriis cineraceo irroratis, poctorc latcrali concolori cineracco ; subalari]nis bruuiieis, albido marginatis, majoribus interioril)us pallidius ciucrascenti- hruuueis ; remigibus infra saturated brunneis, secundariis intus pallidioribus ; rostro nigro ; tarso, digitis ad basin, et membrano interdigitali dimidio basali carnescenti- albis, pedibus alitor nigris. Long. tot. 12-5, culm. 1-2, akc 10 -1, caudal 5*85, tarsi 1-45. For description of the yomagbirdand variations in the jilnmage, Dr. Coues' article in the Philadelphia ' Proceedings ' must be consulted. As will be seen by Mr. Eaton's note below, he believes that he saw this species, wliich was procured in Kerguelen Island by the German expedition. The British Museum contains specimens of tliis Petrel from South Australia {Sir George Grey) ; and the South Atlantic, Lat. 36° 50' S., Long. 27° 50' W. ; Lat. 31' 43' S., Long. 40° W. {J. JIacgilUvray). [Off Cape Sandwich and the neighbom'ing low land, and out at sea during the first few days sail from. Kerguelen Island, I noticed a Petrel very like (E. lesson i, but differing from, that species in having a dark colom-ed tail and back. This may have been (E. brevirostris ; but when I was looking through the collection in the S. African Museum on my return to the Cape, I was led to believe the species I had seen to be (E. mollis, Gould, which is represented in the collection mentioned by a specimen taken in Lat. 31' 26' S., Long. 30° 26' E., exhibited as Brocellaria mollis, Gould. The Uce of (E. mollis, according to Bulow, are Trabeculiis schilingii and Colpo- cephalum furcatum of that author. — A. E. E.'\ BIKDS.— R. 15. SilAKPE. 129 Procellaria nereis. Thalassidroma nereis, Gould, P. Z. S. 1840, ^;. 178 ; id. B. Austr. vii., pi. 64 ; Bennett, Gath. Nat. p. 240; Abbott, Ibis, 1861, i?. 164; Grai/, Ibis, 1862, p. 245 ; Buller, B. N. Zeal. p. 322; Finsch, J.f. 0. 1872, p. 257, 1874, p. 213; Sharpe, Voy. Ereb. 8f Terror, Birds, App. p. 34. Procellaria nereis, Gray, Gen. B. iii., p. 648 ; Bp. Consp. ii., p. 196 ; Coues, Pr. P/iilad. Acad. 1864, p. 81 ; Goidd, mnidb. B. Austr. ii., p. 476 ; Giglioli, Faun. Vertebr. Oceano, p. 36 ; Gray, Handl. B. iii., p. 104, ??o. 10852 ; Scl. and Sale. Nomencl. Av. Neotr. p. 148 ; Coues ^ Kidder, Bull. U. S. Nat. litis, ii., p. 31; iid. op. cit. iii., p. 16; Cab. ^- Beichen. J.f. O. 1876,;;. 329. Oceanitis nereis, Cab. J.f. O. 1875, j;. 449. ? Ad. supi*a sordid^ cinerea, iiropygio et svipracaudalibus clarius cincreis, his sub- basaliter albis ; caudu quoque clare ciuei'ea, uigTO terminata, rectricibus extimis nigricantibus ; pileo et collo undique et collo postico cinerascenti-funiosis, inter- scapulio clariore cinereo lavato ; scapularibus dorso concoloribus, longioribus nigri- cantibus ; tectricibus alarum cinerascenti-fumosis, marginalibus anguste cinereo lavatis, medianis clariiis cincreis anguste albo tcrminatis ; remigibns sordid^ cinereis, primariis extus ad basin nigricantibus, secundariis etiam magis nigricantibus ; gutture, prsepectore, et pectoris summi lateribus fumosis ; corpore reliquo subtus pure albo, liypocbondriis imis et subcaudalibus cinereo lavatis ; subalaribus et axillaribus pure albis, niinoribus et marginalibus fumoso-brunneis, his imis albo terminatis ; remigibus infra cinerascentibus, intus versus basin albidis ; rostro et pedibus nigris ; iride nigra. Long. tot. 7, culm. 0-55, ala? 5'0, caudse 2*8, tarsi 1-2. Compared with an Australian example in the collection, the Kerguelen skin has rather a larger bill, and is more du.sky brown on the head and throat, and has the brown colour extending lower down on to the fore-neck ; these characters are, how- ever, exhibited in another Australian example, so that there seems to be only one species. The Sea-Nymph Petrel was not previously known to inhabit Kerguelen Island, but specimens were collected by the English, American and German expe- ditions. The British Museum contains specimens from New Zealand {Sir George Grey) ; oflf the eastern coast of New South Wales {J. Macgillivray) ; Lat. 43° S., Long. 140° E. {J. Macgillivray), and the following : — a. $ ad. Royal Sound, Kerguelen Island. Rev. A. E. Eaton. [Late ia the night of the 6th of November the faint cry of a strange bird in the distance roused me from sleep. Calling for a dark lantern, I proceeded witli George Wilson, the sapper on watch, to search for the Petrel, guided by its call, wliich was vittered at intervals until we were quite close to it. The light Ijeing now turned on in the right direction, the bird was discovered sitting upon the open ground withiu a £ 130 ZOOLOGY Ol- KERGUELEN ISLAND. yard or so of us, and it -vvas so dazzled that it made no attempt to escape, being cauLjlit by band. It proved to be a female of P. jicrcis. Its call was very similar to the crake-like cry of P. oceauica. Dr. Kidder* said that this Petrel was common at Molloy Point. As no other example was either seen or heard by me besides that mentioned above, I suspect that this species rarely ventm-es so far uj) ihe Sound as Obscrvatoiy Bay. Fron\ the specimen of P. uereis whose casual capture is mentioned above, six examples of its parasite, Zipeitrus clypcatus, Giebcl, were obtained. — A. E. E.'\ Oceanitis tropica. Procellaria grallaria, Licht. Verz. Douhl. p. 83. Proccllaria oceanica, Bp. Zool. Joiirti. iii.,^. 89 {nee Kuhl.). Thalassidroma tropica, Gould, Ann. N. M. xiii., p. 366. Thalassidroma melanogastra, Gould, Ann. N. JI. xiii., p. 307 ; id. B. Anstr. vii., 2)1. 02; Layard, B. S. Afr. p. 358; Finsc/i, J. /. O. 1872, i;. 257, ISli, p. 212; Buller, B. N. Zeal. p. 319; Uutton, Ibis, 1871, i?. 42; Buller, I. c. p. 121; Sharpe, Toy. Ereh. and Terr. Birds, App. p. 31; Cab. Sf Meichen. J.f. 0. 1870, ^J. 329. Fregetta tropica, Bp. Consp. ii., p. 197 ; id. C. E. xiii., p. 709 ; Cones, Pr. Fhilad. Acad. 180-1, ^j. 87. Fregetta melanogastra, Bp. C. B. xiii., p. 709 ; Coues, Br. Bhilad. Acad. 1801, p. 87 ; Gould, Kandb. B. Anstr. ii., p. 179 ; Giglioli, Faun. Vert. Ocean., p. 38. Procellaria melanogaster, Schl. Mtis. P. B. Procell. p. 0; Gray, I. c. p. 105. ? Ad. Fumoso-nigricaus, m-opygio nigerrimo, plumis imis albo tcrminatis ; supracaudalibus couspicu6 albis, fasciam latam cxhibentibus ; tectricibus alarum cinerascenti-brunneis, majoribus pallidioribus vix albido tcrminatis; tectricibus prima- • riorum et ala spuria saturatius f umoso-nigricantibus ; remigibus nigris vix versus apiccm cinereo lavatis, intiis pallidioribus, fumoso brunneis; secundariis clariiis cineraceo lavatis et anguste albo limbatis ; cauda nigr&, vix cinereo lavata, ad basin extremam alba ; guttm-e toto, faciei et colli lateribus, prsepectore et pectore summo fumoso-nigricantibus, guise plumis basaliter albis ; corpore rcliquo subtus pur6 albo, medialiter fumoso-nigricante, fasciam latam per pectus et abdominem ductam cxhibentibus ; subcaudalibus albis longissimis basaliter albis, ad apicem dimidialiter fumoso-nigricantibus ; subalaribus et axillaribus pur6 albis, his versus basin fumosis, tectricibus marginalibus brunneis, majoribus pallide cinerascenti-bruimeis anguste * Since the preceding paragraph was written Dr. Kidder's Eeport has been sent me from the Smitli- sonian Institute. At p. 32 he writes of this species as follows : — " The first specimens were taken on the " 28th and 29th of October, being dug out by the dogs from small burrows under clumps of Azordlu. A " pair captured on the latter date were found under a tussock not two yards above high-water mark, on the '• beach, under a high cliff. No eggs were found at that date. Eggs were first found, December 12, under " the overhanging margins of clumps of grass and ' Kerguelen tea' {Accena ascendens), in a bit of swampy " lowland near the sea. Strange to say, I have only found the male with the egg. In this locality there " were no burrows, the overhanging herbage seeming to afford sufficient protection to the nests." BIRDS.— R. B. SHARPE. 131 albo terminatis ; rostro ct podibus nigris ; iricle nigra. Long. tot. 7'0, culmen 0'6, alee 9*4, caudee 3"15, tarsi 1'55. ^ Ad. siniilis foeminai adulta^. Long. tot. 7"2 culm 0-6, alte G'8, caudse 3-0 tarsi VG. The differences between 0. melanogastm and 0. tropica are extremely slight, consisting in the white throat and the greater amount of black on the abdomen and centre of the body in the latter bird. I believe it possible that O. leucogastra is also only a stage of plumage of the same species, the four specimens in the Museum being apparently immatiu*e, Avith narrow whitish edgings to the feathers of the upper surface. Whether this is the case I am, however, unable to prove at present. The following are the series of measurements of the specimens in the British Museum. Total Length. Culmen. Wing. Tail. Tarsus. a. (J 0. melanogastra, Kerguelen Island 7-2 0-6 6-8 3-0 1-6 *• ? » >, • - 7-6 0-6 6-4 3 15 1-55 c. S 0. tropica, Lat. 6° 33' N., Long. 18° 6' W. 7-0 0-6 6-7 30 1-65 d. ? „ Lat. 12' S., Long. 30^° W. 7-0 0-6 6-55 3-1 16 . 90, />/. {\Q,Jig. G {ncc L.). Proccllaria oceanica, Kiihl. Bcitr. Zool. p. l'M'>, lab. x., Ji(j. 1 ; Gray, Gen. B. iii., p. 6J.S ; Schl. Mas. P. B. Procell. />. (>. Procellaria wilsoni, Bp. Jonrn. Acad. Bliilad. iii., pt. 2, ]>. 231,/;/. i\.,fi/. (55 ; 3Iacgill. Br. B. v., p. i'56 ; Barm. Til. Bras, iii., p. 140; Cass. U. S. Rrpl. Kvp. p. 402. Thalassidroma oceanica, Schinz. Europ. Faun. p. 397, j^^- 1 '■> Gray, List Anseres, Brit. Mas. p. IGl ; id. Gen. B. m., p. G^S; id. List Br. B. p. 225; Pelz. Beis. Nopara, Vog. p. 115; Degl. §' Gerbe, Orn. Enrop. \\.,p. 380. Oceanitis wilsoni. Keys. 8f Bias. TFirb. Ear. p. 238 ; Bj). C. M. xlii., p. 769 ; Salvad. Ucc. Ital. />. 301 ; Gigl. Faun. Verlebr. nelV Oceano, ^J. 38. Oceanitis oceanica, Bp. C. B. xlii., p. 709 ; Gould, Sandb. B. Anstr. ii., p. 478 ; Salcad. Cat. Ucc. Sard. ^j. 132; Giglioli, Faun. Vertebr. Oceano, p. 37; Cones and Kidder, Bull. V. S. Nat. Mus. n., p. 30; lid. op. cit.m.., p. 10. i Ad. fidiginoso-niger, pileo undique aliqnot cincrasceute, regione auriculari magis nigricante ; tectricibus alaru.m fumoso-nigricantibus, majoribns versixs apicem pallidc brunneis, plagam f ormantibus ; tectricibus primariorum remigibusque nigris, intus brunnescentibus, secundariis quoque pallidioribus extus brunnescentibus ; plumis uropygialibus imis nigris albo terminatis ; supracaudalibus pur6 albis ; Cauda nigra, recti'icibus ad basin albis, externarum albcdine magis extensa, ; corj)ore suljtiis fiunoso-bruuueo, lateraliter saturatiore, subcaudalibus brunneis ad basi)i albis ; ci'issi lateribus conspicuc albis, quibusdam funioso lavatis ; subalaribus fumoso-brunneis, intimis yix pallidioribus ; rostro nigro ; pedibus nigris, membranis flavis ; iride nigra. Long. tot. 6, culm. 045, alse 5'35, caudte 2'45, tarsi, 1"35. The above is a description of a Kerguelen Island skin, and after a compai-ison of our series I am unable to find any grounds for separating a northern and a southern species. Some examples are blacker and some are grayer, as is the case with other Petrels. The British Museum specimens arc from the following localities : — Yarmouth ; Atlantic Ocean {Beo. W. Kennali) ; South Africa (Sir A Smith) ; South Australia {Sir G. Grey) ; Lat. 36|° N., Long. 121° ^^ _ (j-. Macgillivray) ; and from the ice off Louis Philippe Island {Antarctic Expedition) . To these is uoav added the following : — BIRDS.— K. 15. SIIARrE. 133 a. ad. Royal Sound, Korguek'n Island. Rev. A. E. Eaton. [From the 10th of October, when we passed Cape Sandwich, until the middle or third week of November, we completely lost sight of the Storm Petrels. About the period last mentioned, however, they began to frequent Observatory Bay in large numbers. Theu- fii'st appearance in it took place during a strong breeze which lasted several days. When this was succeeded by more moderate weather, we saw little of them in the day-time; but towards evening they used to fly over the water like Swallows, and some of them might be observed flying near the ground far away into the country, following the coui'se of the valleys, or playing round the inland cliffs. We tracked them along the lower hUl-sides and the margins of lakes over rocks and bogs ; but om* efforts to learn what became of them were unattended with success. Probably at that time they were not preparing to breed, and the birds were merely going overland from the bay to other inlets of the sea. At length when we went to Thumb Peak their mode of nesting was discovered. Carefully Avatching, with Lieut. Goodj'idge, R.N., the birds flying to and fro about the rocks, we observed that they occasionally disappeared into crevices amongst piles of loose stones, and crept under loose masses of rock. Having meanwhile ascertained their call, we were able by listening attentively to detect the exact positions of several of these hidden birds. They were easily caught when the stones were rolled aside ; but they were in couples, merely preparing for laying, and therefore we did not find any eggs. On our way back to Observatory Bay after the Transit we called at the American Station, and were informed by Dr. Kidder that he had observed this Petrel on the shore near Molloy Point. The sea-shore in the neighbourhood of Observatory Bay is of a different character (for the most part) from that which is adjacent to the American Station, and, being less favourable than it, was seldom resorted to for nesting by the Petrels. But the country in general about our bay afforded them un- limited accommodation. For, provided that they can find a slope of shattered rocks with suitable chinks and crevices, or dry spaces under stones or large bou.lders shel- tered from draughts, whether they be near the Sound or on the sides and summits of high hills, they readily appropriate them. The egg is laid upon the bare ground within the recess selected by the birds, either in a chance depression formed by con- tiguous stones or in a shallow circular hollow excavated in the earth by the parent. Having found numbers of their nesting- places I will describe my method of searching for them. "Whenever there was a calm night I used to walk with a darkened bull's- eye lantern towards some rocky hillside, such as the Petrels would be likely to frequent. It was best to shut off' the light and keep it concealed, using it only in dangerous places where falls would be attended with injury, and progress in the dark was hardly possible, lest the bii-ds seeing it should be silenced. On arriving at the ground selected it was probable that Storm Petrels Avould be heard in various directions, some on the wing, others on theii- nests, sounding their call at intervals of from two to five minutes. Those on nests could be distinguished from others 134 zo(»loi;y of kehguelen island. living- In ilu'ii- (.Ties procoodiiig' Irom lixinl positions. iiaving settlcil wliii-li ol' the birds s1k)u1(1 bo soarchod attor, a i-autioiis advance had to be made in her direetion, two or three steps at a time, Avhen she was in lull cry. As soon as she ceased an abrui)t halt was imperative, and a pause ol' some minntcs might ensue before she recommenced her cry and permitted another slight advance to be elTected. In the course of this gradual approach the position of the bird might he ascertained approximately ; but it had to be determined precisely, and to learn exactly where she was she had to be stalked in the dark noiselessly. No gleam could be permitted to escape from the lantern. Loose stones and falls over rocks, — to avoid them it was sometimes necessary to dispense Avith slippers, and feel one's way in stockings only, for should the Petrel he alarmed once with the noise or the light, she would pro- hahlv remain silent a consideral)le time. Now and then it would happen that upon the boulder beneath which she Avas sitting being almost attained the bird would cease calling. When this occurred, and many minutes elapsed without her ci'y being resumed, it was advisable to make a detoiu- and approach the rock from the opposite side, as her silence might be attributed to her seeing a person advancing towards her, aud she would probably recommence her call so soon as he Avas out of sight. If she did not, a small pebble thrown amongst the rocks AA'ould usually elicit some sounds from her, as she would most likely conclude that the noise was being made by her mate retiu'ninsr to the nest. When the stone beneath which the bird was domiciled was gained at last, redoubled care had to be exercised. By stooping doAAni and listening very attentively her position could be accurately ascertained. Then the lantern was suddenly turned upon her before she had time to creep out of sight, and her egg could be secured with the hand, or Avith a spoon tied on to a stick. Sometimes I worked Avithout a lantern, and marked the positions of the nests Avith piles of stones so that they might 1)e revisited, by day. Several eggs were obtained in February from nests which had been thiis marked early in the previous month. The first egg taken by us was found by a retriever on the 22nd of January, on an island in Swain's Bay. Captain Fairfax sent me a nestling a day or two before we sailed for the Cape. Two of the eggs were laid in unu.sUal situations. One of them, was found by a man imder a Fringlea plant ; but this may have been an agg of Pro- cellaria nereis. The other was deposited just above the tide-mark in a cavity of a rock rather open to the air and light. I had found the bird there one night, had taken her up into my hand, and had gently replaced her in the hollow, nearly a month before the egg W'as laid. The voung bird in the egg has the tarso-metatarsal joint short. In the S. African Museum there is a specimen of F. oceanica from the S.E. coast of Africa, another from the S. coast of Africa, and two from Table Bay. A. E. E.^ BIRDS.— R. 1!. SIIARPE. 135 Prion vittatus. Blue Petrel, Fors/er, Voyage, i., pp. 91, 153 ; id. Drawings, No. 87. Petrel bleu, Buffon, S. N. Ois. ix., p. 31G. Broad-billed Petrel, Lath. Gen. Syn. iii., pt. %p. 414 ; id. Gen. Hist, x., p. 194. Procellaria vittata, Gm. S. N. i., p. 560 ; Knhl. Beitr. Zool. pi xi., fig. 13 ; Forster, Bescr. Anim. p. 21; Sc/il. Miis. P. B. Procell. p. 16; id. 8f Poll. Faun. Madag., p. 144. Procellaria forsteri, Lath. Ind. Orn. \\., p. 827. Prion vittatus, Lacep. Mem. de VInst. 1800, p. 514; Gray, Gen. B. iii., p. 649, pi. 178, fig. 1 ; id. List Anseres, ^c, Brit. Mns. p. 165 ; id. Voy. Ereh. ^- Terror, Birds, p. 18 ; Gould, Ann. N. S., xiii., p. 366 ; id. B. Anstr. vii., pi. 55 ; Bp. Consp. ii., p. 192 ; Beichenh. Handb. Longipennes, pi. x.,figs. 771, 772 ; Gray, Ibis, 1862, 2). 247; Pelz. Beis. Novara, Tog. p. 147; Gould, Handb. B. Anstr. ii., p. 474; Cones, Pr. Philad. Acad. 1866, p. 169; Giglioli, Faun. Vertebr. Oceana, p. 44; Gray, Handl. B. iii., p. 108 ; Bnller, B. N. Zeal., p. 312 ; FiMsch, J. f. 0. 1872, p. 256, 1874, p. 211; Sharpe, Voy. Ereb. 8f Terror, Birds, App. p. 38. Pacbyptila vittata, lUiger, Prodr. Syst. Mamm, Si'C., p. 274 : Temm. PI. Col. 528 ; Bnrm. T/i. Bras, iii., p. 444; Cab. 8f Beiclien., J.f. O. 1876, 7;. 328. Procellaria latii-ostris, Bonn, et Vieill. Enc. Ileth. i., p. 81. Pacbyptila forsteri, Stepli. Gen. Zool. xiii., ^;. 251; Less. Traite, p. 618; Jard. 8f Selby, III. Orn. i., pi. 47 ; Swains. Classif. B. ii.,p. 374. Pacbyptila banksii. Smith, III. Zool. S. Afr. pi. 55. Prion banksii, Gould, Ami. N.H., xiii., ^. 366; Gray, List Anseres Brit. 3fns. p. 165; id. Gen. B. iii., p. 649; Bp. Consp. ii., p. 193; Gray, Ibis, 1862, p. 247; Kirk, Ibis, 1864, p. 338 ; Layard, B. S. Afr., p. 862 ; id. Ibis, 1867, p. 460 ; Giglioli, Faun. Vertebr. Oceano, p. 44 ; Gray, Handl. B. iii., p. 108 ; Finsch ^ Hartl. Vog. Ostafr., p. 815 ; Finsch, J. f. O. 1872, p. 256, 1874, p. 211 ; Buller, B. N. Zeal., p. 311 ; Sharpe, Voy. Ereb. 8c Terror, Birds, Ajyp. p. 34. Procellaria banksi, Schl. Mus. P. B. Procell. p. 17 ; id. &j- Poll. Faun. Madag. p. 145. Prion magnirostris. Gray, Handl. B. iii., p. 108. Prion australis, Potts, Ibis, 1873, j5. 85. (J Ad. clar6 cinereus, pileo obscuriore ; vitta, uropygiali indistinct^ nigricante ; tectricibus alarum brunneis vix cinereo lavatis ; medianis et majoribus clard cincreis ; ala, spuria et tectricibus primarioruni brunneis vix cinereo limbatis ; remigibus cincreis, intiis albis, primariis longis extus nigricantibvis, secundariis intiniis subterminaliter nigricantibus, albo terminatis ; cauda clarc!' cinerca, versus apicem nigricante, rectrice extimA, omnino cinerea ; regione lorali cincrcii albo variegate ; vitta latS, superciliari alba ; macula ante-oculari nigricante ; regione parotica sordidt^ cinereS,, plumis versus basin allns ; genis et coi-pore subtus albis. 136 ZouLU(iV OK KKUGUKLKN ISLAND. colli ct pectoris latoribus clarr cinorois ; liypocliondriis rt sul)c;ui(lalibus ciiuuvn lavatis, his loiigioril)us apicalitiM- nigiicaiitil)us ; subalaiil)us I'l axillaribus albis, liis vix cineivo lavatis; roslro prdibusiiuo ccerulcis ; iridc satxxrati' briinncfl. Loni>'. tot. K>(»; rulni. II; ala>, 8-3; caiula, I--'.") ; tar.si, 1-2."); (liij,i(. iiunl. c. u., 1C5. ? . mari similis, scd rostro angustiore ot laininil)iis inimis conspicuis. Altlunigh !Mr. Eaton did not briny back a skin ol' this species, its occun'once in Kerguelen Island was certilicd by the discovery of a head in lh(> stoniacli of a Uiant Petrel. The American natnralists did not notice it, bnt the German expedition is stated to have brciiyht liaek this species of Prion I'nmi tlic island. Prion vlttatus is a bird easily recognisable by the ahapc of its l)ill ; in colour it exactly resembles P. desolatus, and a I'cw Avords on the subject of these birds may not be ont of place here, seeing that my conclusions are of a very difTcrent nature from those of recent Avriters on the broad-billed Petrels. To take lirst the paper by Dr. Cones in the " Proceedings" of the Philadelphia Academy for 1866 Qj. 162), \ve find that he separates Prion (type P. vittatiis) as a distinct genus from Psendo- 2)rio>i (type P. turtur), keeping in the latter genus four species, P. banlcsii, P. turtur, P. arid, P. brcvirostris. The chief dilTercnees between Prion and Pseudoprion are to be found in the shape and laminations of the bill ; and indeed the latter form almost the only specific characters for the distinguishing of the above-named four species. Mr. G. E Gray, in the " Handlist," follows the arrangement of Dr. Cones, keeping his genera as sub-genera, but he adds to P. vittatd a second species, P. magnirostris, Gould (ubi ?) ; and to P. hanksii and its allies ho adds P. desolala of Gmelin and of Kuhl, concerning which species see below. The chief points to be noted are the reference of P. hrevirostris of Gould (1855) to P. ariel of Gould (1844),' in which I agree with Mr. Gray, and the reference of P. rossii of Gray to P. banksii, in which I do not agree with him, for an examina- tion of the types shows mc that if P. ariel were a species, it is also P. rossii of Gray. But P. ariel is not a species, according to my studies of the genus, and is only P. turtur, when not full grown. Again, I consider P. hanksii to be also no species, but to be a stage only of P. rittatus, the laminations in the bill being developed wath age, and not being specific characters. The solution of this question has been much simplified for me by the examination of three birds presented to the British Museum by Sir George Grey, of which the following are illustrations {Plate VII., Figs. 2-5). They were sent as exemplifying the old male, old female, and young male of P. vittatns. There are no differences in colour or markings. Then, again, any number of intermediate links are to be found in a series of specimens, and as a variation in the extent of the laminations accompanies a difference in size of bill, it follows that the birds must either be regarded as stages of one species, or must be divided into BIRDS.— K. 1}. .SlIAHPE. 137 several species, which has been the plan adopted by many recent writers. In the plate are figures of the largest and smallest bills in a scries of P. vittatus {Figs. 3-7) . Oil comparing tliese figures with those of P. desolatus given on the same plate {Figs. 8-10), it will be seen that though the size varies, the shape is constant, the l)ill being bowed out from the base and gently incurved towards tlie tip. The variations in the dimensions of the wing and middle toe are shown in the accom- panying list of specimens in the British Museum : — a. ? . Cape Seas {A. Smith ; type of P. banksii). Wing, 7"55 ; middle toe, 1'45. b. $. Coast of Australia {Sir G. Greg). Wing, 7*2; middle toe, 1-45. c. c?. Ad. Australian Seas {Sir G. Grey). Wing, 7"6 ; middle toe, 1*5. f/. ? . Ad. Australian Seas {Sir G. Grey). Wing, 7'15 ; middle toe, I'l. e. ^ juv. Lat. 35° 1' S. ; long. 6° 15' E. {Fd.) Wing, 7-0 ; middle toe, 1-45. f. S. Eastern entrance to Bass' Straits {llacgillicray). Wing, 7*4;; middle toe, 1-5. g. New Zealand {Sir G. Grey). Wing, 7*2; middle toe, 1-35. //. New Zealand (»S'//' G. Grey). Wing, 6-8; middle toe, 1-35. i. New Zealand {Dr. Hector). Wing, 6-8; middle toe, 1'6. k. (?. Ad. Pitt's Isl., Chatham Isl. {IF. L. Travers). Wing, 7-85; middle toe, 1-65. /. Auckland Islands (Antarctic Expedition). Wings, 6'65 ; middle toe, 1"45. ?». n. <3 . South Seas (Antarctic Expedition ; types of P. rossii)'. ?. Skeleton. Menado ; Dr. A. B. Meyer. The soft parts in the description are copied from an original label attached to Mr. Travers' specimen from Pitt's Island. Mr. Gould gives them as follows : — " Bill light blue, deepening into black on the sides of the nostrils and at the tip, '• with a black line along the sides of the under mandible ; ii'ides very dark brown ; " feet beautiful light blue." The plate represents the bill with a yellow " nail " at the tip, an important feature when considered along with the occasional appearance of a nail in specimens of P. desolatus {nide infra). Dr. Buller, in describing the bill, does not notice this yellow nail, and further information is desii'able on this point. Prion desolatus. Brown-banded Petrel, Lath. Gen. m..,23t. 2, p. 409; id. Gen. Hist, 's.., p. 187. Procellaria desolata, Gm. S. N. i., p. 562; Kuhl, Beitr. Zool. pi. -ai., fig. 7; Gray, Gen. B. iii., p. 618. Daption desolatum, Shato, Gen. Zool. xiii., p. 244. Procellaria turtiu', Kuhl, Beitr. Zool. p. 143, pi. xi., fig. 8 ; Smith, III. Zool. S. Afr. pi. 54; Gray, List Anseres, S^c., Brit. Mus. p. 165 ; id. Gen. B., iii., ^;. 648 ; Schl. Mus. P. B. Procell. p. 17 ; Layard, B. S. Afr. p. 361. s 13S ZO0I.(XiY OK KEKGUELEN ISLAND. Prion turtur, GoitU, Ann. N. JI. xiii., 1811, p. ac.ll ; id. B. Aiislr. yn., pi. 51; Bp. Consp. ii., p. 193 ; Bcichcnh. Ilandb. Lougipcnncii, pi. x..,Ji(js. 771, 775 ; Coues, P/-. Bhilad. Acad. ISGO, jIj. lOG; Gow/t/, Uandh. B. Aitstr. ii., i^. 172; Pe/;;. Pm. Isovara, J'og. p. 117; Glglioll, Faun. Vcrlehr. Oceano, p. 16; G^rav/, 7/a?»f?/:. P. iii., p. 108 ;' i>'«//6'r, B. X -^<'('/. y>. 309 ; Flnsch, J. f. O. 1872, p. 250, 1874, J3. 311 ; Sharpe, Toy. Erebus S)- Terror, Birds, App. p. 3-1, p)l. 29. Prion m-iel, Gould, Ann. N. JI. xiii., p. 306 ; id. Intr. B. Austr. p. 117 ; Bp. Consp. ii., p. 191 ; Grai/, Ibis, 1802, y;. 217 ; Giglioli, Faun. Vertcbr. Oceano, p. 45 ; Gould, Ilandb. B. Austr. ii., p. 173 ; Grai/, Uandl. B. ni., p. 108. Prion rossii. Gray, List Anseres, Sfc, Brit. Mus. p. 165 ; Bp. Consp. ii., p. 195. Prion brevirostris, Gould, P. Z. S. 1855, p. 88, i?/. 93. Haloba?na typica, Bp. Coiisp. ii., j^. 194. Procellaria arid, Schl. Mus. P. B. Broccll. p. 19. yEstrelata dosolata (pt.), Coues., Br. Bhilad. Acad. 1800, jij. 155. Pscudoprion banksii, turtiir, arid, brevirostris, Coues, t. c. p. 166. Prion desolata. Gray, Sandl. B. iii., p. 108. Pscndoprion desolatus, Coues, Bull. U. S. Nat. Mus. n.,p. 32. PacbyptUa arid et turtur, Cab. J. f. O. 1875, p. 449 ; id. 8f lleichen. J. f. 0. 1870, p. 328. Ad. clar6 cinercus, pilco vix satiu-atiorc, vitta m-opygiali nigricante iadistinct^, ; tectricibus alarum maximis brunneis, paullo cinereo lavatis, reliquis darfe cinereis ; ala spm-ia et tectricibus primariorum cinerascenti-nigris vix cinereo lavatis ; remigi- bus cinereis, intus albis, primariis longis extus uigricautibus, sccundariis intimis subterniinaliter nigricantibus, albo terminatis ; cauda clar6 cincrca, versus apiccm nio-ricante, frontc canescentc, plumis albido terminatis ; loris et vitta lata supcr- ciliari albis ; facie lateral! alba, rcgionc paroticu supcriorc cincrca ; plaga anteoculari fuscescente, albo variegata ; corpore subtris toto albo, colli et pectoris summi lateri- bus clarfe cinereis ; bypocbondriis et subcaudalibus paullo cinereo lavatis ; subalari- bns et axillaribus albis ; rostro coeruleo ; pedibus lilacino-ccErulcis, unguibus nigris ad basin lilacinis aut albis ; iride cyanescenti-cinereo. Long. tot. 10-5, culm. 1*1, alec, 0-95, caudjB 3-7, tarsi, 1-25, digit, med. c. u. 1-5. Tbe name of P. desolatus was originally conferred by Gmelin on Latliam's "Brown-banded Petrel." Tbe latter is stated to bave been in tbe collection oE Sir Josepb Banks, from tbe " Island of Desolation." Tbe description is apparently taken from a dried specimen, as tbe colom's assigned to tbe soft parts show : — "The «' bill is black with the tip yellowish, the legs brown, webs yellow, claws black." These are the colours which dried skins exhibit, but they are not found in any species of Prion when alive. It is, therefore, most probable that Latham's description was taken from an actual skin, as in 1824 he reproduces it almost verbatim in his " General History of Birds," with the habitat "Island of Desolation; Sir Joseph " Banks." In many instances Latham appears to bave drawn up his descriptions BIEDS. — K. B. SHARPE. 139 from the paintings and drawings made by Eoi-ster, Parkinson, and Ellis, for Sir J. Banks, but, when he has done tliis, he generally states the fact in liis latest work. We may therefore conclude that the type of his " Brown-banded Petrel " existed as a skin in Su- J. Banks' collection, probably as late as 1824, Init it is not now in the British Museum. Ellis' drawing is unmistakeable, and was perhaps taken from the actual type-specimen Avhen in the flesh ; it was from access to this drawing that the late Mr. Gray was enabled to make oiit that P. desolatus was a Prion, and the collection of specimens by the recent expeditions has led to the re-discovery of an interesting species. On looking over the series of Frion skins in the Museum, I have found a specimen from Christmas Harbour, Kerguelen Island, collected diu-ing the Antarctic Expedition, and I have carefully compared it with the other birds in the Museum collection. It is of the species usually called by naturalists Frion tiirfnr, and is, I believe, the true P. tiirhtr of Kuhl. At first I was inclined to consider this a different species, as it has such a distinct yellow nail at the end of the biU, but whether this is a specific character or not, I cannot at present say for cer- tain. It is very plain in some individuals, and in others very indistinct, so that it may be merely the fading of the bill after death. In the allied species, P. vit- tatns, Mr. Gould describes the bill as entirely blue, but then he figtires it with a yellow nail to the bill : this character, therefore, shoiild be looked after by any one who may have the opportunity. On mature consideration, I believe that Prion turtur is only the male of P. desolatiis, and is distinguished by its larger bill. This organ alone defines it from P. vittatus, which has a differently shaped bill, more bowed sideways from the base to the tip. P. ariel, P. rossii, and P. hrevirostris, are only young birds apparently with the bills not fully developed. The following is a list of the specimens of P. desolatus in the national collection. a. Madeira. J. Gould, Esq. ; type of P. hrevirostris ; wing 6-55, middle toe 1'4. b. Cape Seas. Su- A. Smith ; fig. I. c. as P. tiortur ; wing 7"3, middle toe 1'5. c. Eoyal Sound, Kerguelen Island. A. E. Eaton; wing 6-95, middle toe 1*5. d. Christmas Harbom*. Antarctic Expedition ; wing 7*15, middle toe 1-6. e. Australian Seas. Wing 7'1, middle toe 1-4. /. Cook's Straits, New Zealand. Dr. Lyall ; wing 6-9, middle toe 1-5. ff. ^. Indian Ocean, Lat. 40f° S., Long. 123^° E. {J. 3Iacgillivray). Capt. Stanley ; wing 6"G, middle toe 1'5. h. Indian Ocean (as above). Capt. Stanley ; wing 7'3, middle toe 1-5. The figm-es {PL vii.,^;^;. 8-10) illustrate the variation in size of bill in this species. They show the constancy of the shape, notwithstanding a slight variation in size. On comparing these figm-es with those of P. vittatus, it will be seen that the outline of the bill in the present species is different from that of the last-named bird. When seen from above, the sides of the bill are nearly straight, and this form is constant, even when the size varies conspicuously. [The burrows both of this species and of S. ccerulea resemble rats' holes. They s 2 140 ZOOLOGY OF KERGUELEN ISLAND. are usually iiindt> in AzoreUa or amongst Adctui iirowiug ii])()n dry rocky slopes or stony i^romul ; but a few of tlir birds took ])ossossion of sonuMil' tlio burrows out of ■wiiii'h H. C(Pnile« bad boon ('\ icicd, dropeninii' 1 hem to adapt flicni to tlirir own requirements. A well marked tia(d< l(>adin[>' to llic inoiitli ol' the hole is worn by tlie birds runnini!^ down the slope to gain impetus for their start on lakint;' ilii;lit, wliieli patb they also use in retm-niny; to the nest if they ebanee to I'all short of the entrance. Sometimes the burrows arc brancdied, and have two or three entrances ; occasionally their sidings are ctds de sac, and only abandoned " leads." Tiie egg is laiil upon loose d(>l)ris of Azorelht, &c., or on tlio bare ground constituti)\g the lloor of the terminal chamber, as is tliaf of II. C(crnlea* When we disembarked in Observatory Bay, P. desolatiis was pairing. Eggs obtained on the 29th of November were rresh. Most of the nestlings bad flown before we left the Island. It has already been stated by Dr. Kidder, and also in the Proc. Hoy. Soc. 1875, that during the breeding season, the various species of burrowing Petrels arc found iu their holes in pau's until the egg is laid ; after that, imtil the young is hatched, onlv one bird, at a time remains in the hole bv dav, the other returning with food at intervals during the night, and that when the nestling issues from the gq^, the parents leave it by itself the Avhole of the day, and visit it only in the night. The Storm Petrels have the same habits. The call uttered by this species in its flight may be denoted thus : u-u, vi -u, u- u, and so on. Now and then, it also (as I suppose) uses another call, which is repeated only at distant intervals. It consists of three short notes slurred, and the inter- mediate note is three tones bigber than the other two. At a distance, this has a resemblance to the mew of a cat. The multitudes of Blue Petrels Avhich breed in Kerguelen Island are hardly conceivable. Every dry hillside and knoll in the neighbourhood of Boyal Sound was populous with them to a remarkable extent. During the day the birds Avere silent, excepting when a noise happened to disturb them and cause them to coo. But on calm nights at the end of October and beginning of November, their mingled cries produce a low continued murmur like the sound of distant street trafSc in a large tow^n, in which the calls of only the nearer bnds can be dis- tinguished, and the rustling of their wings as they fly by is almost incessant. Eatlier Sidgreaves thought it would bave been worth while to ascertain how many of them on an average crossed the disk of the moon viewed through a teleseoj)e in the course of five minutes ; but more important Avork prevented this being done. * According to Dr. Kidder (Bull. U. S. Nat. jMus. 187o, p. 36) the egg is covered with the debris. 1 :im inclined to suspect that his finding it so was either dependent upon the style of the implement used by him iu digging the birds out, or was attributable to the vivaeity of his dogs when they drew the Petrel ; for I employed a sharp shovel-headed spade, and no dogs, and usually saw the egg when I stooped down to look into the chamber. BIRDS. — R. J). SIIARPE. 141 Thpy left the land before the Expedition sailed, so that towards the end of February scarcely a bird could be found anywliere in a burrow, and rarely could even one be heard at night, perhajis only one in ten or twenty minutes. They had withdrawn from the sheltered sound to the more open sea. When we were steaming past the Prince of Wales' Foreland en route for the Cape, large flocks of Blue Petrels were in close attendance upon dense shoals of fish between us and the shore, which Avere playing here and there at the surface of the sea, beating the water into foam. This Petrel burrows rapidly, loosening the soil with its beak, and shovelling the earth backwards with its feet like a domestic fowl engaged in dusting itself, whUst its wings are held just a little apart. Several examples of this species are in the S. African Museum under the names of Prior banksii and P. turlur, which have been captured at Green Point, near Table Bay. The louse obtained from F. desolatns is Lipeurus clypeatus, Giebel (one example only.) — E. A. JE.I Halobsena coerulea. Blue Petrel, Forster, Voyage, '\.,j). 91; Lath. Gen. Syn. iii., ^^. 2,^;. 415. Procellaria coerulea, Gm. S. N. i., ^j. 5G0 ; Bonn, et Vieill. Enc. Meth. i., p. 80 ; Gray, List Anseres, 8fc., Brit. Mus. ^j. 165 ; id. Gen. B. iii., p. 648 : id. ll)is, 1862, p. 247 ; Layard, B. S. Afr. j). 361 ; Buller, B. N. Z., p. 306 ; Finsch, J. f. O., 1872, p. 255, 1874, p. 208 ; Shat-jje, Voy. Ereb. 8r Terror, Birds, p. 33. Pachyptila ccerulea, Illiger, Prodr. Syst. 3Iamm., 8fc., p. 275. Procellaria similis, Forster, Draioings, No. 86 ; id. Descr. Anim., p. 59. Procellaria forsteri, Smith, Lit. Zool. S. Afr., pi. 411. Halobsena ccerulea, Bp. Consp. ii., p. 193 ; id. C. R. xlii., p. 768 ; Gould, Handb. B. Av.str. ii., p. 457 ; Coues, Pr. Philad. Acad. 1866, p. 163 ; id. ^ Kidder, Bull. U. S. Nat. Mus. ii., p. 34 ; iid. op. cit. iii., p. 17. Halobajna typica, Bp). C. B. xlii., p. 768. Zaprium coeruleum, Coues, Bull. TJ. S. Nat. 3Ius. ii.. p. 34. Ad. supra satm-ate cinereus, pileo distinct^ saturatiore ; fronte albicante ; tectri- cibus alarum cinereis, minimis nigricanti-brunneis, majoribus albo terminatis ; tectricibus primariorum nigricantibus ; remigibus cinereis, inttis albis, primariis extiis nigricantibus, secundariis intimis albo terminatis, subterminaliter nigrican- tibus, scapularibus secrmdariisque concoloribus et eodem modo coloratis ; cauda quadrata, cinerea, conspicui^ albo terminate ; facie laterali et regione siipraoculari all)is, regione parotica summa cinerea ; corpore subtiis toto albo, pectoris et colli lateril)us et hypochondriis imis cinereo lavatis ; subalaribus et subcaudalibus albis ; rostro nigro ; pedibus nigris ; ii-ide saturate brunneA. vel nigrt\. Long. tot. 10, culm. 11, alse 8-5, caudae 4-0, tarsi 1"25. The description is taken from Sii- A. Smith's specimen from the Cape Seas, 1J.2 ZOOLOGY OF KKRGUELEN ISLAND. fiLjuvcil bv liiin as P. fursfcri. The single spoc-imou brought by Mr. Eaton is uot adult, as it lias still a tVw remains of down attached to the sides of the neek. It is niueh darker than the one described, but 1 am nnable to judge Avhcthcr this is a sign of ao-c, or arises from tlu> exposure of Smith's specimen to the light, Avbich may have bleached it. The ]?ritish Museum now possesses the following skins : Glohiuccphalus was exposed upon the beach in Swain's Bay, a couple of dozen of these bii'ds \xsed to laki' possession of the carcass during the absence of the ilcnsing ])arty. It was amusing to Avateli I Iumu assembling to feed u])on it. Those which happened io Ik; flying about in its immediate neighbourhood began to circle i-ound it as soon as the men retired; and after passing and repassing it a few times they settled one by one upon the water, swam to the shore, and waddled quickly towards the krang. Others attracted to the spot by their movements might now be seen hurrying up from various directions to partake of the banquet. The largest of the birds meanwhile had taken her stand iipon the krang with outstretched wings. There she stood tugging at loose ends of sinews, and with difficulty tearing off with her beak morsels of the tough tiesh. If any of the others V(Mitured to approach too near before she had allayed her hunger, she rau open-mouthed at them with wings half spread, and drove them off with loud croaks. Seldom did any dare to withstand her attack ; if he did she allowed him to get Avliat be could from off the tail of the carcase where he could make very little impression upon it, wliilst she returned to the more fleshy pbrtion of the trunk. When the stronger bu'ds grew tired of eating, the others were permitted to feed ; and this they did greedily, quarrelling from time to time amongst themselves. A few yards away upon the slopes of Azorella, small groups of Skuas were standing waiting impatiently for an opportunity of gorging them- selves, but not daring to associate with the Giants. When the boat was rowed towards them, the Petrels alarmed waddled off in haste to the water, and swam away at full speed, looking like prototypes of the roe. We chased them, and they tried to rise from the water, running with their great feet splashing along the sui'face and flapping heavily with their wings (feet keeping stroke with Avings), making quite as much noise as Swans starting to fly. Where it was perfectly calm they could hardly take flight ; but where the breeze was blowing they easily rose into the air by running to windward. We drove some on shore uj) a hill ; it was a most exciting chase as we gained rapidly upon them ; but becoming fatigued with their climb they turned roiind and rushed past us down the .slope with an impetus that sufficed to start them on the wing. The Giant Petrel is troubled with an undescribed Lipeuriis (Giebel), and with Docophoi'Hs colo7'atiis, Eudow, neither of which were found by me. There is an example of O. gigantea in the South African Museum from the Cape Seas.— yf. U. K'] 15IRDS.— R. 15. SHARPE. 145 Diomedea exulans. The Albatross, Edioarcls, Not. Uiist. B. ii. pi. 88 ; L'iUbatros, Bviss. Orn. vi., 1). 12C. Diomedea exulans, Linn. S. N. i., p. 214: ; Gm. S. N. i., p. 506 ; Vieill. Gal. Ois. p. 231, pi. 293 ; Gray, Toy. JErebiis and Terror, Birds, p. 18 ; id. Gen. B. iii., p. 650; Gould, B. Austr. yn., pi. 38; Bp. Consp. ii., p. 184; id. C. B. xhi., p. 768; Later, in. N. Amer. B. p. 821 ; Cass. U. S. Expl. Exp. p. 397 ; Schl. Mus. P. B. Brocell. p. 333 ; Peh. Beis Novara, Vog. p. 147 ; Gould, Uandh. B. Austr. ii., p. 427; Coues, Br. Bliilad. Acad. 1866, p. 175; Becjl. et Eerhe, Orn. Eur. ii., p. 366 ; GiglioU, Faun. Vertehr. Oceano, p. 49 ; Gray, Handl. B. iii., p. 109 ; Anderss. B. Bam. Ld.p. 355; Finsch, J.f. O. 1872,^. 254, 1874, ^j. 206; Buller, B. N. Zeal. p. 289 ; Scl. and Salv. Nomencl. Av. Neotr. p. 148 ; Sharpe, Toy. Ereh. and Terror, Birds, App. p. 32 ; Bree, B. Eur. 1876, v., j?. 90 ; Kidder Sf Coues, Btdl. U. S. Nat. Mus. w..,p. 11, Cab. ^ Beichen. J.f. 0. 1876,^9. 328. L'Albatros clu Cap de Bonne Esperance, Buff. Bl. Enl. 237. Chocolate Albatros, Lath. Gen. Syn. iii., i)t. 1, p. 308. Diomedea spadicea, Gm. S. N. i.,p. 595. Diomedea albatrus, Pall. Zoogr. ii., ^9. 308; Forster, Descr. Anim. p. 27. Diomedea adusta, Tschudi, J.f. O. 1856, ^j. 157. Ad. \)\\xh albus, dorsi plumis, scapnlaribus et corpore subtus plus minusve cinereo fasciatim irroratis ; tectricibus alarum nigricanti-brunneis, basaliter allns ; vemigibus nigricantibus, secundariis et scapularibus albis, ad apicem cinerascenti- brunneis ; cauda alba, cinereo versus apicem marmorata, pogonio externo terminaliter nigricante ; corpore subtus toto cum subalaribus et subcaudalil)vis pm'e albis ; rostro albido, vix coccineo tincto, versus apicem flavicante ; palpebris pallide viridibus ; pedibus albis, coccineo tinctis ; iride satiu-ate alba. Long. tot. 38, culmen 7*6, alse 25'5, caudse 10-0, tarsi 4"9. Juv. brunneus, alis caudaque saturatioribus ; pUeo dorsoque dilute brunneis, basaliter albis et pallidius marginatis ; facie laterali, regione parotica, gulaque pm"^ albis ; subtus pallid^ brunneus, subcaudalibus saturatius brunneis. Considerable variation is seen in the plumages of the Albatross from youth to maturity, the young birds being brown, and gradually becoming whiter and Avhiter with age. According to Sir J. Hooker's " Journal" none of this species were observed at Christmas Harbom-, but above Cape Fran5ois " the nests were huddled together, as " many as 50 or 60 of them, and were built on the grassy slopes above the precipice " 700 or 800 ft. above the sea. A good deal of straw and stubble was mixed with " them, or rather plastered up with the clay to give it consistency. Theu" height was " about 1^ ft., and their breadth much the same. From a distance they looked " like so many Cheshii-c cheeses." T Lie ZOOLOGY OF KEimUELEN ISLAND. Tlic Museum contains spocinicns from the Cape Seas {Capt. naryy, E. M. Laiicjicorfhii, &n:.) ; Soutli Australia {Sir George Grei/) ; Noav Zealand; and one from Kerguelen. o. Ad. Kcrguelen Island. Antarctic Expedition. [The Great Albatross bred on the flat ground near Shoal "Water Bay. The birds captured on thcii" nests -were destroyed by liberty men from the IT.S.S. Monongahela for the sake of their wing-bones and feet (just as the Sooty Albatrosses were by our men), much to the regret of Dr. Kidder and the American Astronomers. Capt. Fuller of the whaliug licet said that a few pairs of the birds build near Sprightly Bay. Their occurrence in the neighbourhood of Mount Campbell AA-as reported by H.M.S. " Challenger." The vicinase of the Trincc of "Wales Foreland would have been worth a visit ; but there was no means of getting there from Observatory Bay by boat. The pre- valence of sudden and violent squalls makes boat navigation in the open Sound extremely dangerous. An American boat's crew sailing from Three Island Harbour was once detained nearly a fortnight at the Foreland by strong winds bloAving out oL' the Soimd ; and the men were almost starved before their schooner could rescue them. An adult Wandering Albatross can breathe without much difficulty with a weight of about 130 lbs. upon its back. "When specimens had to be killed we employed largo men to sit down upon them, holding their beaks to prevent the birds from biting. The pink stains on the sides of the neck mentioned in letters from the " Challenger " and in Dr. Kidder's report were well marked in our adult examples. Two new species of lice Avere obtained from this Albatross, namely, Docophorus dentatiis and Nirmtis anguUcollis, Giebel. Previously Lipeurm thoracicus, Eudow, was known to occur upon it. — A. E. E^ Diomedea melanophrys. Diomedea melanophrys, Temm. Fl. Col. 456 ; Gray, Gen. B. iii., p. 650 ; Gould, B. Atistr. TO.., pi. 43; Beichenb. Ecmdb. Zotigijmmes, pi. xvi.,figs. 797,798; Bp. Consp. ii., p. 185 ; Sehl. Mm. P. B. Procell., p. 33 ; Gotild, Uandb. B. Austr. ii., p. 438 ; Pelz. Beis. Novara, Vog. p. 148 ; Coiies, Br. Bhilad. Acad. 1866, p. 181 ; Gray, Saudi. B. m.,p. 109; Finsch, J.f. O., 1872, i?. 254, 1874, ^x 206; Anderss. B. Bamara Land, p. 355 ; Buller, B. N. Zeal. p. 292 ; Scl. 8f Salv. Nomencl. Av. Neotr. p. 148 ; Sharpe, Voy. Ereb. Sr Terror, Birds, App. p. 32. Thalassarche melanophrys, Bp. C. B., xlii., p. 768 ; Giglioli, Faim. Vertebr. Oceano, p. 57. Ad. supra branneus, dorso paullo cinerascentc, scapularibus dorso concoloribus, imis nigricantibus ; alls brunneis, tectricibus majoribus versiis basin cinereo lavatis ; remigibus satm-at^ brunneis, intus cinerascenti-brunneis, versus basin albis ; m-opygio et supracaudalibus pur^ albis ; cauda cinerascente, versus apicem brun- nescentiore, ad basin albida ; pileo et collo undique albidis, hoc vix brunneo lavato ; BIRDS. — li. B. SHARPE. 147 plumis snpvaocnilarihus saturate hrunncis ; corpore subtus toto purt' albo ; subalavi- bus et axillaribus f umoso-lirunneis, majoribus versus apicem ciiierasce7iti-albis ; remigibus infra pallido fumoso bruimeis, intus basalitcr albis ; rostro f ulvescenti- albo, basalitcr nigro marginato ; peclibus flavicantl-albis, pallide coeruleis ; iride pallide brunneS., saturatius marmorata. Long. tot. 28, cuhnen, 5*2, alae 20*5, caudse 8-0, tarsi 3-3. [Not observed by us. The Challenger Expedition believed that they saw this Albatross on the eastern side of the island, but it was not seen by Dr. Kidder or l)y the German expedition. Its louse is L'qienrm prox. — A. E. E.'\ The specimens in the British Museum are two in number, a. ad. Cape Seas. Purchased. h. s ji^iii- Off the entrance to Bass' Straits, July 11, 1847 {J. MacgilUvray). Capt. Stanley. Diomedea culminata. Diomedea culminata, Gould, P. Z. S. 1843, p. 107 ; id. Ann. N. H. xiii., p. 361 ; id. B. Austr. vii., pi. 41 ; Gray, Gen. B. iii., p. 650, ]il. 179 ; Bp. Consp. ii., p. 185 ; id. C. B. xiii., p. 768 ; Schl. Mus. P. B. Procell., p. 35 ; Coues, Pr. Philad. Acad. 1866, iJ. 183; Gotdd, Handb. B. Atistr. u.,p. 'kSl ; Gray, Hcmdl. B. iii., p. 109; Cah. ill Von der Decken's Eeis. m.., p. ol; Finsch, J. f. 0. 1872, j). 254, 1874, p. 206; Buller, B. N. Zeal. p. 295; Sharpe, Voy. Ereh. 8f Terror, Birds, App. p. 32 ; Cab. 8r Beichen. J.f. O. 1876, i?. 328. Diomedea chlororhynchus, Atidub. B. Amer. 8«o. viii., p. 79. Thaiassarche culminata, Giglioli, Faun. Vertebr. Oceano, j). 59. Ad. pileo colloque undique pallida cinereis ; plumis ante-ocularibus et regions oculari saturate cinereo lavatis ; fascia infra-oculari alba ; dorso cinerascenti- brunneo, scapularibus alisque nigricanti-brunneis ; remigibus nigricantibus, intiis chocolatinis, scapis flavicantibus ; uropygio et supracaudalibus pur6 albis ; cauda saturate cinerascenti-brunnea, scapis flavicanti-albidis ; corpore subtiis toto pui"6 albo, facie et colli lateribus vix cinereo lavatis ; pectoris lateribus intimis celat6 et axiUaril)us intimis satm'at^ brunneis, his longioribus et subalaribus piu'6 albis, subalaribus alse margini proximis brunneis ; rostro nigro, culmine comeo, mandibula, ima et gonyde aurantiaca. Long. tot. 30, cuknen, 4*7, alae 20, caudse 8*0, tarsi 2'8. Juv. similis adulto, sed saturatior, pileo colloque saturatius cinereis. This species occm*s in the German list as obtained at Kerguelen Island, a skeleton of the bird having been preserved. Dr. Kidder in the American Report also states that it was common along the coast, and was occasionally seen in Boyal Sound. The following specimens form the series at present in the British Museum. a. ad. Lat. 36^° S., Long. 95i° E. Capt. Stanley. June 5, 1847 {J. Mac fjillivray). T 2 148 ZOOLOGY OF KERGUELEN 1SLA>;D. h. ad. Australian soas. .1. Gould, Esq. c. juv. Oil" Van DiiMiicn's Land, August 3, 1839. rf. juv. Olf Voragua. Mr. lU-ydges. Diomedea fuligiiiosa. Albntross with a white ovohi-ow, Coolca t'oy(i()e., i., p. 38. Sooty Albatro.ss, Lath. Geii. Si/n. iii., pt. i.. p. 309. Diomodoa fuliginosa, Gm. S. N. i., p. 595; Groi/, Gen. J!, iii., //. 050; Bp. Coiisp. ii., /;. 18(5; Lairr. B. N. Amer. p. 823; ScJd. IIiis. P. B. ProcelL, p. 35; Feh. Beis. Nomra, T'og.p. It9; Grot/, Ilandl. J3. iii., p. 109; Finsch, J. ,/". O. 1872, p. 25 J., 187-1, p. 206; Bidler, B. N. Zeal. p. 290; Sharpe, Voij. Ercb. c^- Terror, Birds, App. p. 33; Cab. ^' Beichen. J.f. O. lS7(i, p. 328. Diomodoa spadicoa, Less. Man. d'Orn. ii., p. 391. Diomedea fusca, Andith. B. Amer. pi. ccceliv. ; id. op. cit. Svo. viii., p. 83, pi. 454. Diomedea palpebrata, Forster, Descr. Anim. p. 55. rhoebetria t'uliginosa, Bp. C. B. xlii., p. 768; Goidd, llandh. B. Anstr. ii., j;. Ill; GiglioU, Faun. J'ertehr. Oceana, j}. GO ; Coues, Br. Bhilad. yicad. 1866, p. IbG ; id. and Kidder, Bull. U. S. Nat. Mus. ii., p. 21 ; iid. op. oil. iii., p. 12. Ad. supr;\ fiiliginoso-brunneus, plumis dorsalibus sscpfe pallidiiis marginatis ; loris, facie laterali, mentoqne saturatius brunneis ; annulo supra- et post-oculari argentes- centi-albo ; alis fuligiuoso-bruuneis, remigibus saturatius brunneis, scapis albicantibus, secundariis basalitcv cinerascentibus ; cauda saturatt^ brunnea, rcctricum seapis eon- spicue albis ; corpore subtus cinerascenti-brunneo, plumis obscuri^ fulvescenti- brunneo marginatis ; subalaribus et remigibus infra chocolatino-brunneis ; rostro nigro ; pedibus paUid^ carneis ; iride purpurascenti-griseo. Long. tot. 27, culm, 4-7, alfe 20-5, caudae 105, tarsi, 3*0. Besides tlu-ec examples from South Australia {Sir George Grey), and Lat. 38° S., Long. 30° E. {J. 3IacgilUvray), the British Museum has the following : — a. ad. Royal Sound, Kcrguelen Island. Capt. Inglis, R.N. [The Sooty Albatross is common in Royal Sound. The hills near the sea on the mainland and islands present occasionally places suitable for its nidification. As a rule the nests are built in the most sheltered situations that can be found at the foot of the precipitous terraces of volcanic roct which are so characteristic of the neigh- bom-hood. Here and there recesses hollowed out at the base of these terraces and cliffs are thoroughly protected by the overiianging rock from wind and rain. In dry nooks of this nature I>. fuliginosa constructs its nests of pieces of adjacent plants (especially Festuca erecta) disposed in the form of alow truncated cone hollowed out at the top. The nests appear to be used many years in succession, as the original materials of several that were examined seemed to have been reduced by age to BIRDS.— R. B. SHARPE. 149 vegetable mould. These old fabrics are relined with fresh dry grass when the birds return at the commencement of the breeding season. The position of her uest is liable to be betrayed to i^crsons walking mthin sight of the female when she is sitting, for every now and then while she is observing their movements she Avill utter her cry, and thus reveal her situation. If anyone goes near her she assumes a rather formidable attitude, and rufflmg up the feathers of the neck snaps fiercely and loudly with her beak at the intruder, the noise resembling that made by a large dog in catching flies. But notwithstanding her menacing gestures the egg can be secured (if it be desired) ^rithout displacing her from the nest. A pocket hand- kei-chief presented to her with the loft hand, or a hat placed gently upon her head, will completely engross her attention while the egg is being abstracted from beneath her with the right ; and she will afterwards remain in the nest complacently watching her visitor's retreat. Nearly a dozen of their nests were taken by the English Expedition. From one found near Thumb Peak a female was brought by Staff Commander Inglis, of H.M.S. " Supply," from whom the skin preserved was obtained. On the 23rd of October a female was killed, while she was sitting with the male in her nest, by Lieut. Dowding, R.M.I., and Mr. Edwardes, Assist. Surgeon of H.M.S. " Volage." They found that she would have laid her egg in the nighty its shell being spotted already. Another female subsequently laid in the same nest. The Eev. J. B. Budds found a nestling about a week before we sailed from Royal Sound ; a newly born kid was sitting upon it. On our way to the Cape we saAV a Sooty Albatross the pale band on whose neck was of a dirty white instead of the usual ash colour. A single specimen of a Lipeurits was found upon a Sooty Albatross in Royal Sound (probably a moulted skin). — A. E. JE.'] PELECANIDiE. Phalacrocorax verrucosus. Graculus carunculatus, Schl. Miis. P. B. Felecani,p. 20 {spec. h.). Halieus (Hypoleucus) verrucosus, Cab. J.f. O. 1875, 7;. 450,^:;/. l,y?^. 1. Graculus carunculatus, Coues Sj' Kidder, Bull. TJ. S. Nat. Mns. ii., j). 7 ; iid. oj). oil. iii., J9. 8. Halieus verrucosus, Cab. 8f Beichen. J.f. O. 1876, ^j. 329. Ad. supra purpurascenti-nigcr, pileo pauUo cristato ; intcrscapiilio vix virides- centc ; scapularibus et tectricibus alarum distincte viridescentibus ; rcmigibus bnuineis, extus viridescente lavatis ; cauda nigra ; facie laterali et genis anticis purpvu'ascenti-nigris ; genis posticis, colli lateribus, et corpore subtiis toto, pm'6 albis, hypochondriis imis tibiisque purpixrascenti-nigris ; subcaudalibus albis, longioribus piu-purascenti-nigris ; pectore laterali celatim viridescenti-nigro ; subalaribus sordide viridescenti-nigris, his imis et remigibus infra brunneis ; rostro nigro ; carunculis lull ZODLOGY DF KEKtiUELKN ISLAND. rostri basal is hctcUavis; poililms ll:ivis ; iridiM-yanoa. Long', tot. 25, cnlin. "J^S, aliv 11-2, caudiv 5-0, tarsi 2-1'. JiiF. purpurasccnti-brunneus, dorso obscurf^ viridescento ; subti\s brunncus, plumis basalitor albidis. iaciem striolatam cxbibontibiis ; t^enis posticis gulAquc \mvi> albis. Professor Scblogol, \\\\o bas a KerL;uolcn siipcmiuoii in tbo Loiden Museum, considers it to be speeilically tbe sanie as tbe Falkland Island Cormorant, " tr6s " rcconnaissable il uue lai-ge raic blaucbc s'otendant sur Ics plumes de I'aile couv- "■ rant I'avant bras." Dr. Cones, after examining tbe specimens brougbt by Dr. Kidder from Kcrguelen, wi'ites : " I liave no besitation in identifying tbis species " as above, altbougb tbe single adult specimen collected does not sliow tbe wliite " trans-alar fascia spoken of by autliors." lie is, bowever, evidently influenced l)y Scblegol's determination. On tbe otber band. Dr. Cabanis considers that tbe Kcrguelen bird is different on account of its smaller size, especially of tbe feet and bill, and from tbe want of the white band in tbe wing. The other diirerenccs men- tioned by Dr. Cal)anis do not seem to me to be of any great importance, but the Kercriielen skins in the Museum certainlv do exhibit the difTeronces enumerated bv him. I consider that the material at my disposal is too limited to decide the question, and I therefore follow Dr. Cabanis in his determinations, a. ad. ; h. juv. Kcrguelen Island. Antarctic Expedition. [C] [Tbe habits of tbis Cormorant are so similar to those of the common British species, that it is needless to describe them. Mention has already been made else- where (Proc. Roy. Soc. 1875) of their remarkable tameness. Another strongly marked trait in their disposition is inquisitiveness. Not only will they direct thcii' flight towards a man walking by the shore, in order to have a good look at him ; but if he chances to be standing still, the Shags will not unf requently alight close to him to stare at the stranger. Sometimes they are attracted by noise, not bowever quite to so great an extent as tbe Spitsbergen Guillemots. On the day of our first landing in Observatory Bay, whilst standing amongst a flock of Cormorants Avhich were basking on a point, I fired at a Teal. The Shags fled precipitately at the sound of tbe report ; but three minutes had hardly elapsed before five and twenty of the birds were standing round me in a circle almost within reach of the gun, mute with astonishment, looking at me first with one eye and then with the other. There is much virtue in a mere name. Our men called these Cormorants " Shags," and would not touch them. Some of our American friends (not the astronomical party) having designated them " Shag-ducks," shot a few dozens of them with rifles, to eat them. They could have killed as many as they pleased, for in the intervals of fishing the Cormorants rest in the cliffs, and do not readily take flight unless thoroughly alarmed. The Shags in Observatory Bay were commencing to build on the 16th of October. The first eggs were found by us abovit the middle of November. BIRDS. — R. 15. .SHARPE, 151 The British Cormorant (P. carbo.) has a Bocophorm, but the louse of the Kerguelen species was not discovered, though I procured a carpet bag full of the nestlings for my Skuas, Dr. Kidder found "a tick of prodigious size " upon some young birds. — A. E. JB^ Tach3rpetes aquila. The Man-of-War Bird, Edwards, Gleanings, vi., p. 209, ^;^. 309. La Fregate, Briss. Orn. \\., p. 506, jil- xliii.,^^. 2 A. Pelecanus aquilus, Einn. S. N. i., j). 216. Le Grand Fregate de Cayenne, Etiff. El. Enl. vii., pi. 961. Frigate-Pelican, White-headed Frigate-Pelican, and Palmerston Frigate-Pelican, Lath. Gen. Sijn. iii., pt. 2, pp. 587, 591, 593. Pelecanus leucocephalus and Pelecanus palmerstoni, Gm. S. N. i,, j)- 572. Fregata aqmla, Illiger, Erodr. Si/st. 3Iamm. 8f Ac. p. 279 ; Gray, Eist Anseres Brit. Jfus. ]). 190 ; Gosse, B. Jamaica, p. 477 ; Eeich. Randh. Steganopjodes, pi. xxxi., figs. 372 ; G. C. Taylor, Ibis, 1859, p. 150 ; Schl. Mas. F. B. Eelec. p. 2 ; Buller, B. N. Zeal. p. 339 ; Gigl. Eaun. Vertebr. nelV Oceano, p. 63 ; Sharpe, Voy. Erebus 8f Terr., Birds, Apip.p. 35. Tachypetes aquila, Vieill. N. Bid. d'Sist. Nat. xii., p. 143 ; id. Gal. Ois. ii., p). 187, ^J^. cclxxiv. ; Kittl. Kupf. Vbg. p. 15, taf. -s.^.,fig. 1; Eess. Traite, p. 606; Audiib. B. Amer. pi. ecccxxi. ; id. B. Amer. 8to. vii., p. 169, p)l- 421 ; Bp. Consp. ii., p. 167; Lawr. B. N. Amer. p. 873; Eartl. Orn. W. Afr. p. 260; Burm. Th. Bras, iii., p. 549 ; Cass. U. S. Expl. Exp. p. 358 ; Newton, Ibis, 1859, p. 369 ; Blasias 8c Baldam. in Naum. Vog. Eeutschl. xiii., p. 287 ; Gould, Handb. B. Austr. ii., p. 499 ; Finsch 8^ Earth Faun. Central-Folyn. p. 265 ; Finsch, J. f. 0. 1872, p. 260, 1874, 1). 216; Cab. Sr Beiclien. J.f. O. 1876, i;. 329. Tachypetes leucocephalus, Kittl. Kupf. Vog. p. 15, taf. xx.,fig. 2. Attagen aquila, Gray, Gen. B. iii., p. 669 ; Gould, B. Austr. Intr. p). c ; Jet^d. B. Ind. iii., p. 853. Fregata leucocephala, Beichenb. Eandb. Steganopodes, pi. xxxi., ^(7. 373. Tachypetes palmerstoni, Cass. U. S. Expl. Exp. Birds, p. 359. Tachypetes minor, Eartl. Orn. Madag. p. 87. Ad. Niger, dorso toto a^neo vu'idi et piu'pureo nitente, plumis lanceolatis ; pileo saturatiore, saturate viridi ; alis caudaque nigris vix bronzino lavatis ; gula nudft rubra ; corpore reliquo subtus nigro, pectore medio purpureo, lateraliter teneo- viridi ; subalaribus nitid6 nigris ; rostro cinerascente, versus apicem nigro ; pedibus carneo- bninneis ; iride nigra. Long. tot. 34, culm. 4*0, alse 23*0, caudae 16*0, tarsi circa 2"85. Juv. brunneus, purpureo et viridi nitens ; tectricibus alarum pallidC' brunneis albicante marginatis ; pilco nigro, collo postico brunuco, plumis pallidius margiuatis ; gula nuda miniis extenst\, gutture et facie laterali saturatin brunneis ; pectore albo, abdomine brunneo, subcaudalibus et subalaribus nigris ; alis caudflque nigris. 152 ZOOLOGY OF KKKUUKLKN ISLAND. The German cxpodition liroui^ht back a lu-.-uI (if tliis species obtained at Ker- Jiuclon Island ; it bas not been mot witb bv any otbor of tlio visitors. 1 bav(> not assured nnsolf ibat T. minor is really a distinct species, I'or allboiigb sonic oT tlie l)irds are evidently less bulky in their proportions, on measuring* tbcm, tbn sup- posed dilVerences of size arc found to bo very sligbtly pronounced. As the Museum series is not sufficient to enable me to determine Ibc question for certain, i have refrained from giving a list of the specimens. IMPENNES. Aptenodytes longirostris. Patagonian Penguin {pt.), Peiiii. F/ill. Trans. Iviii., j). 91, pi. 9; Lal/i. Gen. Si/ii. \i.,j). 5(53. Aptenodyta patachonicha 0^/.), Gm. S. JV. i.,]). 556. Le ]\[ancbot do la Nouvelle Guinee, Sonii. Toy. N. Giiin.p. ISO, pi. 113. Le Mai\ehot des Isles Malouines, JSnJJ'. J?l. Enl. x., J?l. 975. Apterodyta longirostris, Scop. Del. Faun, et Flor. Insuhr. ii., /j.91, no. 69. Pinguinaria patachonica {nee Forster), Shaio in Miller's Cimel. Phi/s. p)l- 4;5 ; id. Nat. Misc. xi., pi. 409. Hairy and "Woolly Pinguin, Lath. Gen. Hist, x., p. 392. Aptenodytes pcnnantii, Gray, Ann. N. H. xiii., p. 315 ; id. Gen. B. iii.,^>. 612; Beichcnh. JIandlj. Fij(jopodcs, pi. 1, figs. 1, 2 ; Bp. C. B. xlii., p. 775 ; Gould, P. Z. S. 1859, iJ. 98; Scl. P. Z. S. 1800,;;. 390; Hi/atl, Cat. Orn Coll. Boston Soc. N. IE. i., p. 11 ; Cah. 8f Beiehen. J.f. O. 1876, p. 330. Spbenisctis pcnnantii, Sehl. Mus. P. B. JJrinatores, p. 3 ; id. Dirent. p. 268. Aptenodytes longirostris, Cones, Pr. Philad. Acad. 1872, p. 193 ; Sharpe, Voy. Erebus and Terror, Birds, App. p. 37, ;;/. 32 ; Coues and Kidder, Bull. U. S. Nat. Mus. ii., p. 39 ; lid. op. cit. iii., p. 18. Supra cinerascens, plumis omnibus cincreo apicatis, supvacaudalibus majus distincte tenninatis ; alis cinercis, pennis remigialibus seriatim cincreo terminatis, margine alari suninio nigricante ; cauda rigida nigra ; pileo summo usque ad nucliam nio-ro ; facie laterali gulaque tota nigris ; plaga lata aurantiaca j\ regione parotica postica per coUnm lateralem angustante et gulam uigram marginante ; colli lateribus cinereis dorso concoloribus, antic6 late nigro marginatis : jugulo medio aurantiaco ; corpore reliquo subtiis sericeo-albo, pectoris lateribus dorso concoloribus ; ala subtiis alba, versus basin et apicem nigricante, margine alari late nigricante ; rostro nigro, versus basin mandibulse carneo ; pedibus nigris ; iride Iset^ brunnea. Long. tot. 34, culm. 3-4, alse ll'o, cauda 44. The Kerguelen Island specimens in the British Museum are of larger bulk than those from the Falkland Islands. 15IRDS. — R. B. SIIARPE, 153 (I. ad. Kerguelen Island. Antarctic Expedition. h. ad. Swains' Bav, Korsruelen Island. Rov. A. E. Eaton. c, (I. ad. Falkland Islands. Antarctic Expedition. [The King Penguin does not breed in Royal Sound nor in Swain's Bay. The master of the schooner " Roswell King," Capt. Fuller, stated that it breeds in very few ])la('es upon the island ; that a large community occupy a position on the hills west of ]\[<)unt Ross near Tal)lo Bay (Kcrguolen Island), and that there arc others near Capo Sandwich. They were seen from the " Volage " in December at this last-named locality. The eggs are laid about the beginning or middle of October. In December and January small parties of these Penguins come into sheltered inlets to moult. "Wo used to find them in Swain's Bay, Carpenter's Cove, and in a bay near Vulcan Cove. Usually they were standing amongst the herbage within a few yards of the shore ; occasionally tliey were between the tide marks. The officers of H.M.S. " Volage," who had more opportunities than I had of seeing these birds, were of opinion that they remained on shore without food until the moulting was completed ; because if the Penguins while the change of plumage was progressing came from the sea every day, their breasts would in all probability be denuded of feathers, for they are then so easily detached from the skin that they could hardly fail to be stripped oS in the efforts of landing ; whereas their breasts were well clothed with old loose feathers until the new plumage was matured. Tliei-e are so few land animals in Kerguelen Island that the unwonted sight of people walking itever failed to attract the notice of the King Penguins. Standing at their ease in their sheltered lioUows they uttered as it were derisive cries from time to time while the strangers laboured through the Azorella. Seldom did they take the trouble to stii" when anyone approached them, but remaining in a group, some standing still, others lying down, they quietly awaited the progress of events. Their unconciousness of danger was singularly shown by the following incident. One day while grappling for Algce in Swain's Bay I came with one of the men upon six Kings in a group. Seeing that some of them had finished moulting and were well coloured, we walked up to them, seized the two finest by their necks, and sat down upon their backs. The others stayed beside us unconcerned at the fate of their companions, though they were beating the ground beneath us with their wings and gasping for breath within a yard or so of them. "What shall be will be :" so they made themselves comfortable, and they were not molested. Meanwhile my bird was becoming moribund ; and happening to look at its eyes I noticed that the colour of the iris was a very dull hazel, and that the pupil in contracting assumed a quadrangular form. The eyes of the survivors of the party presented the same peculiarity, which appears to l)e a characteristic of the species. V 154 ZOOLOGY OF ki:kguelen island. There are Tour examples of this Penguia from tlic Crozcttcs in the South African Museum, 2 youn^- and 2 adults. The louse of the Kiny; Penyuin is Gouiodos hrcoipca, n. sp., uf \vJiieh i found only one example. — A. E. E^ Pygoscelis taeniata. . Le Manchot Papon, Sonnerat, Toy. N. Gitiii. p. 181, jsZ. cxv. A])tenodytes papua, Forster, N. Comm. GOlthxj. iii., p. 140, pL 3 ; Gm. S. N. i.,p. 550 ; Jlci/f. Gal. Ois. il, p. 2iG, pi. 299 (var.). Papuan Penguin, Loth. Gen. Syn. iii.,^j^. 2, p. 665. Apterodyta papua?, Scop. Del. Faun, et Flor. Insubr. ii., p. 91, No. 71. Chrysoeoma papna, Steph. Gen. Zool. xiii., p. 59. Pygoscelis papua, Gray, List Anseres, Sfc, B. M., p. 153 ; id. Toy. Ereh. 8j- Terror, Birds, pi. 25 ; Ileichenh. Handh. Pygopodes, pi. ii., fig. 738 ; Jlyatt, Cat. Orn. Coll. Boston Soc. N. U. I, p. 13 ; Cab. ^ Bcicl/en., J.f. O. 1870,^;. 330. Eudyptes papua, Gray, Gen. B. iii., p. Oil ; Cass. U. S. Expl. Exp. p. 261< ; Goxdd, P. Z. S. 1859, jj. 98; Abbott, Ibis, 1860, i^. 336. Apteuodytes tseniata, Peale, U. S. Exjil. Exp. p. 264. Pygoscelis wagleri, Sclater, P. Z. S. 1860, p. 390. Spheniscns papua, Schlegel, litis. P. B. TIrinatores, p. 5. Pygoscelis taeniata, Cones, Pr. Acad. N. Sci. Philad. 1872, p. 195 ; Scl. Sf Salv. Nomencl. Av. Neotr., p. 151; Sharpe, Voy. Ereb. Sf Terror, Birds, Ajyp. p. 38; Cones 8f Kidder, Bull. U. S. Nat. Mm. ii., ^j. 41 ; iid. op. cit. iii., p. 18. Ad. supra nigricans vix cinereo lavatus ; alis magis cinereis, margine alari conspicua et remigum apicibus fasciam terminalem latam formantibus albis ; supracandalibns rigidis nigricantibus cinereo lavatis ; rectricibns nigris, marginalitcr brnnnescentibus ; fascia lata verticali all)a. ab utroque oculo per verticcm ducta ; facie laterali et guttm-e cinerascentibus, gntturis plumis albido variis ; corpore reliqno subtus sericeo-albo ; ala inferiore alba, remigibus extimis apicaliter cinereis plagam conspicuam exhibentibvis ; pectore subalari et plaga altera ad ortum alae posita cinereis ; rostro Isete am'antiaco, culmine nigro ; pedibus aurantiacis ; iride Isete bmnnea. Long. tot. 31, culm. 24, alae 8-5, caudae 5-5. a. ad. Kerguelen Island. Antarctic Expedition. b. ad. Kerguelen Island. Antarctic Expedition. c. iuv. Falkland Islands. Su- W. Bm'nett and Admu'al Fitzroy. [The Johnnie (as the -whalers call this bii-d) is common in Royal Sound. It builds in commvmities, some of only a dozen, others from 70 to 160 families. A more populous colony ixpon the mainland was visited by six officers from the ships, who estimated the nvimber of nests in it to amount to 2,000 or more. These laro-er communities are approached from the sea by regular paths, conspicuous at a BIRDS. — H. B. .SHAHPE. 155 distance, like well-worn sheep tracks, which lead straight up the hill from the water. Their formation is due to the Penguins being very particular about wliere they land and enter the sea. A small party of the birds occupied a position upon the neck of a low promontory within an liour's Avork of Ol^servatory Bay. Their nests were nearest to the farther side of the isthmus ; but when they were approached the male birds used to run to the water, not by the shortest route where it was deep close to the rocks, but by the longest to a place where the shore was shelving. It Avas amusing to see tliem start off in a troop as fast as theii" legs could carry them, holding out their wings and tuml)ling headlong over stones in their way, because as they ran they would keep looking back instead of before them, and to hear their outcries. Panic and consternation seemed to possess them all ; but the females (possibly because they could not keep up with their mates) seldom went far from their nests ; and, if the intruder stood still, soon retui'ned and settled down again upon their eggs. Not many weeks had passed before a change was effected in their conduct. The young were hatched, and now the mothers anxiously endeavoured to persuade them to follow the example of their fathers and run away to sea. But the nestlings preferred to stay in their nests ; they did not mind if the stranger did stroke them; although their anxious mothers ran at him with oj)en mouths whenever he dared to do so. Only a few of the older chicks could be prevailed upon to stii' ; and they after waddling a few yards became satisfied with then- performance and turned to go home again. The mothers, who had straggled to a greater distance, began to return too. It was now that the more tardy youngsters began to experience the ills of life. Every Penguin that had reached its place before them aimed blows at them as they passed by towards their oavu abodes. One of the little birds certainly did seem to deserve correction. It saw its neighboiu"'s nest empty and sat down in it. The old female Johnnie, the rightful occupier, presently returned in company with her own chick, to whom, having put her head well into his mouth, she began to administer refreshment after his rim. Seeing them so pleasantly engaged, the small vagrant, thoughtlessly presuming upon her generosity, went nearer and presented himself to be fed also, as if he had a right to her attention and care. She looked at him while he stood gaping before her with drooping wings, unable for the moment to credit what she saw. But suddenly the truth flashed upon her, and provoked by his consummate audacity slie gave vent to her indignation, pecked liis tongue as hard as she could, chased him out of the nest, darting blows at his back, and croaked ominously after him as he fled precipitately beyond the range of her beak, leaving trophies of down upon the scene of his unfortunate adventure. The whole of this community of Penguins was subsequently boiled down into " hare soup " for the officers of H.M.S. "Volage;" and very nice they found it. The nests were composed of dried leaf-stalks and seed-stems of Pringlea, together with such other suitable material as happened to be at hand. There were two eggs u 2 I-jG zoology of Ki:i{(,ri;i.i:N island. in ovpvv no-it. and one of (luMn ^v:ls invarinbly lari;t>r tlian tlio otlior. Most likely the birds liatolunl from the larii:tM' ciri^s aro ol" tli(> opposilo sox to those Avliicli aro produced from the smaller, \\hether the big or tlie little egg is the lirst to be laid was not ascertained. As is the ease witli many dther Iviiuls of I)inls, Johnnies are very regular in llicir habits. Every afternoon at nearly the same time they repair to tlie shore when th(>v have done tisliing, landing in small i);irti<'s at llieir aeeustomed plac(>s at the heads of shallow inlets. On issuing from llic water they dispose tluMnselves to vest, seldom proceeding beyond the verge of the shore. Those which are inclined to sh-ep put their heads behind their flippers; the others stand amongst them with the neck shortened so as to bring the head down close to the body with the beak slanting upwards and forwards, somewhat in the manm>r of a very young thrush during repose. Their eyes present a rather tearful appearance, and resemble bits of dull black glass s(>t in their heads, — perhaps the nietitatiug membrane may be kept drawn over tluMu. At frequent intervals a kind of watery fluid is ejected from their mouth l)v a shake of the head. 1 was led to suspect that these Penguins ;uv liable to be attacked l)y seals,* for in places not much frequented by man, if they once effect a landing they do not readily return to the water on being alarmed, but run away froni the sea up hill as fast as they can go. After they have gone some distance they turn round and look back while they take breath ; but as soon as they are rested sufficiently they willingly resume the ascent. It is not until they have been driven so far as to become thoroughly tired that they refuse to proceed further; but Avhen this stage has been reached it is useless to urge them to advance without a pause. As they face about, the sight of the boot ready to push them over is greeted with deprecating sighs, and should these be disregarded, and they be sent over upon their backs, as soon as they regain their feet they rush at their driver, launch their bill at his knees, beat their wings furiously against his calves and shins, and make a dash on all fours down the hill at full speed to regain the sea. TTlien they became accustomed to being chased by men, the Penguins acquired the habit of betaking themselves to the water at the first alarm. A small party of these bii-ds used persistently to land in Observatory Bay every evening at the very time when the men erecting our huts were returning to the ship after their work. Such of the Johnnies as managed to escape being caught one day vceve sure to reappear the following evening just at the critical time, dragging themselves out of the water to afford sport to the men. By the time that the huts were completed the survivors were reduced in number to a couple of birds ; and there can be little doubt that these would have followed their late comjjauions into the soup-kettle had the putting up of the Observatory occupied one more dav. * lu the Arctic Regions Loon? are occasionally caught on the water by Walrus. ]$IRDS. — R. H. SHARPE. 157 The cry of the Johnnie distantly reserahlcs the short hark of the fox. In the South African Museum there is an adult specimen from the Crozettes. Its louse is unknown. A full-grown bird sent me by Captain Fairfax M'as infected with a tick.—,/. E. -£".] Eudyptes chrysolophus. (Plate VIII., fig. 2.) Catarractes chi-ysolophus, Brandt, Bull. Acad. Sci. St. Felersh. ii., p. 314 {iiec aiict. recent.). EudA'ptes chrysocome (nee Forster), Abbott, Ibis, 1860, ji- 337 ; Set. P. Z. S. 1860, iJ. 390; Cab. cj- Beichen. J. f. O. 1876, ^j. 330. Eudyptes diadematus, Gould, P. Z. S. 1860, p. 419 ; Schl. Mas. P. B. TJrinatores, p. 8 ; Coues, Pr. Philad. Acad. 1872, p. 206 ; id. §• Kidder, Bull. U. S. Nat. Mus. ii., p. 47 ; iid. op. cit. iii. p. 20. Eudyptes catarractes. Gray, Sandl. B. iii., p. 98. Ad. supra nigricans cinereo lavatus, alis cinereo nigricantibus, margine alari summa vix albicante, margine remigiali medialiter alba ; cauda rigida, dorso concolori ; facie laterali gulaque dorso concoloribus ; pilei plumis nitidis nigris elougatis cristam formantibus, froutis plumis basaliter aurantiacis ; fascia superciliari cristali a loris supra ociilum per latera capitis ducta ; corpore reliquo subtus piu'e all)o, pectoris lateribus dorso concolori ; ala subtus alba, margine alari nigricante, plaga nigricante etiam prop6 ortum alae et ad apicem remigialem posita. Long. tot. 24, culm. 2-35, alse 7"15, caudse 4*8. Juv. similis adulto, sed minor et fascia superciliari sulpburea nee aiu'antiaca distinguendus. I never commenced the study of a bird under greater disadvantages than in the present instance. Dr. Coues and Drs. Cabanis and Reichenow both record a species of Penguin from Kerguelen Island under the names of JE. diadematus and B. chry- socome respectively, and I have no specimen before me wherewith to test these identifications. This is the more unlucky as I do not agree entirely with the identifications of recent authors, and a good series of the Kerguelen Penguins would have been of great service. The latest writers on these birds have been Professor Schlegel, Prof. Hyatt, and Dr. Coues, and they all agree in recognising four species, where at the most I can only find three distinct ones, as follows : — a. superciliary streak golden yellow, commencing above the eyes, not from the base of the bill ; forehead golden yellow at the base of the feathers, black at the tips chrysoloi)hus^ h. superciliary streak sulphur yellow commencing at the base of the bill. 158. /AKlLOCiY OF KEUtiUKLEN ISLAND. o. fovohcad erostctl, no yellow hasos to (ho loalhors, the siili)lmr-c'olourod eyebrow produced backwards, ioniser than the black plumes oi' the head . chri/socome.* h. forehead crested, no yellow bases to llic feathers; the sulphur-coloiu-ed eyebrow vcrj^ long and drooping and coterminous with an inner black crest sdUalor. Dr. Cones in his usual painstaking- manner has worked out these Yellow-crested Peni^uins from the material at his disposal in America, and after examining them care- fully and describing the differences in plumage in detail he observes : — " Although I " am able to distinguish the three currently accredited species, in the few specimens " examined, yet the distinctions are not of a very satisfactory nature, and I strongly " suspect that Avhen specimens enough shall have been compared, the supposed " specifie characters will melt insensibly into each other, so that at most only varietal " distinction can be reasonably asserted. Indeed I am not sure that differences of age " or season or special conditions of plumage may not be the sole basis of the supposed " species." These remarks apply to E. catarractes, E. chrysocome, and E. chrysolopha of liis paper. As regards the first of these, " at once distinguished by the shortness " of its tail .... also known by its inferior size, &c.," I am convinced that these are only signs of immaturity, and Von Pelzeln's plate in the ' Novara ' Voyage opens our eyes to the way in which these birds progress from the nestling to the adult. Size alone appears to me to be of no value as a character, and it is cm'ious to see how some of these Penguins {E. chrysolopJms, i.e. E. diadematus, for instance) differ in their bulk, though apparently in full, richly-crested, adult plumage. With * The synonymy of the true E. chrysocome appears to be as follows : — Eudyptes chrysocome. (Plate VIII., fig. 3.) Aptenodytcs chrysocome, Forster, .V. Comm. Gotting. in., 2>. 13o, pi. 1 ; id. Dcsci: Anim. p. 99. Pinguinaria cirrhata, Shaw in Miller Cimel. Phys. pi. xlix. Pingiiinaria cristata, Sliaw, Nat. Misc. pi. 437. Eudyptes chrysocome, Gould, B. Auslr. xn., pi. 83. Eudyptes pachyrhynchus. Gray Voy. Erebus and Terror, p. 17; id. Gen. B. iii., p. 641 ; Finsch, J. f. O. \%-i2,p. 261, 1874, p. 217; Gray, Handl. B. \n.,iy. 98. Chrysocomu catarractes, Bp. C. R. xl'ii., p. 775; Gould, TTandb. B. Austr. n., p. 517. Chrysocoma pachyrhynchus, Bp. C. B. xlii.,p. 775. Spheniscus chrysocome, Sclil. Mus. P. B. Urinatores, p. 6. Eudyptes catarractes, GiglioU, Faun. Vertebr. Oceano, p. 28. Eudyptes chrysolopha. Gray, Handl. B. iii. p. 98. Eudyptes chrjsocomus, Bullcr, B. X. Zeal., p. Mb, pi. 'ii,Ji I)ird was not found aiiywluMc near the soutlieni end of Ker- guelen Island. The crests obtained were broui^ht from Herd's Island. Sir Wyville Thomson in one of his letters to " Good AVords " (op. cit. 187 t, Nov., p. 311) slates that the " Macaroni" occurs at Christmas llarbouv in small numbers in company with the Eock-hopper; and that on the outer cliirs beyond the mouth of the harbour there arc some strong Penguin rookeries consisting almost exclusively of the " Macaroni."— .i. E. E-l Eudyptes saltator. (Plate VIII., fig. 1.) Le Manchot hupe de Siberie, Buff. PL Enl. 98-E Chrysocoma saltator, SfcpJi. Gen. '/jooL xiii. p. '')^,j)f. So Eudyptes chrysolophus (iicc Bmudt), Gray, Gen. B. iii., y>. 041; Abholl, Ibis, 1860, p. 338; Scl. P. Z. S. 1800, p. 390; Sc/d. Mas. P. B. Urinatores, p. 7; Gray, Sandl.B. m., p. 98; Scl. ^- Salv. Komeucl. Av. Neotr., p. 151; Coties, Bull. Z^. S. Xaf. JIxs. ii., p. 45; Kidder, op. cit. iii., p. 19; Cab. ^- Beichen. J. f. O. 1870, p. 330. Catarractes ehrysolopha, Beicheiih, Sandb. Pygopodes, pi. la. figs. 12-14 {pt.)- Chrysocoma ehrysolopha, Bp. C. B. xlii.,7;. 775. Eudyptes chrysocome, Pelz. Beis. Novara, Vog. p. 140, p)l- 5- Ad. supra sordid^ cinereus, pilei phimis rigidis, elongatis, cristam frontalem cxhibentibus, vcrticis lateralis plumis quoque elongatis, cum fascia lata superciliari cristam duplicem formantibus ; facie laterali cum colli lateribus gnlaque tota brunuesccnti-eiuercis : corporc reliquo pure albo ; pectore laterali, hypochondriis, imis, ct tibiis postice ciuereis; ala supra satiu'ate cinerea, margine alari summa vix albida, secundariis etiam albo terminatis ; cauda rigida dorso concolori ; ala subtus alba, ad basinet juxta marginem alarum summarum cinerea; remigibus primariis versus apicem cinereo-nigricantibus ; rostro aurantiaco ; pcdibus albicantibus ; iride cocciuea Long. tot. 23, culmen 2-0, alae 7-0, cauda^ 3-5. Juv. similis adulto, sed crista absente et gutture allndo brunueo mixto distiuguendus. I have fully discussed my reasons for changing the name of chrysolophus for this species under the heading of the previous bird, and it seems curious that Stephens' name of saltator has not been applied to it before, as the plate, though apparently derived from Buffon's illustration in the " Planches Colorizes," leaves no doubt as to the species represented. BIRDS. — R, B. SHARPE. 161 The following specimens constitute the series in the British Museum : — a. ad. [Cape of Good Hope]. Sir A. Smith. h. c? ad. Tristan d'Acunha. J. Maegillivray, Esq. c. S juv. d. min. Bounty Island. Purchased. e.f. min. Falkland Islands. The Admiralty. g. ad. Kerguelen Island. Capt. Fairfax, U.N. [On some parts of the coast Avhere the interstices of fallen rocks piled up at the base of cliffs afford them suitable shelter, the Eudyptes abound. Their colonies in Royal Sound were smaller than those iu Swain's Bay. The most populous of their communities visited by us were situated on the shores of the promontory to the eastward of Vulcan Core. There were there some thousands of the birds, aud they were very noisy. Their cry is a kind of guttural cackle somewhat like the syllables " Gurougha, gurougha, gurougha," pronounced rapidly. The designation " Bock- hoppers " applied to them by the whalers is extremely appropriate ; for although they occasionally walk a few paces at a time over a plane surface of rock, with the confined gait of competitors walking in a sack race, their ordinary mode of progres- sion is a series of bounds executed with much apparent ease and with an elasticity of motion such as is exhibited by Kangaroos. Standing amongst them silent (but most certainly not in silence) it was interesting to watch their proceedings. Those birds which had eggs far advanced in incubation remained in their nests, scarcely noticing the hand which stroked their backs unless they saw it move. Others stooping low peeped out of tbe neighbouring crevices beneath the boulders, or jumping a few yards away stood upon the rocks to gaze at their visitor, leaving their fresh eggs or empty nests unguarded. Bmls more confident than these then began to come near, Sheathbills for the eggs, carefully avoiding the Penguins who croaked at them as they sauntered past them ; and other Penguins who were ready to risk an approach for the sake of choice materials for nests so unexpectedly left at their disposal. Two or three hard pecks with its bill, and the Chionis is happy over its new laid Penguin's egg ; two or three journeys to and fro, and the thieving Bock-hopper has carried off to his own nest the choicest portions of his timid neighbour's. The despoiled birds soon see what is going on at their homes, and come bounding back to the rescue. Satisfied with the dispersion of their depre- dators, this most hopeful of mothers seats herself upon her broken eggs demurely, while her neighbour with an air of resignation betakes himself once more to the task of collecting dried Prinf/lea seed-stalks and other rubbish for the repair of his riiined nest. Sometimes tbe eggs are laid upon the bare rock without anything else to rest upon. There is the same sexual difference in size between the two eggs in a nest of this Eiuhjptes as there is in those of the Pygosceles. The crest on eacli side of the head is separated into two divergent drooping plumes one above the X 162 ZOOLOGY OF KEUGUELEN ISLAND. other, oxccptins: in swimminp;, wIkmi \hc lunl visin<» to tlio surface has the crest closely flattened down npon the sides of the head, where it forms a yellow streak, 'llic eyes are reddish oranu:e, with circular hlack pupils. lu the S. African IMuseum this species is represented from the Crozettcs hy two specimens (and juM-hajis a third) under the names Endyptes nigrivestis and E. chri/socoma* — A. E. E.'\ * In tho same collection is nnotlier specios fi-oin the Crozettes represented by ono epeciiiu'ii iiiiirki'd Eiidi/plrs chri/solophiis, wliich wiis iidt found by us in Kcrguclcn. A second spociuion under the same name E." c/irt/soloj>/iiis, vur." iVom tiio same islands appeared to be an albino of the same siiecies. Explanation of the Plates. (Plate VI.) Querquedula eatoni, Sharpe. Figure of an adult specimen procured hy the Antarctic Expedition. (Plate VII.) Eig. 1. Stei'covarius antarctieus. Head of a specimen, collected by Mr. Eaton in Kerguelcu Island. Fig. 2. Stercorarius catarractes. Head of a British specimen. Fig. 3. 'Prion vUlatns. Bill of an adult female from South Australia. Presented by Sir George Grey. Fig. 4. P. vittatus. Bill of a young male from the same collection as the foregoing. Fig. 5. P. vittatus. Bill of adult male from the same collection. Fig. 6. P. vittatus. Bill of a very old male from the Chatham Islands. Fig. 7. Bill of the type of P. hanksii, from the Cape of Good Hope ; probably of the same age and sex as fig. 4. Fig. 8. Prion desolatus. Bill of the type specimen (probably a young male) of P. hrecirostris, Gould, from Madeira. Fig. 9. P. desolatus. Bill of a male specimen from Christmas Harbour, Kerguelen Island, with a very distinct yellow " nail " to the upper mandible ; pro- bably an old male bird. Fig. 10. P. desolatus. Bill of a female bird brought by Mr. Eaton from Royal Sound, Kerguelen Island. (Plate VIII.) Fig. 1. Endyptes saltator. Head of an adult bird from Tristan d'Acunha. Fig 2. Endyptes chrysolophus. Head of an adult bird, procured by the Antarctic Expedition. Fig. 3. Endyptes chrysocome. Head of an adult bird from New Zealand. (Type of E. pachyrhynchus. Gray.) EGGS. — H. SAUNDERS. 163 Eggs. — By Koxoard Saunders, F.L.S., F.Z.S, Chionis minor, Uartl. (Lesser Sheathbill.) The general character of 19 eggs is a dirty white ground, splashed and blotched with brown. At the first glance there is a startling superficial resemblance in coloration, and sometimes in shape, to a very common dark form of the egg of the BazorbUl {Alca torda) ; other specimens are in shape and markings like boldly blotched examples of eggs of the CEdicnemus group of Plovers. On shining them to the light, the eggs show a green membranous lining. Unfortunately the egg of the other species Chionis alba (Gm.), of the Falkland Islands, is not known, for Capt. Abbott, who wrote an account of the birds of the Falklands (Ibis 18), did not obtain it, and the statements made to him by the sealers as to its egg being white, must be received with doubt. The first eggs were obtained on 23rd Dec, and in stating that none were found by jVIi". Eaton until 10th January, the American Naturalist, Dr. Kidder, must have been labouring under an error. The complement of eggs seems to be one or two, and rarely three. The average dimensions of the egg are 2"2 in. x 1"5 in. "s"- Querquedula eatoni, Sharpe. Thu'ty eggs of this species present remarkable variation, the general hue being of a pale green or greenish buff. Laying appears to commence early in December. The average dimensions are 2 in. x 1" i in. Laras dominicanus, Licht. Thirty eggs of this bird present the usual characters of the eggs of the larger species of Gulls, being of various shades of olive-green, ranging to stone-colom*, and occasionally brown, spotted and blotched with darker shades of brown and streaks of black. Eggs of this species in my collection from the Crozettes, and from the Ealk- land Islands present precisely similar characters. The first e^^ was obtained on 14th October ; the complement is three, as is usual with Gulls. Average dimensions, 2-85 in. X 1-9 in. Stercorarius antarcticus. Lesson. In five eggs of this species, three taken about the middle of November are of a pale olive-green indistinctly blotched with brown, and two others, obtained on 8th December, apparently from the same nest, as that is the usual complement, are X 2 164. ZOOl-OGY 01' KKUGUKLKN ISIAND. of a brownish bu(T i^round witli rathor bol(l(>r in;iv1vinii,s. Although sliglitly larger than tho i>ggs of Sicrcorariiis catarracfcs, the ri'presontativp s])('('ios ol" \.\\v Northern lltMnisphere, there is no noticeable diiVereuco in character. Length 3 in., breadth 21 in. Sterna virgata, Cah. Eight eggs of this s|HM'ies are of an olive colour blotched with black, the marks tending to form a zone. They do not dilVer much from eggs of typical Sterun, but there is a tendency to greenish, and an absenc(> of the white in the ground colour, which approximates theni to eggs of S. antarctica of New Zealand, the near ally of this species, the eggs of which, so far as I may venture to judge from only two specimens, average somewhat less. It appears to lay but one q^^^ on the terraces on the sea-side hills, commencing towards the end of November. The dimensions are 1"75 in.Xl'2 in. Pelecanoides urinatrix, Gm. Ten eggs are all pure white, except where peat-stained, nearly equal at each end, or but very slightly pointed. Dimensions, 1'5 in. Xl'l in. Majaqueus aequinoctialis, Linn. Tfl^elve eggs of a pure white colour (except where the granulations of the shell are filled up with the yellowish dirt from the burrows where they are deposited) have a more repulsively musky smell than any other eggs of the group. Length 3-2 in., breadth 2-1 in. .fflstrelata brevirostris, Less. Two eggs of this species, badly cracked owing to their fragile texture, are of a pm'e dull white, nearly ovoidal in shape, and measure 2*15 in. X 1"7 in. iEstrelata lessoni, Garnsi. Eleven eggs of this species, of the characteristic dead white colour, have some- what less of the musky smell than those of most of the burrowing Petrels. Length 2'75 in., breadth 1-85 in. Slightly pointed at smaller end. Procellaria oceanica, Kuhl. Nine eggs average 1-3 in.X'9 in., and are of a duU white colour, with minute purple-red spots which generally form a zone, usually, although not invariably, at the larger end ; they are however at times distributed sparingly over the whole surface. There is not much difference between the shape of either end. EGGS.— H. SAUNDERS. 165 Prion desolatus. Six eggs of the same character and texture as the preceding species, and dis- tinguished by their greater length in proportion to breadth, the average of the former being 1-8 in ; nearly as much as in Halohtena ccertilea, whilst the breadth is only 1-3 in. Halobaena coenilea, Gm. There are ten specimens in the collection, of the rough granulated texture, blunt ends torn, and dead-white coloiu-, characteristic of the eggs of this family. There is no sign of a zone of rust-coloured markings, but there is the usual musky smell about the shell. They vary in dimensions a good deal, and as from the delicate nature of the shell, several specimens are total wrecks, it is not easy to take an exact average, but I am inclined to think that one of the above eggs belongs to the fomier species. lO in. in length, and 1-5 in. in breadth, is tolerably close. The single e^§ is laid as early as 23rd October, onwards. Diomedea fuhginosa, Gm. There are two examples of this egg ; white, spotted towards the larger end, so as to form a zone, with minute reddish marks. The e^^ is long in proportion to its breadth, being 4i in. X 2-65 in. Phalacrocorax verrucosus, Cab. The six eggs are of the usual pale greenish blue, with the chalky incrustations characteristic of those of the Cormorant family. Average length 2'3 in., breadth 1-45 in. Aptenodytes longirostris, Scop. (Eggs noticed in MS., but none appear to be in collection.) Pygosoehs tseniata, Peale. A solitary egg of this species is of a pale blue thickly coated with calcareous matter. It measures 2'5 in. X 2 in. Eudyptes saltator. Two eggs ascribed to this species differ considerably in size, the larger measm-ing 2-7 in. X 2 in.; the other 2-4 in.xl"65in. The colour is very pale blue with a white calcareous coating irregularly disposed over the surface. In shape they are somewhat pointed at one end. 166 ZOOLOGY OK KKRGUEl.KN ISLAND. FisuES.— j5// Br. A. Giinlhcv, F.B.S. The species collected arc fom* in number, three of which wore previously known to occur on tlie shore of the Island, viz. : — Harpag'iferhispims, Chwuichthys rhinoce- rafus, and Noiothenia coriiceps. The fourth is a Ray, apparently undescriljcd, Mliieli may be characterised thus : — Raja eatonii. Allied to R. smithil. Snout of moderate length, the anterior margins meeting at a right angle ; the width of llic iiitcr-orhital space is two-sevenths of the distance of the eye from the end of the snout. The anterior profile, from the snout to the angle of the pectoral fin, is slightly emarginate, the outer jiectoral angle being rounded. The greater part of the upper surface of the body is smootli, minute spines arc distributed between the eyes, and in a narrow stripe along the margins of the body ; a broad band of minute spines along the median line of the back and the upper surface of the tail ; a single larger recurved spine in the middle of the back ; a series of nine or ten rather small spines placed at a considerable distance Ironi each other along the median line of the tail; no spines on the side of the tail. LoAver parts smooth. Upper lip fringed on the side ; teeth pointed, conical, in about 30 series in the upper jaw. M(de with a patch of claw-like spines on each pectoral tin. Brownish black above, with indistinct round whitish spots ; whitish below, with some irregular brownish-black spots ; lower part of the tail brownish-black. A single adult male was obtained in Royal Sound. It is 264 inches long, the tail measuring 14 inches ; its greatest width is 18 inches. A specimen of the female sex was obtained during the visit of the " Challenger " to the island. It differs very little from the male, but, of course, the claw-like spines on the pectoral are absent. The lower part of the body is entirely white. A second species of Ray {Raja murrayi, Gthr.) was found in Kerguelen's Land also by the naturalists of the " Challenger." This will be described in connexion with the collection made during that expedition. MOLLUSCA.— E. A. SMITH. 1G7 MoLLTJSCA. — By Edgar A. Smith, F.Z.S., Smior Assistant in the Zoological Department, British Museum. (Plate IX.) Of the thirty-three species treated of in the following paper, and representing twenty-five genera, twenty-five were obtained by the Rev. A. E. Eaton, the remaining seven having been collected during the Antarctic Expedition under the command of Sir James Ross. Of this number, 18 are new to science, and nearly all are men- tioned for the first time as inhabiting this locality. Some of the species are of very great interest, especially the new genus Neo- buccinum, the Struthiolaria, and the new genus Eatoniella among the Gastropods, and of the bivalves the Saxicava, Lissarca, and the magnificent Solenella, by far the largest known species of that genus. The Malacological fauna resembles generally that of the Falkland Islands and South Patagonia. More than half of the genera and seven or eight of the species found at Kerguelen Island are known to occur at those localities, and further research will probably discover a still greater number of genera and species to be common to these two, longitudinally, so widely separated localities. With respect to their latitudes the difference is unimportant, since they both range between 49° and 54° S. lat. As the Cape of Good Hope, Tasmania, and South West Australia, are the nearest points of mainland, it might be expected that some resemblance to the fauna of those countries might be observable. However, it is not so, as far as our present knowledge extends. Many of the shells from Kerguelen Island have the generally unattractive appearance as regards coloration which so frequently obtains in species found in cold climates. Indeed, some of them seem to be southern re- presentatives of boreal types. The Neobiiccinum, Trophon, Saxicava, Kellia, Yoldia, Badula, and Doris, are remarkable instances of similarity to northern forms. The following table shows the great afiiuity Avhicli exists in the fauna of South Patagonia and Kerguelen Island. Only those species are quoted from Patagonia which arc identical with or nearly allied to Kerguelen forms, and a blank signifies that the genus has not yet been recorded from there. 1G8 ZOOLOGY OF KEUGUEl.KN ISLAM). CJt'iiorn. PH(ii);oninn Species. Kcixuelcii Species. Ncobuocinuni N. catoiii, Sm. Troplion T. philippiaiiiis, Diinktr T. nlliolat)nitii.«, Sm. Strutliioliii'ia S. liiiraliilis, Sm. Piirpiiru P. siriata, Maili/ii, (American E.\p.) Ailmi'to ( — ? ) A. ( — ?) liimiiiwroruiiH, Sm. Littoiina L. setosa, Sm. llytlrobia 11. pumilio, Sm. ?* " " II. i'uli';iin)sa, Gould. Entouiolln E. kcigmlciiL'iisis, .S/M. i» ~ ' E. (.'alif^iiiosa, Sin. »» " " E. sid)riil'csc"cns, Sm. Skenea S. siibcaiialicMilata, liMii, .S'h(. Si'issurelln - S. siipraplicata, Sm. Trochus (Photinula) T. (P.) expansus, Soivb. 'V. (P.) cxpaiisus, Soivl). ra!(>lla P. a;nea, Marti/ii - P. kcrfjiK'lcnonsis, Sm. '» " P. fiicjiiensis, Iirrvc P. I'uejjiensis, /I'/v. i» * P. inytiliiiii, Giiifliii P. iiiytiliiia, Gmcl. Siplionnria - S. iiiii-ji'llaiiicii, I'liilijipi - S. rcdiinicidnin, lire. Hi-miurtlinim 11. sotulosuin, f /jr. (Amer. Exp.) Doris - D. tiibcrcul.'ita, Ciiv. Il-lix - II. hookc'ii, Jlccve. Saxicava 8. antnrctica, Philippi S. liisiiicata, pears to be convex. This species ditVei's from normal Hjidrohia', whose spii-c is generally rather elevatcHl, in having the apex but very slightly cxsertcd above the last whorl, and consequently the contoui' of the spire is unusually convex. Eatouiella, Dall. Ealoiiia, Smith, Ann. and Mag. Nat, Hist. 1875, xvi. p. 70 (name preoccupied). Eatoniella, Dall, liulletin 11. S. Nat. Mus. 1876, iii. p. 42. Testa forma) rissoideaj; apertm-a subcircularis ; peristoma simph'x, continiimn, mavginc labrali haiul incrassatiun. Opcvcnlmn ovatum, pauci- vel unispirale, nucleo subterminali a latere columcUari paiihdum remoto, svibtus ossiculo promincnti a nucleo cxsm-gentc et marginem colimicllarcm versus du-ecto munita. There are two genera which have affinity to the present one — Jeffreysia and Bissoina. With the former it agrees in the form and character of the aperture, but ditfers in having the nucleus of the operculum not lateral, but situated within the margin and towards the lower end. It agrees in this respect with Missoina (see Adams, "Genera of Eecent Mollusca," vol. iii. pi. 35. f. 1, «&/>»), but is dis- tmguished from that genus by the absence of the slight basal channel of the aperture and the lack of any incrassation to the labrum. The operculum of Jeffreysia is composed of concentric layers (as in FMrpnra), commencing from a nucleus situated on the margin of the inner or columellar side ; and the ossicle or rib proceeds " from the nucleus in the direction of the outer margin " (Jeffreys, " Brit, Conch." iv. p. 58 ; in the figure, I. c. pi. 1. f. 3, it is apparently the reverse). In Eatoniella the operculum is spiral, consisting of one or mow; whorls, the nucleus is situated within the margin and about one fourth the entire length from the lower end, and the ossicle is directed towards the inner margin. The name originally given by me to this genus having been prcoccupi(!d, has been modified to that in use, by jVIi\ Dall. Eatoniella kerguelenensis. (Plate IX., fig. 10.) Eatonici kerguelenensis. Smith, Annals and Mag. Nat. Hist. 1875, xvi. July, p. 70 ; Eatoniella Jcerguelenensis, Dall, Bulletin TJ. S. Nat. Mus. No. ui. p. 42. Testa ovato-conica, tenuis, olivaceo-nigrescens, labnmi versus pallidior semi- pellucida, vix rimata ; anfractiis 6, convexi, Iseves, parum nitidi, liiieis incrementi striati, sutura simpHci sejuncti ; apertura fere circularis, longitudinis totius -^ sequans ; peristoma simplex, continuum, ad regionem umbilicalem leviter incrassatum et vix reflexiun. Long. 3 mUl., diam. If. MOLLUSCA.— E. A. SMITH. 175 Operculum ovatiuu, intus concavuni, nucleo posteriore non tamen terminali, crassiusculum, super mai'gincm exteruum lira inciassatum, uuispii'ale, supra iucre- menti liueis valde striatum, infra ossiculo elongate a nucleo exsiu'gente munitum. Sab. — On a sponge {Teihya antarctica). Royal Sound, in 40 fathoms. This species was found in company with Rissoa Kergueleni. It is of a very ditferent form, the spu'c heiug conical, the last Avhorl shorter and a trifle hroader ; and it also difl'ers in colour. In general aspect it A'cry much resembles several species of Sydrohia ; hut the operculum at once distinguishes it. EatonieUa caUginosa. (Plate IX., lig. 9.) Eatonia caliginosa, Smitb, Annals and Mag. N. H. 1875, xvi. July, p. 71. Eotoniella caliginosa, Dall, Bulletin U. S. Nat. Mus., No. iii. 43. Testa ovata, modice tenuis, nigra, vix rimata ; anfractus 4^, couvexi, Iseves, vix nitidi, sutura simpHci discreti, lineis incrementi obsolete striati : apertm'a fere circularis, superne paixlulum acuminata, longitudiuis totius ^ fere sequans ; peri- stoma continuum, levissime incrassatum, ad regiouem iimbilicalem albidum, ali- quanto reflexum, et basin versus parum effusum. Long. 2 mill., diam. 1. Operculum ei E. kerguelenensis fere simile. SLah. — Swain's Bay. Found with the preceding species. Tills minute shell, with a simple style of sculptiu-e, is of a very black olive-colour, with a nearly circular aperture, the peritreme of which is black outwardly, and whitish in the columellar region. Eatoniella subrufescens. (Plate IX., fig. 11.) Eatonia subrufescens, Smith, Annals and Mag. N. H. 1875, xvi. July, p. 71. Testa ovata, leviter couica, tenuis, semidiaphaua, vix rimata, sulirufescens, labrum versus albida ; aufractus 4^, lente accrescentes, convexi, sutura subprofunda divisi, laeves nisi striis incrementi tenuiter sculpti ; apcrtura subcircularis, longitudiuis testae \ paulo superans ; peristoma continuum, ad marginem columellarem leviter incrassatym et reflexum, rimam umbilicalem indistinctam efllngens. Operculum ei E. kerguelenensis subsimile, ossiculo tamen fortissimo munitum. Long. 1^ mill., diam. |. Hab. — Royal Sound, on a sponge {Tethya) at a depth of 7 fathoms. The reddish colour of the upper whorls is attributable to the dried remains of the inhabitant. Skeuea subcauahculata. (Plate IX., fig. 15.) Skenea subcanalicnlata, Smith, Annals and Mag. Nat. Hist. 1875, xvi., p. 71. Testa minuta, orbiculata, deprcssa, tenuis, subcUaphana, albida, late profundeque 17(i ZOOLOGY oi- Ki;iJ(;rKi,i:N isi.and. lunbilii-ata ; spira mininu' (uoc P. inugellanica, Gmcl.). Testa oralis, antiee iwululuin anj^ustata, convcxc satis elovata, apice promineuli beneqvio ante verso pnesertiiu in excmplis juninrihus, hili' radiatini eostata, costis pariim prominentibus, et siope aliis iniii()ril)us inlerji'ctis, lincis incrementi coneeu- tricis undiilatis eleganter crebreque inseulpta ; extus cocrulco-cinerea, costis soepis- simc saturatioribus, et apiccm versus in excmplis detritis ferruginca; intus jucunde icneo-fuscescens, plerumque margincm versus pallidior (liac parum undulata) ; cicatrix muscularis pcrspicua. Long, (exempl. max.) 82 mill., lat. 70, alt. 45. The above description is based upon the examination of about twenty specimens. Animal (in sjiirit) with the sole of the foot grccnish-ash colour, inky-black on the sides, encircled around the middle by a bluntly serrated frill which is interrupted in front of the head ; brauchiic pale buff-colom* ; margin of the mantle blackish, furnished with alternately small and smaller cirri along the edge, the smaller black ; bead and tentacles black ; lips pale buff. Lingual ribbon very long and narrow ; teeth not hooked, in diverging pairs ; median pair two-pronged, the inner prong much the larger, and resembling in form a flat spear-head ; lateral pairs alternate with the median pairs, four-pronged, the second prong from the centre and the outermost considerably larger than the other two, which are of nearly equal size, the innermost however rather the larger. Hab. — Swain's Bay usually on rocks about 1 fathom below the surface ; but one specimen was also obtained at the extreme verge of low water. Dead shells were occasionally found on some of tbe islands scattered over the recreation grounds of Cormorants and Gulls (Eaton) ; also lloyal Sound (Kidder). The tigiu'c in the Con. Icon, of this species (as J? . ferriujinea) is very good as regards form, but does not show the prominent apex, and only represents a small specimen. This name adopted by Uecve being preoccupied by Gmelin cannot be retained, and consequently I have substituted for it kerguelenensis, as in all proba- bility the species is exclusively indigenous to Kerguelen Island. The P. ferruginea, Gmelin, founded on a figure in Martini's Conchylien Cabinet, I., pi. viii., fig. G6, is well refigurcd by Pieeve-(Con. Icon, viii., f. lia-b) under the name of P. costoso- plicata, which is the fia'st adjective in Martini's brief description. Gmelin's name must therefore be retained for the same species. The T . ferruginea, var. (no author mentioned) figured by Sowerby in his Genera of Recent and Possil Shells, fig. 4, is one of the endless varieties of P. cenea, Martyn. Prom this species P. kerguelenensis differs in having the shell less prominently costated, differently coloured, and in the apex being very prominent and much cm-ved over so as to give it a capuliform apjiearance, a character constant in all specimens, yoimg and old, elevated or depressed ; it also diff'ers in the coloration of the animal. Now in some depressed varieties of P. cenea the apex is somewhat iMOLLUSCA. — E. A. SMITH. 179 curved over, but it is not prominent and lias not the appearance of an umbo, as ex- hibited by the apex of P. hergtielenensis. In the determining of tlie present species, having had occasion to examine the Patagonia shell, I Avill give the results of my investigation. The synonymy is as follows : — Patella (Patinella) senea. 'Patella upper surtaco near the tips. A frill similar to that of /ccrt/Ke/eiu'iisis cncireles the foot. Patella (Patiuella) fuegieusis. (Plate TX., ligs. 11, 11«.) Patella fKegioisis, Eeeve, Conehol. Iconica, viii., sp. 7S. The description given by E^evc is very good, but he do(^s noi lay sulTlcicut stress upon tlic beautiful raised concentric ridges. He calls them stria;, which term scarcely gives the idea of thread-like lirations such as those. They are very closely packed and undulate A-ery prettily on and between the numerous radiating ribs. The tigure, except in outliiu^ and the position of the apex, gives but a poor idea of this beautifully sculptured Patella. It represents the number of ribs at about forty, whereas there are usually about sixty. The specimens from Kerguelen's Island are a tritle narrower and much more depressed than examples from the Falkland Islands ; in fact, it is only near tho apex that they are at all raised, and towards the margin they are up-turned, so that the dorsal surface is concave, and this form of the shell certainly prevents the animal froni entirely concealing itself when adhering to a flat surface. But this iieculiarity of form only exists in adult specimens, for several small ones are like ordinary species in this respect. The radiating ribs are almost obsolete in the flat examples, but the undulating concentric lirations, which are more prominent and farther apart than in the type form of the species, define their position ; in young shells they are more pronounced. Colour generally uniformly purplish slate, with the apical region ferrugineous ; interior similarly tinted, but rather more deeply. One shell has a white border. They are all very thin and fragile, and the edge is very liable to break off in a line with the concentric raised lines of growth. The animal has the sides and sole of the foot greenish -grey, the edge of the mantle and gills pale buff, the tentacular filaments on the margin of the mantle blackish except at their tips, tentacles short and thick, pale buff, with a black spot above. The friU-like expansion of the mantle, similar to that of P. (vnea and P. kergue- lene7isis, is a little above its edge, is bluntly serrated, and interrupted beneath the head. Teeth of the lingual ribl^on slightly hooked, in pairs, scarcely diverging ; the central pairs two-pronged, the inner prong mYich the larger, spear-head shaped ; the lateral pairs alternating with the central ones are four-pronged, the innermost prong smallest, the next two subequal, and the outside one situated nearly at right angles to the rest of the tooth, about the same size or a trifle larger. Sab. — Royal Sound and Swain's Bay, everywhere very common on the sub- merged fronds of long floating kelp {Macrocystis) bordering the shore. MOLLUSCA.— E. A. SMITH. 181 Patella (Nacella) mytilina. Patella mytilina, Gmelin, Syst. Nat. p. 3G98. Patella coucliacea, Gmelin. 1. c. p. 3708. Nacella mytiloides, Schumacher, Syst. Vers Testae, p. 179. Patella cymhularia, Lamk. Anim. S. Vert. ed. 1, vol. vi., p. 335 ; ed. 2, vol. vii., p. 541 ; Philippi, Al)hild. & Beschr. iii., p. 1, f. 2. Patella cymbium, Phil. Archiv. Naturgesch, 1845, p. GO ; Abbild. p. 7. Nacella mytilina, Gmel., Dall, Annals & Mag. Nat. Hist. 1871, vii., p. 289. Nacella cymbularia, Lamk., Adams Genera Eec. Mol. i., p. 467. Var. = Patella vitrea, Phil. 1. c. p. 9, pi. i., f. 4 a, h. Var. =: Patella hyalina, Phil. 1. c. p. 8., pi. i., f. 3 a, h. (Not. Patella cyinbularia, Dclessert, Recueil. Coq. Lamk. pi. 23, f. 8 a-c, which is one of the many varieties of Patella cenea, Martyn.) On carefully examining a large series of specimens (about 40) of this species, I can arrive at no other conclusion than that the two forms described by Philippi, above quoted, are but variations of Gmelin's shell. The principal ditrerence is in the position of the apex. In a long series every position is found ; in some shells it is quite marginal or nearly so (P. hyalina) ; in others it is slightly more remote from the margin, and again still more so, until it takes that position which it occupies in the typical mytilina. The character of the sculpture varies to no material extent. All examples are more or less radiately costatccl, sometimes conspicuously, and at other times almost obsoletely, the crenula- tion or undulation of the margin varying in coarseness with the strength or feebleness of this costation. The concentric lines of growth are of the same Eceneral nature in every specimen, and all possess minute scratch-like striation (visible only under a lens) radiating irregularly from the apex. The form of the marginal outline is considerably altered in those specimens whose apex is quite marginal, being con- siderably narrowed at the apical end and somewhat acuminated. The coloration varies in shells of similar form, some being of a general greyish tint, varied at intervals with darker concentric rings and often a few radiating palish stripes on the ribs. Others are uniformly yellowish-brown, others pale lutcous broadly striped with black, and finally, others are of a uniform pale horny colour ; but all have the apex cupreous. The six Kerguelen specimens obtained by Mr. Eaton are uniform in shape, oval, but rather acuminate towards the apex, which is only slightly removed from the margin, rather depressed, more so than Magellan specimens. They are of a brownish- red colour for the most part, and gradually blend into olive towards the margin, coppery within. Thus it will be seen that they only differ from P. mytilina proper, in form somewhat and coloration. Animal similar to that of P. kergnelenensis. 182 ZOOLOGY 01" KKUGUKLKN ISLAND. JJah. — On YOiiny; and short I\[acroci/sli^ at a di'ptli of two lallionis in Swain's I3av, at the enil ol' the same promontory as Littorina sclosa. From the descriptions of the four jireceding species of PatcUhl(V it Avill lu: ohservcd that the animals olVcr \w partieular dilVerences exteriorly except in coloration. In all " (t scalloped frill, iii/erriqjled only in front" encircles the sides of the foot, and this peculiarity induced Dall to ini])ose upon f1ios(^ spocicvs possessing it the subgeneric name of Fittiitclla. As P. mi/tilina possesses a shell dillering somewhat in loiin and texture, perhaps it may he well to retain the subgeneric title Nitcella proposed by Schumacher for this species. With tills group Messrs. Adams place P. pellucida, Linn., but since this animal has the branchial cordon interrupted in front of tlie head, it becomes necessary to locate it in another section, which Leach has styled Fatina. Siphouaria redimiailum. Siphonarid redimiculum, Reeve, Conchol. Iconica, ix. sp. 21., 1856. Siphonaria magellanica, FhUippi .'' ]VIalall)U)ng trssclkitions, wliosc intrrsticcs arc grey, whicli is " separated hy a thin pak' inegukir line from tli(> more liiiely retieukited upper " jiortion of the sides and hai-k ; the iuterspaecs of the retieukxtion ol' tkesc kist are " slightly raised and hiaek, and cause the surface to he somewhat granulated. " Some of the lines of growth in the shell arc occasionally straw colour." Hah. — Common in the neighbom-hood of Royal Sound, especially on the teiTaccs of basalt. CONCHIFERA. Saxicava bisulcata, sjj. iiov. ( Plate IX., fig. 21.) Testa transverse subrhomboidalis, mediocriter crassa, A\alde ina'quilateralis, tnmida, alba, epidermide tenuissima lutescente partim iuduta, sulcis duobus minimc profundis ab umbonibus usque ad medium marginis ventralis, ibique productis dentes duos etfingentibus arata, lineis incrementi concentricis sul)tenuibus inscul[)ta; umboncs parvi, contigui, fere terminales, incurvati ; margo ventralis minime arcu- atus, medium versus bidentatus ; latus anticum extra umbones vix j^rominens, posticum a marginc dorsali decliviter arcuatum ; ligamentum distinctum, flavescens. Diam. transversa 6 mill. Alt. 3^. Crass. 3J. Hab. — Kerguelen's Island. This very curious species of Saxicaca is the only one with which I am acquainted having dentate ventral margins to the valves. These dentitions are the prolongations of two shallow sulci wliicli radiate from near the umbones, not quite down the centre of the valves but a trifle posteriorly, to their edge. When the shell is closed the ventral margin has an undulating aspect, which is produced by the interlocking ol these little tooth-like projections. Besides these two snlcations in some specimens there are very faint traces of others, or raised ridges radiating down the posterior dorsal slope. Many of these shells have several porcellaneous tubercles which adhere to the inner surface of the valves, and all of them have a single perforation at their base. Kellia consangidnea sp. iiov. (Plate IX., fig. 20.) Lasea rubra, Dall, (non Montagu), Bulletin U. S. JS'at. Mus. 1876, No. iii., p. 45. Testa transversa, valde ina?quilateralis, paululum oblique ovalis, postice leviter angustior, tumida, baud perfragilis, flavescens, prope cardines ct latera purpureo- roseo tincta, intus dilute rosea, ad umbones plerumque erosa, epidermide mediocriter crassa, flavescente induta, concentrice tenuiter striata, hie illic fortius concentrice plicata ; umbones aliquanto prominentes, contigui, ad circiter longitudinis J a latere antici siti ; Unea cardinalis arcuata, purpureo-rosea ; valva dextra dente conico infra MOLLUSCA. — E. A SMITH. 185 apiceni umbonis (qui in fossa parva in valva sinistra accommodat) munita, ct utrinqiie dente latei'ali promincnti (qui in sulco prof undo in valva sinistra inter dentem similarem marginemquc dorsalem accommodat) instructa ; valvarum mar- gines generalitcr uhique arcuati. Diam. transversa 3 mill. Diam. long-itud. 2 J. Crass. 2. Sab. Royal Sound ; abundant under stones between tide-marks at Observatory Bay. At a fii'st glance this species might easily be mistaken for the European Lascea rubra, to which it has a very great resemblance. It is, however, of a rather stronger structure, the epidermis is thicker, the form too transversely rather more elongate, the lunbones always, in the seven examples at hand, considerably eroded, and there is not the faintest trace under a powerful microscope of that minute (apparently punctate) radiating striation which is observable in L. rtibra. The dentition is a little different also. The Kellla miliaris described by Philippi in Wiegmann's Ai'chiv fib* Naturgeschichte, 1845, p. 51, from the Straits of Magellan, is another closely allied species but cUfferently coloiu-ed, it has no epidermis, and the umbones are rather more prominent. Lepton parasiticum. (Plate IX., fig. 22.) Ball., Bulletin U.S. Nat. Mus. 1876, iii., p. 45. Kab. — Parasitic on a Sea-urchin {Uemiaster cavernosns), living in the deep ambulacra, and also on the surface of the test. This species is remarkable for its parasitic nature and being viviparous. On opening an adult specimen I found it to be filled with about a dozen very small ones. Lissarca, siibgen. nov. Testa sequivalvis, subrhomboidalis, valde insequilateralis (umbonibus fere termi- nalibus), concentrice striata; linea cardinalis utrinque paucidentata, in medio Isevis; margines valvarum intus dentati. Tills sub-genus of Area is distinguished from Bavbatia of Gray, which is the nearest allied group, in having the shell concentrically (and not radiately) striated, with the umbones nearly terminal, and consequently it is very much more intequi- lateral than is usually the case in Barbatia. Area (Lissarca), rubro-fusca, sp. nov. (Plate IX., iig. 17.) Pectunctilus miliaris ? Bhil. Wiegman's Archiv. Naturgesch. 1845, p. 56. Testa valde imcquilateralis (umbonibus fere terminalibus), ventricosa, irregulariter subrhomboidalis, autice oblique aliquanto truncata, postice latior, arctiata, medio- A A 18G ZOOLOGY OF KKUGUELEN ISLAND. critcv crassa, iiliiqno vubro-rusfa ("pidcnuidc lutoo-olivacra ((MUiitorquo t- one on trice lauiinata amic'ta ; area ilorsalis angustissima, linearis; nniboncs magui, mcdiocriter pi'omincntes, fere contii:;ui ; liuca cardinalis in medio rectiuscula, kevis, utrinquo leviter arciiata, dentibus al!)idis, obliquis tribus vel quatuor (posticis quani anticis longe niajoribus) munita; niarujo ventralis parum arcuatus, latus versus anticum. levissime sinuatus ; valvarum margines (])ra>tcr prope si mini levissiinum et ad medium lateris postici) intus fortiter denticulati. Diani. transversa 1 mill. Alt. 2;-'. Crass. 2.\. ILah. — Kerguelcn Island (Anlarelir llxpcd. and Transit Expod.). It might be thouglit that this remarkable little species has been described from young slieUs on account of their smallness, however their comparative solidity, the strong teeth on the hinge and on the margin, give thorn the appearance of bei7ig adult. These marginal dentations are interrupted at the posterior extremity of the valves, and also towards the anterior end of the ventral margin, where it is faintly sinuated, and Avhere also the little byssus protrudes. The hinge line is moderately straight, and botAveen it and the dorsal lino, which is slightly arcuate, there is an extremely narrow linear area. The umbones are rather tumid and project slightly above the dorsal line. Prom the umbo to the posterior end a little above the middle of the valves there is the faintest depression. The posterior muscular scar is sub-pear-shaped, largish ajid well defined, the anterior is small and indistinct, and the pallial line is simple and continvious. The hinge- ligament is central, and so small as to be scarcely traceable. Its teeth are strongly developed, especially the three or four posterior ones, which are conspicuously stronger than the anterior and more oblique. I have quoted PhUippi's Fectunculus millarls with a note of interrogation, as I cannot reconcile certain peculiarities in the present species with his description. At all events they are congeneric, and therefore P. miUaris must be placed in Lissarca, which is distinguished from Fecliinculiis by its transverse trapezoidal form and the subterminal position of the umbones, which in the latter genus are almost central; the teeth are fewer and the valves not radiately striated or ribbed. L. rnhro-fusea apparently differs from L. miUaris somewhat in form, the umbones not being very acute, the margins of the valves are crenulated only in certain jilaces, and not in others, the number of teeth on the hinge-liue is smaller, and a dorsal area exists although it is extremely narrow. These are, it is true, but small distinctions, and had Philippi's description been more copious, possibly these shells might have been referred to his species without doubt. The habitat of his species is the Straits of Magellan, and this is favourable to the identity of the two species ; however, until an opportunity is offered for the comparison of authentic examples, it seems to me that it will be the safest course to apply a distinctive name to the Kerguelen Island form. MOLLUSCA.— E. A. SMITH. 187 Two siiocimcns collected by Mr. Eaton differ from tliose obtained by the Antarctic Expedition many j'-ears ago in being rather shorter, and wider posteriorly, in having the nmbones less terminal, and the ventral margin of the valves being without deuticulations ; the last characteristic may be due to immature age. Yoldia subajquilateraHs. (Plate IX., fig. 18.) Yoldia stihcequilateraUs, Smith Annals & Mag. N. H. 1875, xvi., p. 73. Testa ovalis, postice acuminata, suba;quilateralis, postice paululum brevior, con- vexinscnla, opidermide olivacca vcl flavo-olivacea iuduta, concontrice rugose striata, ntriuque umbouibns ad margiuem subvcnti-alem striis panels su.bgranosis radiantibus insculpta, utrinqne leviter hians, intus cteruleo-alba ; margo dorsalis utrinqne nmltum declivis, antice levissime convexo-arcuatus ; postice fere rectus ; margo ventralis ubique arcuatus ; latus anticum late rotundatum, posticum subacuminate productum ; fovea hgamentalis parva triangularis ; dentes cardinales utrinque 11 ; sinus pallii latissimus parum profundus. Lat. 34 mill., long. 23, crass. 9. The animal resembles that of Solenclla gigantea in all resjiccts excepting that the edge of the foot is bluntly serrated or scalloped. Hab. — Swain's Bay. Dredged in 7-10 fathoms in very sheltered water. I know but one species which approaches the present one somewhat closely, namely Y. eightsii of Couthouy. Erom this, however, it is well distinguished by its different form. By reference to Jay's figvire upon which Y. eightsii is founded (for no description is given; Cat. Shells, 1839, ed. 3, pi. i. f. 12 & 13), it will be perceived that a very inequilateral shell is there represented, with a much excavated posterior dorsal slope ; on the contrary, Y. stibcequilateralis is almost equilateral, with a straight posterior dorsal acclivity. Solenella gigantea. (Plate IX., fig. 19.) Solenella gigantea, Smith Annals and Mag. N. H. 1875, July, p. 72. Testa elongato-ovalis, postice subrhomboidalis, parum inasquilateralis, postice longior, aliquanto ventricosa, marginem versus posticum compressiuscula, epider- mide nitidissima (vel fusco- vel flavo-olivacea) induta, lincis incremcnti concentricis (interdum j)TOminentibus) ornata, et striis paucis tenuissimis et confertis ab um- bonibus usque ad medium lateris antici radiata, intus alba, iridescens ; margo dorsalis utrinque leviter declivis, ventralis vix arcuatus ; cxtremitas lateris antici brevioris paululum medium supra leviter acuminato-rotundata ; postica superme A A 2 188 ZOOLOCY OF KERGUELEN ISLAND. subrostrata, iiim^Hlio lovitor simiata; don tes card inalos postico circitor 32, anticc 11 ; impressio pallii poi-prol'undo sinuata. Lat. 02 mill., long. 32, crass. 19. Animal furnished with a large foot of an oval rnmi beneath, acuminated at Ixtili ends, deeply cleft down the ciMitro, and dcejily striated across, surrounded by a nearly even margin; siphons small, united, retractile, very nn(M|unl in size; the upper or exhalant one very slender, the low(>r considerably larger ; gills small, ter- minating at the side of the body ; mouth very small; palpi vei-y large, lamellately wrinkled within, the terminal aj)pendagcs towards the siphons with undulating margins; the margin of )uantle treble-edged, simple, only fringed near the siphonal extremity. Hab. — Royal Sound, both at Observatory Bay and near the eastern shore of Swain's Harbour, on mud in about 10 fathoms. This magnificent species is by far the largest yet described of this genus, and is at once known from the other three species by its different form. The posterior end pouts in the same manner as in the Xortli American Yoldia t]irackpfori)i,is. 'V\\c epidermis in young and half-groAvn specimens is of a bright yellowish olive coloiu' ; but in the adult shell it becomes of a dark olive-brown, and is much eroded in the umbonal region ; it is shglitly reflexed within the margin of the valves, and is held between the two outer edges of the mantle. The few radiating contiguous striations towards the anterior end f urnisli another very distinctive character. Mytilus magellanicus. M. magellanlcus, Chcmn. Con. Cab. viii., pi. 83, f. 742 ; Knorr, Vergniigcn iv., pi. XXX., f. 3 ; Eeeve, Con. Icon, x., pi. G, f. 22 ; Dall, Bulletin N. S. Nat. Mus., Ko. iii., p. 1". M. bidens (Linn. ? part.) et auctorum, vide Born. Mus. Vindobon. p. 128 ; Gmelin, p. 3351 ; Dillwyn, Descript. Cat. p. 313. Hah. — Eoyal Sound, on roots of kelp and on rocks at a depth of 3 fathoms, obtained by means of a grapple. This common Magellan species varies in coloiu' very considerably. The largest specimens are generally of the form and dark purplish-black colour, as represented by Beeve's figure ; others are of a more purplish-slate tint, clothed with a rich olive- ■hrown epidermis, and again others are altogether bright yellow. In the British Museiuu there are several examples of this species with the locality New Zealand -.attached to them. I cannot trace any character whereby they can be separated specifically, and the habitat is certainly correct, as the shells were received from a reliable authority. !Mr. Eaton says that " the specimens brought up by the kelp often had extremely thick sheUs, and occasionally measured upwards of five or six inches in length ; the MOLLUSCA.— E. A. SMITH. 189 thin yellow examples appeared to be young, and those which were olive-brown to be less aged than the purple-black shells, as a rule. The animals were in great request among the blue jackets (when the expedition first landed), who used to haul up the kelp and collect them by buckets-full ; but early in November they went out of season and became uneatable." Mytilus edulis. Mytilns edulis, Linn. Syst. Nat. ed. 12, p. 1157. Mytilus canaliculus, Dall (non Hanlcy), Bull. U. S. Nat. Mus. 1870, 3, p. dl ; (non M. caualiciilafus, Martyn). Probable varieties are M. uugulatus (Linn, part), Lamarck ; M. chilensis, Ilupe; M. chilensis (Philippi), Reeve ; M. obesus, Dunker ; M. trossulus, Gould. Hab. — Abundant on rocks between and a little below tide-marks at Kersuelen's Island. After a careful consideration of this species, I cannot arrive at any other con- clusion but that the Kerguelen shells undoubtedly are specifically the same as the common edible mussel (JLT. edulis). No definite distinction can be traced in the shells (unfortunately only eleven in number) collected at Kerguelen, from specimen from the Dutch coast bought in the London market. The form of the shell (always more or less variable), colour of the exterior and interior, the hinge vdi\\ the few irregular teeth, muscular scars, and the punctures in the interior towards the ventral margins, are precisely alike in both local forms. Mi*. W. H. Dall, who has given an account of the Mollusca obtained by the American Transit party {see Bulletin U. S. Nat. Mus. p. 48) remarks that " the shell of this species closely " resembles some varieties of Ili/tilus edulis, but the soft parts are quite different. " The foot is large and quite flat beneath. The viscera and branchite are white, " the foot and mantle edge streaked with dark brown." I have closely examined the soft parts of four Kerguelen specimens, and contrary to Mr. Dall's assertion that they are quite different, I find them to be exactly the same as in European specimens, excepting that the foot is smaller if anything, and not " larger," but this may possibly be due to contraction, since the foot in some Dutch examples is larger than in others of similar size. In coloration not the slightest difi'erence is discoverable, both forms having the foot and the mautle- m.ar£:in more or less brown. Several other species, for instance, M. chilensis, Hupe, described in the zoological portion of Gay's History of Chile, M. chilensis of Philippi, published in Reeve's Conchologia Iconica, and obesus, Dunker, in the same work, apparently do not oifer any appreciable specific characters whereby they may be separated from 31. edulis, and I do not feel convinced that the large 31. nngulatus is anything more than a gigantic form of this species. And again Jeffreys, speaking of II. trossulus, Gould, says that it " probably differs in no other respect than being called a 'representative ' 190 Z(1(M.()C.Y OF KEKGUELEN ISLAND. (< species." Thus it uould api)eav that these so-called species probably arc hut locality-spoeios, but uutil an oppoituiiily occurs of studying- the Mylilidie oi niausc, it would be liazardous to allirui so dcliuitcly. None oi' tiic speciaieus from Kcr- ffuelou Island exceed two inches and a hall' iu Iciitrth. Modiolarca trapeziua. Modiola trapes'tiia, Lanik. Anim. S. Vert. cd. 2, vol. vii., p. 21. Jloiliola f rape:: ilia, Kiister, Con. Cab. viii., heft. 3, pi. 0, L 10 & 17. JTodiolarca trapeziua, Gray, Synopsis Brit. Mus. 1810, p. 151 ; Proc. Zool. Soc. lSi7, p. lOi); Adams, Gen. Eec. Moll, iii., pi. 122, i'. 1 and la. ]?haseoUcama trap)ezina, llupe, Gay's Uist. Chile, Malacologia, pi. 8, f. 9. JPhaseoUcama inagellanica, Kouss., Voy. au Pol Sud, ]\roll., p. IK!, pi. 2(!, f. 2a-f/. Gaimardia irapesiiia, Gould, Atlas United States Explur. Expcd. pi. -XI, f. 568. Uab. — Kcrguelen Island. This species is also found at the Falkland Islands. Messrs. H. and A. Adams describe " the hinge with two small oblique teeth in the right valve, Avhich receive " two corresponding ones ou the left." All the specimens which I have examined, l)e they adult or young, have but a single tooth in each valve, or more strictly speaking a single rounded tubercle situated just below the apex of the lunbo, and some- times the faintest indication of a second. The ligament is very slender, only just visible exteriorly, and placed in a narrow' elongate groove posterior to the umbones. The shells present various grades of colour, some being as Lamarck states, " luteo-fulva," others gradually passing into dark purplish-red. Modiolarca exilis. (Plate IX., fig. 21.) 3Iodiolarca exilis, H. & A. Adams, Proc. Zool. Soc. 1863, p. 435. Testa parva, inajquilateralis, irregulariter transverse ovata, antice breviter ros- trata, mediocriter convexa, sub cpideiinide tenui fugaci flavo-olivacea (plerumque partim detrita), fusco-purpurca, coneentrice tenuiter striata, umbones prominentes, incurvati, contigui, propius ad latus anticum siti ; margo dorsalis antice subrecte declivis, postice arcuate declivis ; latus anticum angustum paululum rotundate rostratum, posticum late arcuatum ; margo ventralis leviter arcuatus antice levis- sime sinuatns ; dentes 2 sub apicem utrseque valvae ; ligamentum fere omnino internum, in sulco elongate postico sitimi. Diam. transversa 5 mill. Alt. fere 4. Crass. 2i. Hah. — Kcrguelen Island and " Falkland Islands," Mus. Cuming. The onlv specimen of this species from Kcrguelen Island agrees precisely %vith others from the Falkland Islands, of which there is a large series in the Cumingian Collection, and also with the types in the collection of Mr. Henry Adams. MOLLUSC A. — E. A. SMITH. 191 In every example tlie tliin yellowish-olive epidermis is worn off from a large portion of the sm-face of the valves, and is retained only near the ventral margins and on the sides. Modiolarca niinuta. (Plate IX., fig. 23.) Kidderia mintita, Dall, Bulletin U. S. Nat. Mvis. 1876, No. 3, p. 46. There are three sijecimens of this speeies obtained by the early Antarctic Expe- dition in the Museimi. At first they appeared to me to differ so slightly from M. jntsilla of Gould from Terra del Fuego that I had labelled tliem as a variety of that species ; but Mr. Dall, who possibly has a larger series at hand, has pointed out certain differeuces, which although slight may be siifficient to separate the two forms. However, I cannot retain the genus Kidderia as described by him. He says that "it diifers from Modiolarca in its single anterior muscular scar, the " presence of strong mjmphce for the sub-internal ligament, and in thefull devc- " lopment of the cardinal teeth." Of the three examples at hand I have opened two, and in both distinctly observe that there are two anterior muscular scars as in Modiolarca. The second, the upper one, is extremely difiicult of observation, because it is situated deep within the shell and under the apex of the umbo. In M. pusilla two scars are also present. The second character referred to as distinguishing Kidderia from 3Iodiolarca, namely, " tho presence of strong iii/nqyhce," is merely one of degree, and equally untenable. In the present species they are less strongly develoj)ed than in j^i^silla, and rather more so than in the type of the genus (M. hxipezina.) The third distinction is likewise one of degree of development, the teeth of the Patagonian species being rather stronger than those of the Kerguelen shell, and both more developed than those of the common trapezina, in which they are represented by a very small tubercular tooth in each valve, and sometimes a faint trace of a second one. Eadula (Limatula) pygmsea. (Plate X., fig. 16.) Lima pygmaa, Philippi, Archiv. f. Natui-gesch., 1845, p. 56. Limatula falldandica, A. Adams, Proc. Zool. Soc, 1863, p. 509. Hah. — Swain's Bay. There can be no doubt, I think, that the above two names have been applied to the same species. Both authors describe their shells as ovate, equilateral, and the costation obsolete on the sides. In the Museum there is a specimen from the Strait of Magellan, the locality cited by Philippi, rather larger than his example, which was evidently a young shell, but answering to his description in every respect, and agreeing, excepting in size, perfectly with the type of Falldandica, which measures ISJ: ZOOLOGY (IF KKRGIIKLKN ISLAND. 15 mill, long and 11 in Invadth and 10 in tliicknoss. Nothing need Lo added io Adams's oxcellont description c>xcei)ting the numlun- of tlio i-il)s, which average about 21 in the largest specimens. BRACHIOFODA. Waldheimia dilatata. Tercbrofida dihitoto, Lamarck, Anim. sans Vert., ed. 2, vol. vii., p. oSO; Sowcrhy, Tliesaurus Conch., i., p. 352, PI. Lvx., figs. 18, 19. Ti-rchralithi yaudichamli, Blainville, Diet. Sci. Nat., 1828. WahUieimia dilatata, Gray, Cat. Brachiopoda Brit. Mus., p. 59. Tercbmtida {iraldheimia) dilatata, Reeve, Con. Icon., xiii., PI. II., lig. 2. if^fi.^Observatory Bay. Ou rocks at 1 fathoms, obtained Avith tlu; aid oF a grapple out of a cleft in the rocks. Reeve questions the correctness of the hal)itat attributed to this species by Gray. But considering how many species of animals found at Kcrguelen Island are also indigenous to Patagonian seas, there can be little doubt that Gray Avas quite correct in this instance. Exi'LAXATKIN OF PlATE IX. Fig. 1. Neobuccimim Eatoni, a. operculum. 2. TropLon albolabratus. 3. Struthiolaria mirabilis, a. operculum. 4. Admete limiireiv'formis!. 5. 5«. Scissurella supraplicata. 6. Litforina setosa. 7. Hydrobia pumilio. 8. „ caligiuosii, and opercuhim. 9. Eatoniella caliginosa. 10. „ kerguclenensis, and operculum. 11. „ subrufescens, and operculum. 12. Rissoa kergueleni. Fig. 13, 13«. Patella (Patinella) kerguclenensis. 14. 14a. „ „ fuegieusis. 15. Skenea subeanaliculata. 16. Kadulu pygmsea. 17. Area (Lissarea) rubro-fusca. 18. Yoldia subiequilateralis. 19. Solenella gigantea. 20. Kellia coiisanguinea. 21. Saxicava bisulcata. 22. Lcpton parasiticum. 23. Modiolaicu niinuta. 24. „ exilis. POLYZOA.— G. BUSK. 193 PoLYZOA.— ^y G. Busk, F.B.S. (Plate X.) Owing to the absence* of any published accounts of the Polyzoa collected at Kerguelen Island by the Challenger, the American Transit of Venus, and the German Surveying and Transit of Venus Expeditions in 1874-75, the subjoined list treats exclusively of Mr. Eaton's collection. The 26 or 27 species comprised in it are all of them inhabitants of the littoral or Laminarian zone, and were obtained with the grapple in Swain's Bay and Observatory Bay. Of the whole numljcr 17 or 18 belong to the suborder Cheilostomata, 9 to the Cyclostomata. No represen- tative of the Ctenostomata was collected. The collection affords nine or ten forms previously undescribed ; the remainder belong to a fauna which ranges from the southern extremity of S. America (which may be regarded as its " centre ") to New Zealand in a westerly direction, one or two species extending even farther, to AustraKa and the Cape of Good Hope. It is observable that no Arctic form has been brought from Kerguelen Island, althoufji some have been met with further south, two instances of the occurrence of the Arctic Hornera lichenoides obtained during the voyage of H.M.SS. " Erebus " and " Terror " having been communicated to me by Sir J. Hooker. Mr. Eaton suspects their absence may be attributed to the shallowness of the areas searched by hii)i, the greatest depth being not more than 10 fathoms. CHEILOSTOMATA. Salicornaiiiid^. Salicornaria malvinensis. Busk, Brit. Mus. Cat. Polyzoa, part i., p. 18, PI. Ixiii. 1, 2, and Ixv. (bis), 1. Hab. and Dist.— Swain's Bay. Also East Falklands and S. Patagonia (Darwin). Onchopora. Ann. and Mag. Nat. Hist. 1876, xvii. 116. The genus Onchopora, as originally constituted, embraced Tubicellaria of D'Orbigny, but in the place cited above I have proposed to restrict it to those forms which have no tubular prolongation of the mouth. They certainly constitute a very distinct type. Onchopora sinclairii. (Plate X., figs. 1, 2.) Busk, Quart. Journ. Micr. Soc. v. 172, PI. xv. 1-3. Hab. and Dist. — Swain's Bay. Also New Zealand (Sinclair). The Kerguelen Island specimens have afforded me an opportunity of givini'- better figures of this species than the earlier drawings. * This account was drawn up in 187(j. B B 194 ZOOLOGY OF KERGITELEN TSLAND. CELLULARIIDiE. Cellularia cirrata. Celliilaria cirrata, Ellis and Solaiidcr, Zoopli. 29, tab. iv. D. Men'ipca cirrafa, Lamx., Exp. Meth. p. 7, pi. iv., fig. D.U 1 ; Bk. Brit. Mus. Cat. Poly, i., p. 21, pi. XX., 1, 2. llab. — Swain's Bay. Also S. jU'rica, Krauss. Meuipea fiiegensis.* Busk, Cat. Polyz. i., p. 21, pi. xix. Hal), and Dist. — Swain's Bay. Also Falldand Islands (Hooker) ; Ticrra del Fucgo (Darwin). Meuipea patagonica. Busk, Cat. Polyz. i., p. 22, pi. xxiii., 1, xxv., xxvi., 1, 2. Uab. and Dist. — Swain's Bay. Also Falklands (Hooker) ; Port Desire, Pata- gonia, at low water (Darwin). CjUJEREIDiE. Caberea boryi. Caberea boriji. And. Savig. Descript. del Egypt. Explic. tab. xiii., partincnt of tbe Fauna. Tbe Antarctic Expedition under Capt. Sir James Ross visited tbe island in tbe winter (May and June 1810), and of tbe species of Crustacea in tbcEritisb Museum, apparently broucjbt back by officers of that expedition from " Kerguclen Laud," two ai'c new to science. A brief summary of tbe results obtained by tbe Challenger Expedition by dredging in tbe neighbourhood of Kerguelcu Island, was giA^en by the late Dr. R. von AYillenioes-Subm in a letter to Professor C. Th. E. von Siebold, published in tbe Zeitscbrift f. Wissenscliaft. Zool. xxiv., 187 1. Prom it w(! learn that upon tbe island itself nothing was found except a small Bracbyurous Decapod (probably tbe Salicarcinns planatus) . The inhabitants of the 1st zone (not deeper than 40 faths.) comprised several species of Serolis, Sphceroma, Arctiirus, some Gammarida;, several species of Caprella, and some Pycnogonida. Richer and more interesting results were obtained in the 2nd zone (10 to 120 faths.), where Tauais and Praniza, very remarkable jlniphi^wda, Mijsidcp, and a Nebalia were discovered. The German Surveying and Transit of Venus Expedition dredged off the open coast, and collected, for the most part, along the rather exposed shores in tlie neiglibourhood of Betsy Cove. They entered also Vulcan Cove and Foundry Branch, but whether they searched at all for Crustacea in these sheltered inlets is uncertain. The collectors who accompanied respectively the English and American Transit of Venus Expeditions, were advisedly unprovided for dredging in deep water, and consequently their operations were confined to the beach, the laminarian zone, and depths external to the latter, not exceeding 10 fathoms. Mr. Eaton obtained 10 species, of which 7 were new to science. Dr. Kidder with more limited oppor- tunities for work, obtaining 7 species, added 3 to the fauna (2 of them were new to science), and thus raised the total to 15. No species of terrestrial Isopoda, of the family Oniscidcc, have been discovered on tliis island. Among the 15 indigenous species, several are characteristic of the Antarctic region, which in its widest sense embraces Tierra del Fuego, the Falklands, and the lands and islands of the Antarctic Ocean. Halicarcinus planatiis and Sphceroma gigas are known to inhabit the seas of Patagonia and New Zealand, and are especially abimdant in the former area ; Cassidina emarginata is indigenous also to CRUSTACEA. — E. J. MIERS. 201 the Falklands and Patagonia ; Jtcra piihescens occurs at the Falklauds, and SeroUs lotifrons has been obtained from New Zealand and the Aucldands. In addition to the Kcrguelon Island species, I have included Scrolls sejitem-cari- no fa froixi the Crozets in the present paper. DECAPODA. PlNNOTHERID^E. Hahcarcimis planatus. Cancer planaius. Fab. Ent. Syst. 1793, ii., 446. Lencosia planata, idem, Ent. Syst. Suppl. 1798, 350. Hymenosoma leachii, Guer.-Men., Icon. R6g. Anim. iii. Crust. 10, pi. x., 1 ; id., Voy. Coq. 1828, ii. 22. Halicarcimis pilanatus. White, Ann. & Mag. of Nat. Hist. 1846, xviii. 178, pi. ii., 1 ; idem, List Crust. Brit. Mus. 1847, p. 33 ; Dana, U. S. Explor. Exped. 1852, xiii., Crust, part i., 385, pi. xxiy., 7 ; M. Edwards, Ann. Sc. Nat. 1853, ser. 3, xx, 223 ; Heller, Reise der Osterr. Freg. Novara, 1865, Crust, p. 66 ; Miers, Cat. N. Zeal. Crust. 1876, p. 49 ; Smith, Bull. U. S. Nat. Mus. 1876, iii. 57. Hymenosoma tridentatmn, Jacq. & Lucas, Voy. del' Astrolabe, 1853, Zool. iii. Crust, p. 60, pi. V. 27. ^f'6.— Kerguelen Island, very common everywhere on the Macrocystis (Eaton) ; on rocky beaches and at 5 fathoms (Kidder). Also Tierra del Fuego ; the Falk- lands, abundant ; New Zealand ; the Auckland Islands. It is to be noted that in White's figure of this curious little flat crab, the three frontal teeth are not made sufi&ciently prominent : they rise from under the raised marginal line bordering the front of the carapace. The largest specimen in the series collected by Mr. Eaton is a little over 13 mm. in length. ISOPODA. ASELLID.E. [Jaera piibescens. Dana, U. S. Explor. Exped. 1853, xiv., Cmst. part ii., 741., pi. xlix., 9; Smith, Bulletin U. S. Nat. Mus. 1876, iii. 63. Hah. — In company with Sphceroma gigas on rocky beaches. Kerguclen Island (Kidder) ; also Tierra del Fuego (Dana).] ^GIDiE. Mg& semicarinata. (Plate XL fig. 1.) Miers, Ann. & Mag. of Nat. Hist. 1875, xvi. 115. Corpvis punetatum, elongato-ovatum ; coxis oblique bilincatis; ultimo segmcn- tonim post-abdominis postice trimcato Icvissimeque emarginato, carina longitudiuali c c 202 ZOOLOGY OF KKKGUELEN ISLAND. la^vi ;ul mav^incm ]nistic:nii liaud attiiigonto ; iiciiiiltinio ( xlcridriluis rainormn apptMidii'ium subovatis, intcrioribus truncaio-trian^iilai-ilius, band us(|ii(' ad a|iic<'m sogmonti soquoiitis attingoiitibns. Body clongato-ovato, modoratoly (.•onvox, punctate ; tbe piiiict illations very s])arsc or wautiug; on 1b(> anterior halvos of the segments of the peri'ion, Imt niDi-c niinun-ous and coarser ui)on tliose of the pleon, especially the last of them. Eyes large. Each segment of tlic iicri'ion is travcrs(-l(^ small specimen of this species, but states that it is very common under stones on the shore, and in the kelp. A large series is in the collection of the British Museum, and many specimens are stated by Mr. Smith to have been obtained by the officers of the American Expedition. It is worthy of remark, that all of the specimens from Kcrguelen Island collected by the American naturalists, as well as those in the British jNIuseum, are the true »S'. ff/gcs of Leach ; and that there is no example among them of the variety named by White lanceolata, which occurs, as well as the normal form of the species, at the Palklands, where it appears to be as common as the typical »S'. gigas alone is at Kcrguelen Island. The variety mentioned is characterised by the acutely lanceolate rami of the appen- dages of the penultimate segment of the pleon. I was at first disposed to suspect that this difference between the two forms Avas merely one of sex ; but I have since observed adult males of both of them. Dynamene eatoni. (Plate XL, fig. 2.) Miers, Ann. & Mag. of Nat. Hist. 1875, xvi. 73. Corpus late ellipticum, lateribus subparallelis, convexum ; ultimo segmentorum post-abdominis convexo, lateribus subrectis convergentibus, postice emarginato, excisura, fere eadem longitudiue ac latitudiue, rotundata ; pennltimo ramis appen- dicium suba'qualibus, integris, ovalibus, usque ad excisuram segmenti sequenti vix attingentibus. Body broadly elliptical, convex, smootli, naked, the sides of the pereiou almost parallel. Cephalon transverse, closely encased within the first segment of the pereion, bordered in front Avitli a thin raised line. Eyes very small. Segments of the pereion short, of equal length above ; the first three with the posterior edge nearly straight, the next four bent slightly backwards at the sides. Segments of the pleoiT (the last excepted) coalescent, with the lines of union marked by incised lines at the sides ; last segment convex, with the lateral margins almost straight, and with a roimded emargination, which is about as wide as deep, at the distal extremity. Upper antennae with the basal two joints dilated, the first about twice as long as the second ; the third joint very slender, and about as long as the second ; the flagellum, Avith about li joints, reaches as far as the posterior margin of the first segment of the pereion. Lower antenna) with the first four joints slightly dilated, the first two short, the next two longer, the fourth usually the longest ; the fliagellum, composed of about 24 joints, reaches to the posterior margin of the tliird segment of the pereion. Pereiopoda slender, almost naked ; c c 2 204 ZOOLOGY OF KKHGUELEN ISLAND. (lac'tyli witli two flaws. Kami ol" llic lateral appendag-cs of the ploon snboqual, oval, outiiv, not ciliated, reacbiug almost to the terminal uotek ol' the I'ollowiuy segment. Colour reddish or i^reyish broMii, Avitli darker spots. Length of the largest (S) 17 ; of the smallest 3 mm. JIab. — Swaiu's Bay and Observatory Bay ; common. Mr. Eaton collected a good series of examples of dilfercnt ages and sizes. In contour 2). eatonl somewhat resembles D. d/imerili, Aud.,* whii-h has been recorded Jiom the coast of Natal ; but in that species the terminal notch is mucb deeper, aud the ranii of the penultimate segment are much shorter. Plate XI., fig. 2, D. eatoiii (enlarged) ; a, cepbalou (enlarged) ; b, mandible ; c, maxillipes ; r/, Mli pereiopus. Cassidina emargiuata. Cf'ssid/iia emoi-ginata Guerin-Menev., Icon. Il(>gne Anim., Texte Crust, p. 31 ; Cuningham, Trans. Lin. Soc. 1871, xxvii., part iv., 499, pi. lix. 4. Cassidina latistylis, Dana, U.S. Explor. Expcd. 1852, xiv. ; Crust, part ii. 784. pi. lii. 12. Sab. — Eoyal Sound and Swain's Bay. Also "W. coast of Patagonia and the Straits of Magellan (Cunningh.), plentiful; Falkland Islands (Gu(5r. M6n.). The form of tbis species clianges considerably with tlie advance of age. In young examples (those seen by me are nearly all of them females) the body is more convex, and is proportionally narrower than it is in the others; resembling the form figiu'ed as C. latistylis by Dana. In older specimens (males) from Kerguelen Island, the body is more depressed and is much broader than in the others ; resembling Guerin-Meneville's description and Cuningham's figure. There are specimens of this latter form in the British Museum from the Straits of Magellan, and it is probable that the two may be identified as conditions of the same species. ANISOPODA. SEROLID-E. Serolis latifrons. Serolis latifrons (White, List Crust. Brit. Mus. 1847, p. 106 : — name only, with- oxit description) ; Miers, Ann. & Mag. of Nat. Hist., 1875, xvi. 74 ; id. Cat. N. Zeal. Crust. 1876, p. 117, pi. iii. 7 ; Smith, Bull. U. S. Nat. Mus. 1876, iii. 63. Corpus culminatum itaque convexum, acute ovatum ; ultimo segmentorum post * Dynamene dumerili. Sphtsroma dumerilii, Aud. in Sav. Desc. Eg. 1809, Explic. tab. Crust, i. 95, pi. xii. iv. ; Krauss Siiclafr. Crust. J 843, p. 65. S. savigniji, M. Edwards, Hist. Nat. Crust., 1840, iii. 208. CRUSTACEA. — E. J. :\IIERS. 205 abdominis subtviangvilari, postice emavgiuato, supra tricarinato, carina, in medio un4 e basi ad apicem altt\, recta, alteraque utrinquc minus cxpresst\ prius ad margincm anticum et adjacentc atquo parallola, deinde jiixta latus rctrocurvatii, non tamen ad marginem attingeute : penultimo ramis appendicium acuminatis, imparibus, ex- teriore brevissimo. Body roof-sbaped, with a series of impressed lines and punctulations near the posterior margins of the segments. Segments of the pereion sinuated, but not (as in some of the species in this genus) much prolonged posteriorly. Terminal segment of the pleon large, sub-triangular, with the apex semicircularly emarginatc ; an elevated keel extends directly from tlie middle of the base to the terminal notch, and on each side of it one less prominent runs outwards close to and parallel with the base of the segment, and towards the lateral margin is cm*ved backwards so as to terminate eventually not far from the same. Upper antennse short ; the joints of the peduncle dUated, the last of them less so than the preceding. Lower antennae Avith the last joint of the peduncle as stout as but about twice as long as the pen- ultimate joint. First pan* of gnathopoda with the carpus very short, slightly prolonged and acute at its distal extremity ; propodus broadly ovate, articulated with the carjius at the middle of its lower edge, which in front of the joining is armed with a close-set series of short spines ; dactylus acute, arcuate. Second pair of gnathopoda (in the male) slender, the propodus not dilated, the dactylus when retracted fitting into a cavity in the lower surface of the propodus. Pereiopoda slender. Rami of the appendages of the penultimate segment of the pleon narrowly acuminate, the outer ramus very small and not half as long as the other. Colour brown, with irregular paler blotches. Length of the largest example 30 mm. Sab. — Royal Sound ; common about rocks in shallow water, e.g. in a cove full of reefs adjacent to Observatory Bay, and along the rocky beach near the other English station at Swain's Haulover. Dr. Kidder also obtained a specimen near MoUoy Point ; and it was obsei'ved seemingly by Sir J. Hooker at Christmas Har- bour {fide Eaton). Also Ptcndezvous Cove, Aucklands (Brit. Mus.). The present species as well as jS*. septem-carinata are clearly distinguished from all that are enumerated by Professor Grube va. his Monograph of the genus* by the form and direction of the ridges upon the last segment of the pleon. [Serohs bromleyana. Will.-Suhm,, Zcitsch. f. Wiss. Zool. 1874, xxiv., App. p. 19. llah. — In 1975 faths., S. of Kerguelen Island (Ch. Exped.). This species is very large, and has the segments of the pereion produced at the sides into very long spines.] * Beitrage zur Kenutnis der Gattung Serolis, in Archiv. fijr Natnrgcsch. Berlin, 1875, xli. 208-234, pis. V. & vi. 206 zooLOGi OF ki:rgup:lkx island. [Serolis septem-carinata. (Plato XI., Iv-. 3.) Serolh qiiodncarhKita, White, MS. List Crust. l?rit. ^lus. 1*^17, |>. IOC: — iiaino only, no description : — iiomcii hieptitvi. Scrolls scptem-cctrhtatd, Miers, Ann. & Ishvj;,. of Nat. Hist. 187"), x\ i., IKi. Corpus depressum, i'Ui;'osuiu ; ultimo segmeutoruui ])ost-abdonunis laicrihus siuuatis, apiee lente cmarginato, dorso in lougitudineni recte sc])iiMn-eaiiiKiluiu; pemdtimo raniis appendieium parvis lamellifonuibus subacutis, horum intcriore l)aulo niajore. ]iody depressed, and (esj)ecially at the sidi-s) rugose. Segnieiils of the ])ereion ■with the postero-lateral angles prolonged backwards and acute ; the corresponding angles of the i)enultiinate segment of the pleon similarly produced, so that their apices lie almost in a straight line ^vith the distal extremity of the ultimate segment. The last segment of the pleon very slightly concave at the sides, with a shallow ajiical emari^ination ; dorsum traversed longitudinally by soven cavintc ; of these one in the middle extends to the terminal emargination but is indistinct, Avhile the carina next to it and next but one are somewhat thickcmcd posteriorly, and tenuinate before meeting the lateral margins. Upper antenna3 small and very slender; peduncle with the first two joints short and dilated, the third joint longer and slender, the foiu'th A-ery small ; llagcllum hardly reaching to the postero-lateral angle of the first segment of the pereion. Lower antennte witb the terminal and subterminal joints of the peduncle long and subequal to one another (flagcdla im- perfect). Gnathopoda almost similar to those of S. laUfrons. Pcreiopoda slender, the claws small. Rami of the appendages of the j'Lmultimate segment of the pleon small, lamelliform, subacute ; the inner ramus rather the larger. Length of tbe largest example 3 3 mm. Huh. — The Crozets. Three specimens obtained by Lieut. A. Smith, E.N., arc in the British 3Iuseum. The catalogue name applied to this species by "White, W'as probably given to it on accoimt of the prominence of four of the ridges (the nearest two on each side of the median ridge) of the terminal segment. Plate XL, fig. 3. S. septem-carhwta (enlarged 2x2); a, appendage of the penultimate segment of the pleon.] AMPHIPODA. Orchestiid^. [Hyale villosa. Ilyale villosa, Smith, Bull. U. S. Xat. Mus. 1876, iii. 58. Sab. — Kerguolen Island, on rocky beaches (Kidder). Mr. Smith described this species from a single somewhat mutilated male example, nearly 10 mm. long.] CRUSTACEA. — E. J. MIERS. :207 [Lysianassa kidderi. Zi/siaiiassa Ichhleri, Siuilh, Bull. U. S. Nat. Mus. 1876, iii. 59. Long. 3 to 1- mm. Uab. — Kerguelen Island, on rocky beaches, with Hi/ale cillosa (Kidder). According to Mr. Smith (loc. cit.), "all the specimens received are apparently " immatm-e, and the males evidently, and very likely the females also, have not " attained the adult characters." He fm-ther observes that " the characters " assigned to the genus Lijsianassa (as restricted by Boeck) would require con- ■*' siderable modification to admit our species."] Anonyx kergueleni. (Plate XI., fig. i.) Lysianassa kergueleni, Miers, Ann. & Mag. of Nat. Hist. 1875, xvi. 71. Cicca, corporc la?vi ; angulis capitis antero-lateralibus acutis, prorsus productis ; iis tertii segmcntorum post-abdominis postero-lateralibus in lobis angustis et acutis reflexis, ad apices siirsum leviter curvatis ; segmento terminal! lamelliforme bipartite. Smooth ; eyes invisible ; antero-lateral angle of cephaloii acute, produced below and beyond the base of the upper antenna. Third segment of the pleon with the postero-lateral angles prolonged as narrow lobes, which are more or less cmwed upwards at the tips, and which at their base are nearly at right angles with the posterior margin of the segment. Upper anteunte subpyriform ; their first joint large and stout, the next two short ; flagellum composed of about 14 joints, the first of which is longer than the next ; accessory appendage 5-jointed, with a slender hair at its tip. Lower antennae slender, rather longer than the upper; the last two joints of the peduncle longer than the one preceding them ; flagellum composed of about 21 joints. The mandibles have the slender palpus inserted on a level with the strong molar tubercle. First pair of maxillte with the inner lobe slender, ovate, and armed at its ajiex Avith two setae ; the outer lobe strong, truncate, armed at the apex with three or fom* spines ; the last two joints of the palpus lamellifoi-m, ovate, finely denticulated along the apical margin. Second pau- of maxillae with the lobes rather narrowly ovate, ciliated at the apex ; the outer lobe the larger. The maxilli- pedes have the inner lobe long, reaching nearly to the extremity of the antepenulti- mate joint of the palpus ; the outer lobe extends almost to the apex of the penulti- mate joint of the same, and is minutely denticulated at its rounded apex and along its inner edge ; terminal joint of ])ali)us unguiform. Tirst pair of gnathopoda witli the carpvis about as long as the propodus ; the propodus more than twice as long as •wide, with the sides parallel up to the ciliated and obliquely truncate distal extre- mity; dactylus acute, reversible. Second ])air of gnathopoda longer than the first, slender and weak ; propodus shorter than the slender carpus, and with dense long hair at its distal end ; dactylus obsolete. Coxae of the second pair of pereiopoda 208 ZODLOGY OF KERGUELEN ISLAND. deeply eniavG^inate behind, and with the postero-hilcral angk^ shortly produced hack- Avards at the apex into a l)rt)a(l ohluse lobe. Tliird, fourth, and Ufth pairs of pcrei- opoda with the basa broadlv i)l)long, and their straight ])()stcrior marg-ins very minutely serrated. All of the pleopoda are biraniose ; the rami slender, acutely lauccolatc. Telsou lamelliform, h)iii;( r tlian l)road, slightly narrowed towai'ds its apex, cleft almost to its base ; the lobes mueronatc. Length of the largest example about 15 mm. Hab. — Eoyal Sound. Common. In the form of the antero-lateral angles of the eephaloii, and of the postero- latci-al angles of the tliird segment of the pleon, this species to some extent resembles *(1) Ilippomedoii holbuUi, Kroyer, as described by Boeck, as well as (2) II. abi/sii, Goes, and (3) Anonyx immilns, Lilljeborg, — all from tlic Northern Seas. But it dift'ers from these species in having the inner lobes of the maxillipedcs proportionately much longer ; aiul in this respect it approaches more nearly to the type of structure exhibited in Orchomeite, Boeck. The eyes also, which are well marked in the species just referred to, are not visible in any of the specimens of A. kergiteleni. On account of the subcheliform character of the first pair of the gnatliopoda, and the divided telson, I refer this species to the genus Anonyx, as defined by Mr. C. Spencc Bate, instead of retaining it in Lyslanassa, where I placed it at first. I cannot refer it with certainty to any one of the numerous genera recently esta- blished by Boeck in his systematic arrangement of the Scandinavian and Arctic Amphipoda ;-\ I believe, indeed, that it will be found necessary to introduce im- portant modifications of the systematic arrangement and generic characters proposed by tliis author into any general revision of this difficult order, which may hereafter l)e undertaken, based upon the comparison of species from foreign as well as the Em-opcan and Arctic seas. Plate XI., fig. 1, A.ker(jucleni {exASiY^adi, 2x2); a, end of pleon (side view); h, telson ; c, 1st maxilla ; d, 2nd maxilla ; e, maxillipes ; f, 1st gnathopus ; g, 2nd guathopus. Atylus australis. (Plate XI., fig. 5.) Taramcsra australis, Miers, Ann. & Mag. of Nat. Hist. 1875, xvi. 75. Atylns australis, id. op. cit., p. 117; Smith, Bull. U. S. Nat. Mus. 1876, iii. Gl. * (1.) Hippomcdon holhoUi, Krijy., Boeck Forliaudl. Videiisk. Selsk. 1871, p. \02.:=AnoHyx cknticu- latus, S. Bate Cat. Amphipod. Crust. Brit. Mus. 1862, p. 74, pi. xii. 2. (2.) Hippomedoii ubysil, Goi-s, v. Boeck, op. cit., p. 103. (3.) Anoiiyx pumilus, Lilljeborg, v. Boeck, op. cit., p. 110. f De Skandinaviske eg Arktiske Amphipoder, 1872-76. 4to. Cbristiiinia. CRUSTACEA.— E. J. MTERS. 209 Corpus Itevc, carinis spinisque (lovsalil)iis carontibus ; oculis subi'oniformibus ; ' antcnnis subiX!qualibus cxappendiculatis ; pedum primo et secundo suljparibus, manu latoribus subparallelis apiceque oblique truucato, dactylo brcvi ; tortio scgmentorum post-abdomims marginf^ posticA, subrectil, angulisque postoro-lato- ralibus rotundatis ; ultimo scgmentorum longo, bipartito, lobis ad apices emargi- natis. Body smootb, without cloi'sal carina?. Ccplialon with a small subtriaugular median lobe, and broad obtuse slightly prominent lateral lobes ; eyes sub-rcuiform, black. First, three segments of the pleon with the inferior margins rounded, forming a distinct angle with the posterior margin in the second segment, and appearing to be minutely serrulate, owing to a series of small submarginal spines. Antennsc about half as long as the animal, sub-equal, slender, without an acces- sory appendage ; the upper pair- with the first two joints of the peduncle each about as long as the ccphalon, and the third joint short ; the flagellum witb its johits increasing in length but diminisliing in thiclcness towards the extremity ; lower antennai with the first three joints short, the fourth and the fifth joints longer ; the flagellum as in the upper antenuse. Mandibles spinose at the apex ; palpus inserted on a level with the strong molar tubercle, and triarticulatc, the second joint much the stoutest, furnished like the third joint with long cilia towards its distal ex- tremity. Palpi of the first pair of maxillae 2-jointed ; the second joint ciliated at the apex and more than twice as long as the first. Lobes of the second pair of maxillae oval, ciliated at the extremities ; the outer lobe rather the larger. Maxillipedes with the palpi 5-joiuted, and ciliated, their third joints the largest, and then" apical joints uuguiform ; inner lobe ciliated and at the apex spinose, reaching as far as the distal end of the second joint of the palpus, outer lobe denticulated on the inner edge, rcacliing to the apex of the antepenultimate joint of the palpus. Gnathopoda with the carpus shorter than the propodus, narrow at the base, enlarging distally ; propodus with subparallel sides, even margins, and obliquely truncate distal extremity ; upon which the acute, slender, and slightly arcuate dactylus can be closed. Thhd, fom'th, and fifth pairs of pereiopoda with small transverse coxae ; the basa longer, with the inferior margins rounded. Pleopoda with acute, slender, subequal rami. Telson lamelliform, reaching beyond the peduncles of the last pair of the pleopoda, cleft nearly to its base ; the lobes slightly emarginate at the apices. Antennae, gnathopoda, pereiopoda, and the rami of the pleopoda, fringed with short hairs. Length 17 mm. Rab. — Swain's Bay and Observatory Bay. Fom- specimens were obtained ; two of them adult. D D 210 ZOOUKiY OK KKROUELEN ISLAND. Tho present spot'ies resembles Atiilus fissieaxda* Dana, I'rom Valparaiso, in having renilbrni eyes, and tlu^ lobes ol' the tclson emarginate ; but it is distinguished from it by the greater length of the tclson, by the lobes of the same being somcAvhat naiTowed towards tlu> apices, and by their cmarginations being very small and placed a little on one side. Also the gnathopoda arc subequal ; the fourth segment of the pleon is slightly produced over the fifth, and its posterior margin is straight ; and the postero-lateral angles of all ol' the segments ol' th<> pleon are rounded, not acute as in A.finsicauda. It is probable that a separate genus Avill eventually have to be formed for the reception of the two species just mentioned, and A. austrmus,f Spcnce Bate. They differ from the normal species of the genus Atylus, as restricted by Boeck, in being destitute of dorsal carinations, and in some other particulars. For A. amtralis I originally founded a new genus FarcwKcra, allied to Melita in having the inner rami of the posterior pair of pleopoda short or rudimentary, but differing irom it in the absence of an accessory appendage to the upper antcnnoe. A subsequent examination of a series of younger cxamj)les showed, however, that my original types had sustained injury, the rami in question having been broken off and lost ; and that in reality the inner rami are as well developed as the outer in A. mistralis. Yet though the genus Paranuera is unavailable for A. mistralis, it will hold good for the reception of Melita tenuicornis,X Dana ?, and Gammams Fresnelii,^ Audouin, mentioned at the time of its publication as apparently included in it ; Tinless, as is probable, there be some error in the figures and descriptions published of these species. The specimens obtained by Dr. Kidder at Kerguelen Island, and doubtfully referred to A. australis by Prof. Smith, differ from the typical form principally in possessing a minute accessory appendage to the upper antennae, which does not exist in any of the specimens examined by me. Plate XI., fig. 5. A. atistralis (enlarged) ; a, end of pleon (side view) ; b, telson; c, mandible; d, 2nd maxilla; e, maxillipes; /, 1st gnathopod ; ff, 2nd gnathopod. Podocerus ornatus. (Plate XI., fig. 6.) Miers, Ann. & Mag. of Nat. Hist. 1875, xvi. 75. Coi"pus Iseve; tribus prioribus segmentorum post-abdominis posticc utrinque eraarginatis, ct angulis postero-lateralibus rotundatis; ultimo eorundem simplici * Atylus fissicauda. Iphimediajissicauda, Dana, U. S. Explor. Exped. 18.53, xiv. 929, pi. Ixiii. 4. t Ati/lus austrinus, S. Bate, Cat. Amphip. Crust. Brit. Mus. 1862, p. 137, pi. xxvi. 4. X Melita tcnuicornis, Dana, in U. S. Expl. Exped. 1853, xiv. Crust, i. 963, pi. Ixvi. 5 g-m. § Gammarus Fresnelii, Aud. in Sav. Descr. Egypte, 1809, Crust, texte i. 93, pi. xi. 3. r CRUSTACEA. — E. J. MIERS. 211 conico; pedum anticis parvis, carpo baud manu minori, secundis magnis, carpo pai'vo, manu valida ovat5, integr^. Ccphalon small, its anterior margin forming almost a right angle with the inferior jnargin in fi-ont of the small round black eyes. Posterior margins of the first three segments of the pleon notched above their rounded and not prominent postero-lateral angles. Ujiper and lower antennaj subequal, very robust ; pedimcles with the last two joints subequal, much longer than the one before them, densely fringed beneath with long flexible hair ; the upper with a small accessory appendage. Palpus of the mandible ciliated, very stout. Inner lobe of maxillipcs very short, reaching to the apex of the antepenultimate joint of the palpus ; outer lobe of the same reaching beyond the distal extremity of the penultimate joint, and denticulated along the inner margin ; the joints of the jjalpus are ciliated, and the penioltimate is the longest joint. Pu'st iiau* of gnathopoda small and weak ; merus, carpus, and propodus somewhat dilated beneath, with long hairs on their inner margins ; dactylus arcuate, acute. Second pair of gnathopoda large and well developed ; merus very small, carpus inserted toward the middle of the inferior margin of the propodus, propodus large, ovate, entire beneath, dactyl strong and arcuate. Last three paii's of pereiopoda mth the propodus shortly spinose, and the dactylus arcuate, acute, and reflexible. Pleopoda bu-amose, the last pair- the shortest ; each ramus has a series of short spines. Telson small, conical, simple, cUiated at the apex. Colour pale, varied with numerous small black spots. Length 13 mm. Rob. — Swain's Bay. Only two examples ; females with ova, in a mutilated condition. This species is distinguished by the long hairs of the antennse, the form of the second pau- of the gnathopoda, whose propodus is not dentate, the shape of the segments of the pleon, &c. Plate XL, fig. 6. P. meatus (much enlarged) ; a, end of pleon (side view) ; b, maxillipes ; c, 1st gnathopus ; d, 2d gnathopus. PYCNOGONIDA. Nymphonh)^. Nymphon antarcticum.* (Plate XL, fig. 7.) Ni/mjihoii (jracilipes, Miers, Ann. & Mag. of Nat. Hist. 1875, xvi. 7G. Corpus gracillimum, pilis brevissimis sparsis ; capite longitudine collo cequali, tuberculo oculigero obtuse subconico ; prolationibus segmentorum thoracis laterali- * It is necessary to alter the designation of this species; the uumo gracilipes, which I appHed to it In 1875, having been adopted in the same yesir by Dr. Heller, for a species collected by the recent Austrian Expedition to the North Pole. I have no means of ascertaining to which of the two species priority of publication, in the year referred to, belongs. D 1> 2 212 ZOOLIKIY 01-' KKKGIKLKN ISLAND. bus loiigc clistantibiis ; pedibus sccundo arliculorum tcrtio duplo longiovi, scptimo octavi ivquilongo. Very slender, sparsely elot bed \\itli very sborl bairs, wbieb l)ecoine more crowded towards tlie extremities of tbe legs. Head and neck subequal in Icngtb, and togctbcr abont as long as tbe rest of tbe body ; tbe former eylindrieal, stout ; tbe latter somewbat constricted in tlic middle. Oculigerous tubercle prominent, situated in front of tbe foremost pair of legs. Body terminated bebind by a sbort obtuse eylindrieal process. First (mandibula)-) pair of ap]iondagos triartieulate ; tbe iirst joint long and slender, tbe second and tbird bairy, forming a eompk;tc cbela ; dactvlus or tbird joint slender, arcuate. Second pair of appendages slender ; tbe fii'st joint very sbort, tbe second tbe longest, tbe last tbree joints bairy. Tbird (ovigerous) jiair of appendages 11-jointed ; tbe first joint very sbort, tbe second scarcely longer, tbe tbird and fourtb again longer, tbe fiftb tbe longest, tbe sixtb to tbe teutb gradually decreasing in size, tbe elevcntb minute. Legs very long and slender; tbe first joint and tbe tbird very sbort, tbe second joint ratber longer, tbe fom-tb, fiftb, and sixtb very long, tbe seventb and eigbtb (first and second tarsal) subequal, straigbt ; claws two, one very small. Lengtb 13 mm. ]Iab. Observatory Bay. On roots of Macrocystis grappled in 5-7 fatboms ; a single example. Tbis species is allied to N. grossipes, O. Fab., as described by Kvoyer, from tbe nortbern seas, but differs somewbat in tbe lengtb of tbe neck, and in tbe proportions of tbe joints of tbe legs and appendages.* Plate XI., fig. 7. N. antarcticum (natural size) ; a, body (enlarged) ; h, mandible ; c, tarsus. Nymphon brevicaudatum. (Plate XI., fig. 8.) Micrs, Ann. & Mag. of Nat. Hist. 1875, xvi. 117. Coi-pus robustum pilosum ; capite sessili crasso, tuberculo oculigero elato obtuse cylindrico; prolationibus latcralibus tboracis parum distantibus, pedibus secundo articulorum tertio parum longior, scptimo octavi sequilongo. Eatbcr stout, bairy. Head sessile, tbick, subcylindrical ; oculigerous tubercle bio-b, slender, subcylindrical. Body sbort ; tbe lateral leg-bearing lobes not remote from one anotber as in tbe preceding species, nor yet so closely contiguous as in TanijsUjlum, terminated bebind by a sbort subcylindi-ical process. First pair of appendages (mandibles) well developed ; tbe first two joints ratber long, tbe cbelae slender. Second pair of appendages witb tbe first joint very sbort, tbe second tbe • Pijciiogonum (/rossipcs, O. Fub. Fauu. Grocul. 17S0, 229. Nijmplwn grossipes, Kroyer, Naturh. Tidsskr. 1844-45. ii. K. i. 108; id. Voy. eu Scaiul. Crust, tab. CRUSTACEA.— E. J. MIEES. 213 lon£vcst, the third long, the fourth and fifth shorter, suhequal to each other. Third (ovigerous) pair of appendages 11-iointed ; the first tlu-ec joints very short, the fourth and the fifth subequal in length and the longest, the sixth rather shorter, the next four very short, the eleventh minute, uuguiforni. Legs with the first three joints very short, the second very slightly the longest of them, fom-th, fifth, and sixth long, seventh and eighth (fii-st and second tarsal) subequal to each other, straight and slender ; claws two, one very small. Length about 7 mm. Hah. — Kcrguelen Island (Antarctic Expedition). Several specimens, mostly females bearing ova. This species is allied to the boreal N. brevitarse* Kroycr ; but it is distinguished by its more robust form, its long and slender oculigerous tubercle, its longer tarsal joints, &c. Plate XI., fig. 8. N. brevicmulaUim (natural size) ; «, body (enlarged) ; h, appendage of second pair ; c, ovigerous appendage ; d, tarsus. ACHELIED^. Tanystylum, gen. nov. Caput sessile, crassum. Appendicium primge 1-articulatae, non chelif ormes ; secundge S-articulatae (?) ; tertitB lO-articulatse. Abdomen postice processu longo styliformi desinitiun. Head sessile, thick. First (mandibular) pair of palpiform appendages inarticu- late, simple, not eheliform ; second pair five-jointed (?) ; third (ovigerous) pair ten- jointed. Abdomen terminated by a long styliform process. The family to which this genus belongs was characterised by Dr. Semper f in 1874 as distinguished by the possession of simple mandibular appendages from all others of the Pycnogonida. It occupies a position intermediate between the Ni/m- phonidce (which have the mandil)ular palpi fully developed, triarticulate and eheli- form) and the restricted Fycnogonida; (which are completely destitute of mandibular palpi. Tanystylum differs from all the other genera of the Acheliida; in having the mandibles reduced to a single joint, and in the slender styliform termination of the abdomen. Tanystylum styUgerum. (Plate XL, fig. 9.) Nymphon stylUjerum, Miers, Ann. & Mag. of Nat. Hist. 1875. xvi. 76. Corpus robustum, hirtum ; capite sessUi ; prolationibus segmentorum thoracis * Nymphon brevitarse, Kroycr, Naturli. Tidsskr. 1844-45, ii. U. i. llo; Voy. cu ScanJ., &c. Crust. tab. xxxvi. 4. ■f Aclieliidm, Semper in Verb. Phys.-Med. Gesellsch. Wurzburg, 1874, vii. 274. 2ii ZUULOGl OF KERGUELEN ISLAND. lateral il)us contiEfuis ; avticulorum pedis soo\iii(lo o1 lovtio lonii-itiuHno suhirquaUbvis, soptiiuo Inrvissimo, octavo i-urvato. Katlior slout. Head very thit-k and soiucwbat barrol-sliaped, widest nearly in Iho middle. Body with the lobes of the leg-bearing segments in close contact with one another, and so constituting a broad mass or plastrum. Eirst (mandibular) pair of palpi-form appendages iini-articulate {i.e. of a single piece) ; second pair apparently 5-joiuted, with the first joint and the third very short, the second, fourth, and lU'th longer; third (ovigerous) paii' lU-joiuted, the first three joints very short, the fourth and liith longer, the next four very short, the tenth minute unguiform. Legs Avith the lii'st tlu-ec joints very short, the next three longer, the seventh joint (first tarsal) very short, the eighth (second tarsal) longer and curved ; claws two, unequal. Lemcth of bodv about 3 mm. Hah. — Observatory Bay ; on roots of Macroci/stis grappled in 5-7 faths. (two specimens only.) The hairiness of the palpi makes it extremely diflicidt to ascertain positively the number and proportionate lengths of their component joints. The styliform termi- nation of the body appears to arise from the dorsal siu-facc of the abdomen and from between the bases of the last pair of the legs. Plate XI., fig. 9, T. styligenim (nat. size) : a, body (enlarged) ; h, oculigerous tubercle ; c, ovigerous appendage ; d, leg. ENTOMOSTRACA.— G. S. BRADY. 215 Entomostraca.— J?y George Steicardson Brady, M.D., F.L.S., C3LZ.S. (Plate XII.) [OSTRACODA. All of the Ostraeoda collected by mc in Kei'guelen Island were lost by a breakage. — A. E. E.] COPEPODA. The Entomostraca submitted to me were taken in the following localities : — One surface-net gathering, in lat. 35° 9' S., long. 45° 30' E. ; another gathering from a freshwater lake, and a thu"d from a pool above high-water mark, both in Kerguelen Island. The oceanic species were Calauus Jimnarchicus and a Sapphiriiia, either identical with or very closely allied to S. dance, Lubbock ; those from Kerguelen Island were a freshwater species, apparently new, described by me briefly in Ann. and Mag. Nat. Hist., Sept. 1875, under the name Centropages brevicaudatus ; and a species from brackish water, JELarpactictis fulvus. No species have yet been recorded by the other Expeditions. Harpacticus fulvus, Fischer. Hab. — Royal Sound, Kerguelen Island ; abundant in pools above high-water mark. It was first noticed by Mr. Eaton at the American station (Molloy Point), in pools by the landing place ; but the specimens preserved came from Observatory Bay. The occurrence of this species in Kerguelen Island is particularly interesting from the fact that it is found all over the European shores in precisely similar situations, that is to say in brackish pools, at or above high-water mark, which are liable to become warm through exposiu'e to the sun's rays. These are in no respect distinguishable from European specimens. Centropages brevicaudatus, Brady, Centropages bremcatidattis, Brady, Ann. & Mag. Nat. Hist., ser. 4, Sept. 1875, vol. xvi. 1G2 ; Eaton, op. cit., vol. xvii. p. 264 Long. ? ^V inch. Male not observed. Female robust, rostrum short and blunt, last segment of cephalothorax produced on each side into an acutely angular oblong ala-form process. Upper antenna as long as the first two segments of the •21 r. zooLonv of kkkouelkn island. (•oi)halotliorax, sliiylilly tapering; toAvavds (ho oxiivinily, 1)osot. witli short setro, 25-joiiitod; joints suhoqual, somewhat increasing in length as far as the penultimate joint. Lower foot-jaw ratlier short, of modin-ate strength, armed at the apex with Uxo slender faleate claws. The otliev month organs present no distinctive pecu- liarities. Swimming-feet with hoth of llieir hrnnches 3-jointcd, the inner branch short ; first pair much smaller than the next thr(H>, with the terminal spines of the out(>r branch smooth, and of moderate length and strength ; iifth pair A«th the second joint of the outer branch produced internally into a strong denticulated spine, and the marginal setic of both of the brandies extremely short. Abdomen short, composed of two segments exclusive of the caudal segments ; the vulva forms a large rounded protuberance on the first segment; the caudal setae are short, sub- equal, half as long as the abdomen, and plumose. Ilab. — This species is very plentiful in freshwater lakes in the neighbourhood of Observatory Bay. It swinis slowly, and with an even motion; and. lience the f(>malos have the appearance of small brown seeds borne along with the water. Their abdomen is somewhat jecinoreusin colour, paler than the ccplialothorax. ]?y some accident they were at first stated to be oceanic (Ann. & Mag. Nat. Hist., Sept. 1S75), but the tube containing the specimens bore a record of their true habitat, and this statement was afterwards corrected (op. cit. March 1870). No males were jDroserved. Plate XII., figs. 11—19, C. brevicaudatm ? (enlarged) : 11, female (side view) ; 12, upper antenna} ; 13, mandible and palp ; 14, maxilla ; 15, upper foot-jaw ; 10, lower foot-jaw ; 17, a swimming-foot of the 1st pair ; 18, one of the 5th pair ; 19, abdomen. Calanus finmarchicus, Gunner. CetocUhis septentrionalis, Goodsir, Edinb. New Phil. Journ. 35, p. 339, PI. vi., figs. 1-11 ; Baird, Nat. Ilist. Brit. Entom., p. 235, T. xxix., figs. 1 a-(/. Mob. and Disf. — Two specimens were taken in the surface-net in lat. 35" 9' S., long. 45° 30' E. It is found also in the British, European, and Greenland seas. Sapphu'ina danae, Luhbock. Trans. Ent. Soc, London, N. S. 1850, vol. iv., part ii., p. 23, pi. xii., 9-11. Dimensions c? J^V "^ch long, j^^o i^cT^^ broad; ? tV(7 i^ich long, y^_ inch broad. Male. — Outline from above sub-elliptical, rather widest in front ; first cephalo- thoracic segment considerably broader than long, equal in length to the four following conjoined. Superior antennse short, stout, moderately setose, 5-jointed, with the second joint about as long as the four others united ; inferior antennae somewhat longer, 4-jointed, geniculated between the second and the third joints ; fii'st and second joints by far the longest, third very short, fourth long, terminated ENTOMOSTRACA.— G. S. BRADY. 217 by a small claw and a bristle about as long as the claw. Mandible (?) slender, with a single apical tooth and two stout plumose lateral filaments ; no palpi. Upper foot-jaw short and stout, 3-jointed, ending in a simple claw ; lower foot-jaw larger, stout, 3-jointed, bearing a large curved apical claAv which is fully as long as the thi'ee preceding joints taken together. Swimming- feet in fom* pairs, all nearly alike, 2-branched, the branches 3-jointed and subequal in length to each other ; marginal spines of the outer branch sharp and slender, the other setae of moderate length and very distinctly plumose. Caudal laminae ovate, about twice as long as wide, armed with 1 very short seta?, of which two are apical, one is on the middle of the outer margin, and the fourth is intermediate between tliis and the other two ; at the extremity of the inner margin is a minute spine. In the median line, behind the bases of the first pair of antennae, is a very distinct nervous mass in the form of a ganglionic ring elongated backwards, which emits numerous diverging lateral filaments, and separates posteriorly into two large nerve trunks. From the fi'ont of this ganglion are given off two short processes, supporting at their extremities two bulb-like " conspicilla " or lenses, slightly in advance of which are two nebulous spots which seem to be of the nature of " ocelli." Behind the bifm'cation of the cephalo-thoracic ganglion lies a glandular organ of considerable size, covering the upper portion of the alimentary canal ; and from it, or from its immediate neighbourhood, two ciu'ved tubes are prolonged backwards to the sixth body-segment, where they converge towards the intestinal tube and terminate in coecal expansions ; these are the " vasa deferentia." The alimentaiy canal is a straight funnel-shaped tube, which extends along the median line to the apex of the abdomen. Female. — Very different in shape from the male ; the abdomen being abruptly narrower than the cephalothorax. Cephalothorax of 4 segments, the fii-st of which is as long as the three remaining together. Abdomen 6-jointed (exclusive of caudal segment), only about half as wide as the cephalothorax, from which it is separated very distinctly. The fij'st segment is constricted anteriorly, and gives attachment to the last (fifth) pair of feet; these feet are rudimentary, and consist of a single joint armed with two terminal setae. The other abdominal segments are about twice as broad as long, and are subequal to one another. The caudal segments are rather wider in proportion than those of the male. Hab.— Taken by the surface-net in Lat. 35° 9' S., Long. 45° 30' E., March 1875 (Eaton) ; Lat. 27° 30' X., Long. 20° W. (Lubbock). Avery similar species, perhaps the same, was obtained by Mr. Eaton on the 25th of September 187 !• on his outward voyage, within 60 miles "W. of the Crozets ; but none of this earlier gathering could be preserved. E E 218 ZOOLOGY OF KERGUELEN ISLAND. Mr. Eaton says of this species : — " The larsre oval llattenod Entomostraea in the waicv relleeteil from their surface a bhio light which changed to opal in certain positions, and were conspicuous like phosphorescent spots from the deck of the vessel. The colour was entirely due to rellection, tlieir actual substance being colom'loss and diaphanous. They SAvam back downwards, and occurred mostly in narrow l)aii(ls ol' l)n)\vnish water, some of which from 50 to 200 yards in width extended for miles as far as we could see on both sides of us. The water derived its tint from jelly-like gi-anulated oval flattened disks easily injured by removal from the sea, which resem])led in appearance spawn of Mollusea; but having nothing better than a Coddington lens at hand to examine it with, I am unable to vouch that it was spawn. There were also various kinds of small jelly-fish in some of the bands,^ — here and there a miniature Velella of ultra-marine blue, — most of them the produce of Sertularians ; and the brilliant Entomostraea entangled in their trailing tentacles could be distinguished from among the others by the constancy Avith which their position relative to one another ■was maintained. As a rule these ccelenterata did not contribute largely to the formation of tlie bands, although they were abundant ; but now and then bands were crossed by us which consisted of little besides these Acalepha). When it is considered that the individual jelly-fish are colourless, with the exception of their proboscis (when there is one), eye-specks, and ovaries, — so transparent indeed that often eye-specks are the only parts of them visible in a white basin — the difficulty of conceiving the innumerable myriads of them in a strip of water miles in length coloiu-ed brown by them alone, may be imagined. The Scqjphit'itKS are less plentiful in these bands than they are in those which are due to the ' spawn.' The smaller Entomostraea were mostly of a beautiful ultramarine blue." The small form here referred to is, I presimie, the female, as the gathering sent to me contains no other species except Calanus finmarchicus, which, so far as I know, never exhibits any bright coloiu-ing. These Sappliirinse even after their preservation in spirits with glycerine present a very vivid opalescence. Plate XII., figs. 1-10, S. dance (enlarged) ; 1, female (from above) ; 2, male (from beneath) ; 2«, upjier antennce ; h, cerebral ganglion ; c, lenses ; d and d, nerve filaments ; e, testis ; f and y, vasa deferentia ; y, alimentary canal ; 3, upper antenna $ ; i, lower ditto ; 5, mandible (?) ; 6, uj)per foot-jaw ; 7, lower ditto ; 8, a swimming-foot of the fixst pair; 9, cerebral ganglion with lenses; 10, caudal lamina ?. ARACHNIDA. — O. P.-CAMB RIDGE. 219 Arachnida. By Bev. 0. B.-Camhridge, M.A., C.M.Z.S. (Plate XIII.) The few examples of Ai-aclmida found during the Transit of Venus Expedition to Kerguelen's Land, were all apparently new to science. One, indeed, at present seems to me incapable of inclusion in any previously recognised order of Arachnids. The whole collection consisted of but five species ; one of Araneidea, and three of Acaridea, the fiftli, being that upon which I have founded a new genus, family, and order. At fii'st sisrht this delicate little Ai-aclmid eave me the idea of beins: a Chelifcr O CO deprived of its forcipatcd palpi ; but a subsequent examination with a stronger lens showed me that it possessed palpi of an entii-ely different character from those of the pseudo-Scorpiones ; and a final scrutiny under a still liighcr power led to the detec- tion of the eyes : in the number and position of these there is displayed a remarkable similarity to the Solpugidca, while there are not wanting some general indications of afiinity to the Araneidea. Its small size and general appearance when alive woiUd probably induce one to refer it to the Acaridea ; but the structure of the mouth-parts, the distinct cephalothorax and abdomen, and especially the character of the eyes, seem to preclude this allocation. It is possible, however, that when the Acaridea shall be more thoroughly worked out by some future arachnologist, the present anomalous little creatui-e may become the type of a suborder, or perhaps of only a family of that order. Meanwhile in forming a distinct order for its reception, I desire to obtain the free criticism and opinion of arachnologists * more conversant than myself with some obscure groups of Acaridea, as to its true systematic position. The following descriptions are reprinted from the Proc. Zool. Soc. Feb. 1876, pp. 259-265 ; and the plate is almost the same as in op. cit. pi. xix., the difference consisting cliiefly in the addition of the figures of Tiillbergia, and the greater enlargement of some of the others. ACAEIDEA. Acaribes. Torynophora. Body oval ; a slightly indented transverse line towards the fore part on the upperside appears to mark the junction of the cephalothorax and abdomen. Mouth- * Dr. T. Tborell incliues to the opinion that this Ai'achnid is an Acarid. E E 2 220 ZOOLOliY OF KKKOUELKN ISLAND, parts almost soldcivcl iogotbor, loaving only the short palpi and the extremities of the falccs tracoal)lc. I/Cffs 8, slender, in 4 pairs (1-2 and 3-1 on each side), 5-jointcd, and terminating- with two soniewhat S-curvod claws springing from a small supormmierary or heol-join(. Ui/ca i'onv, in two pairs, one pair on either side of the capnt. F(ilccs armed on the underside Avith serrated opposed edges. Palpi short, strong, 4-jointed, w ilh a single strong cm'vcd jaw-like claw springing from its base on the up[)crside. Toryuopliora serrata. (Plate XIII., fig. 2.) Length about • 5 mm. This minute Acarid is of an oval form, tolerably convex above, and of a miiform pale lutcous colour. From the fore part of the cephalothorax four pointed processes project, each one terminating with a very small joint, from which sin-ings a curious clavate or spoon-shaped bristle or tag; a few with a somcAvhat similar tag arc dispersed thinly over the upper sm'facc of the body, Avhich is closely wrinkled, the ■nTinkles taking dilferent, but regular, directions on the dilTercnt parts of the body. The ei/es ai"e very minute, in two pairs, one on either side of the caput ; those of each pair are near together but not contiguous. The legs are 5-jointed, slender, and not very long ; they are armed witli line spines, bristles, and bail's, and terminate with tAVo tai'sal S-shaped claAvs, springing from a small terminal joint, and furnished beneath with some slender prominent clavate hairs. The legs are in pairs, the first and second, and third and fourth legs on each side haA'ing their basal joints respectively contiguous to each other, as in the genus Trojnbidium, and articulated to the fore half of the lower surface of the body. The palpi are short strong, 4-jointed, and to the upperside of the base of the digital joint is articulated a strong curved claw. The niaxillcc, labium, and falces coalesce and form a kind of suctorial apparatus, towards the fore part of which on the underside are two o^jposed curved saAV-edged processes. Several examples of this curious Acarid, found under stones, were contained in in the Hev. A. E. Eaton's Kerguclen's Land collection. Being so very minute and delicate, they had suffered considerably by being preserved in strong spirit. Fig. 2. «, upperside, highly magnified ; b, underside Avithout the legs ; c, profile ; d, leg of first pair ; e, extremity of tarsus of ditto ; f, palpi ; g, extremity of one of the cephalic projections, highly magnified ; Ji, one of the clavate hairs on abdomen, liighly magnified ; k, mouth-parts on underside, highly magnified ; o, natural length. ARACHNEDA.— 0. T.-CAMBRIDGE. 221 X7W ACAUUS. Acarus neglectus, sp. u. Adult female. Length one third of a line. The body is of an oval form, broadest towards the fore part, whence it narrows quickly to the extremity of the caput. The Cephalothovax is very short and coalesces with the abdomen, being scarcely traceable by a fine suture. TJie whole is tolerably convex above, and of a dull yellowish, whitey -brown colom-, fm'nished with a few fine bristly hau's on the upper side, and two longish divergent tapering plumose hau's projecting horizontally from the hinder part of the abdomen. Eijes, none. The^jfl^pi and falces are short, but of equal length. The former are 5-jointed, the basal joint being the longest, and the latter have a strong denticulate forcipated claw. The legs are short, articulated to the under side of the fore half of the body, and do not differ much in length, their relative length being apparently 4, 3, 2, 1, or 4, 3, 1, 2. They are strong, tapering, and 6-jointed, the basal joints being, ap- parently, soldered to the under sui'face of the body, and the terminal or tarsal joint is undivided ; theii" colour is a dull vel- lowish brown, and they are furnished with some strong spines beneath the two last joints, as well as with a few hairs of different lengths ; the longest of the latter being one or two near the extremity of each of the tarsal joints, which last terminate with a strongish sickle-shaped claw. The genital aperture is placed just behind the basal joints of the fourth pair of legs, and consists of an elongate brown slip on a circular pale yellowish-l)rown patch. The basal orifice is situated just beneath the liindcr extremity of the abdo- men, and has the appearance of a simj)le brown longitudinal line. The examples from which this description has been made were contained Fig. 1, from above ; 2, from beneath ; 3, falces and palpi beneath ; 4, left leg of the second pair. 222 ZOOLOGY OF KERGUELEN ISLAND. amonirst a mass of fibrous woh-likc substance rocoivod from ]\[r. Eaton Avitb the other Keri,nielcu Ishiiul Arachnids, but were overlooked at the time when those were described and figured in the Zoological Society's Pi'oceedings for Pebruary 1S7G, pp. 259-2G5. Bdellides. Scirus. C. Koch f Scirus pallidiis. (Plate XIII., lig. 3.) Leugtli about 1 mm. As far as I could ascertain trom the single example contained in the Kerguelcn Island collection, this small Acarid is an undescribcd species of the genus Scirus. Its colour is a dull yellowish white, and there are a few obscure blackish mai'kings in two parallel longitudinal lines along the upperside of the abdomen. The body and legs are fiu-nished with a few longish pale semidiaphanous hairs. The ci/cs are in two pairs, those of each pair contiguous, and in the position indicated by the two small oval markings in figure 2 b. The only example received was injured by the action of the spirit in which it had been preserved, so that the exact details of its structure could not be satisfactorily observed ; in the general appearance, however, of the beak-like mouth-parts there seemed to be but little difference from the genus Bdella and others nearly allied. Hab. — Under stones near the lakes at Observatory Bay, in rather dry places. Eig. 3. a, upperside ; 6, upperside of caput and mouth-parts, more enlarged, showing the position of the eyes ; c, natural length. IXODIDES. Hyalomma, C. Koch. Hyalomma puta. (Plate XIII., fig. 4.) Length about 1*25 mm. Body oval. Cephalothorax yellowish brown, strongly tinged with red on either side of its fore part and on the fore part of the caput. Abdomen dark yellow-brown, and (as well as the cephalothorax) thinly clothed with short pale haii's ; the hinder part of the abdomen is of a pale dull yellowish hue, and its margin is indented Avith four small elongate notches. On each side of the underpart, jvist behind the basal joints of the 4th pair of legs, is a roundish patch, whose svu-f ace appears to be covered with minute points, which may possibly be the plates of spiraciolar organs. The legs are moderately long and tolerably strong, 7-jointed, the last or tarsal joint being veiy small ; they are of a pale yellowish colour, marked underneath with patches of a bright shining orange red, and furnished with a few short haii-s ; each ARACHNroA.— O. P.-CAMBRTDGE. 223 tarsus terminates with two curved cliaplianous claws springing from a small claw- joint ; and beneath them is an oval sucker-like pad. The palpi are five-jointed (including the hasal joints or maxillae) ; these latter are of a reddish colour and soldered to the labium ; the colour of the palpi is similar to that of the legs ; the terminal (or digital) joint is short and small ; the next to it (or radial) is large and tumid, the other two joints short. The length of the palpi slightly exceeds that of the falces. The falces are porrected in the form of a beak, and are armed beneath with numerous sharp hooks or teeth directed backwards. Several examples of this Acarid were found by the Rev. A. E. Eaton on a Penguin {Pygosceles tceniattts) in Kerguelen's Land ; and he also found some on a reef of rocks between tide-marks, lurking in crevices. Fig. k a, upperside ; b, underside without legs ; c, caput, palpi, and falces, more enlarged ; d, natural length. Poecilophysidea. JExternal characters of the Order: — Cephalothorax and abdomen covered with a continuous epidermis of a rather slight textvu-e, unsegmented, and united to each other throughout their whole breadth, the point of junction being clearly indicated by a traverse line or sutm'c. Falpi filiform, and terminating with a single minute claw. Legs eight in number, their basal joints closely grouped together on the sternal siu'face of the cephalothorax, the tarsi terminate with two claws, between which is a slender pectinated style. Falces didactylous. Maxillce coalescing at their base. Labium (properly so called) wanting. Eyes two. POECILOPHYSIDES. In its general appearance this curious little Arachnid seems to be a compound of the Spiders, Solpugids, Chelifers, and Acari. On its npperside it reminds one strongly of the Solpugidea, both in the massive falces, and its two eyes on a small tubercle at the fore extremity of the caput ; its underside bears a strong resemblance in the maxillse and palpi to the Araneidea ; its profile resembles that of the pseudo- Scorpiones ; Avhile in its small size, continuous, delicate epidermis, and closely approximated thorax and abdomen, it shows a strong likeness to the Acaridea. P(ECILOPHYSIS. Eyes two, closely grouped on a small tubercle at the fore extremity of the caput, just between and above the base of the falces. Falces massive, as long as the cepha- lothorax, two-clawed, the upper claw fixed, the lower movable, and both curved. Maxilla; large, coalescing at their base, and produced at their inner extremity into a strong angularly pointed projection. Labium none, and sternum none, properly so called, the basal joints of the legs being articulated to the inferior surface of the 224) ZOOLOGY OF KERGUELEN ISLAND. cophalothovax. AlnJomoi longer than the coplialothorax. A small elongate oval aperture towards tlie hinder jiart of the underside is probably the genital opening, while a still more minute orifiec beneath its extremity is probably the anal aperture. Legs moderately long and tolerably strong ; they arc G-jointed, furnished with long bristles, and terminating in two S-curved claws, beneath which is a longish, slender, slightly uptiu'ued style, plumose or finely pectinated along its underside. Ffilpi l-jointed, similar to the legs in armature; the terminal (or digital) joint ends Avith a small hooked claw, and the bristles or hairs on it are long and plumose. Poecilophysis kergueleueusis. (Plate XIII., fig. 5.) Adult female. Length about -75 mm. The cephaJothorax is of a somewhat quadrate form, narrower before than at its jimction with the abdomen ; it is moderately convex above, and has a few long pale hairs or slender bristles directed forwards on its upperside ; its profile line is nearly level ; and the colour of the ccphalothorax and abdomen is pale yellow, the legs and other parts being of a whitish hue. The eyes are small, but close together near the hinder part of a small roundish tubercle or eminence, at the middle of the fore extremity of the caput. The legs are 6-jointed, rather long, tolerably strong, not greatly differing in length, their relative length being 1, 4<, 2, 3 ; they are furnished with long pale bristles ; and the tarsi, which are undivided and with two claws, arc curved some- what in the form of an S ; beneath them is a largish bristle or style, pectinated or plumose on its inferior side. The joints do not difi'er greatly in. length, the first two or basal ones being the longest, and the rest nearly equal. The palpi arc similar, in their general armature, to the legs. The digital joint is longer than the radial, and of an ovoid form ; its hairs are plumose, and the single terminal claw is sharply hooked and minute, The f dices are as long as the cephalothorax, very massive at the base and didactyle, the lower claw being movable and opposed to the upper one ; both claws are cm-ved, but project in the same straight line and in the same plane as the cephalothorax, which the falces equal in length. The maxillcB are long, their inner extremities considerably produced into an angularly pointed form, and extending close beneath the falces, to about two-thirds of their length. The abdomen, looked at in profile, is higher and more convex than the cephalo- thorax, and about double its length ; its fore extremity is conterminous in its breadth with the cephalothorax, but is constricted laterally near the middle, the hinder extremity being rounded and obtuse ; its upper surface is furnished with a few long pale hairs or bristles. Several examples of this minute but most interesting Arachnid were found under stones in Kergnelen's Land by Mr. Eaton. Unfortunately, from their extreme ARACirNTDA. — O. P.-CAMBEIDGE. 225 delicacy and small size, they bad suffered severely from the action of the strong spirit in which they were immersed. Fig. 4. a, npperside ; h, profile without legs or palpi ; c, extremity of tarsus of leg of second pair ; d, digital joint of palpus ; c, portion of falces and maxilla? in profile ; f, natural length. ARANEIDEA. AGELENIBES. Myro. Ceph(dothorax oval, roundly truncated before, and moderately constricted on its lateral margins at the caput. Upper convexity moderate ; profile-line slopes very gradually in a slightly cm-ved line from the hinder slope to the ocular region ; 'chylous imusually narrow, almost obsolete. Spinners short, those of the inferior j)air longest and strongest. Eyes eight, unequal in size and forming a rather large and somewhat oval area, enclosed by two longitudinal curved rows of three eyes each ; the ciu'ves du'ccted from each other ; within this area, and towards its fore part, are two minute eyes near together in a transverse line. Leys not greatly different m relative length, which is 4, 1, 2, 3. Each tarsus termiaiates Avith three claws. MaxillcB large, curved towards the labium, much and roundly protuberant on the outer sides towards their extremity, which is rather obliquely truncated ; the palpi issue from unusually near their lower extremities. Labium rather more than half the length of the maxillae, very difiicult to be seen clearly ; but its form is apparently oblong, slightly rounded at the apex. Myro kerguelenensis. (Plate XIII., fig. 6.) Adult male. Length nearly 4" 7 mm. The eephalotJiorax is of a yellow brown coloiu-, the margins surrounded with a fine black line ; the normal grooves and indentations are well marked, and suffused with dusky black, giving the thorax somewhat the appearance of radiating markings ; the ocular region is furnished with some bristly black hairs ; and some longer and finer ones are distributed along the central longitudinal line to the hinder slope. The eyes are unequal in size, and form a largish hexagonal area on the fore part of the caput close to its fore margin ; they may be described either as in two longitudinal cm-ved rows of three eyes each, with two minute ones in a transverse line towards the fore extremity of the enclosed area, or as four pairs, a hinder pair, two fore lateral pairs, and a fore central pair ; those of the hinder pair are separated by an interval of an eye's diameter, and each is also divided by an equal interval from the hinder eye of the lateral pair- on its side ; the eyes of each lateral pair are separated by a slight interval, not exceeding half the diameter of one of the fore F p 226 ZUOLOCY (M- KKKliLKLKN ISLAM). central eyes ; tho hind laterals arc tlu> larii:i'.st of the (M'i;lit ; and the interval between the Tore laterals is equal to ahout IJ, (lianicter ; tlie interval l)(>t\V(MMi the eyes of the lore central pair is equal to a diameter, ihe disiaiKH" ul' each Irdui the lore lateral on its side heinu; rather i^reator ; and that of each from llic liind lateral is equal to the diameter ol' the lallci-; llic iiitci'\al hetwiMMi the I'dre lateral eyes and the Tore man,'in of the caput is M-ry sliiilit indeed, lieiiii;' less lliaii the diameter of one of the central eves. The A'/7.v are niiidcratfly loiii;' and strong-; tlieii' relative leni;th does not differ greatly, being- \. I. 2. .">. They are of a brownish yellow colour, faintly and imper- fectly annulated with dusky brown ; the amiulations are scarcely perceptible in some examples. They are furnished with spines, bristles, and hairs, some of the latter being- at right angles to the dill'ercnt joints; each tarsus terminates with three ciu'ved claws, of which the inferior is much the smallest, and sharply bent downwards. The palpi are similar in cdloiu- to the legs; furnished with hairs and a few bristles, and of moderate length and strength. Tlic cubital joint is short and bent; the radial is slightly longer and spreads out gradually on its outer side into a some- what irregularly shaped, prominent, l)ut not very large apophysis at its fore extremity ; the digital joint is long and rather narrow, similar in form to that of some species of Te\v. Eiiluu's tHiUeflioii rcprrscnl 7 species, ul' w liicli only one was previously known. Of these G l)eh)ni^- to the Rhynchophora and 1 to llie Bruchelytra. The M/iynchopIiora are very peculiar, ami possess an ass(Mnblage of characters not to he met with in any other species of this tribe. Cons(>quently their systematic position is very douhtl'ul. The s])ecies oriijjinally deseri1)ed und(M' tlic name Eclem- uorhiiiKS viridis (G. R. Waterh.) as an ally oi" Fhijllohiiin is placed n(>ar Jiliinomacer by Lacordaire in his A\()rk on the Genera of the Coleoptera, .'ind on iirst inspection was referred by me to the Bvaclunleridime. I am still of o])inion that it accords better with either P//y//oii/w or Bvachydcfes than it does whh lihiuomacer ; but thou2:h I liave examined it and the other allied species very closely, T a)n unable even now to assign to them a deiinite place in the series of genera. The Brachclytron is a Pliylosuti, which dill'ers only as a species from ijie norinal British representatives of the same genus. .Vll of Mr. Eaton's species are apterous ; but Dr. Kidder, of the American Titmsit of Venus Expedition, states that a large brilliantly coloured specimen of a beetle flew to a light in his tent one night at Molloy Point. (Kidder, Bulletin U. S. Nat. Mus., 1876, iii. I.9.) BRACHELYTKA. ALEOCHARID.i;. Phytosus atriceps. (Plate XIV., fig. 15.) C. Waterhouse, Ent. Mo. Mag., 1875, xii. p. 55. Rufo-testaceus, breviter pubescens ; capite abdominisque in quatuor segmentis basalibus nigrescentibus. Long. 32 mm. General form that of P. nigriventris, but a little broader. Reddish-yellow, scarcely shining. Antennae scarcely longer than the head and thorax together, slightly thickened towards the apex, the three basal joints elongate, the first a little longer than the second, the third shorter than the preceding, elongate-obconic, the fourth subquadrate, the following joints blackish, the fifth to tenth short, the four penulti- mate transverse, the last ovate. Head rounded, thickly and finely punctulate. COLEOPTEKA. — ('. (). WATERIIOUSE. 231 Thorax a little broader Ukui the head, scarcely broader than long-, depressed, very finely coriaceous, slightly narrowed towards the base, all the angles roiuided. Elytra scarcely narrower than the tliorax, narrowed towards their base, one-third broader than long, very finely coriaceous, shoulders oblique. Abdomen less dull, blackish, the apex reddish, the sides nearly parallel. The claws in this species are distinctly longer and more slender than in Ph. ni- griveiifris, but the tarsi present no other ditt'erence. Hab. — Under stones near high-water mai-k, Observatory Bay. Only two speci- mens seen, one of wliieli lias been lost. January. Sir C. "V^'yville Thomson in " Goofl "Words " for November 187i, p. 750, enume- rates 3 species of the CurculionidcB and 1 " small Staphylimis " (? Phytostis) fomidon the island by the Challenger Expedition. PHILHYDRIDA. [Ochthebius, sp. P Dr. Kidder, in Bulletin U. S. A'at. Mus., 1876, iii. 49, records the finding of a very few specimens of a small black Ochthebius (sp. undescribed) on rocks near the sea and about the roots of wet tufts of grass. It was not observed by Mr. Eaton.] KHYNCHOPHORA. ECTEMXOR ItlNID-^. Canonopsis. C. Waterh., Ent. Mo. Mag., 1875, xii. 55. Bostrum short, thick, truncate at the apex ; maxilhe visible ; antennal scrobes small, deep, and somewhat reniform, open in front. Antenna; placed very near the apex of the rostrum, long, but not very slender, scape just reaching the front margin of the thorax, slightly enlarged at the apex ; funiculus with the 1st and 2ud joints elongate, the 1st rather longer than the 2nd, 3rd joint short (about two-thirds the length of the 2nd), the 4th, 5th, and 6tli joints globular, the 7th a little broader than the 6th, the club concrshaped. Eyes round, moderately prominent. Thorax aslon^ as broad, scentlv constricted in front and behind the middle, truncate in front and behind. Scutellum very small and triangular. Elytra united at the suture, not broader than the thorax at their base, convex, gradually enlarging to the middle, and then again narrowed to the apex. Wings absent. Legs moderately long, femora strongly clavate ; anterior tibiae nearly straight ; apex of the posterior tibia? truncate, hollowed, the margins ciliated. Tarsi spongy below, the 1st joint as long as the two following together, the 4th joint a little longer ; the claws curved, separated, ^tletasternum very short. Intercoxal projection of the abdomen wide, nearly straight in front and at the sides ; 3rd and 4th segments equally short, much shorter than the 2nd. Pvgidium slightlv visible from above. 232 ZOOLOGY OF KEKGUELEN ISLAND. Canonopsis sericeus. (Plate XIV., lig. 9.) C. Watorh., op. cit., p. 55. Subi'usil'ormis, convexus, ttavo-griseo-scricous ; antennis piceis, rlava nigra : I'rontc fovea magutl impressii, ; thoracc sub lequo latitudiiie ac longitucliue, supra in longitudiuem canaliculato, canalicnla ante medium cxpansi\ et tuberculo obsoleto utriuque instructs ; elytris elongato-ovalibus, apud bases thorace baud latioribus attameu 3-^ longioribus, punctato-striatis, instcrstitiis vix convexis, quorum sccundo vittis brevibus velutinis nigris duabus ornato ; femoribus apud bases piceis. Long. 12; elytr. lat. 1 uini. Hab. — Cat Island, Tlirec Island Uarbour, Royal Sound. Common under stones, 11th October. Ectemnorhiuus. Ectemnorhbiits, G. K. Watcrh., Trans. Ent. Soc. London, series 2, ii. p. 184. Agonchjtro, C. 0. Waterb., Ent. Mo. Mag., 1875, xii. 55. M. Lacordaire (Gen. des Coleopt., vi., p. 562) says of this genus, " Corj)s oblong, aile." The type specimen -o-hich Lacordaire had for examination is now in the British Museum collection, and is undoubtedly without wings. Ectemnorhinus viridis. (Plate XIV., fig. 10.) Ectemnorhimis viridis, G. R. Waterh., loo. cit. Agonehjtra longipennis, C. Waterh., Ent. Mo. Mag., 1875, xii. p. 56. Elongatus, nigro-piceus, cinereo-sericeus ; capite anticc angustato, fronte foveola leviter impressii, rostro supra bicarinato, antennis piceis, clava, nigrescent! ; thoracc leviter conrexo, sub aeque longitudine ac latitudine, ante et postca paulo angustato, lateribus in medio leviter rotundatis, in dorso carina longitudinali nitida ; elytris apud bases thorace 1 latioribus, et fere quadruple longioribus, paulo in medio ampUatis, leviter punctato-striatis, interstitiis planis ad apices rotujidatis hume- ris obtusis; corpore subter pedibusque piceis nitidis, femoribus supra tarsisque nigrescentibus. Long. 7 ; lat. 2*5 mm. Antennae moderately long and stout, 1st and 2nd joints of the funiculus slightly elongate, subequal, the 3rd joint shorter, the 4th to 7th becoming gradually shorter and slightly transverse. The silky pubescence which covers the head, thorax, and elytra is fine, and not very close on the former. The elytra strongly embrace the abdomen, and are broadest about the middle, truncate at the l)ase ; the shoulders angular, but obtuse. The tibise are very sHghtly flexuous. Var. — Silky pubescence green, appearing golden in some lights. Hab. — Common in the neighbourhood of Royal Sound. COLEOPTERA.— C. O. WATERFIOUSE. 233 Ectemnorhinus angusticollis. (Plate XIV., fig. 11.) Agonehjtra angusticollls, C. Waterh., op. cit., p. 56. Ec. viridi affinis, distiuguenclus tamen a tertiis ad septimos articulos fuui- culonim antcnnarum transversis, a thorace supra baud carinato, elytrisque postice latioribus. Long. 7 mm. This species is very closely allied to the preceding ; hut the anteuntc are shorter, owing to the 3rd to 7th joints of the funiculus being transverse, the 6th and 7th very strongly so. There is no distinct carina on the thorax, and the elytra are broadest behind the middle, very broadly rounded at their apices. The silky pube- scence is yellowish iu the female, with indications of two paler, broad, abbreviated bands across the suture, one next the scutellum, the second about the middle. The male is narrower, darker in colour, and has the bands narrower, shorter, and less distinct. Hab. — Near Royal Sound. Ectemnorhinus gracilipes. (Plate XIV., fig. 12.) Agonelytra yracilipes, C. Waterh., op. cit., p. .56. Niger, griseo-pubescens ; capite supra fere piano ; thorace capite latiori, in medio parum ampliato, paulo in latitudinem minus quam longitudinem ; elytris ad bases thorace f latioribus, gradatim usque ad medium ampliatis, postice angustatis, convexis, supra depressiusculis, leviter punctato-striatis, interstitiis planiusculis ; antemiis piceis, clava nigrescenti ; pedibus gracilibus, longis, tarsis articulo tertio bene dilatato. Long. 3-5— 4-7 ; lat. 1-25— 1-8 mm. The elytra are gently roimded at the base; the shoulders are distinct, but very blunt. Hab. — Near Royal Soimd. Ectemnorhinus brevis. (Plate XIV., fig. 13.) Agonelytra brevis, C. Waterh., op. cit., p. 57. Niger, senescens, viridi-grisco parce pubcsccus ; rostro in longitudinem supra leviter impresso ; thorace in longitudinem baud minus quam latitudinem, convexo, ante et postea augustato, lateribus bene rotundatis ; elytris apud bases thorace f latioribus, 2^ eodem longioribus, medium tenus gradatim ampliatis, apicem versus angustatis, convexis, sat fortiter punctato-striatis, interstitiis planiusculis ; antennis G G 23-4 ZOOLOGY OF KKRGUELEN ISLAND. piccis, clav;\ inatrnA niirroscniti ; ]i(MliInis nii^TOseontibus, fomovilnis nd bases libiis- qur intus piccis. Long. 3"2 — I'T; lit. 1'7 — '2 mm. The pubescence is short and tliick (mor(? like narrow scales), and is so arranged as to give the elytra a slightly mottled ai)pearance, a spot on each side of the sntiire being particularly distinguishable. 11 ah. — Near Eoyal Sound. Ectemiiorhinus eatoiii. (Plate XIV., fig. 11.) C. Watcrh., Ent. Mo. Mag., 187G, xiii. p. 51. Pp'iformis, convexiusculus, nigcr, i)arcc ct brevissime A-iridi-griseo-pubcscens; capite rostroque in longitudincMU fortiter impressis, h6c brcvi, baud apiccm versus angustato; antennis nigris ; thoraec paulo capite latiori, vix in lougitudinem juinus quam latitudinem, antice posticequc angustato, subtilissirae coriaceo, subopaco, in medio dorso longitudinaliter carinato, post utrinquc obsolete tuberoso ; elytris ad bases thorace paulo latioribus, post bene ampliatis, supra depressiusculis, distincte striatis, striis lateralibus obsolete punctatis, interstitiis plauis, transversim subtilis- simc strigosis ; pedibus longis, I'emoribus ad bases piceis, et in medio iuflatis, tibiis intermediis paulo cm-vatis. Long. 5-5 ; lat. 2-05 mm. This species is most nearly allied to U. hrecis, but is much less short, and has the shoulders of the elytra effaced. The thorax is distinctly carinate above ; the forehead and rostrum are deeply and broadly impressed, the antenna) are rather slender ; the apex of each elytron is broadly and bluntly rounded, and rather expanded on the outer side ; the femora are somewhat strongly inflated in the middle, the tarsi arc rather narroAver than in the other species of the genus, and the claw joint is very long. A single example ; jirobably a female. Hab. — Near Royal Sound. LEPIDOPTERA.— A. E. EATON. 235 Lepedopteba. — By the Bev. A. JS. Eaton. Larvae of at least two, perhaps of three, species of Lepidojitera were obtained in the vicinage of Observatory Bay. One sort was occasionally brought off to the ships by the men in gatherings of Pringlea in. October and the early part of Kovcmber. It would probably develop into a moth about as large as an Agrotis of medium size, and was conjccluied to belong to the Nocttiina (Etn., Proc. Hoy. Soc. 1875, xxiii. 354). Its affinities, however, may be very different, as it is likely to have been the larva of the insects referred to by Dr. Kidder as " lepidopterous •' insects of moderate size, with very imperfect and abbreviated wings, active in " their movements, and . . . mth . . . antennie . . . long and thread-Hke "... [and labial palpi] "pectinate, and curling backAvard over the top of the head." These were obtained by him " on the evening of December 18 ... . from the " roots of grass" (Kidder, Bulletin U. S. Nat. Mus. 1876, iii., 50). If the larvse above mentioned and these moths are the same insect, the condition of the palpi of the imago seems to indicate that it is related to the Gelechiidce. And this sup- position is quite consistent with the form of some other larva? (believed to be younger examples of the species obtained in October) which were foimd commonly in moss in December and February near Observatory Bay. They could be identified on the island, in situ, by the folloAving brief description. Larva GelechiidiCorm with 16 legs, grey, paler beneath and along the spiracles ; head and dorsum of first segment pale corneous, the usual raised dots of the other segments dark grey, shining, each bearing,— some a short, others a long, — testaceous hair. Length 13 mm. and upwards. Common in wet places on the hill sides near the sea, making galleries in moss. All of these larvae died before our arrival at the Cajie. (Etn., Proc. Eoy. Soc. 1875, xxiii. 501). In Sir J. Hooker's MS. Journal mention is made of a moth with rudimentary wings clothed with mouse-colom'ed hair found at Christmas Harbour, which may have been Dr. Kidder's insect. The other larvae observed belonged to the moth described below. Dr. Kidder reports •• a single flying tineid moth was obseiTcd soon after our " landing, but supposed to be a clothes-moth from our own boxes." This could not have been the Emhryonopsis because that is unable to fly ; and as no other instance is on record of a moth being seen on the wing in Kerguelen Island, there is good reason for concuriing in the opinion that it was an introduced species. Endrosis fenestrella had been transported to Cape Town in some boxes belonging to the English expedition a few weeks earlier. Embryonopsis, Eaton, 1875. Imago. Labial palpi long, recurved, smooth; the first joint very short, the second of moderate length, the third longer than the other two together, tapering G G 2 230 ZOOLOGY OK KKRGUELEN ISLAND. to an acute point. Antennae simple, iiliform, tlir hasal joint slightly larsjer than any of the others ; in the ^ as long as the ahdomen, in the ? rather shorter. Anterior -wings snbcorneoiis, aeumiuately ovate, convex, extending almost to^thc apex of the abdomen in the S, and only as far as its middle in the $ ; longi- tudinal ncrvures 5, simple, the suhcosta extending nearly to the middle of the costa, the next two nervm'es subequal and just falling short of the apex of the ■ning, the fifth terminating about as far from the fourth, as the fourth from the third ; no transverse veinlets ; the posterior wings extremely minute, not reaching even to the base of the abdomen ; posterior femora thickened ; genital appendices of c? abnormal, very broad, comprising two lateral pairs arising from the pleuro-ventral region ; ? the ovipostor 2-jointed, extensile. Larva Gelechiidiform, with 16 legs. The smoothness of the head, and the conformation of the tri-articulate labial palpi, together with the form of the larva, attest the relationship of Embryonopsis to the Geleclnkhc. The exceptional shortness of the wings, ami tlie presence in the male of an outer pair of lateral appendages eusheathiug an inner paii', and also the broadness of these last, readily distinguish it from other genera in that family. Embryonopsis halticella. (PI. XIV., fig. 8.) Eaton, Ent. Mo. Mag. 1875, xii. 61. See also E. Doubleday, Ann. Ent. Soc. Fr. 1818, Bull., p. Ixiii ; Thomson, Good Words, 1874, November, p. 750 ; Eaton, Proc. Roy. Soc. 1875, xxiii. 354 ; Moseley, Journ. Lin. Soc. Botany, 1876, xv. 54. Imago c? ? , sooty black varied with ochraccous ; antenn?e black, theii" basal joint, the vertex, and the palpi, sprinkled with ochraccous ; forewings with a wide longitudinal ochraccous streak through the middle; hind wings pale; legs somewhat ochraceous ; abdomen ochraceous at the sides ; in the ^ the last dorsal segment is triangular and is bifid at the apex, the last ventral segment is semi- elliptical ; the outer pair of lateral appendages parabolic, papyraceous, about half as long as the inner, externally clothed with scales, internally glabrous and smootli ; inner pair of lateral aj)pendages corneous, very closely invested by the outer pair, and very broad, subeirciilar, externally flatly convex, lutescent, polished and glabrous, traversed with a few fine raised lines, and slightly incrassated towards the margin, furnished at the extreme base with a series of very long appressed hairs which spread fanwise over them ; internally they are very abundantly furnished with an ochi-aceous tomentum disposed in slender toothlike bimdles ; lower penis cover (?) linear lanceolate, longitudinally concave above, narrowed suddenly to a slender point whose extreme apex is slightly tm-ned upwards, piceous ; penis sheaths strong, slender, tapering, finger-like, slightly arcuate, connivent towards theij tips, and strongly bearded beneath with rigid obliquely set testaceous bristles ; upper penis cover scaphoid, carinate, obtuse, testaceous. Long. corp. 5 — 5'5 mm. Larva. — Dull pale yellowish, with the paler dorsal line bordered by a pale brownish lilac stripe on each side extending from the fii'st to the penultimate LEPIDOPTERA.— A. E. EATON. 237 segment. Head dark piceous; first segment with a transvei'se dorsal fuscous blotch which posteriorly is incised by the dorsal line ; spiracles, tlic usual dots and their hairs, two dorsal spots on the penultimate segment, the dorsum of the last segment and a spot at the base of each of its prologs on the outer side, black. Frons pale green. Long. corp. " 5 lines," probably a note-book error for 5 mm. Hab. — Coramou in the vicinage of Royal Sound. The larva was fovmd on the 17th of November (the first day the imago was noticed), feeding upon and residing within the sheathing leaves of the yoimg shoots of Festuca cookii. Several speci- mens were obtained in patches of that grass close to the shore in Observatory Bay. It was afterwards detected in other places, and in shoots of F. erecta. The moth also occiu's on the slopes of the inland hUls. It is unusual for larvae belonging to this family to burrow witliin grass shoots. The homologies attributed to the elements of the genital armature above, are different from those which were given in the original description. Subsequent examination of the genitalia in recent representatives of most of the principal groups of Lepidoptera has led to a modification of the views entertained in 1875 concerning the nature of some of the parts. What was then taken to be the termination of the anus is now believed to be in no way connected with that outlet, and is considered to be a portion of the accessories of the genital apparatus ; and similarly that which was specified as the penis in the earlier diagnosis, is now described as the lower penis cover. Many persons would doubtless be disposed to qiiestion the validity of this last correction, and maintain that the earlier view. was the right one ; but so far as has been ascertained the penis in the Lepidoptera is in most, if not in all cases, an erectile, membraneous introvertible tube, which is invisible in dried examples. This tube, which can be protruded by means of pressure in the living insect or one freshly kiUed, is closed at the apex by a sphincter, which is relaxed spasmodically at intervals to discharge a coloured fluid supposed to be meconium, until a doubt led to its being examined imder a high power, when it was discovered to be in all cases crowded with spermatozoa. The lower penis cover projects immediately below this membraneous tube, and in the present insect is longitudinally concave above, as if with a view to its accommoda- tion ; and there is some reason for suspecting that it enters the vagina simulta- neously with the penis. The upper penis cover is used in the introversion of the penis after it has been distended, thrusting it inwards as it were with the pecks of a bird's beak by its up and down motion. Plate XIV., fig. 8, Emhryonopsis halticella (from above), a, head (from the side) ; b, anterior wing ; c, appendages of S (from the side) ; & and c^* inner lateral appendages (from without and fi'om within) ; c' genitalia , antennae, from within and from without ; b, proboscis ; x, mandible ; c, thorax and 1st segment of the abdomen, from above ; d, wing and halter ; e, appendices S . DIPTERA.— G. H. VERRALL. 241 EPHYDRINiE. Amalopteryx, Eaton, 1875. Head transverse, subquadrate, frons inclined, rather tumescent about the ocelli, bristles long, 2 ascending separately near the inner orbit, 2 subcrect slightly diver- gent near the posterior orbit, and 2 prorect ascending and divergent by the foremost ocellus ; epistoma very steep, below very prominent, widely convex and strongly bristled, above keeled in the middle with a concavity on each side ; gence large, bristled below ; back of the head wide, descending below the eyes ; antennae short, second joint with spinulose bristles toAvards the apex ; third joint cu'cular com- pressed; arista very shortly pubescent ; proboscis stout, the second joint of the stipes somewhat callous and pubescent beneath ; palpi very short, subcylindrical ; lamella? deficient altogether. Thorax convex, well bristled; scutellum semi- elliptical, the four bristles long ; wings almost as long as the abdomen, linear, with five simple longitudinal nervures, the subcosta becomes confluent with the costa at the transverse fold, the radial and cubital are united to the discoidal near its base, the radial also joins the costa just before the apex of the wing, the cubital joins the apical margin immediately beliind the apex and is thickened towards its termination, the discoidal is confluent with the margin near the middle and thickens it near their junction, it is united to the apex of the thickened anal vein by the second transverse vein ; halteres of moderate size, not very long, clavate ; legs of moderate length, strong, without spines, the posterior femur somewhat thickened. Abdomen 5-jointed, ovate oblong; the apex in c? slightly obtuse, genitalia concealed, in ? produced into a short stout tooth-like ovipositor directed downwards and forwards. This genus resembles Secamede in the depth of the back of the head, its round naked eyes, the carination of the upper part of the epistoma, the five-jointed abdomen, and the spinose bristles of the second joint of the antenna. Amalopteryx maritima. (Plate XIV., fig. 2.) Eaton, Ent. Mo. Mag. 1875, xii. 58. See also Thomson, Good Words, 1874 (Nov.), p. 750 ; Eaton, Proc. Roy. Soc. 1875, xxiii. 355 ; Kidder, Bulletin U. S. Nat. Mus. 1876, iii. 52. Fuligineous, with legs and setse deep black ; body completely invested with an extremely minute very closely appressed fuligineous pubescence, and with numerous short fine appressed black haii-s, which are very sparse on the frons ; epistoma somewhat caerulescenti-griseus ; antennae dull black ; wings pale nigrescent with piceous nervures, slightly spinulose along the costal and apical margins ; pulvilli whitish. Long. Corp. lon^•us fulvus fusco-pictus ; c'lyi)oo oxcisuris latoralibus dolinito ; antonnis i;:racilil)us ; piothoraoo trapczoidali, motatlioraco loiigiore ; abiloiuine angustc oblongo, margiaibus profimcle creuatis, nigro fuscis, iemiuai Faseiis fuscis. Long. Corp. 2-5 mm., capit. 0-2 mail., thorac. 0"35 mm., abd. 1"75 mm. ? Head elongate, narrowed gradually in advance of the antenna; to about | of its breadth, ending -with the very convex fore margin of the clypeus, which itself is sharply limited on each side by an acutely defined marginal notch ; in this notch stand 2 bristles of equal length, and lieneath in front oT the mandibles is 1 on each side, while posteriorly at the winged margin are 3 — 5 bristles half as long. The pai-allel temporal margins, wanting seta\ posteriorly turn inwards at right angles at the slightly concave occipital margin, and each cm-ve has one long marginal bristle. Antennte inserted after the middle of the liead, each in a deep sinus whose anterior angle is acute ; their first joint equal in length to the second, the third and fourth considerably shorter, the fiftli longer than the penultimate and terminated with tufts of bristles. Prothorax somewhat transverse, and like the metathorax trapezoidal with the sides almost straight and without bristles, only the metathorax has foui- imequal bristles at the angles. Anterior legs as usual the shortest and stoutest, the other two pairs more slender, the hindermost reaching backwards as far as the middle of the abdomen : coxa? long ; trochanters sharply defined ; femora slender, longer than the tibia?, which like them have but few hairs ; ungues slender, strongly curved. Abdomen long and narrow, very slightly widened in the middle : segments somewhat transverse, all of equal length from the fii-st, their apical margin slightly convex, their posterior lateral angles obtusely rounded but pro- jecting so as to give a crenate outline to the side ; the said angles of the first segment with one seta, the following with three very long unequal seta?, the last two segments with short and minute bristles : above and beneath, the surface has only very scattered bristles. ? Head pale yellow, laterally bordered with brown ; the clypeus projecting in front, clear and transparent ; ocelli black ; antennge without markings. Thorax edged with darker at the sides ; legs without markings. Abdomen pale brown bor- dered with black brown, the edging becoming paler at the last two segments. (J Antennse longer than in the female. Abdomen whitish, with blackish brown marginal spots and pale stigmata. Hab. — ? Eive examples were eaptiu-ed in the feathers of the neck and breast of Procellaria nereis on the 6th Nov. 1871 at Observatory Bay. Another example was previously taken there on Prion desolatus, 14th Oct. ^ One example from the neck feathers of Diomedea exulans, in March 1875. As only one specimen lies before me, I am unable to decide whether the differences above described which it presents are indicative of specific distinction from the females with which I have associated it, or are merely sexual peculiarities. I'lIILOPTliRA.— C. GIEBEL. 257 L. clypeatus, first described from several examples from HalohcBua cccmlea in the collection of the Mviseum of Ilalle, cannot be confounded with any other known species occurring on the Storm Petrels, owing to the peculiar form of the elypeus, the short antennae, the trapezoidal segments of the thorax, and the abdominal markings. Lipeurus sp ? A white skin of a Lipennis 2 mm. long, from the head of Diomcdea fuliginosa, was obtained in Dec. 187J; ; but no systematic position can be assigned to it. It is true that lludow gave a diagnosis of a species from the same bu-d, vinder the name Lipennis merklionalis (See Giebel, Insecta Epizoa, 255) ; but that was stated to have an octagonal metathorax, and a broad pale-dull-yellowish abdomen, which peculi- arities, to say nothing of the markings, are not shared by our insect ; besides the head of that species is compact, whereas in ours it is elongate. K K 258 ZOOLOGY OF KERGUELEN ISLAND. INLvKiNE Annelida.—^// fr. C. M'lntosli, M.D., F.ll.S. (Plate XV.) Tho collection of Marino Annelida made diu-ing the stay of the English Transit of Venus Expedition comprises seven species, representing five families, one of which, however, is Nemertean. Six appear to be new. like the Polyzoa and Ccvlontcrata they were procured liy a grapple in the Laminarian region, from depths of 10 fathoms and under. Tlie llev. A. E. Eaton states that the shore Avherc he was stationed was somewhat unfavourable for collecting between tide-marks, as it con- sisted for the most part of ledges of rock without loose boulders, or of a coarse and barren shingle. The mean temperatui-e of the w^ater between tide-marks was 30"^ E. Mr. Eaton found ilie same paucity of Annelida in the littoral region at Spitzbergen. The ^Vmerican Ti-ansit of Venus Expedition obtained 4 species. The tubicolar forms and Pohaioidai occm-red on the roots of Macrocystis, and some of the young Nereids in the usual silken tubes on the fronds of Delcsseria. None of the Annelids were found under stones, excepting the Earthworm described by Professor Laukester. POLYNOID^. Herinadion longicirratus. (Plate XV., figs. 1— t.) Hermadiou lonoicirratus, Kbg. Fregatten Eugcn. Rcsa, &c., p. 22, taf. vi. 33. This form seems to be identical with Kinberg's species from Yoi'k Bay, Straits of Magellan, though the scales and bristles differ slightly from the published figures — the former being densely covered with minute spinulose papillse (fig. 1), and the latter (fig. 2) showing dorsally a less expanded distal region, with a close series of oblique rows of spines (fig. 3). The tip in some is slightly dilated. The ventral bristles, again, have the curve of the terminal hook pronoimced, while the spinous region is rather narrow and short (fig. 4<). All the bristles are of a deep brownish yellow hue. The antenna?, tentacular cirri, and dorsal cirri have a filiform tip attached to a bulbous region, the latter and the rest of the cirrus beneath being furnished with small clavate papillae. Much more minute clavate papillae occm* on the palpi. The brownish scales generally have a few whitish touches : the fii'st is circular, the succeeding reniform, and the posterior elongated from before backwards. It is a large and broad form, one specimen being about 2^ inches long. Hah. — Swain's Bay and Eoyal Sound, Kerguelen Island {Eaton) ; rork Bay, Straits of Magellan {Kinherg). MARINE ANNELIDA.— -W. C. M'INTOSII. 259 Eupolynoe mollis. (Plate XV., figs. 5—9.) M'lntosh, Ann. & Mag. of Nat. Hist. 4tli ser., 1876, xvii., p. 319. This species superficially resembles Alentia gelatinosa, Sars, tliougli a close examination shows many points of diflercnee, and leaves a general impression that the form is intermediate in character hetwcen the latter and such tj^ies as Earmothoe imbricata, L. The head is proportionally larger, and docs not exhibit the nuchal process so characteristic of A. gelat'nwsa ; while instead of the closely approximated pair of largo eyes on each side, the lateral pairs are widely separated, a large one occupying the anterior prominence and a small one being situated at the posterior border. Moreover they nearly constitute a square, whereas in A. gelatinosa they lie in the processes of a V. The tentacle is absent ; but its basal segment is very large in comparison with the antenna? and tentacular cirri. In A. gelatinosa they do not dift'er much. The scales appear to be fifteen on each side, and they arc nearly as soft as those of A. gelatinosa, which they further resemble (tliou.gh smaller) in shape and smoothness. With regard to the latter, however, a high power shows that there is a limited area, near the outer and anterior border, covered with distinct papillce which are low and truncate (fig. 5). The dorsal cirrus has a very slight enlarge- ment below the tapering tip (as in A. gelatinosa) ; but, in addition, it has a few minute clavate papillae. The latter also occur on the ventral cirri. The feet are as distinctly marked as in ^Zera^m; but there is a much greater disproportion between the dorsal and ventral bristles, both of Avliich are pale. Tbe dorsal fascicle consists of a short series of somewhat translucent bristles with distinct spinous rows (almost as well marked as in Uvarne), and gently tapering to a smooth portion at the tip (fig. 6) the fine longitudinal lines being somewhat wavy. The long ventral bristles, again, consist of two groups, more evidently separated than in Alentia or Eupolynoe anticostiensis. The superior tuft arises behind the spine, and is composed dorsally of slender bristles (fig. 7) Avith very elongated and delicately tapered spinous regions, ending in minutely bifid tips like those in Eitpolyuoii anticostiensis.* A gradual change ensues toward the lower l^ristles (of this tuft), which have a stouter shaft, a shorter spinous region, and a strong hook with a secondary process at the tip (fig. 8). The bristles of the next series have still stronger shafts, shorter spinous regions ; and the hook at the tip increases in size, while the secondary process diminishes (fig. 9). Inleriorly, again, there is a tendency to repeat the elongated spinous region and slender forked tip of the upper series. * Ann. & Mag. Nat. Hist. ser. 4, vol. xiii. p. 265, pi. x. f. 3. E K 2 2G0 ZOOLOGY OK Ki:KOrKLEN ISLAND. There are nine papillae on the dorsal border of tl\e extruded proljoseis, and as nianv on the V(>ntral surface. A filiform cirrus occurs nndcr each iul'erior maxilla. llab. — lloyal Sound. Nereidt:. Nereis eatoiii. (IM. XV., tl-s. 10—12.) M'Intosh, Ann. & Ma-, of Nat. llisl. Ith ser., 1S7G, xvii., p. 320. This species somewhat resembles Nereis dumeinlii, And. & Ed. The head has four largo eyes, the anterior pair being- somewhat ovoid and by lai' the larger. When turned backward the long tentacular cirri reach to the fourteenth segment. The uiaxilltu have about eight distinct teeth behind the point. The paragnathi lorin, near each maxilla, live long rows and i'our shorter; and there arc besides several interrupted transverse rows between the foi'mer on the ventral surface. All are composed of denticulate liorny processes of microscopic size. The anterior feet have blunt processes ; their cirri are shorter ; and the bristles have on the wliole shorter tips than in N. dumenlli, ranging from those with long tips (fig. 10) to those with short terminal processes (fig. 11). The articulating end of the shaft in the latter organs has also a somewhat wider pit for the terminal process. At the twenty-fifth foot (fig. 12) the superior lingula is rather larger than in N. dumerilU, and the outline of the other processes also differs. Towards the posterior cxtremitv {c.(j. the sixtieth foot), again the superior lingula forms a very prominent elongated process, Avhich is much thicker and less pointed than in the British form ; and it also differs from N. polyodonta, Schmarda, in this respect. Tlo.h. — E/Oyal Sound. [Nereis antarctica. Nereis antarctica, Verrill, Bulletin U. S. Nat. Mus., 1876, May, iii. 64. Hah. — Eoyal Sound, on the beach (Kidder).] TE11EBELL1D.E. Amphritrite kerguelenensis. (PI. XV., fig. 13.) M'Intosh, Ann. & Mag. of Nat. Hist. 4th ser., 1876, xvii., p. 321. A laro-e form with seventeen setigcrous tubercles. The cephalic region shows four lobes, viz. the ventral anterior lobe, a large process in front and beneath the first branchia, a fan-shaped lobe, and finally a large fold running from the root of the last branchia downwards. The long branchiae spring from three short trunks on MARINE ANNELIDA. — W. C. M'INTOSH. 261 each side. There is a prominent papilla below each setigerous tubercle in the first six segments, and in addition a similar process below the second branchia. The ventral scutes appear to be twelve. The hooks (fig. 13) somewhat resemble those of A. affinis, Mgrn., but differ in the anterior curvature. The colour of one specimen was purplish brown. This species forms a heavy tube of tine mud, lined by a thin chitinous secretion ; and, from the flattening of the ventral surface, it would appear to lie on the bottom. Huh. — Royal Sound. Neottis antarctica. (PI. XV., figs, li, 15.) M'Intosh Ann. & Mag. of Nat. Hist, ith ser., 1876, April, xvii., p. 321. A very large member of the family, differing from Thelep^ts in having three groups of branchiae on each side, and from Grxjmcea by the fact that the bristle-tufts commence on the thu'd segment, and also by the structure of the hooks. The cephalic lobe is furnished with numerous ocular specks. The bristles resemble those of Thelepm, as also do the hooks, which are home on a thin lateral lamella marked by a band of dark pigment. A single process only appears in profile (fig. 14) above the large tooth of the hook, though two are very evident in oblique views (fig. 15). The brownish body is peculiarly streaked posteriorly by pale transverse lines. The animal constructs a large chitinous tube of a dark brownish colour, on which Polyzoa, Zoophytes, and Algae flou!rish. Sal). — Eoyal Sound, very common. [Neottis spectabiUs. Verrill, Bulletin U. S. Nat. Mus. 1876, May, iii., 66. Kab. — Royal Sound, in 12 fathoms (Kidder).] SEBPULIDiE. Serpula, sp. (Plate XV., fig. 16.) The softened specimen resembles S. vermicular is, L., in external appearance; but the operculum is absent. The branchiae appear to be about forty in number on each side. The anterior hooks (fig. 16) are larger than in S. vermicularis, and form a triangle of quite a different shape. The uncini along the edge of the organ are seven or eight in number, the iuforioi-, as usual, surpassing the rest iti size. The posterior hooks present the same structure, and are accomi)anied by the brush-shaped bristles as in ^S*. vermicularis. 262 ZOOLOGY OF KERGUELEN ISLANll. The tube resembles that of the Litter, even to th(^ (ioiibk> i'unncls so often seen in front. The :\l)senee of the opercuhun prevents further defuiition. Tlic undeveloped left opereular process resembles lliat in IS. vennicularis, though it is somewhat longer. Hab. — Swain's Bay. NEMERTINEA. LINEID.E. Lineus corrugatus. (Plate XV., %. 17.) M'Intosh, Ann. & Mag. of Kat. Hist, dth ser., 1870, xvii., p. 322. Body (in spirit) flattened, rather abruptly pointed anteriorly, and more gradually posteriorly. The oesophageal region is marked externally by a scries of prominent and somewhat regular rugtc, which sweep from the mouth dorsally and ventrally ; so that the dorsal view recalls that observed in Arioii ater. Coloiir dark olive throughout, with the exception of a white band, which crosses the anterior border of the snout, and passes backward to the posterior third of the lateral fissm-e, where it bends dorsally and terminates. The special characters are the very large mouth, with the prominent rugse, which show that the animal j)robably possesses umxsual powers of oesophageal protrusion — a supposition borne out by the great development of the external circular muscular fibres (fig. 17 cm), the dorsal longitudinal coat, and the other fibres of the organ. The internal glaiuhdar lining y is also very firm. The outer layers of the proboscis correspond with the type in the Lincidae ; but the internal longitudinal layer e, observed in an imperfect condition in Micrura fusca,* is largely developed. JIuh. — Swain's Bay. Explanation of Plate XV. Fig. 1. Portion of the scale of ILermadion longicirratus, Kbg., the inferior edge being slightly tm-ned so as to show the papillae in profile. X 210 diam. Fiff. 2. Dorsal bristle of the foregoing form. X about 20 diam. Pig. 3. Tip of another dorsal bristle. X 90 diam. Pio-. 4. Ventral bristle from the middle of the fascicle. X 97 diam. Pig. .5. Portion of the outer and anterior border of the scale of Miipohjnoe mollis. On the inferior margin the papillse are seen in profile. X 210 diam. Pig. 6. Tip of a dorsal bristle of the same. X 210 diam. Pig. 7. Tip of a slender bifid bristle from the superior ventral series. X 350 diam. Pig. 8. Tip of one of the lower bristles from the same tiift. X 350 diam. * Brit. AnneHda, Ray Soc, Ft. i. p. 103, PI. 20, fig. 4. MARINE ANNELIDA. — W. C. M'INTOSIT. 263 Fig. 9. One of the smaller bristles from the middle of the next ventral series. X 210 diam. Fig. 10. Bristle with a long tip, from the ventral series of Nereis eatoni. X 350 diam. Fig. 11. Bristle with a short tip from the ventral scries of the same. Fig. 12. The twenty-fifth foot of the foregoing form. X about 12 diam. Fig. 13. Anterior hook of Amphitrite kerguelenensis. X 350 diam. Fig. 14. Anterior hook of Neottis antarctica. X 350 diam. Fig. 15. Tips of two of the former seen obliquely. X 350 diam. Fig. 16. Anterior hook of Serjmla ? X 350 diam. Fig. 17. Vertical transverse section of the ventral body-wall in Linens comi(/a(its, showing the thick circular muscular layer {cm) enveloping the oeso- phageal region ; d", pigmentary layer divided (as in Lineus marinus) by a definite black band {2) ; 3, curious translucent stratum cut into some- what regular spaces ; e, external longitudinal muscular layer of the body-wall ; e', circular muscidar coat ; e", inner (longitudinal) muscular layer ; J, firm glandular lining of the oesophagus ; v, vascular meshes around the oesophageal region. X 55 diam. Fig. 18. Transverse section of the proboscis of the same : a, external coat ; b, great longitudinal muscular layer ; c, belt of circular muscular fibres ; d, base- ment layer ; e, internal longitudinal muscular layer, specially developed in this form ; f, glandular lining of the organ thrown into various folds ; g, lozenge-shaped portion of longitudinal fibres formed by the crossing of two bands from the circular muscular coat ; v. A. E. Eaton, on his rotuni from the Expt>dilioii ol' lh(> Transit observers to KerguekMi's Land, ])hu'ed in my hands For description two small earth- worms obtained by him in the island, and preserved in stronjif spirit. The speci- mens were small and inrmaturc, not exceeding 1 J ineli in length; l)nt by cutting transverse sections of one, and slitting the other up the median dorsal lino, staining with carmine, and mounting in Canada ])alsani, I bav(> succeeded in nialcing onf the affinities of the s])eeies. The study of the various species of Earthworms (Lvimbricidoc proper) has only recently been attempted a\ itli due attention to anatomical detail. Tlieir excessively complicated generative glands, ducts, and ])oucbes present the greatest diversity of arrangement, so as to enable us to establish a series of strongly marked genera, which, while differing in the arrangement of these parts, yet present but slight difFerences in external form, or in the arrangement of their seta\ Professor Edmond Perrier, availing himself of the very fine collection of exotic Lumbricidae in the Jardin des Plantes, has been the pioneer in this branch of investigation, and in his memoir " Rechercbes ])onr servir a I'histoire des Lombriciens terrcstres," published in the " NouvcUes Archives du Museum d'llistoire Naturclle, 1S72," he has established a series of genera on the only possible characters in modern zoology — namelv those derived from thorough anatomical examination. M. Perrier has stiidied earthworms I'rom North and South Africa, from the East Indies, from the West Indies, from North and South America, and a number of scattered islands, and lias rendered it evident that he has tapped a rich storehouse of zoological facts of first-rate importance. Presenting, as they do, a considerable number of genera, and occurring as they do almost universally on the earth's sm-faee where there is vegetable soil — being moreover absolutely destitute either of means of transport or of power to resist deleterious agents whilst being passively transported (earthA\'orms and probably their eggs are rapidly killed by sea-water), the Ltimbrtcklce promise to yield, when fully investigated, a mass of information bearing upon the problems of the causes of geographical distribution and the connections of continents and islands in past epochs — more decisive and indisputable in its character than that presented by any similar small group of the animal kingdom. The essential feature of their oi'ganisation which gives to the Lmnhrickhe so interesting and important a position, is the possession of a most sensitive generative apparatus — sensitive, that is to say, in the sense of responding l)y innumerable modifications of its highly-developed male ducts, prostatic glands, seminal reservoirs, penial setae, copulatory pouches, and other accessory glands, to those slight differences of environment which whilst thus afPect- TERRESTRIAL ANNELIDA. — E. RAY LANKESTEK. 2G5 ing the genitalia so as to create generic distinctions, have yet left the external form and character imaffected. The two small specimens from Kcrguelen's Land are the first Earthworms of special interest Avhich I have received, though for some time, through the kindness of Sir J. Hooker, Earthworms, found in the Wardian cases sent from abroad to the Royal Gardens at Kew, have been forwarded to me for examination. I may take this opportimity of saying that persons who may wish to preserve specimens of exotic earthworms for examination in this country should either send them home alive, which is easy and the most satisfactory to the student, or should kill them with chloroform, by which means they are prevented from shrinking, and then place them first for tM^enty-four hours in weak spirit, and afterwards in the strongest which can be prociu'ed. The Earthworms brought from Kerguelen by Mr. Eaton are small specimens of a species of AcantJiodrihis. The genus Aeanthodrilus is established by Edmond Perrier for the reception of three species, two of which come from New Caledonia {A. ohtusus and A. uiigulatus), whilst the third {A. verticillatus) is an inhabitant of Madagascar. The addition of Kerguelen's Land to the distribution already indicated by PeiTier for Aeanthodrilus, is a matter of some consequence, though tmtil our collections of Ltimbricidce are more exhaustive than at present, it would be very rash to discount the conclusions to which we shall be ultimately led. I propose now to give the characters of the genus Aeanthodrilus as indicated by Peri'ier in his classical Avork, and then to point out the distinctive characters of the Kerguelen species. Characters of the genus Acakthodriltjs, icith notes on the neio species. The LtimbricidcB are divided by Perrier into three sections according as the male generative apertm-es are in front of, within, or behind the clitellum. The genus JjtimbricKS alone is Prceclitellian, the genera Anteus, Tit anus, Rhinodrihis, Uro- chata, and Geogenia are Intraclitellian, whilst Pontodrilus, Eudriliis, Moniligaster, Acanthodrihis, Digaster, Ferionyx, and Ferichceta are jPostcUtelUan.* The genus Aeanthodrilus is especially characterised amongst the FostcUtelUan Lumbricidce by the possession of two pairs of male generative orifices which are placed in the 17th and 19th, 18th and 20th, or 19th and 21st scgmcnts.f These orifices are so placed as to give exit each to a bundle of greatly elongated and specially modified " penial " or " genital " setae. The term "penial" proposed by Pemer is more appropriate than that which I had previously used in describing * It is impossible to determine the true value and position of the genera of LumbricidiB establislied by Kinberg, since he has not furnished the necessary anatomical details, ■f The cephalic lobe and the buccal ring form i\\oJ!rst segment. L L 266 ZOOLOGY OF KERGUELEN ISLAND. similiu'ly motlitiod seta) in Cluotog-astcr and Nais (sec my paper " On distinct larval and sexual ibrnis in the Oli-oclKeta." Ann. & Ma-;'. Kat. Hist. 1870). The existence ol" these peuial setie is Avhat has sii!j,'t;ested the name ol' the i^euus, since they appear to be unique amongst the Lumhric'uhe, though Ave ihid similar setae in the Naididce, and in Lumhricus an enlargement and elongation ol" the setae in several of the segments connected with the reproductive organs, though not a well marked specialization of form is noticeable. The setie which are tlius modified in Acaiithodrilus are those which correspond to the two ventrally ])hice(l pairs (one on each side the median line) o£ a segment of Liuuhricns, tiic dorsally plaeetl paii-s being unmodilicd. In the new Acanthodnlus the penial setae are in two bundles of foiu' each, or eight altogether to each male genital pore. They arc notched near the anterior extremity as in Perrier's A. verlicillutus (see fig. G.) Perrier gives as a character of the genus that the locomotor setae are arranged as in Luvibricm in four series, each group of bristles containing as in Lumhricus two fimctional seta). This character must be amended, since in the new Acaiilhodrilus of Kerguelen's Land the setae are arranged, not in four series of bundles or groups, each containing two setae, but in eight scries, each seta standing alone, and widely sepai-ate from its fellows of neighbouring series (lig. 4). Thus on each segment we can distinguish, on each side of the median antero-posterior vertical plane, a medio-ventral seta and a latero-ventral seta, a latero-dorsal seta, and a medio-dorsal seta. It becomes quite clear that the double ventral series in Luiubricus and the other species of Acaiithodrilus, is formed by the approximation of two single series such as we see in the medio-ventral and latero-ventral .series of the Accm- thodrilus of Kerguelen's Land, since in certain segments of this species, namely, the 16th, 17th, ISth, and 19th, the two separate ventral series of single setae approach one another, and form a double ventral scries (see fig. 2), exactly comparable to the arrangement which obtains throughout the series both dorsal and ventral in Luni- bricus and most other Ltimbricidte. Acanthodrilus is stated by Edmond Peiuner to possess the median dorsal pores leading fi-om the body-cavity to the exterior, which are wanting in some genera. A full description of the genitalia of Acanthodrilus is still a desideratum. The exact position of the testes and ovaries is not known, nor do my very young speci- mens from Kerguclen enable me to supply the requhed information. Opening close by the side of the penial bristles, and with its orifice covered by a flap of integument (fig. 7), is a tube (one on each side in each of the two penial seg- m.ents) which runs horizontally, and expands into a short, undulated, thick walled caecum. These four caeca have been observed by Perrier in the various species of Acanthodrilus studied by him, and appear to be prostatic glands connected each with a distinct vas deferens, which place the four male genital orifices in continuity with the testes situated about the 11th segment. The four caeca (sec fig. 7._pr.) are TERRESTRIAL ANNELIDA.— E. RAY LANirESTER. 267 well developed iu the specimen from Kergviclen's Land, but I was unable to find the testes or vasa defcrentia in these small specimens. The cingiUum noticed by Pcrrier in his Acanthodnlus imgulattis on the 14th, 15th, 16th, 17th segments, was not developed in my specimens. The copulatory pouches, which in the species described by Pcrrier are placed to the number of two pairs iu the 8th and 9th or 8th and 10th segments, have a similar position in the ncAv species, namely, on the line of the latero-ventral scries of seta; between the 7th and Stli and the 8th and 9th segments. The cephalic protuberance, prostomium, or upper lip (fig. 3 pr) is wortliy of note from its peculiar setting in the buccal ring ; a similar form of prostomium is described by PeiTier in the AcantJirodrilus verticillattis of Madagascar. Distinctive features of the Acanthkodrilus of Kergitelens Land. Male orifices and penial setae placed in the l7th and 19th segments ; orifices of the copulatory pouches between the 7th and 8th, and 8th and 9th segments. Setaj arranged, not in four double, but in eight single rows, viz. : right and left medio- vcntral, latero-ventral, latero-dorsal, and medio -dorsal. The latero-ventral and medio-ventral rows converge in the 16th, 17th, IStli, and 19th segments. The penial setse are formed by eight sctigerous sacs, a latero-ventral and a medio-ventral to each of the fom- male genital pores. The prostomial lobe is short, and sunk within the buccal ring. The genus and species are briefly characterised thus : — Acanthrodrilus, Edm. Perrier. Xymb)'icid(V post-clitelliani, poris genitalibus masculis quattuor, duobus in seg. 17, 18, vel 19 ; duobus in seg. 19, 20, vel 21, prope setas ventralcs positis. SetsB ventrales, poros genitales justa, valdc elongata? et nimaerosfc, pe»i«Z6'5 dicuntur. BurssB copulatrices utrinque duee in seg. 7, 8, vel in scgmentis vicinalibus. Acanthodrilns kergiielenensiSt sp. n. A. poris genitalibus masculis in seg. 17 et seg. 19 positis ; bursis copulatricibus inter seg. 7 et 8, et inter seg. 8 et 9. SetJB locomotores in lineis 8 ordinatnu, utriuque medio-voutrales, latero-ventrales, latero-dorsales et medio-dorsales. Lobus prosto- mialis brevis, rotundus, annulo buccali immersus. LL 2 268 ZOOLOdV OK KKRGUELEN ISLAND. 7inu /J M3.7 mv lig/i. Explnnation of the IVoodcnf. Fig. 1. Young specimen of Acanthodriliis kergnelenensis, natural size. Fig. 2. The anterior twenty-one segments of tlio same magnified and seen from the ventral aspret, to show the position of the seta3 and the genital orifices. Id, latero-dorsal series of seta;. Iv, latero-ventral series of seta;. mv, medio-ventral series of setfe. The medio-dorsal serii's are not visible. Between segments 7, 8, and 8, 9, are seen the orifices of the copulatory pouches. In segment 17, and segment 19 are seen tlie two pairs of male genital pores, which are provided with the penial seta>. Fiij. 3. Dorsal view of the three anterior sea:ments to show the form of the prostomial lohe i^r. md, medio-dorsal set®, kl, latero-dorsal setse. Fig. 4. Diagrammatic section through the thirtieth segment to show the position of the eight series of setaj, and of the openings of the segmental organs. md, medio-dorsal seta:;. ■n;i{Hi:sri!iAr. annelida.— e. ray lankester. 269 Id, latero-tlorsal setce. Iv, latero- ventral setaj. mv, medio-vontral setse. sg, sosfmontnl organs. Fig. 5. Ordinary locomotor scta3. Fig. 6. One of the ponial setie {" genital setae" mild of ChcBtogaster and Nats), drawn to the same scale as fisr. 5. Fig. 7. The male genital orifices and setigerous pores of the left side (diagram- matic). iiif, cut edge of the integument. Jl, flap of tlie integument which overlies the male genital oritice ood. ovd, orifice of the vas deferens. pmv, penial seta and orifice of the medio-ventral series. p/v, penial seta and orifice of the latero-ventral series. ss, penial setae in the setigerous sac. - 2)7', prostatic citcum of Perrier. mv, medio-ventral locomotor setae. Iv, latero-ventral locomotor setae. m longitudinal muscular hand separating the orifices of the medio-ventral and latero-ventral setigerous sacs. The setae, setigerous sacs, and prostatic caecum of the 19th segment are not represented. They would he identical with those of the 17th. 270 ZOOLOGY OF KERGUELEN ISLAND. EcniNODERMATA. — Bif Edgar A. Smith, F.Z.S., Senior Assistant Zoological Deportment, British Museum. (Plates XVI. and XVII.) Opportunity was taken some pacjes hack of exhibiting iho volations of the Mol- luscan-fauiia dl" Iverguelen Island lo llial of llin Falklands and Patagonia ; and it was pointed out that representatives of horeal types entered into its composition. ^Materials for similar eompavisons hetween the Echinodermata indigenous to the same regions scarcely exist, hut such as there are, mak(> ita])parent that what ohtains in the ;^[ollusca holds good also in the Echinodennata with respect to geographical distribution. / The sources of information relating to the animals of this class inhabiting the vicinage of Kerguelen Island, are at present limited to a few relics of the collections made there by the Antarctic Expedition in 1840 ; the collection made in 187 t by Dr. Kiddor of the American Ti'ansit of Veniis Expedition, part of the Challenger's collections, and ^Mr. Eaton's specimens. The remaining portion of the Echinoder- mata obtained by the Challenger Expedition is still in course of investigation ; and t lie collections of the German Ti'ansit and Surveying Expedition are likewise not yet worked out. This mil account for the fewness of the species at present known from the island, which altogether amount in number to 15. Two of them were obtained by the Antarctic Expedition, one by the Challenger, four others by Dr. Kidder ; besides five out of the preceding seven, eight others were obtained at Observatory or Swain's Bay. The species not found by Mr. Eaton are Asferias rnpicola, which was common on the beach near Dr. Kidder's station, and Cidaris mitrix : those which he did get were procm*ed almost exclusively from the Lami- narian zone. This will suffice to show how far dependence can be placed upon the following summai'v of the geographical relations of the little known fauna. The species hitherto recorded from Kerguelen Island comprise 1 Holothurian, 2 Echini, 9 Asteriida?, and 3 Ophiuridse, — in all 15 species referable to 12 genera. Prom Patagonia are known 1 Holothurian, 1 Echinoid, 3 or 4 Asteriidse, and 5 Ophimudfc, — in all, 10 or 11 species comprised in as many genera. Of these the Holothi;rian, the Echinoid, one of the Asteriidse, and one of the Ophiuridse, — that is 4 species, are indigenous also to the coast of Kerguelen Island. A similarity to certain boreal terms is exhibited by some of the species. Thus Porania antarctica strangely resembles P. 2i'ulvillns of the northern seas of Europe ; the Fedicellaster represents another septentrional genixs ; Fferaster affinis imitates closely Ft. militaris of boreal waters. The genera Ophioglypha and Ophiacantha are almost cosmopolitan in distribution ; yet the Kerguelen Island representive of ECHINODEKMATA. — EUGAK A. SMITH. 271 the former, O. hexactis, in colour and iu tout ensemble approaches 0. Sarsii of the Greenland coast. The new form Lcptoptychasto' has been found only at Kerguelen Island. JIOLOTUURIID^. Pentactella laevigata. "* VerriU, BiiU. U. S. Nat. Mus. 1876, iii., p. 68. Hah. — "Kerguelen Island, twelve fathoms " (Kidder). Very common on the laminiM oi Macrocystis m Royal Sound and Swaiu's Bay (Eaton). Also Patagonia (Antarctic Expcd.). The colour of this species during life is dull whitish ; the body is somewhat opaque, but the distended tentacles are semi-transparent white, tinged very faintly at the tips of their ultimate ramifications with pale rose colour. Owing to their whiteness, specimens can be seen on the kelp, sometimes at a considerable depth below the sm*face, appearing to be much larger than their real size thi-ough the refraction of the light. ECHINID.E. [Cidaris uutrix. Wyrille Thomson, Journ. Linu. Soc. 1876, vol. xiii., p. 62, fig. 4. Sob. — Balfom* Bay, Royal Sound, 60-70 fathoms. This most interesting species forms part of the collection of the Challenger.] Heiniaster cavernosas. Tripylus cacernosus, Philippi, Wiegm. Archiv. f. Natm'gesch. 1815, p. 315, pi. xi., f. 2. ^rissojosis cavernosa, Agassiz, Ann. Sei. Nat. 1847, viii., p. 5. Heiniaster cavernosas, Id., Revision of the Echini, 1873, part iii., p. 587, pi. xxi'-., fig. 1, 2. Tripylus aiistralis, Philippi, op. supra cit., 1845, p, 347, pi. xi., fig. 3. Br'issopsis aiistralis, Agassiz, Con. R. Ann. ScL Nat. viii., p. 5. Paorina amtralis. Gray, Ann. & Mag. of Nat. Hist. 1851, vii., p. 132. Heiniaster aiistralis, Agassiz, /. c. 1873, part iii., p. 586, pi. xxi''., fig. 3. Faorina antarctica, Gray, op. supra, cit., 1851, vii. p. 132 ; idem, Cat. Rec. Echin. 1855, p. 57. Heiniaster cordattis, Verrill, Bulletin U. S. Nat. Mus. 1876, iii., p. 60. Heiniaster sp. "Wyvillc Thomson, Journ. Linn. Soc. 1876, xiii., p. 67, fig. 6. The spines when iii-st observed were greenish olive in coloiu", but are gradually acquh'ing in spirit a brownish olive hue. Hub. — Swain's Bay and Royal Sound, on mud, in 5-10 fathoms. Extremely ■2r2 ZODLUCiY ol KllKGUKLliN ISLAND. common; Acoessible liay, 20-50 latlioms {l/'i/ri//c Thomson); also Patagonia {F/iilippi). A comparison ol' tlu> typos oi" Faurina aiittirclicti \\\\\\ \\\r Tlcmias/rl collcciod by Mr. Eaton at Kcrii;ui>lcii Isiaiul, shows (lial llio species nic identical with one another. Dr. Gray in 1855 says, in ret'erence to this species aiul the two I'ovms described by IMiilippi, cited above, — " Perhaps these three arc only one species." Mr. Ycrrill, in a footnote at oji. cit. p. 71, states that M. A. Aj^assiz is oi" t\u\ same opinion, viz., that the Kcrguelon species is identical withPhili^jpi's. After a careful study of Philippi's ligurcs and descriptions, Dr. Gray's types, and ]\lr. Eaton's specimens, I thoroughly concur in believing tlieni to be all of one species (the fonu A\ ith dee]) ambulaeral furrows the female; lliai with shallower furrows the male) . If it should luTeafter be discovered that the Patngonian species is, on the con- trary, distinct from the Kerguclen Island species, the name proposed by Gray {antarctica) will still have to be retained. ASTEKIIU.E. Asterias meridionalis. (Plate XVI., lig. 1.) Perricr, Eevision d. Stcll. 1875, p. 76 ; idem, Ann. & Mag. of Nat. Hist. 1876, xvii., p. 36. The specimens from Kcrguelen Island exhibit, in certain particulars, inconstant variations from tbe type of this species, and also from one another. The following remarks, however, are not confined to these aberrations and differences. Body 6-rayed ; rays rather stout, gradually tapering, convex above, rather more than twice as long as the disk is wide. Ambulaeral spines irregularly biserial, but not constantly so, rather stouter than in the type ; in one example they are at ii'regular intervals imi-serial, and towards the inter-radial angles of the mouth they arc also in single scries. Next to the ambulaeral spines succeeds a series of tenta- cular pores ; and then the lower or ventral margin of the ray bears a double (rarely triple) series of short stout spines. Above these extends a naked band with small groups of papidae in it ; and above this naked band, limiting the dorsal area, is a row of isolated spines similar to the spines in the other series. The dorsal spines of the disk and rays are numerous, irregularly scattered, short, blunt, scabrous to the touch, and striated ; and among them are interspersed numerous small groups of papula? : one example has the spines of the upper and lateral surfaces conformable to those of the type ; but another example bears spines which are acute and shortly conical, particularly so on the disk and thickest portion of the arms ; and in this specimen they are also disposed more regularly than is usual, tending to arrange themselves in longitudinal series down the arms. These spines are about 2 mm. long ^:ClIi^'()UEKMATA.— EDGAR A. SMITH. 273 in the disk of the example last mentioned, but become shorter and less regular in arrangement towards the extremities of the rays. The madreporic plate is placed in an angle of the rays, and about midway between the centre of the disk and the margin. Colour reddish l)rown above, pale bulf beneath. Dimensions. — Length of arm, 3 inches ; width of arm at the base, 7 lines ; width of disk nearly l.j inches. Hah. — On roots of Macrocystis in Observatory Bay. Also obtained by the Antarctic Expedition. M. Perrier in his earlier description likens this species to A. ciminghami, to which he accidentally ascribed 6 rays instead of 5. This might have been mis- loading, had he not subsequently in the Annals and Magazine of Natural History (loc. cit.) corrected the mistake. Asterias perrieri. (Plate XVI., fig. 2, 2a-b.) Smith, Ann. & Mag. Nat. Ilist. 1876, xvii., p. 106. Discus sex-radiatus, modicc amplus, in latitudine ad spatium maximum inter radiorimi oppositorum apices circiter f ajqualis ; radii cylindraceo-attenuati inferne anguste planati. Sulci ambulacrales hand latissimi ; sjjinaj ambulacrales subgraciles, subcylindricaj, non clavata), in scrie unica posita}, in exempl. maxim, circiter 3 mm. longse ; spinse ventrales in scrie duplici spinis ambulacralibus adjaccntes, binatte (nisi rarissimc ternatai), divergentes, spinamm intima subacuminata, major intermedia, sed minor et crassior ambulacrali. Dorsum lateraque spinis brevissimis minutis subconicis diverse aspersis, papulis innumeralibus intei-positis. Tessella madreporiformis pai-viuscula, inter centrum margincmque disci intermedia. Color saturate fusco-rufus. Disci diam. 45 ; radiorum longit. 150 ; ad bases crass. 19 mm. ( Young.) Six-rayed ; the rays very short, nearly as broad as long, A;ith only two rows of ambulacral tentacles bordered by a single series of spines ; the latero- dorsal margin with a single row of large spines, the dorsal area with a similar longitudinal series down the middle of the ray. Hub. — Common on the roots of Macrocystis, and also taken in shrimp pots, at Observatory Bay. The largest specimen has a cluster of some hundreds of young ones clinging to its ventral disk. A. rztfjispinu of Stimpson is allied to this species, with which 1 have had much pleasure in associating the name of M. Edmond Perrier of the Jardin des Plantes, who has recently identified many species of the Asteriidie in oiu- jiatioual collection. M 51 27 J ZOOLOGY OF KERGUELEN ISLAND. Pedicellaster scaler, (riatc XVI., fi-. 3.) Smith, Ann. .^- 31 ng'. Nat. llisl. 1S7G, xvii., ]). 107. Discus quinquc radiatus, iu lat itudiiio ad radii loiigiludincm ciiTitcr scminequalis ; rndii subcylindrici, scnsiiu attcnuati, modicc acuti. Spintp ambulacralos gvaciles a'quilongiv, iii scricbus tribus posita*, duplo longiorcs dorsal ibiis ; aiiguli oris iutcr- radialos spinis parvis ad apices binis. Spintc dorsalos disci radiorumqiie brcvissimsD, obtusa\ scabra\, distantcs, in siibstructurani quasi clathratani diverse dispositiu, eoc prope ambulacra paulo long-iores ctutoris ; inacidtc rcticuli nuda;. Anus subc(>ntra- lis ; tcssella madi'cporil'ormis subrotunda, propc marginem in angulo iuterradiali sita. Disci diam. 9, crass. C ; radiorum longit. 18, ad bases crass 5^- mm. Hah. — On roots of Macroci/stis, Observatory Bay. This species appears to agree very fairly with Sar's description of his genus PediceUasfer, excepting that the ambulacral furrows cannot be said to be " l)road," and that the ambulacral spines are in three rows instead of two. 33ut these differ- ences are more specific than generic ; and therefore I think that this may, notwith- standing them, be properly considered a second species of that northern genus. Ecliinaster spinulifer. (Plate XVI., fig. 1.) Othilia spimdifera, Smith, Ann. & Mag. Nat. Hist. 1876, xvii., p. 107. Discus qvdnque radiatus, ad radii longitudinem iu latitudine scmiicqualis, modicc crassus, supernc leviter rotundatus ; radii cylindraceo attenuati, breves, spinis scabris brevissiniis nunierosis in substructuram quasi clathratam diverse dispositis, inteiTallis clatlu'onim nudis ; prope spinas ambnlacrales risque parallele spatium linearc fere nudum, radii basin versus sensim latius, serie solum unica spinarum minutarimi (louge muiorum etiam spinis ambulacralibus) mimitum adjacet. Spinas ambulacralcs divergentes, transverse super tessellas singulas quaternis dispositse, dua3 interiores paulo longiores ; spina gracillima parva recta est intima3 hariim apud basin. Anguli oris interradiales spinis parvis duabus terminantur. Anus subcen- tralis; tessella madreporiformis in angulo iuterradiali submarginalis. Disci diam. 7 ; crass. 6 ; radiorum longit. 11-, diam. apud basin 4i\ mm. Sab. — Observatory Bay. This curious little species is remarkable for the shortness of the rays, and for the minute slender spine within the ambulacral groove at the base of the innermost spine being straight, instead of heing curved or hooked as the homologous spine is in the normal species of the genus. The spines on the dorsal and lateral surfaces display no regularity in their arrangement ; but parallel with the ambulacral sjiines in the almost naked nan'ow space adjacent to them, the series of very small spines is disposed definitely, one sprue on each plate, and above this the spines are placed two or three on a plate. ECHINODERMATA.— EDGAR A. SMITH. 275 Pteraster affinis. (Plate XVI., fig. 5.) Smith (sp. nov. ?) Ann. & Mag. Nat. Hist. 1876, xvii., p. 108. Discus magnus, quinquc radiatus, subtus planus, supra convexus, modice crassus, ad radii longitudinem in latitudine aequalis. Radii breves, e basibus latis statim ad apices angustati, ibique ita rccurvati ut sulcos ambulacralcs exponant ; subtus utrinque extra sulcos nicmbrana tenui spiuis gracilibus circiter 30 apicibus vix ultra margiucm niembranfc projicicntibus induti; tessella; intcr-ambulacralcs spinas graciles quatuor in mcmbrana tenuissima fere ad apices producta gerentes, intima harum paulo cajteris brcviori ; anguli oris intcrradiales spinis similibus octonis, pariter membraua conjunctis, duabus cxtimis longe brevissimis, duabusque intimis longissimis, spinis at que supra illas crassis duabus sibi parallelis, leviter in medio concavis et apices versus acuminatis, longissimaeque spinarum 30 latcralium in longitudine ajqualibus. Dorsum lateraque supra projccturis spiniferis minimis, foramiuibus minutis baud numerosis interpositis ; illae spinis scabris in membrana comiexiva pgene apicium tenus conditis diverse a quaternis usque ad denas munitis ; foramen centrale modice amplum, fimbria spinarum brevium membrana conjunc- tarum circumdatum. Color (in spiritu vini) sordidus, pallide ocbraceus. Exempl. max. disci diam. 15, crass. 7 ; radii longit. 17, diam. ad basin 8 mm. Exempl. minor, disci diam. 10, crass. 5 ; radii, longit, 9 mm. Sah. — On roots of Macrocystis, Observatory Bay. Tbis species approaches very closely to P. Dance of VerrUl,* which is supposed to have been found at E/io Janeiro. It appears, however, to possess longer arms (though the smaller specimen, it will be noticed, is considerably shorter in the rays than the larger) ; the spines of the dorsal fascicles are everywhere uniform and scabrous ; there are only eight spines at each of the inter-radial angles of the mouth, and the two larger spines above them are not very long, but are stout. In these respects chiefly it differs from Mr. Verrill's species. Porania antarctica. (Plate XVII., fig. 1.) Smith, Ann. & Mag. Nat. Hist. 1876, xvii., p. 1.08. Discus qviinque radiatus, radiis sul)-breviter conicis apices versus acumi- natis, modice crassus, ad radii longitudinem in latitudine oequahs, subtus planus, supra convexus, ubique indutus cutem crassam carnosam, infra transverse inter margines suJcosque ambulacralcs lineariter sulcatam, superne la^vem, nisi apud medium disci et etiam super radios spinis tubercularibus parvis sparsim ornatam ; margines laterales inferno spinis brevibus compressis ad apices latis et truncatis fimbriatoe, spina unicS, super tessellam singulam sulcis linearibus jam supra dictis * Described in Proc. Boston Soc. Nat. Hist. 1869, vii., pp. 386, 387. M M 2 270 zoou)(;y or kekguei.kn island. dcfinitam ; spinrc amlmlacralcs biscriatjiD, cxtcrioirs duplo lonyiorcs inviltoqiic r()T)US- tiores intorioribus, l:ita\ ad npicos abnipt(^ Iruncalio lu^qiu^ an;;"iistat!t% (>xtus Icvltor sulcata^ ita ut duplicos esse vidcantur; tossoUa madivporironnis rot undo ovalis, paulo ad contniiu propior quam marp;incm ; anus centralis, papillis brcvissimis spinirormibus circitor 12 circumjoctis. Color carnous vcl sanguineus. Diam. max. 00, minim. IS mm. jjal, — Dred£!:cd in a retired inlet ol' Swain's Bay, in aboui TO faihs. of water, and outside tbe belt oi Jlacroci/sfis. The furrows on the outside of tbe exterior ambulacral spines are formed ehieily by tbe skin Avbieb clotbcs them. The minute tubercles on the back exbibit no regu- larity in their arrangement ; there arc about a dozen of them on Ihe middle oC the disk and a few on tbe sbort conical arms. This species may be distinguished from tbe rather closely related northern Poraiiia pitlviUus, Milller, by dilTerences in th(> ambulacral spines, and in the number and character of the marginal spines. P. pnUUlus has 3 or 1 spines on each of the marginal plates, which are mucb smaller than the single spines in P. antarctica. Asferopsis is llie name usually adopted byautborsfor the genus to which Midler's species belongs. The genus thus designated was founded by Midler and Troscliel for the reception of Asterias carinifera, Lamk. (sec Wiegm. Arehiv. ISiO, p. 323) ; and in their System der Asteridcn, 1812, pp. 62-1, they include in the same genus A. ctenacantha , A. pulvillus, and otber species in addition to the one just men- tioned. Porania is the appellation given to Midler's A. pulvillus by Gray, who described it as Forania gihhosa, Leach, in the Ann. & Mag. of Nat Ilist., 1840, December, p. 288. It has appeared advisable, to me, to restrict Asteropsis, (Mill. & Troscli. (of which Gi/mnosteria, Gray, is a synonymc) to A. carinifera, Lamk., inermis. Gray, &c,, and to retain Porcmia, Gray, for P. pulvillus and antarctica. Pentagonaster meridionalis. (Plate XVL., fig. 6, Ca.) Astrogonium meritUonale, Smith, Ann. & Mag. Nat. Hist., 1876, xvii., p. 109. Discus quinque radiatus, ad radii lougitudinem in latitudiue circitcr J sequalis, depressus, subtus supraque leviter convexus. Radii modice longi, apud bases lati, apices versus sub-repentc attenuati, spinis ambulacral ibus in seriebus quatuor dis- positis, interioribus compressis apices versus dUatatis tandemque truncatis, exterio- ribus teretis ad apices rotundatis in longitudine subparibus. Auguli oris inter- radiales singiili spina unica conica robustu exstante super angulum inter-ambu- lacralem (similiter ac in Pterastro) et infra banc spinis parvis 6 — 8 in serie reclinantibus. Discvis radiique subtus fasciculis spiuosis in seriebus ex sulcis ECIIINODERMATA.— EDGAR A. SMITH. 277 ambiilacralibus ad margines laterales excun-entibas, fasciculis ipsis ex spinis parvis cylindricis siibaeutis, multo quam spinis ambulacralibus brevioribus, compositis : latcra angusta, fasciculis parvis spino>is quadratis (circitcr 20 super radios singulos) confertim in scriebus augustis duabus margini appositis, iis in serie infcriorc ca3tcris super superficieni ventralem, iisquc in serie superiore coeteris super super- ficiem dorsalem consimilibus. Fasciculi dorsales numerosi, brevitcr pedunculati et fere contigui, singuli ex spinis brevissimis ad apices obtuse rotundatis 10 — 20 constructis; interstitia fasciculorum omnia nuda, pcdicellariis magnis multis munita. Radii ad apices supra tubcrculo magno, unico ; tessclla madreporiformis rotunda, elerata, fere in medio inter centrum marginemque, posita ; anus sub- centralis. Disci diam. 24 ; crass. 10 ; radiorum long. 29 mm. Kab. — Dredged in Observatory Bay in 5-10 fatbs. on mud. This species apparently belongs to the section of the genus Astrogonium defined by Gray for the reception of a form from Port Essington (Proc. Zool. Soc. 1847, p. 79). For it differs from Dr. Gray's short divisional description in no respects excepting that the ventral smface is not entirely covered with granules (or spines), and that the disk is not flat. Probably this last distinction is due to Dr. Gray's example having been dried, whereby it very likely had become shrunken and more depressed than it would have been had it been preserved in fluid like the Kerguelen Island example. This species, J. paxillosum. Gray, is now placed in the genus T?entacjonmter, by M. Perrier. Leptoptychaster. Lepty chaster. Smith, Annals & Mag. Nat. Hist. 1876, xvii., p. 110. Derivation (Gr.) XctttoV 'narroio, tttv^ a plate, and da-Ti^p star; in reference to the narrowness of the ventral plates. Discus quinque-radiatus, depressus ; radii modice longi ; superficies dorsalis fasciculis spinularum minutarum pedunculatis confertim obsita ; radii serie unica laterali tesseUarum tenuium transversarum lameUiformium usque ad ambulacra vix productarum muniti, serieque altera fasciculorum spinarum minutarum (fascicvilis unicis cum tesscllis ordinate dispositis) inter tessellas ct ambulacra interposita; tessella madreporiformis sujier margiuem in angido interradiali locata. This remarkable form of Starfishes is perhaps more nearly i-elated to the genus Luidia than to any other. It differs from that genus, however, in the lateral spiniilose narrow plates not reaching quite to the ambulacra, but leaving a narrow intermediate space occupied by groups of spines in line with the plates, which gradually becomes wider towards the base (in Luidia the plates extend to the ambu- lacra) ; and also it has no elongated spines, and the body is larger in proportion to the rays than it is in that genus. 278 ZOOLOGY OF KERGUELEN ISLAND. Leptoptycliaster kergueleueusis, Smi/fi. (Wato XVII., fig. 2.) Leplijchasto' korguelcucnsio. Smith, .Vnn. & Mng. Nat. Hist. ISTH, xvii., p. 110. Archaster iwcavattis, AVyville-Thomsnii, Journ. Liuii. Soc. 1870, xiii., p. 72j fig. 10. Discus quinqiic-radiatus, modicc ampins, dcprossus, supra et. infra ])lanus, in latitudino ad radii longitudincm circitcr ij-ooqualis ; radii apud bases baud admodum lati, ai")iccs versus sensim attcmiati, spinis ambulacralibus grarilibus, transverse (i-atione ad sulcos babitA) quaternis vel quinis positis, duabus intcrioribiis longi- oribus, caiterisquc gradatim brevioribus ; iidem ad latera subt usque, tessellis minute spinulosis (spinulis ipsis scabris ct apud radioram bases longissimis, indequc apices versus paulatim abbrcAnatis) angustis, transversis, lamcUiformibus, obtecti ; aliqui ex his apices radiorum versus usque ad sulcos ambulacrales pa^nc attingent, bases vero versus ita gradatim ab liis recedunt ut apud augulos interambnlacrales arcam trigonalcm \itrinque rclinquant ; inter tessellas ct spinas ambulaitralcs fere per totara radii longitudinem, in scrie unica, fasciculi parvi spinarum brovium inter- ponuntur, et prncter banc juxta basin, in areis trigonalibus supradictis, series aliaj quatuor singulatim per vices intercalantur, omnes fasciculi cum tessellis ordinibus collocantes ; anguli oris interradiales acutis spinis utrinquc quaternis vel quinis muniti. Superficies dorsalis fasciculis pcdunculatis spinarum brcvium confertim scrta ; tcssella madrcporiformis modice magna, subovalis, fasciculis spinarum obsita, ad angulum inter-radialcm super margincm posita. Disci diam. 23, crass. 8 ; radiorum longit. 38, ad bases lat. 13 mm. JIab. — On roots of Macroci/stis, Observatory Bay ; also obtained in the island by the Antarctic Expedition ; and " Off Cape IMaclcar, south-east coast of Kcrguclen Land, from, a muddy bottom at a depth of 30 fathoms " {IWi/. Thomson). Tl\e small fascicles of little spines on the dorsal surface are borne on short peduncles of skin, and are so closely packed together that their apices constitute an even sui-f ace ; and the madrcporic plate being covered with sj)ines similar to those of the rest of the surface is concealed from view. Ophiuridjj:. Ophiacantha vivipara. (Plate XVII., fig. 3 a-c.) Ljungman, Ofvers. K. Vetensk. Akad. Forbandl. 1870, p. 471 ; Smith, Ann. & Mag. Nat. Hist. 1876, xvii., p. 110. Ophiocoma didelphys, "Wyville Thomson, Jom'n. Linn. Soc. 1876, xiii., p. 78, fig. 13. Examples from Kerguelen Island differ slightly from specimens of O. vw'qjara found at Patagonia, in the oral shields, the adorals, the upper arm-plates, and the uppermost arm-spine. ECHINODERMATA.— EDGAR A. SMITH. 279 (Kcrguelen Island form). — Disk covered on both sides with conical scabrous granules (as described by the author) ; oral shields only slightly longer than broad ; adorals quadranrjidar (not trigonal), about n as large as the orals, almost touching one another on the inner side, their lower margins the longest ; 3 or 4) oral papilkc on each side of an angle, of "which the outermost one (or sometimes two) is broad and flattened ; arms, side and lower arm shields normal {I.e. according well with Ljungman's description) ; arm-spines slender, 9 or 10 in number, the uppermost one is shorter than the next spine, and after this the remaining spines gradually diminish in length ; the upper arm-plates arc not so wide as in the typical form. Diam. disk 12 ; length of the longest arm-spine 3^ mm. Hah. — Common on the roots of kelp in Observatory and Swain's Bays {Eaton) ; entrance to Eoyal Sound ( Wyville Thomson) ; also Patagonia. The diiferences pointed out above are hardly of sufficient importance to con- stitute specific distinction between the Kerguelen Island and the Patagonian forms. Patagonia, as Liitken suggested (Zoological Record 1872, p. 418), is most probably the true habitat of the tj^ncal specimens, and not Altata on the W. coast of Mexico (the locality cited by Ljungman), which is doubtless erroneous. The geographical range of Opliiocantha is very extensive : 5 of its species are from the northern seas, 1 from the Mediterranean, 1 is Portuguese, 1 West Indian, 1 Javanese, and lastly there is this 1 from Patagonia and Kerguelen Island. Ophioglypha hexactis. (Plate XVII., fig. 4 a-c.) Smith, Ann. & Mag. Nat. EList. 1876, Peb., xvii., p. 3 ; Verrill, BuUetin U. S. Nat. Mus. 1876, May, iii., p. 72. Discus hexagonalis, angulis propter radios interruptis, lateribusque leviter con- cavis, depressus ; papillfB orales apud angulos septense vel oetonaj, acute conicse, una ad apicem longissima, caiterai utrinque gradatim breviores ; scuta oralia pali- yel ligoni-formia, parva, manubrio latissimo brevissimo ab ore remoto, et laminA, cordata ad latera utrinque leviter concaviuscula ; scuta adoralia angustissime linearia, oralibus adjacentia ; intra angulos orales prope bases sunt scuta subovalia-oblonga bina, ct pone hos alteram minus transversum apud angulorum apices situs. Radii sex, graciles, circiter triplo-longiores latitudine disci ; scutorum inferiormu sextum post basin transversum, latissimum, margine externa in medio leArissimc angulata, lateri- bus brevissimis rotundato — truncatis, et margine orali externae simillima, nisi angulo magis prominenti ; scuta lateralia subtus conjuncta, commissuris apicem radii versus gradatbn in longitudinem excrescentibus ; scutorum superiora aliqua circiter radii medium subquadrata, paulo ab apicibus bases versus augustata, postice arcuata, ad latera recta, antice concava ; aliqua ad discum propiora subovalia, et his valde disparia, gradatim minora breviora atquc latiora, scutellis parvis multis forma ct amplitudine diversis, ad scuta lateralia conjuncta; scuta »'arfi«/ja subparva, angustc 280 zooLOcY ov Ki:ia;ri:i.KX island. subovalia, sosc longo romola ; papilUr ad latrra inscisunr disci pcr-minul:e al([UO munoi-osa> (ciivitcr 10), supra quo juxta basin bvacliii aliio sox advcrsus sex illorum in ordiuo obstaut ila ut fimbria raciant. Siiina- brai-hialcs trcs, breves, crassiuscula*, baud por-acuta>, pallidir, loquiloiig;!^ (nisi suprcmu sicpo paulo longiori) ; papilla) ambulai-rales super foramina ultima (inlVa-bracbialia) ct lissuris oris eoujunctoe, quatuor, super foramen penult imiun tres ; super pauea sequentium biua*, et super ca^tera uniea^ omues diverse formatiie, alitv; breves, compressic ct squamuli- vcl seuli-formes, alia^quc spiuis braeliialiljus subsimilcs nisi paulo brcviorcs. Discus diverse minutcquc squamulosus. Color supra uljique purpureo-niger, subtus sordide albidus. Disci diam. 21 mm. Jlab. — Very common on roots of JIacroci/slls, Observatory Bay ; also I'atajronia. Tliey require to be killed as soon as tliey arc taken out of tlio sea, as otherwise they rapidly break themselves up. O. hexactis cannot be confounded with any other species hitherto described. The number of arms, the peculiar shape of the oral shields, and the peculiarities of the ray-shields and spines distinguish it at once. The ventral portion of the disk visible between the arms is rather large ; and the oral shields are only half as loiig as the space between them and the sides of the disk. In colour and general appearance it approaches O. Sarsii, Liitkcn, of the Greenland coast, which seems to be its nearest ally ; but these species are so different from one another in detail, that it is needless to specify their distinctions. The genus Ophiogli/pha is very widely distributed ; 2 of its species are from arctic or the northern seas, 4 from the Mediterranean, 2 from between Cuba and Florida, 1 from China, 1 from Paget Sound, 1 from Patagonia, 2 from Kerguclen Island, and 2 from Sydney. Ophioglypha brevispina. (Plate XVII,., fig. 5 a-c.) Smith, Ann. & Mag. Nat. Hist. 1876, xvii., p. 112. Discus in latitudine ad radii longitudinem circiter 3 sequalis, depressus, ad latera rotundatus ; papillae orales ad angulos oris septense vel octonte, quarum ume ad apices cteteris longiores, et extremoe cseteris latiores ; dentes compressi, hastati, ad latera curvati, quatuor ; scuta oralia paulo longiora quam lata, trigono-obcordata, angulis exterioribus rotundatis, et apice subacuto ; scuta adoralia per-angusta, intus scutis oralibus subtus conjuncta, ad juncturas latiora quam extus ; infra scuta adoralia alia duo eis paulo ampliora (duplo quidem latiora) fiunt anguli oris. Ptadii quinquc, modicc longi, paulo latiores quam crassi ; scutorum inferiorum sextima post basin trans versum, margine externa leviter ciu'vata atque in medio parum acuta, marginibus lateralibus rectis per-brevibus, marginibus interioribus leviter excavatis atque apieem acutum versus convergentibus ; scutorum primum {i.e., basale vel intimum) aliis major et dissimile, margine postica excavata in loco anguli levis, et EClil^JODERMATA.— EDGAR A. SMITH. 281 marginc anticu liaud acumiuatA, — secundum tertio amplius, csetcraquc usque ad apicem gradatim minora tandomquc ibi pci-minuta ; scutorum latevalium subtus quinquc vcl sex priora disjuncta, ctutcra tamen ])er transvcrsum conjuncta, commis- suris apicem radii versus gradutim in longitiidiuem excresccntibus, — supra septen- dccim priora disjuncta, ca?tcra super dorsum conjuncta ; scutorum superiorum sextum post basin panun transvcrsum, margine externa rotundata, lateribus antice recte convergcutibus, et marginc antica, concavi-truncato, — caetera radii ' apicem versus gradatim angustata, prajsertim a fronte, itaque tandem quodquo in angulo acuto ante producitur ; squamit- disci diversaj ct imparcs, una in medio aliisque paucis parum rcmotis cii-cumpositis quam ceteris majoribus ; scuta radialia diversa, contigua, prtecedentibus subaiqualia ; papillarum ad latera inscisura? disci (in exempl. max. 22, in exempl. minoribiis circiter 16-17) sex vel septem superiora ca^teris ampliora ; spintu bracbiales tres, incrassatoe, per-breves, paulo tantum squamis ambulacralibus longiorcs, quarum binte sunt tertio, quarto, atque interdum quinto scutorum, atque ca^teris unica) ; papilla? super margines primi forauiinum ambulacraHum (infra-bracbiale) utrinque quatuor vel quiuque. Color albidus. Disci diam. 9 mm. Sab. — Observatory Bay. Several species inhabiting tlie seas of the Nortb bear a superficial resemblance to this form : — such are 0. albida, Forbes, O. robnsta, Ayres, and 0. nodosa, Ltitken. And besides these O. Lijmani^ Ljungman, from Patagonia, is very like it. 0. bvecispina resembles this last species in having very short arm-spines scarcely longer than the ambulacral papillae ; but differences in their size, and in the relative length oftheu' arms, and in the form of their radial shields, &c., afford good specific characters for their distinction from one another. Note on the Actinozoa. — By the Bev. A. E. Eaton. At Kerguelen Island, as in all cold seas, Actiniida; are not usually found between tide-marks, but occupy sites constantly under water. In high northern latitudes they seldom occur in situations less than a fathom in depth, and are usually of a red colour. At Kcrguclen Island the genus Actinia was represented by one if not by two red species frequenting rocks in shalloAV water just below the lowest limits ol' the tide, and on the roots of Macroci/stis in four or five fathoms. Tliey did not seem to be identical with any of the species figured by Gosse in his History of British Sea Anemonics. An Ilyanthus was obtained sparingly by the dredge on mud in Observatory Bay ; but the siiecimens unfortunately perished during my temporary absence from the station at the time of the Transit. N N 282 ZOOLUOV OK KKRGITKLEN ISLAND. nxmiowx.—Bt/ Professor Allnutn, M.B., LL.B., F.ll.S., P.L.S. (i'laic X\ 111.) The species of Hytboida coUoutecl at Kcrguelen Islautl and placed in my hands foi' determination amoimt in number to seven. Oi" these, one cannot he separated i'rom the widely distributed SertidarcUa polyzoiuas ;'* the others were bci'ore unknown. Among them the gymnoblastic hydroids arc represented by a single species only, a Coryne or Sijncoryne, whose nearer determination in the absence of the gonosome is impossible. The new calyptoblastic forms are represented by five species. Of these one con- stitutes the type of a new genus (Uypanthea). The others belong to the genera Sertitlarella, JZaleciuin, and Campauularia (provisional), all of which arc well re- presented by other species in our own latitudes. The species which has been referred to Campaimlaria (though in the absence of a fuller knowledge of its gonosome, only provisionally) cannot be speeiiically dis- tinguished from a hydi'oid obtained last autumn by II.M.S. " Valorous " in Baffin's Bay. It belongs to a common group of campauularian forms ; but yet the fact of identical forms occurring in such widely separated localities, though under con- ditions probably very similar, is one of great interest and significance, more especially as the distribution can hardly be explained, as in certain other cases, by the transporting agency of ships' bottoms. On the whole the hydroid fauna of Kerguelcn Island, so far as it is represented by this collection, exhibits little which can be referred to as impressing on it any- thing of a special or characteristic facies. The only imusual type is offered by the form for which I have constituted the ncAV genus Uypanthea. AU the others, notwithstanding the specific peculiarities by which most of them are distmguished from forms occurring elsewhere, belong to well-known and widely distributed types. The following are the diagnoses of the new Kerguelen Island Hydroida. HYDEOIDA CALYPTOBLASTEA. Skktulauella. Sertularella unilateralis. (Plate XVIII., figs. 10, 11.) Allman, Ann. & Mag. of Nat. Hist., '1th ser., 1876, xvii., p. 114. Trophosome. — Hydrocaulus attaining a height of about an inch and a haLf, alternately and pinnately branched, monosiphonic ; Hydrothecae deep, divergent and somewhat tumid below, slightly cm'ving towards the stem above, strongly four- toothed, all of them deflected towards one side of the stem and branches. * I do not now regard the Sertularella kcrgiiclcnsis of the preliminary notice (Ann. & Mag. Nat. Hist. 1876) as suificiently distinct from S. polyzoiiias to justify its separation from that species. IIYDROIDA.— PROl'". ALLMAN. 283 Gonosome. — Gonangia arising just below the base of a hydrotlicca, ovoitl, with a four-toothed termmal orifice; distal portion marked with wide annulations Avhich become obsolete towards the proximal end. Dredged in Swain's Bay. The mode in which the hydrothecaj, though springing from opposite sides of the stem and branches, are all deflected towards one side, causes them to appear to be monostichous, and thus gives to the species a peculiar and well-marked character. Sertularella lagena. Allman, Ann. & Mag. of Nat. Hist., 4th ser Trophosome. — Hydi-ocaulus springing from a creeping stolon, attaining a height of about an inch, slightly branched ; internodes much attenuated toAvards their ])roximal ends, and there fiu-nished with two or three well marked obhque annulations. Hydvothecse rather dis- tant, borne by the internode close to its distal end, tumid below, becoming narrow towards the orifice which is distinctly four-toothed. Gonosome. — Not known. Dredged in Observatory Bay, Kerguelen Island. This is a small species distinguished chiefly by its flask-shaped hydrotheeoe, and proxi- mately attenuated internodes with oblique annulations 1876, xvii., p. 111. Sertularella lagena. Halecium. Halecium mutilum. (Plate XVIII., figs. 8, 9.) Allman, Ann. & Mag. of Nat. Hist., 4th ser. 1876, xvii., p. 114. Trophosome. — Hycboeaulus attaining a height of about an inch, h-regularly branched ; branches with two or three oblique annulations at their origin ; inter- nodes short, each carrying for the support of the hydrauth close to its distal end a bracket-shaped process which is not produced into a tube, and which is surrounded by a narrow membranous punctate margin. Gonosome. — Not known. Dredged in Observatory Bay, Royal Sound. This species like S. macroceplwliim, Allman, from the Western part of the Gulf Stream, and R. sessile, Norman, from the Hebrides, is remarkable for the utter absence of the tubular prolongation of the lateral orifice of the internode which gives support to the hydranth in most of the species of Halecium.. N N 2 281 ZOOl.OGY OF KinUil'KLF.N ISLAND. Cami»anui-.\i;ia. Campanularia ? cyliudrica. (Plato XVIII., figs, d., 5.) Allman, Ann. & Mag. of Nat. Ilist., Ith scr., 187G, xvii., ]). 1 1 1. Trophosomc. — llyilroid attaining a lioight of ahout a quurier oC an inch ; peduncles springing from a creeping iiliforin stolon, marked with several anmilations at their proximal ends, and having at their distal (uids each a single glohular annu- lation -which immediately supports the hydrothcca ; in their intermediate portion they are slightly corrugated. Ilydrothecoc deep, cylindrical, with the margin deeply cut into about l\\c>Ive strong teeth. Gonosome. — Gonangia on very short peduncles which spring from the creeping stolon, cylindrical above with a flat summit, and then taperiag below towards the peduncle. Dredged in Swain's Bay. Beyond our knowledge of the situation and the external form of the gonangia, we know nothing of the gonosome, and therefore the reference of this species to the genus Campamtlaria is merely provisional. A form which cannot be distinguished specifically from this, has more recently been dredged by H.M.S. " Valorous" from 00 fathoms in Baffin's Bay. Hypanthea. Trop1ioso7ne. — Hydrotheciu pedunculate, inoporculatc, with walls enormously thickened and so far encroaching upon the cavity as to render impossible the complete retraction of the hydranth. Gonosome. — Gonangia enclosing fixed sporosacs. The genus Hijpaiithca is very remarkable from having the place of the ordinary hydi"otheca3 occupied by bodies which maybe said to support rather than contain the hydranths, w^hich are thus almost as incapable of being withdrawn into a jorotective receptacle as are the hydranths of the various species of Halecium, — a most excep- tional condition among the calyptoblastic hydroids. Hypanthea repeus. (Plate XVIII., figs. 6, 7.) Allman, Ann. & Mag. of Xat. Hist., ith ser., 1876, xvii., p. 115. Trophosomc. — Peduncles attaining a height of about a quarter of an inch, springing at intei-vals from a creeping stolon, having each a globular annulus just below the hydrothcca, destitute of annulations in the remainder of tlieir extent. Hydrothcca? obconieal with very oblique margin, their cavity forming distally a shallow cup, whence is prolonged a narrow cylindrical tube backwards through the axis of the hydrothcca. IIYDROTDA.— PROF. ALLMAN. 285 Gonosome. — Gouangia elongated, narrow, passing gradually into a short peduncle wbicli springs from the creeping stolon ; colonics monoecious ; the male gonangia exceeding in height the peduncles of the hydrothecse, fusiform, open- ing on the summit by a xiarrow circular orifice ; the females shorter than the males, scarcely narrowing towards theii- distal extremity where they open by a wide orifice. Dredged in Swain's Bay. The singular form of the hydrothecae, the large naked hydranths, and the greatly elongated gonangia, give to Uypanthea repens a striking physiognomy. Tlie great development of the chitinous perisarc is shown not only in the hydrothecse but also in the peduncles on which these are borne ; for the perisarc of the peduncles like that of the hydrothecse attains a great thickness and narrows their cavity in a manner similar to that by which the walls of the hycbothecse contract the space contamed by them. The lips of the orifice of the female gonangia are inverted for a short distance. In each gonangium of either sex one sporosac only is developed. This forms a greatly elongated sac which occupies almost the whole of the cavity of the gonangium. The presence in this hydroid of male and female gonangia in the same colony is another very exceptional feature. Several of the hydranths in the specimen were sufficiently Avell preserved to admit of their form and their relations to the other parts being determined. HYDEOIDA GYMNOBLASTEA. CORYNE. Coryne (?) conferta. (PL XVIII., figs. 1-3.) Allman, Ann. & Mag. of Nat. Hist., 4th ser., 1876, xvii., p. 115. Trophosome. — Hydrocaulus attaining a height of about an inch and a half ; much and irregularly branched, forming dense tufts ; stems and branches distinctly and regularly annulated. Hydranths with about 20 tentacles. Gonosome. — Not known. Gathered on 3Iytilus at or near low water mark in Observatory Bay (the tide falls only about 2 feet). Scarce. The absence of the gonosome renders the reference of this species to Coryne provisional. It might with equal reason be regarded as a Syncoryne. Its densely tufted stems with their strongly annidatcd perisarc confer upon it a well marked character. This is the only Gymnoblastic species in the collection. 286 ZOOLOGY OF KEKGUELEN ISLAND. SpuNuiuJ-t;.— i>V/ U. J. 0("iosa, H. carnom, and H. sauguiuca. To these we might add a iifth species, II. 2)finicea, for the Kerguolcn varietv diifers from the normal British form only in the possession of spicules twice the size of those of the latter. Of the three species remaining Thalysias is common to the Mediterranean and the seas between the Americas ; the Ute occurs on the N.W. coast of Spain and in the Mediterranean ; and one only, the Tethya, is decidedly antarctic. This last was the only specimen obtained from a considcn'able depth ; all of the others were either collected vni\\ the grapple within the Lami- narian zone, or were the produce of shore-collecting betAvecn tide-marks or amidst the refuse of the beach. Probably more extended research would have brouglit to lio-ht divers of the many peculiar forms which abovmd in the Cape seas and in those of the southern part of Australia. In the com'sc of my examination I have met Avith very few Forammifera, no Glohigerina, and no Coccoliths. Halichondria panicea. Johnst. Brit. Spong. 1812, p. 114, pi. x. & xi. 5. Var. — Dry specimen : Irregularly globose, 4t inches in diameter ; colour on the siu'facc white, interiorly light sponge yellow. Structure felt-like, fibreless : texture lio-ht, delicate, sub-compressible. Spicules of one form only, viz. : — long, smooth, acerate, cmwed and often slightly bent in the centre ; average maximum dimensions losr ^ 6000 iii^JJ- This specimen was picked up dry on the shore, apparently after it had been for a considerable period rolled about in the sea. It afforded accommodation to an extensive menagerie of animals, belonging to Crustacea, Annelida, and Sjjongida. An amorphous fragment of the same species gathered when fresh was preserved in spirits ; it is of a light yellowish sponge-colovu'. ]];ab. (Var.) Eoyal Soimd, Kerguelen Island ; (normal form) British coast. Isodyctya rosea. Bowerbank, Monogr. Brit. Spong., vol. ii. 282. Amorphous, encrusting the malleoloid expansions of a 3Ielohesia. Colour pinkish. SPONGllDiE.— H. J. CARTER. 287 Structiu'c isoclictyal (vide illustr. iu Bowerbk., op. cit. pi, xx., 309). Textui'c paniccous, friable. Spicules of one form only, viz. : — rather short, smooth, aceratc, curved, and often slightly bent in the middle. Average maximimi dimensions ^ inch. The largest specimen appears to have been picked up on the shore after many years exposure. Among other cxamjiles are some amorphous fragments in spiiits, gathered when fresh, of a light sponge-coloui- ; also some dry pieces still retaining the characteristic roseate tint presented by this species. Sab. — Eoyal Sound, Kerguelen Island ; also British coast. Thalysias. Thalysias, Duch. de Eonbressin et G. Michelotti, Spongiares de la Mer Caraibe, Ilarlem, ISei, p. 85, pi. xvii., 1, &c. Schmidtia, Balsamo-Crivelli, Atti d. Soc. Ital. do Scienza, 1S63 {Schmidtia ficiformis) , vol. v., taf. iv., 2, &c. Gulf of Naples. Thalysias subtriangularis, D. S^- M. Mammilerous sessile form. Dry specimen. Mammiform processes rising out of a procumbent mass. Surface even, uniformly and closely pierced with small holes. Colour yellowish brown. Vents large, single, each at the end of a mammi- form process. Pores originally in sarcodc tympanising the little holes on the surface. Structure compact. Texture close, uniformly firm. Spicules of one form only, viz. : — smooth, acerate, curved, and often slightly bent in the middle : average maximum measurements ^^ inch. Hab. — Boyal Sound, Kerguelen Island; also the Mediterranean, and seas of Central America. HaUchondria plumosa, Johiston. Hab. — Royal Sound, Kerguelen Island, on Salichondria panicea : also British coast. HaUchondria carnosa, Johnston. With spicules of one form ouly, viz. : — pin-like, smooth, with globular terminal head, slightly fusiform shaft, and fine point : average maximum dimensions '^^ inch. Hab. — The same as that of the preceding species. HaUchondria sanguinea, Johnston. With spicules of one form only, viz. : — smooth arcuate and curved : average maximum dimensions ^^^ to '^l'*^ of an inch. Tethya autai-ctica. Carter, Ann. et Mag. of Nat. Uist., Ith ser., 1872, ix. 1-12, pi. xx. Part of a large specimen ; in appearance very like grey hair matted together with sand and mud, [in form similar to a small cocoa-nut prolonged at the base 288 ZOOLOGY OF KERGUELEN ISLAND. into a short cyliudritvil pediiiu'lo as thick as ouo's wristj. iS])iciilt's, hnvj; hivg'c stout aceratos, luixcd with apparently an equal quantity ol" delieate anehor-aud- lork-hcaded long shafts. The accrates are iniu-h larger and longer than those of the specimen deserihi'd hy nie in lli(> Amials it Magazine of Natural Ilistory (h)c. cit.) averaging in tluMi- maximmu dimensions .I^ hy ,/;,,, inch. JIoIj. — lloyal ISouud, 10 rathoms, brought up by aiu-hor, Eaton ; also dredged in lat. 71° 30' S., long, (obliterated), ironi a depth of 300 fathoms, and in lat. 77^' 30' S., long. 175'' 2' E., from 20(5 fathoms, Antarctic Expedition. Ute capillosa. Schmidt, Spoug. Adriat. Mecres, p. 17, taf. i., (j-(5 a-b. White ; long and eom])ressed, sac-like with narrowed apcrtm-e ; 2 inches long, ^ inch broad, ?■ inch thick. Spicules tri-radiatc of Iavo sizes, mixed with thick, stout, rudely formed, accrates, which project hair-like above the sm'face. (In spirits ; gathered in the fresh state.) Uab. — lloyal Sound, Kerguelen Island : also Vigo Bay, dredged by Mr. Saville Kent ; and the Mediterranean. THE COLLECTIONS PROM RODRIGUEZ. INTRODUCTORY NOTES. I. The Physical Features op Rodriguez. — By Is. Bayley Balfour^ Sc.B. Situated in the Indian Ocean, 300 miles eastwards from Mam-itius, the island of Rodriguez, like the sister Mascarene Islands, is a mass of volcanic rock. A fringing reef of coral, studded with islets, skirts it on every side, extending on the west ahout three miles from the land, but at the eastern end the edge of the reef is within about a hundred yards of the beach. The island consists of a series of hills. Its extreme length, from a little north of east to a little south of west, is about 11 miles and its breadth from north to south about 5 miles. Within this base the land rises towards the centre of the island, where are several peaks, none attaining any great elevation, the highest point. Mount Limon, being only 1,300 feet above the sea level. A main ridge runs along the island in a direction parallel with its greatest diamctei", and rather nearer the southern shore. Its slopes rise with some abruptness from the sea on the eastern side, but on the west extend more gi'adually seawards, and terminate in a wide coralline limestone plain studded with elevations between Baic Topaze and Anse du Peril. The sides of the ridge as they stretch to the sea are deeply cut into ravines. The slopes on the southern side are shorter, and the ra\dnes deeper and more numerous, than on the north. In their upper parts these ravines are bordered by lofty and iuaccessible cliffs, upon which the volcanic structure of the island is well marked, and coulee is seen to succeed coulee, separated only by thin beds of cinder, agglomerate, or variously coloured clays. In some instances these cliffs are 300 feet high, and as many as 12 successive coul6es may be counted on one cliff. Through these ravines the o o 290 Till': COLLECTIONS FROM RODRIGUEZ. streams as theydoscond I'onu in Ihrir uppor parts a series ol' cascades, and sometimes hii,'Ii falls. One of the liiiest is the Cascade Viefoire at the head of the llivi^re Poiu-suite, where it falls over a eiilV more than a Iiumli'cd feet high. As the sea is approaelu'd, the ravines expand inlo wide \allcvs Ihnikcd hy gently sloping and terraced ridges. These ridges too are in some places marked hy lofty clifTs, on which the columnar hasaltic character of the rocks is well seen. A splendid example of this is Tonncrre ClilV in the valley llivic^re aux huitres, a bold perpendicular face of prismatic columns 200 foct high. At Pointc la Eouchc also this structure is conspicuous. Ou the south-west, the central volcanic ridge gradually descends, tln^ ravines become less deep, and the ground spreads out into a large coralline limestone plain. The demarcation bet^nxt the limestone and the volcanic rocdc is very sharp, but isolated patches of limestone arc met with on the surface of the volcanic region, in the vicinity of the main mass. The caves from which the bones of the Solitaire and other extinct birds have been obtained occur in this limestone plain. Some of them extend for a great distance through the rock, and are rich in stalactites and stalagmites ; others again arc mere small holes. The avIioIo plain is riddled with these caves, and on walking over it one constantly passes small apertures and fissures, evidently "blow-holes" of some subterranean cavern. Wide and deep hollows arc also met with, on the floor of Avhich largo fragments of limestone lie in confused heaps. These are apparently old caves of which the roofs have fallen in, and the continuation of the cavern may be found at either extremity. Tlu; floor of these hollows is composed of volcanic soil, often with large masses of volcanic rock on the sm-face, and commonly clothed richly with vegetation. It is in such places that many of the largest trees on the island are now to be seen. The lime- stone is not found along the northern or southern shores, until we near the eastern extremity, Avhere patches occur at the mouths of the valleys, and even at some distance from the shore. One mass I discovered in valley llivitirc do I'Est, more than a mile from the sea. It is not so abundant at this end of the island. On the southern shore between Eiviere Palmiste and Riviere Poursuite, in- dications of raised beaches are seen, reaching about 20 feet above the sea level. The existence of these masses of coralline limestone indicates clearly a former lower level of the island, and the evidence of raised beaches confirms this. But a consideration of the coral reefs points as clearly to a time when the island stood at a higher level. The present coral reef fringes the coast, extending, as I have mentioned, about three miles on the south-west side, but coming close inshore on the east. An older reef, however, exists now quite submerged in some places to a depth of over 90 fathoms. Upon it the present reef rests, and it extends westwards nearly 15 miles from the present coast, while on the east it stretches about six miles. INTRODUCTORY NOTES.— IS. BATLEY HALFOUK. 291 We have thus proofs of great and intermittent oscillations of the level of the island. Of the islets scattered over the reef, some are volcanic, and the others are com- posed of coralline lunestone and sand. Thoy are all within the compass of the present reef, and only occur on its wider parts ; consequently there are none east of Port Matlumn on the north, and of Port Sud Est on the south. Eight islets are of volcanic origin. Diamond, Boohy, Katrine, Marianne, Desin6e, Frigate, Crah, and Ilcrmitage. Only the last mentioned is on the south reef, the rest range round from the south-west to the north. The coralline limestone and sand islets are more numerous, and arc confined to the southern and Avestern reefs, none occur on the north. Ten of them receive names, as follows : — Gomhrani, Pierrots, Platte, Pantad, Pianqui, Misel, Chat, Zozo, Coco, and Sahle. The last two are mere accumulations of sand and coral debris close to the western edge of the reef, the others are all on the south. It will be seen, then, that whilst the volcanic islets are chieflv on the north and west, the limestone ones arc on the south. The prevailing tjpG of rock composing the island is a dolerite rich in olivine, in many places greatly decomposed. Interspersed with the coulees of rock are found extensive beds of clay. These, which possibly result from the decomposition in situ of the dolerite, are highly coloured, usually bright red or ochre, and form a prominent feature in the landscape. The lava coulees seem to have flowed with great regularity, but there have been marked periods of intermission of volcanic energy. Here and there dykes are seen, one specially well-marked occurs at the seaward end of the Charpentier ridge of the valley Eivi^re Bouteille, forming a conical projection through the layers of rock ; and it belongs to the same period of formation as do the higher parts of the island, — apparently the last outburst of volcanic activity. The exact position of old craters is difficult to determine. Probably there were many foci, but the main ones seem to have been situated about the Grande Montague and Mount Malartic. Many of the small conical isolated hills, such as Montague du Xord, scattered over the island, no doubt also mark the site of old foci. The time that has elapsed since the last exhibition of activity has been sufficient to allow of a considerable amount of denudation, as is evidenced by the fragments of rock and debris with which the more level ground is strewn, and which cover the gentle hill slopes, rendering progression a matter of difficulty ; and the smooth and rounded outline of the hills, only occasionally intermitted l)y a projecting torr or pinnacle, as well as the deep ravines, testify to the same. A curious feature deserves notice. In the Baie aux Huitres are found masses, about a foot in diameter, of shells of a mollusc embedded in or rather cemented by a material resembling tuif. Unfortunately the specimens were lost in transmission, and thus a further determination of their nature is prevented. 00 2 292 THE COLLECTIONS FUO.M KODKIGUEZ, The island is coiuparativcly dry, the soil is parched and avid, ai^l durint;- Iho warm season many of the streams are dried up. ]kit the size of the wat(n'-courses and the enormous hoidders fillinc: their hculs, indicate lar2:(^ torrents in the rainy season. In some places, issuing from the clay, sprini:;s occur, of which the water is hrackish, has a very disagreeahle taste, and is slightly tepid, hut has no smell. As a rule the water of llie streams is good and safe to drink, hut that of some rivers, notahly the Rivere SaumMre, is most unpalatahle and apt to cause slight catharsis. The climate is much like that of ]\[auritius, whore the average annual tem- perature is ^S" Fahr. During the north-west monsoon from Novemher to April, the weather is "wet and warm, and frequently in the iii'st months of the year llie island is Ansitcd hy severe hurricane's. From May to Octoher the soutli-east monsoon prevails, and then the weather is cool and dry. The rainfall is exceedingly irregular, the hills heing hardly high enough and not sufficiently wooded to arrest cloud ; hence also fogs arc rare. 293 II. Heports of Proceedings op the Naturalists. 1. Report of Dr. Is. B. Balfour. I LEFT England upon the l.st Jul_y 1874, and embarked at Marseille.s on board the Messa- gerie.s Maritimes steamer. I reached Mauritius with my confreres on the 4th of Aun^ust. On our arrival there we met Lieut. Neates, the head of the expedition to Rodriguez, who had come by the troopship " Elizabetli Martin " some days previously. H.M.S. " .Shearwater " with the other astronomer.'! and the asti'onomical in.struments an-ived at Mauritius on the Otii, and as there was no cause ibr delay there, Captain Wharton decided to sail for Rodriguez on the 11th. This was carried out, and after a week's voyage the .ship anchored in Matliurin Bay, Rodri-niez, on the afternoon of Tuesday the IStli. Owing to tiie difficulty of carrying heavy baggage over the reefs, rendered doubly so by the neap tides which then occurred, .some days elapsed ere we could take up our abode on the island. These days I spent in makiug an examination of the general features of the island. On the 2Sth of August I quitted the ship, and went to live on shore, and at once commenced my botanical and geological work. As I have already explained in niy letters to the Royal Society,* the flora is very scant owin"' to the frequency of the fires which have swept across the i.sland. Indeed fully one half of the island is entirely bare of vegetation, save the few social weeds which now occupy the place of the destroyed indigenous plants. The only parts of the island where any amount of vegetation is to be found are the upper regions of some of the valleys, round which the fires have evidently passed, leaving them comparatively unscatlied ; but even here the plants are few and the dryness of the soil precludes any luxuriance. I explored thoroughlj^ all those valleys which promised any botanical treasures, and I regret to say the result is very poor. In accordance with the wishes of Captain Wharton, I had all my specimens packed and on board the "Shearwater" by the day of the Transit. On the loth of December we quitted Rodriguez and sisrhted Mauritius on the 18th. I remained in Mauritius until the 5th of February 1875. During my stay I spent much time at the botanical gardens examining the native flora, and I also made excursions over the island for the purpose of investigating the species of indigenous Pandani. Specimens, the residt of my work, I have transmitted to England. Leaving Mauritius by the mail steamer on the 5th February, I reached Bourbon next morninf^. Here I remained investigating its rich and little known flora. Unfortunately bad weather durin"- ray stay rendered botanical and geological explorations both unpleasant and diflicult. I amassed a considerable number of specimens illustrating the vegetation of the island, and I specially devoted some time to the examination of the Pandani. On the Gth March the mail came in, and I had reluctantly to leave Bourbon. I reached Maraeilles on the 31st March and London on April 2nd. Is. B.vYLEv Balfour. * See Proe. Roy. Soc. 1874, p. 135. 294 Till-; (OI.LKC'TIONS KKOM KODKIOUKZ. 2. Eepout ov 11i;nuy TI. Siateu, Esq., J5.A. I LANDKH in Alauritius Aiisxiist 4tli, ls74. Tlio Hon, Edward Kowton kindly uskod me 1o stjiy with him. .My instnictions directed mo to take lii.s advice as to my mode of iirocediirc in Rodriguez, and accordingly I rciiuested George Jcnner, Esq., Director of Eniigiauts for the (iiand Port District, to select for me nine good men and a cook. This took sometime and 1 regretted much that .1 was notable to leave for Ivodrigue/. until September 9th, by the second tri]) of the " Slioarwater." I was wishing at first, nevertheless, to sail Avithout my men, had not Mr. Newton and Mr. Jenner (10 years magistrate of R!)driguc/.) as.sured me that I could do nothing without my men, and that the natives of Rodriguez wi'ie quite unreliable. We landed in Rodriguez on Sejjtember 14th, with our 10 men. Settled in Tort Matliurii. anil stayed there for a day to allow the men to recover from the voyage, which had been stormy ; Oioverument House had been allotted to the naturalists as a residence and head-quarters, but two tents had been supplied from the " Shearwater," one of which was to be my habitation at the caverns. After this the weather was stormy for some days, .so much so that it would not Lave been prudent to attempt the sea voyage to the caves, and no native would pilot one there ; but afterwards the weather changed and I secured two boats, in which, upon the 18th of September, I carried my men, tent, stores, and tools down to the caverns. Owing to the roughness of the country, which is very hilly, it would have been impossible to have carried the stores, &;c. overland, without many porters and great expense. TlielOth was occupied in establishLng camp, in building a hut for the men, and in j)ros]iccting for caves, of which I found several. Next morning we started work : the caves had all been previously dug ovei-, but I determined to dig them again, and to a depth of three feet, and we found a good many bones (mostly tortoi.se, but with some of Pezojjhaps) though they were iu a bad state of pieservation. I used to go out every evening to take my daily exercise and to shoot game (no fowls or goats could be got at the caverns, which part of the i.sland is nearly uniidiabited, so that it was neces.sary to shoot a supply of game), and on these walks I found all the caves. This sort of work, i.e., digging over caves second band, continued till the 6th of October. I used to go home always on Saturday afternoon with the week's bones, stay over Sunday, and leave Government House for the caves on Monday morning in my canoe, carrying a week's rations for the men and my own supplies. On October 7th I found some new caves. These formed a branch of a cavern already dug but as entrance to them was extremely difficult, they had hitherto escaped notice. In the.se we found a large quantity of Solitaire bones and the almost perfect skeletons of a male and female mingled, including the rings of the trachea; very few tortoise bones were intermixed, and I found it an almost invariable rule, that where Solitaire bones were found in large numbers and apparently occupying the spots where they died, there were few or no tortoise bones amonliinij' lo dilVeivnt varieties, For they pass throiigli interniecliate stages. The relsjjar iu most of the roeks hecomes proiuiiieiit in indi- vidual erystals, and iin])arts a micro-pori)li\ ritic cliaraeter to the roek, iiidei)endeutly of the Olivine. Of L;roiind-mass, strictly spcakiiin. tlicsc dolcrites are devoid. Tlic mass oi" the roek is made u\) of the interlaced felspar crystals, the iuterstiees heing filled with a hrown, and in liner grained varieties a nearly opacpie, mineral, which sometimes crystallises in distinct forms, and in its less distinct variety is only to be recognised as crystalline by e.xaminatiou with a high power between crossed tour- malines. It is nndoubtedly Augite. In some instances the Augite is to be recog- nised in quite distinct and transparent short pal(> greenish l)rowu erystals, quite imattacked by decomposition. The Olivine which often porphyrises the rock by its abundance and distinctness oF crystallisation rises into an inijiortant ingredient of these dolcrites. Some of them might indeed be termed Olivine dolcrites, from the abundance of this mineral. The Olivine belongs to a variety rich in iron, and has undergone the usual characteristic decomposition; the mystals being encrusted by, and sometimes entirely converted into, a dark reddish brown mineral, which lias also penetrated the crvstals wherever fissures have affordcfl lines of entrance bv which acid or thermal waters could attack them. In one or two of the rocks of iiner grain a mineral closely resembling Chlorophteite can be recognised. Minute grains of magnetite, the rectangular or nearly rectangular sections of which can often be recognised, are abundant in some specimens of the rock, and are seen very evenly distributed in the mici"Oscopic sections. A very characteristic accompaniment of the felspar is Apatite, a mineral which presents itself in long transparent needles and microlites. The presence in some little abundance of phosphoric acid is attested by treating a solution of the mineral, freed from its Silica, in niti'ic acid by Ammonium Molybdate, confirmatory of the anticipations formed with the microscojjc. The Momrt Grenade Dolerite is liighlv vesicular, and like that from the east side of English Bay contains more Augite than that from Pitou, or from what would seem to be the older laA"as from the bottom of the Cascade Valley. Passing to the other extreme of the rock series, that, namely, in which the grain is finest and the texture most dense, we have a few specimens which, as regards these characters, may be classed as somewhat coarse basalt, or as fine-gTained ana- mesite. Their colour is generally dark in contrast with the coarser grained lavas. The eye detects no distinguishable minerals when aided by an ordinary lens, except that a few crystals of felspar and numerous crystals of olivine are to be seen enclosed in the formless mass, and that in one dense rock from the Tonnerre Cliff a certain silky sheen can be detected in the fractured surface, which is due to a stream-like distribution of the felspar crystals. In the microscope the appearance of those finer rocks is quite similar to that of Is'. S. .MASKELYNK. 209 the dolerites. The Tonnen-e Cliff rock, dark grey in hue, is under the microscope a beautiful congeries of minute blades of Labradorite streaming in directions that run along and round the large crystals of Olivine, and a great number of very small black points of magnetite mingled with some pale yellowish green Augite. These basalts, which are at once the finest in grain and hardest in texture of all the Rodriguez rocks, seem to form the material of the columnar masses of the island, though not confined to these. The localities noted by ^Er. Balfour on some of the more characteristic of them are, Ijcsides Tonnerre Clifl", the locality of the finest columns in the island ; Beline high fall, the rock in the bed of the cascade brook, a columnar inass in tlie Cascade Valley, and the valley on the E. side of English Bay Point. The banks of the river Poursuite yield also a fine basaltic rock ; sometimes vesi- cular and charged with Aragonite, a Magnesian Lime- carbonate and Chloropha?ite in fissm-es, and Zeolites, among which striated little crystals of Chabazite are distinctly recognisable. The felspar in this basalt ofiers indications of commencLng decomjiosition. And similar minerals occur in other vesicular varieties, chiefly of the finer- grained rocks. A curious feature of some of the Rodriguez rocks (and among those collected it is confined to the Anamesites and fine-grained and more strictly basaltic varieties) is seen in the tendency of the rock to break into unevenly roundcxl fragments ; this variety of rock possessing but little tenacity, and being without dilficulty separated into knob-like pieces. This nodular texture is traceable in the more pronouncea ■cases to a great number of curved fissures traversing the mass, which are perhaps the result of the conditions under which it cooled. Not improbably they may be due to an imperfect development of a kind of pyromerid structure, in the produc- tion of which the felspar may have been the formative ingredient. Another peculiarity of certain of tliese Rodriguez Anamesites is probably due to, and in part only a dift'erently developed effect of, a similar cause. These show a spotted or variolitic siu'face, in which very light spots are seen on the darker slaty gray ground. Usually they are about an eighth to a quarter of an inch in diameter, and are rarely confluent. Microscopically examined the rock is seen at the spots to be nearly devoid of Olivine, whereas this mineral is very plentiful in the other parts: while microscopical inspection reveals in some of the spots a peculiar structure in the felspar, which is here and there grouped with some of its crystals rtidely tan- gential to a circle, at some distance from a larger crystal or crystals of the same mineral, which forms as it were the centre or nucleus of the spot. It would seem that here also the spot is due to the tendency of the felspar to originate an orbicular rock structure, analogous to, but by no means resembli]ig, in the grouping of its ciystals, that of the orl)icular corsite. It is to be noted also, as bearing on this point, that the spots in the true variolite seem to be due to a peculiar segregation of p p 2 300 PETKOLOGY OK KdnK'ICl'KZ. Labi-adoritc. The vaviolitic strvicture is especially noticea1)l(' in a rock froin llu- GriMiailo Bay. .\. variety ol' the basalt I'idiu the bed ol' the torrent in tlie Cascade Valley is re- markable for a peculiarity converse to that just described. Here the base of the roek pi'cscnts the pale grey hue of that of the spots in the rock I'roui the Grenade Bay, but is flecked with tortuous stripes, and with spots of darker hue. At these parts it is seen in a microscopic section that the felspar is less, aiid the augite more abundant in the dark spots and stripes than in the paler ground. The felspar is here in somewhat larger crystals than in that of the Tonnerrc basalt, but presents a similar stream structm-c. There remain for consideration certain substances varying greatly in colour from a dull brick red to a liver-brown, and presenting the features of earthy clays, of which j\Ii\ Balfour collected a very complete assortment. It would seem that they form intermediate beds, intervening between two layers of regular basalt ; as in the case of the Cascade Valley, where the most complete series of them was made. The overlying basalt has evidently exercised on the upper layers of this stratum the colouring action to be expected from a molten lava acting on a clay bed ; it has given the clay a brick-red colour, and so far hardened it. The lowest bed is of a brown hue, and appears unchanged by heat. The inter- mediate beds vary in hue from reddish yellow and grey to a deep chocolate or burnt sienna brown. That these clay strata are the result of the action of water on the rocks of the island seems borne out by theu' composition, which is shown by a qualitative- analysis to be that of a ferric-aluminous silicate, almost entirely free from calcium or magnesium compoimds ; a residue, in short, left after these earthy and the alka- line bases had been removed by the action of carbonated waters. In this respect they no doubt correspond to the red earth of Mauritius, the history of which would appear fi'om the scanty notices (Mauritius, by Rev. F. Flemyng, pp. 18 and 19) of it, to be that of a disintegi-ation of the volcanic rocks of that island ; though, from the description, it would seem that it has a more pisolitic structiu'e, and is richer in iron oxide than the corresponding earth in Rodriguez. Remotely, too, it may be compared with some kinds of laterite. !Mr. Bayley BaKom* made some careful observations on this earth with a view to establish its petrological relations and history ; and the following extract from his journal will show the sort of evidence which he derived on the spot in favour of its being a product of the alteration of the lava-rock of the island : — " This red bed, as I call it, is evidently the same as the one wliich I noticed further down, about a mile from the town, and on the opposite side of the valley. I could trace that one a good way up the valley ; but eventually it became hidden by the boulders and scrub, but its presence was indicated by the fragments at the side of the bum below. This bed, at the point where these specimens were taken. N. S. MASKELYNE. 301 is about a foot thick, and is well seen on both sides of the valley, the intervening mass which has been washed away being seen on the face of the Fall Cliff, the distance being about 30 yards. It is to this soft bed that the fall is due, as it lies below a harder rock, of which 64 is a specimen. My specimens show more or less of a gradation from normal basalt to this clay. This clay may be the same as the Terre rouge of Mauritius, but as I have not seen the latter, I cannot say whether it be so or no. It is evidently decomposed lava, the contained crystals of minerals being numerous ; but I have not yet analyzed them. I cannot think of any cause for this decomposition. I observed at the place about a mile from Mathurin, where I first picked this red clay, a spring of water rising just at the top of this red bed, and coming along a tunnel of about 4 inches diameter out of the side of the hill. The opening being about 3 feet above the burn level, and about 4 feet from the water as it then was ; but in wet seasons the burn must rise fully a long way above its opening. However, the water of the burn was good to drink, and beyond the usual earthy taste, possessed more or less by all the water here, it had no taste ; but this spring water was abominable, warm and brackish, specially noticeable on contrasting it with the bm-n water. It may be that this spring, containing as it clearly does some chemical substances in solution, in addition to the usual consti- tuents of water, may have some influence and should ; as it seems probable from facts I have learnt from inhabitants, as also from observation, that many such exist. We may have here a cause of the rapid disintegration. At the Cascade itself, on the E. side, where I picked my specimens, I noticed a hole in the rock just above the red bed similar in form to that through which, below, the spring water emerged, but there was no sign of any moisture — contrarily, the rocks here were very dry." That it is the result of water-worked disintegration of the basalts and dolerites, the analysis renders more than probable. But whether it does not in fact mark an epoch of pause in the volcanic activity of the island, or at least in the N.E. end of it, during which a long process of denudation and water action supervened over a then submerged area, would need a more minute and extensive topographical study of the district than has yet been given to it. There are veins of a white waxy Aluminium silicate traversing the brown or yellowish brown clays, which probalily fill fissures produced by the drying of the original mass, which in some cases, as in tliat of certain specimens of the red baked upper layer, seems to have assumed a quasi-columnar structure. B 0 T A N Y. By /*'. Buyleu Balfoni'. IXTRODUCTOUY REMARKS. WiiKN in Kim Lo2;uai sitilited tlio island ot" llodrig-uez his eyes fell on mouiilains " richly spread with iiivat and tall trees," over Avhich flowed stn^anis Avilh l)anks " adorned with i'orests," and altogether the scene was such as to call iorth from hinr the designations of "a lovely isle," " an earthly paradise," " a little Eden." lie speaks of it i litis (Xcw Voyage to the East Indies, p. 248) : — " 'Tis, as I have hinted, composed of lovely hills, covered all with fine trees, wliosc ])erpetual verdure is " entirely charming;" and "between these great and tall trees one may walk at " ease and find such refreshing coolness in their shade at noon, so sweet, so healthy, " that 'twould give life to those that are dying. Their spreading and tufty tops, " which are almost all of an equal height, joyn together like so many canopys " and umhrellos, and jointly make a ceiling of an eternal verdure, supported by " natm'al pillars which raise and nourish them." Such is his picture of the aspect of the vegetation of Kodriguez. Is such its character now ? No. The great and tall trees have now almost entirely disappeared, the eternally verdant caiiopy formed by their boughs no longer exists, and the "little Eden" is noAv a dry and com- paratively barren spot, clothed with a vegetation mainly of social weeds, and destitute of any forest growth save in unfrequented and more inaccessible parts in the recesses of the valleys ; and, we may ask, AA'hat has wrought this change ? We find in the history of the vegetation of Rodriguez a case similar to that of St. Helena. The same causes which destroyed the peculiar and most interesting Flora of that fertile island have operated, and I regret to say still operate, in Rodriguez, and have effectually changt^d the face of the island. Goats, introduced long ago, are now found in enormous luimbers, eating the young shoots and leaves of any herb, slirub, or tree witliin their reach ; and now too several thousand head of cattle graze on the island, and effectually keep down the vegetation of the spots they frequent. Fires have occurred with great frequency, and every now and then at present sweep across parts of the island, destroying everything within reach. And then there are introduced foreign plants. These are now in great abundance, and in many cases completely occupy the ground to the exclusion of the native vegetation, Avhicli is driven to the secluded parts of the island. Perhaps one of the best instances of their poAver in this respect is found in the case of " L'AcAicie" {Lucceua glcmca). This plant, introduced about 30 years ago, is now found covering the ground for acres, forming so dense a scrub IS. HAYLEY BALFOUR. 303 that it is impossible to penetrate it, and beneath which nothinsj will grow. Originally planted in the valley near Port ^Mathurin, it is now found in almost every valley in the island, si)reading from the banks of the streams up the sides of the valley. It owes its spreading in a great measure to the cattle and goats, which are exceedingly fond of the leaves and pods, and thus the seeds are carried about. Finally, a certain amount of destructive influence is attributable to the settlers in- discriminately cutting down the trees over large tracts. This, however, has now received a check, as the cutting of timber is forbidden by law. These agencies, then, have diiTctly etfected tlie destruction of a great part of the vegetation of the island, so that over large areas hardly a tree or shrub is seen, and the ground is covered by only a scanty clothing of grass and tropical weeds. But tires and the hand of man, through the alteration in the climate consequent on theii' destruction of the foi'ests, have effected indirectly a more permanent injury on the Flora ; for now we have a bare, parched A'olcanic pile, with deep stream- courses for the most part dry, in place of the verdant well-watered island of 200 vears ago. Can we wonder, then, that we find but a remnant of what we may consider the old vegetation still extant in llodrisj^uez ? That the island had orii^inallv a rich Flora there can be little doubt, judging from its j)osition and from analogy with the sister islands. Uufortunatelv Leijuat, from whose account we derive all our early infornTation regarding the island, docs not enter in any great detail into the native plants of the island at the time of his visit, occixpyiug himself more with those which he and his companions found useful, and with those they introduced into the island, and we thus have really no record of the exact nature of the primitive luxuriant vegetation. He only mentions 10 plants as found on the island, and these, though often rather curtly described, I have been able to identify in eveiy instance save one, at least generically. The folloAving is the list of plants he men- tions, and alongside of each is placed the name of the plant Avith which I believe thev can be identified : — Purslain ... - Porliihtca oleracea. Tree with fruit lik(^ olive - - JLhuodendroii orienfale. NastvTiee . - - - Clerodendi'OH laciiilatitm. Pepper - . - - Capsicum frutescens. Ebony .... Diospyi'os diversifoUa. Plantane ... - Latania Verschaffeltii. J Dicfiiosperma alhn var. aiirca. Pavilion . . - - Paudoinis. Eodrigo Kesta - - - Ficiis.) Flower white as a lily and like a jessamine. \lIyopliorbe Verschaffeltl 304 I^OTANV OF RODTJIGUEZ. 1 know of no jilaiit on tlu' island wliich answers the dcscnption ol" tlic one last mentioned, wliic-h runs as lollows : — " Tlioro's a certain admirable flower in this " island which I should prefer to Spanish jessamine, "tis as white as a lily and " shaped some think like common jessamine, it grows particularly out of th(> t runks " of rotten trees, when they are almost reduced to the substance of mould. 'I'lie " odour of this flower strikes one agreeably at a hundred paces distance." I can only suppose it to be some species of Orchid. The comparative absence of Orchids, as I shall show, is a notable feature on the island, and it is probable that Leguat refers to a species now passed away. The vegetation of llodrigucz at the present time is thus very dillercnt from what ir Avas at a comparatively recent period, and a very potent influence in altering its character has been cultivation. The island, though an outlier of the Mascarene group, at a great distance from the other members, and, as it were, out of the way, has seen many changes m this respect. Leguat gives a curious account of the cultivation of the soil by himself and bis companions, and of tlie plants they grew. He says they sowed "Water-melons, Ordinary Melons, Succory, Wheat, Artichokes, Pm-slain, Turnips, Mustard, Gillyflower, Clover-grass. Some of these seem to have tln-iven, but others did not, and it is curious to read of the wheat, " Of three grains of Avheat that came up, we could preserve but one plant ; it bad above 200 ears, " and we were full of hopes that it would come to sometliing, but it produced " only a sort of tares, which very much troubled us as you may imagine. How- " ever, we should not from hence conclude that wheat corn will always turn to " tares here, since in Em-ope such like degenerations arc often to be met with." His observation rather militates against the plant being true Wheat, — more probably Millet. Some of these plants are now cultivated or occur spontaneously on the island. When the island was in the possession of the French many settlers lived and cultivated large estates, but with the liberation of the slaves cultivation de- creased in amount. It thus happens that at the present day only a relatively small acreage of laud is cultivated. The staple of the cultivation now is the Patate, or Sweet Potato, which is grown very widely, and in almost equal quantity is the Manioc, whilst of other roots Yams are chiefly grown. Of the cereals fair crops of Maize and Millet are obtained, and Rice also grows veiy well, but is not cultivated in quantity svifficient for the use of the inhabitants. Wheat formerly largely sown is now seldom seen, and this mainly because of the parroquets and Java spai-rows which aboimd. Beans {Fhaseolus Inaafus), Lentils {JEi-vum lens). Gram {Cicer orietinum), Dholl {Cajanns Indices), Pistache {Arachis hypogea) are all grown to a certain extent, though the rats are great enemies. Of other vege- tables Garlic {AU'mm sativum), Giroumon (Cuctirbiia Pepo), Margose (Momordica balsamina). Melon d'eau {Citrnllus vulgaris), Oignon (Allium Cepa), Papingaye {Loffa acutangulo^, and Patole {Trichosanthes anguina) are the most common. Ginger, Safran (Turmeric), and Arrowroot are also cultivated. Of economic plants IS. BAYLEY HALFOUK. 305 Coffee has formerly been largely grown, but now is never cultivated, as the hun-i- canes so frequently destroy tlie crops. Fair-sized trees, the remnant of old plan- tations, arc now found fruiting freely. Vanilla grows well, and is tried in some spots. The want of Avater is now a great difficulty in the way of manipulating the Sugar-cane, which is therefore not much grown. Formerly Indigo plantations covei-ed some of the central portions of the island, but now its cultivation has ceased. That the soil of the island is good, and that water was formerly abundant, the fact of so many plants being cultivated clearly proves ; but at the same time their cultivation must have acted very prejudicially on tlic indigenous vegetation. From all these causes then indigenous plants have suffered, and the aspect of the vegetation now is a very peculiar one. The elevation in the island is not sufficient to render possible any marked difference betwixt the vegetation of the higher and of the lower parts, and the relative amount of moisture is about the same in the two regions, as the hill-tops are seldom enveloped in mists. But we do find some plants wliicli only occur in the upper parts, while others are found only on the lower districts or on the shore. Although altitude does not much affect the vegetation, difference of soil does so to a great extent. In most places the soU is volcanic, but there are many wide expanses of upraised coral reef forming a limestone soU, and, as we might expect, there are very marked differences between the vegetation on them. Commencing at the shore we find first of all that we have representatives of a Phoenogamic Flora below high- water mark in two species of Salophila, abundant on the reefs, and which are also found in many other tropical islands ; but we miss the Mangrove Avhich is found in Mam'itius. At the mouth of some of the rivers Riippia maritima and ZannichelUa palustris occui'. The shore at high-water mark is freely strewn with Sesuvium portulacastrum, while Ipomxa pes-caprce, Canuvalia obtusifolia, and Zoysia piingens, are found carpeting sandy flats. Of other shore plants Psiadia Coronopus is sparingly found, and so is Phyllanthiis dumetosus, both only on the south side of the island, whilst Clitoria Ternatea, Teramnus labiaUs, JBoerhaavia diffusa, and Achyranthes aspera are also found abundantly on the shore. Of trees Hibiscus tiliaceus forms dense thickets close to the sea, and with it Thespesia populnea and Pisonia viseosa are also found. Where coralline limestone exists on the shores and on the coral islets on the reef such plants, as Suriana maritima, Pemphis acidula, Oldenlandia Sieberi, Tournefortia argentea, Ipomfjpa froyrans, I. leucantha, I. nil, Lycimn tenue, and Myoporum mauritiannm, specially occur. As we pass inland we meet in the valleys at the embouchures of the rivers -ndth the folloAving plants, — Cardiospermum microcarpum, Ccesalpinia Bonducella, QQ -}- 306 BOTANY OF RODRIGUEZ. Flii/salis perurifDia, Dafiira alba, liiciDiKS communis, Eriillirlna iiidica, Carlca Fapaya, and Coi.v Lachrt/ma, in al)imcIanco; and, on. continuing up the rivers. Nasturtium ofllciuole, Herpcstis 3Loiniiena, Alocasia macrorJiiza, Colocasia anti- quorum, C/nrra Commersoui, and such Algic as species of Batrachospcrmnm, Cla- dophora, and Conferva are found in theii" waters. The banks of the streams in most valleys are covered, especially in the lower parts, for about twenty yards on each side of the stream by a dense thicket of Leuccena glauca, giving place in the upper parts in many valleys to Eugenia Jamhos, and close to the stream may be found O.valis cori/mbosa, Ui/clrocofi/Ie honariensis. Salvia coccinca, Flantago major, and Ixumex crispus. On moist rocks at the tops of valleys Lobelia vagans and Filea Balfouri are abundant, and many Mosses and Algce are to be found coating the rocks along with Hepaticw and Trichomanes cuspidatimi. The imdergrowth is very rank in many places, and the plants which most commonly contribute to its formation are such as Ilalvastrnm tricttspidatum, Sida carpiuifolia, Abut Hon indicum, TJrena lobata, Gossypinm barbadcnse, Ilelochia pijramidata, Corchorus trilocularis, Triumfetta glandulosa, Oxalis corniculata, Crofalaria retusa, Ati/losia scaraboeoides, Rhynchosia minima, Rubus rosccfolius, Ageratum conyzoides, Vinca rosea, Trichodesma zeylanicum, Stachytarpheta indica, Aehyranthes aspiera, Cassytha Jiliformis, Commelyna communis, Nep)hrolepis acuta, and species of Cyperacecv and Graminecc. Rocks and stones are everywhere covered with Lichens, chiefly species of Lecanora, Lecidea, and Pertusaria. On the slopes of the valley are found occasionally such plants as ToddaUa aculeata, Gouania rctinaria, Scutia Commersoni, Indigqfera argentea, Tcp)lirosia p)urpurea, Canavalia ensiformis, Daticus Carota, Danais corymhosa, Eupatorium cannabinum, Phimbago zeylanica, Tanulepis splienophylla, Meliotropium indicum, Solamim sanctum, Barleria Frionitis, Agave americana, Fourcroya gigantea, and A loe loma tophy lloides. The commonest tree intermixed with these is Fandanns Jielerocarpus, and on the higher parts of the island P. temiif alius ; but the following trees and shrubs are also very common : Fittosportim Senacia, Quivisia laciniata, Elteodendron orientate, Alhizzia LebbeJc, Terminalia Benzoin, T. Catappa, Foetidia maiiritiana, Matliurina penduliflora, Fernelia buxifolia, Fyrostria trilocularis, Scyplioclilamys revoluta, Carissa Xylopicron, Ardisia sp., Olea lancea, Securmega durissima, Ficus rubra, F. consimilis, Dracaena rejlexa, Bodoncca viscosa, Eugenia tinijlora, E. cotinifolia, Funica Granatum, and Fhyllanthus Casticum. Confined to limited areas in unfrequented spots a few plants such as the following are found : Apjhloia mauritiana, var. theceformis, Bombeya ferruginea B. acutangula, Zanthoxylum paniculatum, Allophylus Cobbe, Sclerocarya castanea, Eugenia Balfouri, Bandla heteropliylla, Fsychotria lanceolata, Fsiadia rodrigue- idana, Sideroxylon sp., Bnddleia madagascariensis, Eypocstes rodriguesiana, Obetia IS. BAYLEY BALFOUR. 307 Jicifolia, Peperomia Boclrigueziy P. and Bulbophyllum incurvitm. Ill the viciaity of habitations j)lants as : Anona muricata. Aryemone mexicana. Bmssica juncea. Gynandropsis pentaplujlla. Mot'iiiga pteryyosperma . Eriodendroii anfmctiiosum. Tviphasia tvifoliata. Citrus decumana. Indigofera tinctoria. A hi' US precatorius. Mcem a toxy Ion CampecManum. JPoinciana regia. Phaseolus lunatus. Cajanus indicus. Acacia Farnesiana. Frtoms communis. JEitgenia Jambolana. Lagenaria vulgaris. Momordica balsamina. Citrullus vulgaris. Opuntia Tuna. Cqffea arabica. hirta, Viscum tanioides, Oberonia brevifoUa, or old plantations -n-e usually meet with such Barthenium Hysteroplioriis. Siegesbeckia orientalis. Bidens pilosa. Lobelia CUjfortiana. Ipomcca Batatas. Lycopersicum Galeni. Solanum Mclongena. Nicotiana Tabacum. Mirabilis Jalapa. Leonurus sibiricus. Amarantlms tristis. Chenopodium ambrosioides. Basella rtibra. Persea gratissima. Tetranthera laurifolia. Euphorbia peploides. Bliyllanthus Niruri. Manihot ntilissima. Musa paradisiaca. Bavenala madagascariensis. Dioscorea alata ? On looking at the vegetation as it now clothes the island, one perceives at once that a line may be drawn across the island which will separate portions very different in aspect. Thus starting from the mouth of the Rividre Saumatre, and passing up to the head of the valley, thence striking somew'hat south-west to a point on the opijosite side of the island about the mouth of Eivi6re Coco, we have on the east a lofty district intersected by many deep ravines, the slopes coming somewhat abruptly down to the sea. The sides of the hills are in this region covered with a thick undcrgroA\th and scrub, often in great part of Ferns, and dotted over them arc a fair number of small shrubs and trees, notably abundance of Screw-pines. The valleys themselves in their upper parts are here filled with a tolerably dense growth of trees and shrubs. To the Avest of the line stretches a hilly country of lower altitude, sloping gradually to the south-west, cut by ravines, which are not so deep and whose sides are not so steep as those on the east. The higher land on this side is covered with a great number of small stones and debris of volcanic rocks, and is quite barren QQ 2 308 150TANY OF IJOniUOUEZ. of any trees or shrubs, save pcrliajis a stray stunted Vacoa, Palmiste, Latauicr, or Citron, and the ground is over wide ai'eas coated by but u scant covering of Cype- ruccic and Grasses mixed witli social weeds ; whilst the valleys are as a rule quite destitute throughout of a covering of tree or shrub, except it may be a lew sti-airirlers at the margins of the streams. As we pass south-west, however, wo come, on the banks of the Rivicl're Quitorze, to an abriipt line of (hMiiarcation betwixt this bare district and one on Avhich vegetation is relatively more abundant and varied. Crossing the line we enter on an extensive coralline limestone plain occupying tlie whole south-AAVst end from Anse Peril to Anse Topaze. This is covered with a poor vegetation, but one that is very characteristic, sucli herbs abounding as Tridax jn'oriimbciis, Fremna serratifolia, Senecio linearis, Sarcos- iemma viminale, Cossi/ffui fil/formis, Ijiomoea leucantlm, and Boerhaavia diffusa; whilst of trees and shrubs Liidia sessilijloni, Terminalia Benzoin, and Antirrhcea fraiigiilaceo, are specially abundant, and more rarely Gastonia cutispongia ; and sparingly such herbs as Nesoyenes decumhens, Dichondra rcpeiis, Hijpoestes iiicon- spicua, and SehKjinella Balfoiiri, occur. On the north-west where none of this coralline limestone is found and the volcanic rock passes directly into the sea the l)arn'nness continues. The ditTerence in landscape offered by the two regions, eastern and western, of the island are very striking, and the abruptness of the line of demarcation is very remarkable. On sailing round from Port Mathurin to the Anse Topaze, as we did in going to the caverns where the bones of Solitaire are found, we had occasion frequently to observe this. In the valley of Riviere Saiimatre the features are very well marked, for on one side the valley is thickly covered with vegetation, whilst the opposite side is quite barren. The difference is no doubt due to the fires, lighted to clear portions of land for cultivation without due care being taken to prevent their spreading and fanned by the prevailing south-east wind, sweeping across the south and north-west of the island more especially ; they have thus converted the paradise into a wilderness. When we come to consider more closely the Flora of Rodriguez we find that it is composed of 470 species and varieties, belonging to 293 genera included in 85 natm-al orders. This then is a proportion of about 5^ species to each family, and hardly 2 species to each genus. This ratio is, however, brought about by the comparatively large number of Cryptogams, for the 470 species comprise no less than 173 Cryptogams. There are therefore 297 Phsenogams. Speaking of the Flora we shall deal separately with the two divisions. Of Phsenogamic plants then we have 297 species belonging to 214 genera included in 75 natural orders. An analysis of these shows us the extraordinary fact that 108 species or over one-third are introduced plants, in many cases relics of former or escapes from present cultivation. These I have thought it well to retain IS. BAYLEY BALFOUR. 309 in tlic list and record as having been found on the island however sparsely, as at some future period any one of them may occupy a more prominent position in the Flora. In taking cognizance, however, of the extent of the Flora at present they must be expunged. From the total number we must also subtract 14 species which are not finally determinable, owing to imperfect specimens having been obtained. And here I may state that from the same cause, in some cases, the determination of species has been very difficult and can only be considered provisional, for our visit to the island happened unfortunately at a time of year when few plants were in flower, and it was just when we were leaving the island that plants were coming into bloom. In all cases of difficulty, however, I have indicated it under the specific name. Of the 14 species I here allude to some most probably are endemic plants whilst others are perhaps widely spread. The following is a list of them so far as can be determined. Desmodimn, sp. Asclepiacls, 2 sp. Mucioia, sp. Stachys, sp. Eugenia, sp. Asparagus, 2 sp. Olea, sp. AngrcBCum, sp. Ardma, sp. Cyperus, sp. Sideroxylon, sp. Bambusa, sp. This leaves 175 Phsenogams belonging to 119 genera and 57 orders which we consider the indigenous Flora, though even some of the plants in this niunber may have to be exckuled. We have thus a ratio of about 3 species to each order and about 2 to each genus, — a very small proportion. Of the total number, Monocoty- ledons constitute 49 or about two-sevenths, a comparatively large proportion for a tropical island. Endemic species number 35 or one-fifth of the whole, 6 of them being Monocotyledons ; 31 species or about two-elevenths are peculiarly Mascarene, of which number one-fifth are Monocotyledons ; and of the rest, 8 species or more than one-twentieth are African plants which do not occu.r in Asia, whilst 14 species or nearly one-twelfth are found in Asia, but do not reach Africa. The remaining 88 species comprise a certain number which are widely spread in the tropics of the old world, 22 species in all or one-eighth of the Flora being of this nature, whilst 66 species or three-eighths are universal tropical weeds. Thus half the indigenous Phsenogamic vegetation consists of common weeds of the tropics. Amongst indigenous Dicotyledons the prevailing orders are, GraminecB, repre- sented by 21 species ; Legnminosa;, by 14 species ; Convolvulacecs, 11 ; Malvacea, 9; BubiacecB, CyperacecB, and EuphorhiacecB, each 8; LiliacecB, 6; Compositce and AmaranthacecB, each 5. These, it will be observed, are all orders which as a rule compose a great part of the vegetation of any tropical island reached by civilization ; but the number of Bubiacece is specially worthy of attention, more especially when compared with the number of Co»iposi(ce. 310 BOTA^'Y 01-' KODUIGUEZ. Grambieiv constitutes then thrcc-scrcntlis of the whole indigenoixs Monocotyle- dons, and ahout throe twcnty-iiftbs of the whole Phnenogams, a fairly large nunihcr compared with eontineutal areas, — a usual characteristic of tropical islands. The species call for no special mention ; none are peculiar', though a form of Panicum is Mascareue. The majority are widely spread, cosmopolitan, or old world species, a few being Asiatic. LegitmbwscVi represented by fom'tcen species, forms two-twenty-fifths of the whole Flora. It contains as usual a great mmiher of species found along or near the shores, such as Cciiavalia ohtiisifoJia, and Cliforia Tcrnatea, and includes many of the commonest plants. Very few plants of llodriguez besides the endemic ones are absent from Mauritius, but cm-iously this family contains three species not found in Mauritius ; these are Conavalla ensiformis, Mucuna gigcmtea, and Rhynchosia minhna. Species of the genera of the two former are found, but not of the latter. Most of the Eodi-iguez plants of this family are universal in the tropics. Coiivolculacecc is next most numerously represented, chiefly by species of Ipomoea, of which eleven are known. j\Iany of the species arc littoral, such as Iponuva pes- capi'rp, Ip. Jeucantha, Ip. nil. Ip. fragrans. The three last mentioned do not occur in Mauritius. Ulalcacecc, with nine species, includes, with one exception {Uihiscus UUiJlorus), widely spread tropical plants. Cyperacece and Amarantliacem call for no special notice, and the other families are noticed in the account of the endemic Flora. The endemic Flora consists of 35 species, or one-fifth of the whole Phoenogamic plants. Of this number 29 are Dicotyledons, and 6 are Monocotyledons. The following is a list of the species : — Zanthoxylum paniculafum. Blospyros diversifolia. Qnivma laciniata. Tamilepis sphenophylla. Hclerocaryo co.stanea. Sarcostemma Odontolepis. Eugenia JBcdfouri. Sypocstes rodriguesiana. Mcdlmrina ]}enduliJlora. Hypoestes inconsjncua. ' Danais corymhosa. Nesogenes decumhens. Mandia Jieterophylla. Clerodendron laciniaUmi. Fyrostria trilocidaris. Fisonia viscosa. Scyphochlomys revoluta. J^rna congesta. Psychoida lanceolcda. Filea Balfouri. Psiadia Coronopus. Peperomia Mrta. Psiadia rodriguesiana. Peperomia reticulata. Abrofanella rhynchocarpa. Pej^eromia Bodriguezi. Lobelia vagans. Buphorhia dapJmoides. IS. BAYLEY BALFOUR. 311 ThijUantlms dimetosus. Jli/ophorhe VerschaffeUi. L'lsfi'ostaclii/s AjiJu'odite. Pandaniis heterocarptis. Aloe lomcttophylloides. Tandamis tamifolius. Latania Versclwffelti. Of the Dicotyledonous orders iu wliicli endemic species occur Buhiaoece is the most remarkable, presenting the greatest amount of peculiarity. Of the eight species it contains, five are endemic, and they all belong to diiierent genera. One of these, Sci/phochlamys, is peculiar to Eodi-iguez, and the genera of two others — I'l/ros/ria and Danais — are essentially Mascarene, the latter extending into Madagascar. The two remaining species belong to the widely-spread genera Sandia and Fsychotria, of which the former is unknown in the other Mascarene Islands, but may have a reprcsentatiye there in the nearly allied Gardenia, which is absent from E,odi"iguez. Three species of Eiihiacea; arc peculiarly Mascarene, two belonging to genera of wide range, Antirrhcea and Oldeiilandia, but the Mauritian tyjie of the species of Oldenlandia is modified into a distinct variety in Rodriguez ; the tMrd is a species of Fernelia, a Mascarene genus. This comparative abundance and peculiarity of the Ruhiacece is very interesting when compared with similar features seen in other oceanic islands of Hke natm'C ; and it is fm-ther interesting to note that in JEuphorUacce we have an order co-exten- sive, though not presenting so great peculiarity. The eight species composing it belong maialy to Fhjllanthiis and EupJiorhla, of each of which there is a peculiar species. There is also a Mascarene species of Phyllanthus which derives interest as one of the plants collected by Commerson, and originally described from his specimens ; in addition there is a species of Qlaoxylon, and one of Securinega, both Mascarene. Compositce and FiperacecB are equally numerously represented by peculiar species, each containing three. In the first-named order, of which five species in all occur, the endemic plants belong to two genera. Fsiadia, a Mascarene type of AsterecB extending into Africa, but not Asiatic, includes two of the peculiar species. Both are suflPruticose, one, Fs. rodrlguesiana, growing only on the higher levels, and there sparingly, has the velutinous character so well developed in many of the Bourbon species, the other, Fs. Coronopiis is glutinous, occurring only on the shores in small quantity, and is interesting as being one of the few plants brought from Bodriguez by Commerson, and originally described as Sarcanthemum Coronopus by Cassini. The third peculiar Composite is a species of Abrotanella, a small genus of Cotidecs, ranging through Australia, New Zealand, and some Antarctic islands, and unknown in any other of the Mascarene group. A fourth Composite is a Senecio confined to the Mascarene Islands, very variable in its character, and the rcmainiuff one is the world wide Sonchus oleraceiis. 312 BOTANY OF UODRIGUKZ. Tn the Piperoccic wo have tlirec peculiar species of Peperomia with a strong East Indian affinity. Asclepiodacecc, Acanthacciv, and Vcrheiiaceic each contain two endemic plants. The Asclepiadaceous Sarcostemma comprises two species, of which one is a novelty, and there is a climlier, common on the island, tlie type of a new genvis Tanulopis closely allied to the East Indian Brachylrpia. The two peculiar Acanthads arc species of Hypoestes, and in Verbetiacea; one of the novelties is a species of the hitherto monotypic Nesogeues, a genus confined to a few of the Polynesian Islands, whilst the other is a species of the common tropical Clerodendron. Of other Dicotyledonous orders none contain more than one peculiar species. Anaco)'di((CC(V has a single species of the ^Vfricau genus Sclcvocari/a, which dilfers, however, very considcrahlj' from the generic type. Of three irurlacea; one species of Eugenia is endemic, whilst another is mainly Mascarene, though occurring else- where in the tropics, and the third is the Mascarene Fastldia viauritiana. I shall only specially mention another order, Turneracece, of which one generi- cally peculiar plant is found, Mathiirhia pendulijlora, whose nearest congener is a central American genus ErhUchia, and it is worthy of note that this family has hitherto been unknown in the Mascarene Islands. The remaining families with endemic species are Rnto.cecc, Meliacece, Campanulctcea, Ebenacea;, Nyctaginacece, Ainaraiithacece, and Urtlcacece. Looking next at the 6 peculiar Monocotyledons we find that they are included in 5 genera belonging to 4 orders. The Monocotyledons from this island are specially interesting, as they include those plants which give a character to the vegetation. Every visitor to Rodriguez will be struck at once by the peculiar featui'es impressed on the landscape by the prevalence of the Screw-pines. They are indeed the physiognomic plants, and far outstrip in numbers any other species; but it is remarkable that though individually so numerous, sioecifically the family is not rich, there being only two species of Fandanus on the island, P. heterocarpus and P. tenuifolhis, both peculiar and very distinct from any other of the indigenous Mascarene forms. Three other species have been registered by various authorities, they are P. odorafismnKs, P. jit'dk, and P. muricatus. None of these are Mauritian or Bourbon species, the first being a native of Asia and the two latter both Madagascar species, and the evidence of their occurrence in Rodriguez is faulty. Is est in interest amongst the Monocotyledons come the Palms, and they are very peculiar. AVe find three species indigenous, and these belong to different genera which are all Mascarene. They arc Latania Verschaffelti, Jlyophorhe Versclwf- felti, and Dictyospernm olhum, var. aureum. Latania Vershaffeltl has been for some time known to European nurserymen as Latania aurea, and Hyophorbe Verscliaf- IS. BAYLEY BALFOUR. 313 felti has also beca known to liorticultimsts under the name of Areco. Verschoffelti. Each of the genera, LokiHia and Jlyopliorhe, includes three species, which present a remarkable correspondence in distribution. The Latania Verschaffelti of Eodriguez is represented in Maiuitius by L. Com- mersont, which grows also in Bom-bon, and on E,ormd Island a third species, Z. Loddigcsl, is found. Of Jlyophorhe, the Eodriguez plant S. Verschaffelti, is represented in Mauritius and Bom-bon by Uyophorhe indica, and on Round Island M, amaricaulis occurs. The genus Dicfi/osjyerrna is Mascarene, and has been created by Wendland to include the type of Palm, originally described by Bory St. Vincent (Voy. I. 306) as Areca alba. The Mascarene palms, formerly described as species of Areca, are all removed from that genus, the non-spiny forms now constituting Dictyosperma, and the spiny forms, ^ rem crinita and A. rubra, combining to make the genus Acantho- plicenix. Dictyosperma is montoypic, but the species is very variable, and in Eotbigucz assumes a very graceful and delicate habit ; the characters, however, are hardly specific, and it is therefore merely a variety. This palm is the well-known Areca aiirea of nurserymen. Of the other endemic Monocotyledons one is a species of Aloe which is very distinct. Liliacece altogether constitutes one- twenty-ninth of the Flora, and in addition to this peculiar Aloe comprises a Mascarene species of Asparagus. An Asphodelus which has a tolerably wide range is to be noted as occm'ring only on two of the small coral islets on the southern reef, Gombrani and Pierrots, and is absent entirely from the main island. A marked featm*e in the Flora is the paucity of Orcliidacece. Only foui- species have been determined ; a fifth, a species of Angr cecum, was found, but in too im- perfect a state for identification. These belong to genera two of which range into Africa and not into Asia, and one is Asian but not African. One species of Listrostachys is peculiar, an Oberonia and a Bulbophyllum are Mascarene, and a distinct variety of a Maiuitian Aerantlms occm-s. The dryness of the soil and climate no doubt have to do with the scarcity of the family, and also to a lar^e extent the destruction of the old forests, as no epiphytic forms now exist. This paucity is the more remarkable when contrasted with the profusion of this family in the sister islands. Analysing the relationship of the endemic plants, we find that three genera are endemic, of these one, Matlmrina, has a near American aflBnity, another, Tanulepis, has a close Asiatic connexion, whilst the third, ScyphocMamys, has its nearest congener peculiarly Mascarene. Five are IMascarenc genera, — Quivisia, Danais, Fyrostria, Latania, and Uyophorbe. In the case of four, Sclerocarya, Psiadia Listrostachys, and Aloe, — we have genera of peculiarly African range ; and of the E R -»- 311. 150TANY OF RODRIGUEZ. rest, AhrotaneUa is an Antarctic genus and Nosogenes is Polynesian, the remainder being foiu* old world and twelve generally dis(ril)uii'd iropical genera. We may tabulate these as follows : — Cosmopolitan Afeican. Endemic. Mascarene. Mafhio'iiia. Sciihochlamys. Tanidepis. Antarctic. AhrotaneUa. Qui CIS ill. I>aiiais. Pyi'osti'ia. Lahniia. Hi/ophorhe. Polynesian. ' Nesogenes. Sclerocari/a. Psiadia. Listi'ostachi/s. Aloe. Old World. Sarcostemma. Hypoestes. JErna. FandioiHs. IN Tkopics. Zauthoxyhim. Eugenia. Bcnidia. Psycliotria. Lobelia. Diospyros. ClcrodendroH. Pisonia. Pilea. Fcperomia. JEhipliorbia. Phyllant/ms. The African affinity of the endemic Flora thus becomes manifest. We have stated that 31 species of Phajnogamic plants in the Flora arc peculiarly Masearenc, of which 25 arc Dicotyledons, and the rest Monocotyledons. They are, — Pitfosporum Senacia. Aphloia mauritiana. Hibiscus liliijlorus. Dombeya acntangula. Dombeya ferruginea. Toddalia paniculata. Elceodendron orieniale. Gouania retinaria. Fcetidia mauritiana. Terminalia Benzoin. Gastonia cutispongia. Oldenlandia Sieberi. Fernelia biixifoUa. A ntirrhoea frangulacea. Semcio linearis. Carissa Xylopticron. Ptiddleia madagascariensis. Solanum macrocarpum . Myoporum mauritianum. Obetia ficifolia. Pliylla n th us Casticiim. Securinega durissima. Claoxylon parviflorum. Viscum tcenioides. Oberonia brevifolia. Bidbopliyllum inciirmim. Aerantlius arachnites. Asparagus umbellulatus. Dictyosperma album. Andropogon foliatus. Olea lancea. Several of these I have already referred to, and I now mention specially the occurrence of the curious Myoporum mauritiuniam^ very sjjaringly found on the island, Avliicb differs very markedly from the type of Myoporineie, and is perhaps endemic in Rodriguez, as the evidence of its occurrence in other Mascareue Islands is faulty. IS. BAYLEY IJALFOUK. 315 An analysis of the relationships of these plants shows us that, — Six belong to genera which are essentially Mascarene; these arc Aphloia, Foetldia, Gastonia, Fenielia, Obetia, and D'lclijosperma, and. two, Domheija and Aerauthus, are essentially African. Ten of the remainder are old Avorld genera, and the remaining twelve are spread tropical genera, l)ut curiously one of these, An- tirrhoea, is not African. Here we have evidence of the individuality of the Mas- carene Elora and also of its primary relationship with the African rather than the Indian type. Tabulating these w'e have : — Mascarexe. Aphloia. Fa'tidia. Gastonia. Fenielia. Ohetia. Dictyosperma. African. Domheija. Aerantkus. Old Would. Fitlosporum. Toddalia. Olea. Carissa. Ill/ojJorum. Claoxylon. Visciim. Oheronia. Bidbophyllum. Asparagus. Hibiscus. Flceodendron. Gouania. Terminalia. Oldenlandia. Antirrhcea. Senecio. Buddleia. Solaimm. Fhyllanthus. Securinega. Andropogon. The connections exhibited by the plants of other countries indigenous in Rodriguez arc not very strong. There are a few, eight, species which have specially an African distribution, and there are foiu'teen of specially Asiatic range which do not reach Africa : that is, more than one-twentieth of the whole Flora have African and nearly one-tweKth Asiatic distribution. The African plants are, — Cosmopolitan in Tropics. Ludia sessiliflora. Desmodi uni matiritiatmm. Desmodium incanum. Sarcostemma viminale. The Asiatic and non- African are,- Calophyllum Inopliylliim. Toddalia aculeata. AUophylus Cobbe. Erythrina indica. jfluCHua gigantea. Atylosia scarahceoides. 'Eugenia cotinifolia. Lycium temie. Ipom ma fvagans. Draccena reflexa. Andropogon finitimus. Ipomcea peltata. Cassyth a filiformis. Draccena angustifolia. Carex gracilis. Panicum Balfouri. Stenotaphrum stihulatum. Zoysia pun gens. R R 2 15 species 9 5J G 5> 6 )) 5 »> 316 r>OTANY OF IJODRIGUEZ. It will 1)0 observed that the African species arc all of a more restricted range than the Asiatic. Of the 8S generally distributed tropical weeds I need say nothing. They will be seen to be such as one Avould expect to meet in an island which has passed through such vicissitudes as Hodriguez. Of the introduced Phrenogamic plants, 108 in number, of 90 genera included in oG natiu'al orders, the prevailing families are — Lcgumhwsce Solanacea; - - - GniDih/ccc - . - CompositiC - - - Mi/rtace(B - - - Coiivolvulacece CKCurhitacece jRtttacccc ... orders commonly and widely distributed, and including those social weeds which follow man's footsteps over the world. I need not further refer to tliem. Besides those Phtenogams I have recorded as occurring on the island, there are a few others reported as existing, but of none of them have I seen authentic speci- mens from the island, and as I did not find them there, I do not include them in the list. They are the following : — • Achillea, sp. Lantana, sp. Ficus, sp. Pandcmns muricatus. Fandanns odoraiissinms. Fandcnms ntUis. In estimating the relationship of the Ci^yptogamic Flora of Rodriguez, we find more difficulties to encounter from the fact that excepting the Vascular Crypto- gams, little comparatively has been done amongst the Cryptogams of the other Mascarene Islands. The lower forms from Rodriguez have, however, been carefully worked out, and I find that there are now known from the island, 173 species of Cryptogams (this number excluding marine Algce) . Thus the Phsenogams and the Cryptogams are ahnost equal in number. Of this number, 26 species or about one-seventh are Filices, OpMoglossacece, and LycopodiacecB. Musci constitute about two-elevenths, numbering 33 species, whilst of SepaticcB I have 18 species, and there is one species of Chara. Lichens are most abundant of all, there being 75 species, or over three-sevenths of the whole Cryptogams; whilst Fv.tujl number 8 species, freshwater Algce 13, and I have only 39 species of marine AJfjce, IS. BAYLEY BALFOUR. 317 111 Vascular Cryptogams, the Eodriguez Flora contrasts very unfavourably ^vitli that of the other Mascareue Islands. The 25 species known, of which 20 are Ferns, is a number relatively and absolutely very small. The scarcity of this group of plants is accoimted for by the dryness of the island, and in confirmation of this fact, we observe the Tree Ferns of the other Mascarene Islands have here no repre- sentatives, and of the large moisture loving genera Trichomanes and ITi/menojylii/lh'.m, so abundant elsewhere in the groiip, the former is here typified in one species only, T. ctts2)klatt(m, whilst the latter is absent. And when we consider the nature of the species which do occm-, they are those characteristic of dry and arid regions, such for example as Adicmtmncatidatitm,As2)lenmmfurcatmn,As^lemimifalcati(m, NephrocUmn unifitm, N. molle, FolypoiUiim phymatocles, Nephrole-pis acuta. None of the species are novelties ; all occu.r in the Mascarene Islands, and three of them, Trichomanes cuspidatum, Davallia manritianu, and Neplirodium crinitum, are peculiai-ly Mascarene, but most are widely spread tropical species. Whilst some, as Pellcea Jiastata, Pteris flahellata, and Neplirodium elatum, are found in Africa, but not in Asia ; on the other hand we have the Asian but not African Asplenium hirkim, and Folypodiiim adnascens. Three Ferns, Lindsay a acutifolia, another Nephrolepis and another Lastvea are reported from Rodi'iguez, but I have seen no authentic specimens, and as I did not find them myself, I have not inckided them in my list. The LycopodiacecB further exemplify the dry character of the island. Only four species are known. Two are widely spread Lycopodium Phlegmaria and Psilotuni triquetrum, whilst two species of Selaginella, S. Balfouri and *S'. rodriguesiana, are novelties. Mosses may be considered as fairly represented in Rodriguez by 33 species, though this is a small number compared with those found in Maiuitius ; the niunber of species in that island, according to the latest enumeration I have seen, being 101. Of the 33 species found, 17 are peculiar ; and of the remainder, 13 occur in the other Mascarene Islands, or in Africa, a few corresponding with those from the western coast of tropical iVfrica ; and it appears from what little is known of the Mosses of the eastern coast, that some species have an enormous range on the African continent. Two other species are found in Asia, and one, JFeisia oontro- versa, is cosmopolitan. Of the Rodriguez species all except one belong to genera which are represented in the Mascarene Islands. The genus Ectropothecium is most abundantly repre- sented by fom" species, and they are perhaps the commonest on the island. It is curious that Uypnnm, which occurs so extensively in the sister islands, should include but one species in R.odriguez. Of all species, the most interesting is OrthotricJmm ])Ucatum, which is not uncommon on the island, previously known only from the specimens gathered in Bom'bon by Du Petit Thouars, and described and well figured by Schwaegrichcn ; it is ixndoubtedly a member of the family of 318 BOTANY OF RODKICUEZ. the Ot'lhotnclm, but its rharactors as a i;<"iuis aiv, at'i-ording to Mitlrn, iiiloriuc- diate, and its position in that family unique. One of (lie most elegant as it is one of the i-aivst speeies is the endemic JLicroiiiifriitm aslruUlt'iint, found trailing over boulders at the top of the valley of the iliviO'reaux Uuitres, and along with it is also found M. aciculare. Ocioblephewm alhklitm is by no means common, growing in tufts on decayed branches of trees and rich vegetable humus in the same valley, and this is also the only station for Ncckera Ie2)lneana and Jlcteoriiuu 'mvolulifol'mni, and sparingly on trees also occurs Plcrogonium cnrvifoliion. On the moist clay rocks at the sides of the beds of streams in shady places, the various species of Eclropotho- ciion are found most abundantly, E. doloare especially forming a thick covering, and this is probably the commonest Moss on the island ; with it also species of Semato2)hyUum, notably S. ineurc'ifolimn, are commonly found as also Callicostella Iccc'utscula, Rhacophilum afrlcaniim, and occasionally species of Flsskiens. On moist rocks near the stream sources we find species of Bartramia ; Brijnm occurs commonly over the island, and Jfelsia and Calyperes arc also frequently met with. The island possesses 18 species of Uopatica; included in G genera, and of these 13 are novelties. Of the remainder, three, Chiloscyphus ohlougif alius, Frullmiia squarrosa, Authoceros fitciformis, are Mascarene or African, one, Lejeunia minu- tissima, is found in the Eastern Archipelago, and Anlhoceros Icevis is the only cosmopolitan species. Lejeuuia is the most extensive genus, embracing seven of the total nimiber of species, and six of these are endemic. Of the species, Lcjeunia Balfouri is one of the commonest, occm*ring on moist clay rocks at the sides of streams, and in similar situations with it Chilocyphus ohlonyifolius and Radula appressa are found, and in great abundance Anthoceros Icevis and A. fucifonnis. Lejeuuia farva is rare, but in similar situations. The stems of the Screw-pines are frequently clothed with Frullania squarrosa, Fr. Apicctlis, and Fr. ohscurifolia, though they also occur more sparingly on other trees, as does Fhrar/micoma carinata. One species of Chara is found in many of the rivers, and it also occm's in Mauritius and Boiu'bon. Looking now at the Lichens, we find that of all Cryptogams they occur most abimdantly. Altogether, 76 species and varieties were collected in a determinable state. Of these, the large number of 35 or nearly one half are novelties, whilst of the remainder, 11 are known from Mauritius, and 7 or 8 from Bourbon. This large number of Lichens is very remarkable when compared with the number in Mauritius and Bombon. From the former island, 89 species and varieties are enumerated by Weddell (Trans. Eoy. Soc. Arts and Scien. Maur. vii. 1G3) and Nylander (Ann. Sc. Nat. -Ath ser. xi. 218) determines 112 species in Bourbon. But I think this relatively large proportion is due rather to our imperfect knowledge of the Lichen Flora of these islands than to their poverty in species as compared with Rodriguez. In addition to the species determined, there are fragments of other species, mainly IS. BAYLEY BALFOUR. 319 crustaceous, in my collection Avbich are quite xindeterminable, being eitbcr sterile or witb imperfectly developed apotbecia. Of tbe Eodriguez species, all save two belong to genera represented in tbe otber Mascarene Islands. Tbe two exceptions are Heppki, of wbicb we bave a single endemic species foimd very sparingly on tbe island, and Pyrenastrum, a single species of this genus, P. American urn, occurring very abundantly on the bark of trees. The most abundantly represented genus is Lecaiiora, wbicb includes twenty-one species and varieties, and fourteen of these are peculiar; and next to it comes Lccidea, with nine species, of which six are novelties. Of the other genera Rama- Una is most numerously represented, having five species and varieties, of which two are peculiar. Artlwnia has two peculiar species, and Cladonia, Pi/.vine, and Opegraiiha have each a single species, which is endemic. The scarcity of species of Cladonia is ciu-ious, as tbe genus is well developed in the otber Mascarene Islands. Of tbe species, one, wbicb will at once attract the attention of any who visit tbe island, is tbe beautiful TJsnea dasypogioides, only occurring in tbe higher parts of the island ; it there bangs beard-like in great abundance from tbe tree branches, and along with it Ramalina suhfraxinea is usually found. Close to these may be seen on the rocks tbe small tufted Ramalina (jracilenta and R. gracilenta f. nodnlosa intermingling with white patches of Lecanora atra f. succedanea, with its large black apotbecia. Farmelia latissima is a not uncommon species in dry spots along with Sticta aurata, and on tbe boulders JPhyscia sjoeciosa is found in abundance. The stems of the Screw-pines are invariably dotted over with stellate patches of Physcia ixicta, and many species of Verrucaria and OrapMs also find thereon a suitable nidvis, as well as Lecanora achroa ; whilst their withered and dried leaves give a home to Arthonia phylloica and A. dendritella. The features imposed on the rocks in many places by certain Lichens is very strikino-. Where such species as Lecanora oMiquans, L. conizopta, Pertiisaria im- pallescens, Lecidea conUnens, and L. configurans are abundant, one might suppose that tbe rocks had been whitewashed, this character being visible at a long distance. Again, on tbe more decomposed or cindery rocks, Lecanora suhfusca f. pumicicola, L. apostatica, L. cinnabarina, Lecidea spuria, L. achroopholis, and L. immutans unite to give a dull mottled and variegated aspect to tbe rocks. Tbe stems of other trees besides the Screw-pines are favotuite sites for certain species, notably we may mention Bois Gandine and Bois Puant ; of the former, more especially, tbe bark is usually quite concealed by lichens, giving it a very white or grey appearance. On it alone Lecanora leucoxantha occurs with its orange apo- tbecia, and also the pure white PeW?<5rt>'/««^Z«/« ; Lecanora punicea and Coccocarpa molyhdcea are found on trees with rough bark, and so is the scarlet Trypethelinm crucntulum; whilst on decayed and decorticated wood forms such as Opegrapha difficilior, Lecanora conizoea, and Pannaria riihiginosa grow in profusion. 320 BOTANY OF lU)Dl?iaUEZ. Fungi arc poorly represented in my collection by 8 species. I found a few more spooios on the island, but in ooiu'se of transmission to Britain I regret tliey have disappeared. The family is not hoyrever abundant on the island. Of the eight species which hare been determined, three turn out to be noreltios. One of them, Fohiporns aspidolopus, is perhaps the commonest form on the island, growing abundantly on the stems of trees, but apparently yvith a preference for Scrcw-i)iues. The widely spread Sell IzoplnjU urn commune and the common British Hirneola ^k;vo«Z« t7? representatives in adjacent islands of hetero- phyllous species do not wlion I licy produce diverse formed leaves always l\av(> the same kind of \ariation. In tiie Masearene Islands i only know of two i^cncra possessing representative sjjeeies wliieh exhihii heteromorphisni. In one of these, Quirisia, the type of variation is the same in both species; in the other, Clcrodcn- dron, it is ditVerent. Tiie phenomenon is conlined to no special order, though 1 may luitc that in four species of lti(htace(C it is very well marked. For the sake of clearness I siiall consider the kinds of variation observed as of three types. 1st. \"ariation dependent on imperfi'ct or arrested development ol' the whole leaf in the young plant, the lamina developing equally. In plants which exhibit this type the leaves of the juveniles are to a certain extent miniatures of the adult. They are very small, but possess the same or almost the same relative proportion of length of lamina and breadth thereof as is seen in the adult. Tlie lamina is developed equally and is not lobcd, and the margin may be entire, but sometimes the parenchyma towards the edge of the lamina is somewhat delieient, and thus, the veins being left prominent, the leaf has a spinose margin. AVith advance in age of the plant the leaves both increase in absolute and relative size, and also the edges of the lamina fill up in those cases where tlu- margin was s[)inose, so that the spinoseness disappears. Sometimes, how- ever, there is a tendency to the perpetuation of a certain amount of tlie spinoseness in the adult, especially at the apex of the leaves, which may be hard and sharp pointed. Three of the species have this type of heterophylly, — Lttdla sc'ssiUflora . Fernelia hiixifoUa. Ccfrissa Xylopicron. These are all species w'bicli in habit somewhat resemble the Box tree {Btixtis semiyervircns) and their leaves are small and coriaceous. The resemlilance betAveen the two latter species, when adult, is very close, but the first and last resemble each other most nearlv when voung. Fernelia biixlfoUa has the leaves in the voun"- plant entire and not spiny. In the other two they are spiny. The three are Masearene species, but I have record of Ludia sessilijlora only as presenting hete- rophylly in the sister islands. 2nd. Variation arising from non-development of the young leaf in one direction, the transverse, usually but not always accompanied by an increase in the other direction, the iougitudiual, the lamina developing equally. In plants which have variation after this type, and it is the commonest, the yoimg leaves are usually greatly elongated, frequently being two or three times as long as the adult. Thus in Randia hetei^ophjUa the young are usually over a foot long, the adu.lt vary from 2^ to 6 inches. More rarely the juvenile leaves are not elongated and may even be shorter than the adult, but it is only in species in which IS. BAYLET BALFOUK. 323 the adult leaves are not large and have a firm and coriaceous consistence that this is observed, a ciu-ious point when considered along with what I liave noted in the first type of variation. But whether the leaves be elongated or short, the relative proportion betwixt the length and breadth of the lamina in the juvenile is vastly different from what is observed in the adult. The juvenile are usually linear, always very greatly narrowed, often only one-twentieth the breadth of the adult, and the contrast in such a case is, as may be imagined, very striking. For example, take Scyphochlamijs recoluta, the juvenile form is only ^th of an inch in breadth, Avhile the adult averages 1^ to 2^ inches. The lamina developes equally at the margins, which are never spiny, and though any crenatures or dentations wliich characterise the mature form, may be represented, the lamina is never lobed or deeply cleft. As I have said this is the commonest type of Aariation, occurring in no less than nine, that is, in more than half of the heterophyllous species, these are : — Domheya ferruginea. JRandia hcteroiihylla. Elceodendron orientale. Fyrostria trilocnlaris. Terminalia Benzoin. Sci/phochlamys revolula. Foetidia mcmritiana. Diospyros diver s'lfolia. JIathiirina pendvUflora. The fia'st four of these are peculiar Mascarene species, the remainder are endemic. Elceodeudron orientale presents the same variations in the sister islands, and judging from a remark of Cavanilles (Diss. III., 121), Dombeyo ferruginea is also heteromorphic ; I have no information regarding variation in the other two species. Of the endemics, the Pyrostria and Diospyi^os are the only two species in which the linear juvenile leaves are shorter than or at least do not exceed the adult. Another species Eugenia cotinifolia is probably heterophyllous after this type, but the specimens I found represented differences only to so slight an extent as hardly to warrant my including it in this list. 3rd. Variation due to unequal development of the lamina in the young leaves. In the plants included in this group the young leaves are not much less, may even be greater in absolute size than the adult, and the relative proportion of length to breadth of the lamina is the same, but the lamina developes unequally, so that a lobed or cleft, sometimes very deeply cleft leaf is produced. The lobation gradually disappears in the older leaves, though frequently slight traces of it remain, especially in an emargination of the apex. If the mature form of leaf has a pinnatelj^ arranged venation, then the young leaves are of the pinnatifid type ; if the veins are radiate the jjalmatifid type is seen. The amount of lobation in the juvenile greatly varies, and the primary lobes are sometimes again once or twice cleft. Four species exhibit this variation, — Aphloia mauritiana. Quivisia laciniata. Hibiscus UUiJlonis. Clerodendron laciniatum. Of these Hibiscus Uliijlorus has palmatifid young leaves ; in all the others they are pinnatifid. The first two are Mascarene species, the others are endemic. s s 2 321 r.OTANT OF I50DRIC.UEZ. Aphlola JIaitritiaua is as variable iii Mauritius, anil so also probably Uibiscns lilii- florns. Qttivma hxciniata and Clerodeiuh'ou laciniatum arc eacli represented in Mauritius by a hetcropbyllous species. But Avhilst Qidcisia hetcrophijlla ol" Mauritius varies in the same manner as the Eodrigucz plant, C/erodcnilroii hc/cro- phi/llum has the second type of variation developed. Abrotanella rhyncocarpa I may mention here; it is heterophyllous, but after a diiVevent manner. It is a small tufted herb, with very persistent leaves. The young leaves, those at the top of the shoots, are entii-e and oblanceolate, but the older leaves below are markedly piunatifid. We have thus tlie converse, as it were, of tliis last type of variation. Some species, in addition to the dilfercnces in form and size, exhibit variations in habit of the leaves at different periods of their growth. This is most marked in Domheya ferruginea, of which the young leaves arc quite green and glabrous on both sides, but in the adult arc clothed on the under surface with a dense brown tomentum. The converse is seen in some to a slight extent, thus Clerodendroii laci/iiaf/iiii, Termhialia Benzoin, and Mandia helerophijUa have pubescent young leaves, while in the adult the leaves are glabrous. As might be expected, wben the leaves of young plants are not so perfectly developed as in tbe adult, they are often of a more firm and rigid consistence. This is specially apparent in species varying according to the first type, in wliich there is deficiency in parcnchjnnatous tissue, for instance, in Lndia sessiUJlora and Carissa Xijlopici'on ; but it is also apparent in plants varying after different types. In 17 species tben, belonging to as many genera, of 13 natiu'al Orders, the heteropbylly exists ; and it is a fact of great significance that every * one of them is either endemic or Mascarene only, that is, one fourth of the whole endemic and Mascarene species on the island. Eurther, on considering the genera, we find that two of them are endemic, four are Mascarene alone, three extend to Africa,, and of the remainder seven are cosmopoHtan, and one Polynesian. Our mformation regarding the variabiUty of the plants of Maixritius and Bourbon is too slight to enable us to say whether an equal amount of variation occm's there. But certainly some of the Mascarene species, which vary in Rodriguez, present, as I have ah'eady noticed, the same amount and kind of variation in these islands though m other cases the heteropbylly may not be to such an extent. So that I think there can be little doubt that the heteromorphism in foliage is a feature of the whole Mascarene Flora. In concluding this subject it is of interest to note that Ilibisctis {Paritk^n) tili- aceus is not heterophyllous in Eodriguez, though it is recorded as such from Africa (Oliv. Flor. Afr. Trop. i., p. 208). A plant of S. tricuspis, a South Sea Islands species growing in the Saharampore gardens, is described f as having sent up from a * I must qualify this. One specie.^, Tjidia sessiliflora, passes into Africa. t Bell in Trans. Bot. Soc. Edinb. VII. 565. King in Linn. Soc. Jouin. XV. 83. IS. BAYLEY BALFOUR. 325 rootiug dccumhcut branch a plant witli entire leaves very different from the parent, and indeed like those of H. tiliacens. Prof. Thiselton Dyer suggests that S. trl- cnspis is an insular form of //. tiliacens, which I think is not at all improbable. But it is strange that in Rodriguez, where heteromorphism is so prominent a feature, S. tiliaceus does not exhibit the peculiarity ; although I may remark there are two distinct varieties of the tree on the island, which I have noticed under the species. Such is the aspect and nature of the vegetation of Rodriguez, and a considera- tion of the Flora leads us, I think, to the following conclusions :— 1. It is a small Plora, and fragmentary. 2. It is that of a dry rather than of a moist region, as is exemplified in the paucity and nature of some groups, such as Ferns, Orchids, and the abundance of such others as Lichens. 3. It is an insular Flora, as indicated by — a. The relative proportion of species, genera, and orders. h. The almost total absence of indigenous annuals. 4. Its facies is tropical. 5. It is essentially Mascarene, though possessing a fair amount of individuahty. G. It presents affinities with the Floras of many other parts of the globe. Its strongest relationships are with the African, but it has also very strong Eastern connections, and some close American and Poljoiesian affinities. 7. Many species exhibit a great amount of variation, but within certain sharply defined limits. Our knowledge of the Flora of all the other Mascarene Islands is as yet too scant to allow us to adjudge the exact extent of the affinities subsisting between the Floras of the individual islands of the group. But enough may be learned from what I have indicated of the vegetation of this single island to point strongly in the dii-ection of their being fragments of a once more extensive Flora, which has been gradually broken up by geological and climatic changes. How far the geological evidence bears this out is indicated elsewhere. 32G BOTANY OF UODIJIGUF.Z. The follovriug is a list, iu detail, of tlio plants wliirli compose the Flora : - ANONACE^. Auoua muricata, Liini.; DC. Prod. i. SI. Nom. vulii'. Corosol. Cultivatcil. Anoua squamosa, Lhni. ; DC. Prod. i. s.") ; Po/. J!i) '• that he did not find in this island any plant, tree, shrub, or herb, wliicli grows " naturally in any part of Europe, that was known to us, except Purslain, whicli '• is small and green. There's plenty of it in some places of the valleys, and that " which Ave sow'd, having brought some of the seed from the Cape, came up exactly " like the Pm'slain of the Island;" and he records how, when a green caterpillar appeared after a hurricane and destroyed the greater part of their crops, the Purslain was untouched. GUTTIFER^. Calophylliim InophyUum, Linn. ; Planch, ct Tvian. Hon. Giittif. 254 ; Wt. Ic. t. 77. !Xom. vulg. Bois tatamaka malgache. Only a few trees of this found in the higher parts of the island. Yields a soft frum-resinous wood. o MALVACE^. Malvastriim tricuspidatum, A. Gray PI. Wright 10 ; Malva borbonica, TFilld. ; DC. Prod. i. 130. A very common plant. Sida angi;stifoha, Lam. ; DC. Prod. i. 459. This may be merely a variety of S. sjnnosa, Linn.; DC. Prod. i. 460. It was found only at one spot on the island, at the mouth of a valley east of English Bay point. Sida carpinifoha, Linn.fil. ; DC. Prod. i. 461. Nom. vulg. Herbe a paniers. Common everywhere, and assuming very various forms according to its position ; fi"om a dwarfed and stunted plant, with few and small leaves, to a small, freely branching uuder-shrul), about 5 feet high, clothed with large leaves. Sida cordifoUa, Linn. ; DC. Prod. i. 464. Nom. vulg. Mauve. Common everywhere. This, like the other species of Sida, is often used as a demulcent. IS. IJAYLEY BALFOUK. 329 Abutilon indicum, G. Don Gen. Syst. i. 504; Wt. Ic. t. 12 ; Sicla indica, j&iw». ; DC. Frod. i. 471. Nom. vulg. La mauve. A very common weed. Abutilou graveolens, IF. and A. Prod. i. 56; Mook. Comp. Bol. Mag, t. ii. ; Sida graveolens, Boxh.; DC. Frod. i. 473. Only a few plants on the shore in Oyster Bay close to habitations. Urena lobata, Linn.; DC. Frod. i. 441; Dot. Mag. t. 3043. Nom. vulg. Herbe a, paniors. A common Avecd. Hibiscus liliiflorus, Cac. ; DC. Frod. i. 446; H. fragilis, DC. Frod. i. 446; H. Genevii, Bojer in Dot. Mag. t. 3144. This endemic Mascarene plant is not abundant on the island. It is remarkable from the variation in form of its leaves. On the youngest trees I met with, the loaves were rounded at the base, deeply tritid, the lobes being linear acuminate. In older specimens the lobation of the leaf gradually disappears, and we have ovate acute leaves. Finally, by a gradual transition of forms, we reach the leaves of mature plants, which are obovate, obtuse or deeply emarginate, and cuneate at the base. The venation also becomes more distinct in the adult leaves, and they are quintuplinerved. A hybrid from this plant, the male being S. Rosa-sine mis, Linn., is figured, Bot. Mag. t. 2891, from Mauritius, and it seems to possess the variable foliage of one parent ; its flowers, though similar to, are much larger than those of Hibiscus liliijloriis. S. Genevii, Boj. Hort. Maiu*. 28; Bot. Mag. t. 3144, is probably only a form of this plant with more dentate leaves. Hibiscus tiliaceus, Linn.; DC. Frod. i, 4<54<; Paritium tiliaceum, Wi. let. 7. Nom. vulg. Vaur or Var. Everywhere along the shores. Of this plant there are two distinct varieties on the island which are recognised by the inhabitants. Of these (A.) Var Mane is the more scarce, at least I met with it less frequently. It forms a tree about 25 to 30 feet high, but I am informed also in some places forms a thicket, though I did not meet with it in such condition. The wood is very hard, is heavy and close grained, and makes a good timber, though difficult to work from its hardness. The bark of the trunk and large branches is thick, quite smooth, and does not split or crack. The bast layers are light-coloured, and make a capital cordage, which is greatly used, as cattle will not gnaw or eat it. The leaves arc pilose at the junction of the lamina and petiole, and the veins are puberulous. Tlie calyx lobes are cglandulosc. (B.) Var rouge. — This seemed to me the commonest variety, and forms dense tickets, and I do not recollect seeing it as a large tree. The wood is soft, porous, and very light, comparatively useless for carpentry, and only fit for burning. The bark is not so tliick as in Var blauc, is rough, cracking, and splitting after the maimer of that of an Oak. The bast layers have a reddish tinge, and do not make so good a cordage as those of Var T T 330 BOTANY OF ROniMGrKZ. blatic. The leaves arc less coriaceous than those of Var hlanc, and ratlur vclutinous than pilose. The epiealyx seems more deeply cleft, and ciicli lobe of the calyx has a linear median dorsal gland which is very consiiicuous. The forms arc very easily distinjinisJKnl at first sinht hy the bark. Tn no instance did 1 find any trace of heteroinor[)hisni in the leav(>s. This is infArcsting in connexion Avith an observation made in the Sahariuipore gardens hy Bell in 18(53 (Trans. Bot. Soc. Edin. vii. 5(55), and again by King in 1868 (Journ. Linn. Soc. XV. 101) on Hibiscus {Paritium) tricuspis, Cav. A lateral branch of this tree curvinij downwards " entered the soil, and rc-appcaring about 2 feet from its point of entrance, gaA'c rise to a lai-gc leafy bush." The leaves of this sport, unlike tlic parent, were not trifid, but like those of II. {]?(n'ilit('m) tiliuceun ; and in addition to some other peculiarities, the calyx segments had each " a large oblong gland full of a viscid secretion on the back." It must therefore approach the Va7' rouge of Rodriguez. Tliisclton Dyer, in a note appended to King's remarks, suggests that a. tricuspid, which is a South Sea Island plant, may be a local form derived from H. tiUacens, and this I think is not at all improbable. It is, however, curious that in an island where heterojihylly is so marked a character in many of the trees, no trace of it is observable in this species, although a heterophyllous variety is noted by Masters in Oliv. Elor. Afric. Troji. i. 208, as occurring in Africa. Thespesia populnea, Corr. ; DC. Prod. i. 456 ; Wt. Ic. t. 8 ; Hibiscus populneus, Linn. Sp. Fl. 976. Nom. vulg. Mahoe. Frequently met with on the shores, specially towards the western end of the island. Gossypium lierbaceum, imw. ; DC. Prod. i. 456; Wt. Ic. tt. 9,10, 11. Nom. "sailg. Cot on. Gossypium barbadeuse, Linn. DC. Prod. i. 456; Wt. III. tt. 28", 28", 28^ Nom. ATiIg. Coton. Both species of Gossypium are widely distributed over the island; O. harhadense is foimd most abundantly on He Pierrot, a small coral islet near the edge of the reef on the south side, covering it almost entirely to the exclusion of other plants. I believe it was introduced from an American ship, which was wrecked some years ago on the reef. The cotton produced is of very good quality, but the inhabitants are too lazy to pick it clean and make use of it. Eriodendron anfractuosum, DC. Prod. i. 479 ; Wt. Ic. t. 400 ; Bot. Mag. t. 3360. Nom. vulg. Ouat. A few trees of this are found planted about the habitations (now ruins) of the earliest settlers at the top of Soupir vaUey. STERCULIACE^. Dombeya acutangula, Cav. Diss. i. 38, f. 2; DC. Prod. i. 498. i\.n endemic Mascarene species of which I found only one specimen in leaf at the mouth of the Eivi^rc de I'Est. IS. BAYLEY BALFOUK. 331 Dombeya ferruginea, Cav. Diss. 1. 12, f. 2 ; J)C. Frod. i. 409. Noni. vulg. Bois inpc. Is met Avith in the upper parts of the valleys. The wood is heavy, and as the popular name implies is used for making pipes. The leaves are hetermorphic. The adult leaves frequently are more cordate and less crenatc tlian in the type, but all are densely fcrrugineo-tomentose beneath. In young trees the leaves are lan- ceolate and taper to the base, and the under surface is pale green and glabrous, not in the least tomcntosc. Cavanillcs I.e. says of the species " Folia sunt certe diversa," but this he applies only to the amount of lobation ; and then, again, he says, " tomcnto rufescente, in junioribus albicantc." His specimens were derived from Mauritius, so that in that island the species apparently varies in the same manner as in Rodriguez. This species is endemic in the Mascarene islands. Melochia pyramidata, Linn. ; DC. Prod. i. 490. Is natui*alised and grows everywhere. TILIACEiE. Corchorus trilocularis, Linn.; DC. Prod. i. 504. Established, and grows in abundance on the barren slopes of the western end of the island. The Rodriguez plant differs from the type slightly, the pods dehiscing usually by four or five valves. Triiunfetta glandulosa, Lam. ; DC. Prod. i. 506. Nom. vulg. Herbe a paniers. A common weed. An infusion of the leaves is a favourite tisane. GERANIACE^. Oxalis corymbosa, DC Prod. i. 696. Nom. vulg. Alleluia or Oseille. A plant I found in leaf in Riviere des Acacies is referred to this species. The leaves closely resemble specimens of a plant so named sent from Mauritius by Bouton ; and there is also a close resemblance to the figure of O. Martiana, Zucc, in Bot. Mag. t. 3938, a typical American species. As I got no flowers it is impossible to identify the plant witli certainty. Oxalis corniculata, Linn.; DC. Prod. i. 692; Wt. Ic. i. 18. Nom. vulg. Petite oseille. Very common everywhere. RUTACE.ffi. Zanthoxylum paniculatum, Balf. fil. Arborea, ramulis vaHdis tcretibus, spinis paucis nigrcscentibus uncinatis armatis ; foliis imparipinnatis, 1.5-19 foliolatis, ad apicem ramulorum confertis, breviter potio- latis ; foliolis oppositis, sessilibus, subcordato-oblongis, obtusis, basi inequalitcr cordatis, glabris, subcoriaccis, supra nitidis, subtus pallidioribus, costa venis(jue pro- T T 3 332 150TANY OK l!()I)HI(,;UKZ. minentibus; paniciilis siiliscssililnis, pntontibus, I'olio ln'Cvioril)iis, ramis puhorulis, innlicollis brovilnis ; cai^siila j^loliosa. Iiivalvi, t;labra iniimtr tulH>rciilata, hrcvilcr stipitata. Arbor glabra usciiu- ad 20 podos alta, cortico albido. Folio altorna, 5-8 poll, longa; I'oliola supcviora sa'i)o inajora, 2-3^ poll, louga, jJ poll, lata ; gcmmanim tegmenta extus gliitinosa ; pdiolus communis brevis, teres, siib|)uberulus. Flores ignoti. Capsiila fusco-nigroscens, .', poll, diam., nsqne ad l)asim in valvas dnas lissa, stipite ^i'. — I'.j poll, longo. Scwiiui ignola. Nom. vulg. Bois Pasncr. I only found one or two trees of tbis near tbe shore of Ansc Quitorze. The wood is white, fine grained, and very bard. It is not far removed from Z. Bndrunga, DC. Prod. 1. 728, hut differs in the want of large glands at the leaf ercnatures. Z. tomentellum. Hook. fil. I'l. Brit. Tnd., i. 1!)3, has some resemblance, but is more spiny. ToddaUa aciileata, Pcrs. ; DC. Trod. ii. S3 ; Wt. III. t. GO. Only in the higher parts of the island. Toddalia pauiculata, Law. ; DC. Prod. ii. 83. A species confined to Mauritius and Rodriguez. Is not common. Only in secluded parts of the island. In Mauritius, infusions of the leaves of this and of tlio foregoing species are much used as expectorants, whilst the bark is said to be astringent. Citrus medica, Linn. ; var. medica'proper, Hook. fil. Fl. Brit. Ind. i. 514. Nom. vulg. Citron. Grows spontaneously everywhere, forming in many places impenetrable thickets. It fruits very freely, and the fruit is in great demand. The leaves and rind of the fruit are used in preparing tisanes for various maladies. The Citron is distinguished by the inhabitants from the real sour Citron, which, however, I Avas never fortunate enough to find, but I am informed it grows abundantly in Maiu'itius. I suppose this latter is the var. acida, Hook. fil. I.e. The natives say that!if the seed of the sour Citron of Mauritius be sown in Rodriguez, it produces a less sour fruit and one like the Citron, and in fact becomes it. Citrus Aurantium, Linn. ; var. a. Aurantium proper, Hook. fil. Fl. Brit. Lid. i. 515. Nom. vidg. Granger. Only a few of these trees occur which fruit very freely. Citrus Aurantium, Var. h. Bigaradia, Hook. fil. 1. c. ; Wt. Ic. 957. Nom. vulg. Bigarado. This is the commonest Citrus on the island, and is very abundant, forming along with C. medica close and dense thickets. The fruit is not eaten ra^v, but is preserved in various ways ; and the leaves and the rind of the fruit are used for tisanes in shiveriugs and colic. IS. BAYLEY BALFOUR. 333 Citrus deciimana, Linn. ; Eook. fil. Fl. Brit. Ind. i. 516. Norn. wilg. Pam- plcmousses. One or two trees of this grow near some of the oldest hahitations. Besides those species and varieties mentioned the inhahitants speak of the " Vangasaille " which was described to nie as about the size of a mandarin Orani'c, and the "Limon," said to be a little smaller than the Citron. I did not meet with these forms. Triphasia trifoliata, DC. Prod. i. 53G. Nom. vulg. Orangine. Not common, and only near the dwellings of the early settlers, where manv introduced plants occur. SIMARUBACE.E. Sui'iana maritima, Linn. ; DC. Prod. ii. 91. Common along the shores where is coralline limestone and on all the coral islets of the reef. MELIACE^. Quivisia laciniata, Balf.fil. (Plate XIX.). Frutes glaber, ramosissimus ; foliis oppositis v. suboppositis, suhscssilil)us, niti- dis, rigidc coriaceis, reticnlato-venulosis ; adultis obovato-cimeatis, obtusis v. retusis, marginihiis integris, junioribus profunde pinnatifidis, lohis 3-5 ohtusis ereetopa- tentibus, in formam adultam gradatim transeuntibus ; floribus in cymas axillares bifloras brevissime pedunculatas collectis, pedicellis hreribus ralidisque, erectis V. suhorectis; colyce cotyliformi, minute 4-dentato, strigoso, dentil)us deltoideis ; petalis 1 oblongo-ellipticis obtusis, patentil)us ; staminibus 8, tubo brevi ; ovario dense strigoso ; stylo versus apicem subito incrassato ibique strigoso nee non con- stricto ; stigmate capitato, leviter umbilicato, ohscure 4-lobato, laevi. Frutex habitu Buxi. Folia diversiformia, f-l poll, longa, \-^ poU. lata. Pedi- cellus ^l- poll, longus, leviter strigosus. Alabastrus glohosus, subtrigonus. Calyx Jo poll. diam. Petala flava, J-|- poll, longa, cxtus in medio strigosa, ad mar"-ines glabra. Staminum tubus y\^ poll, longus ; anthera ovoidea, filamentis dimidio breviora. Friict. ign. Nom. vulg. Bois balais. This handsome member of an endemic Mascarene genus is very abundant in Rodriguez. It produces a hard fine-grained wood which is greatly used for making spoons. Ir, is nearly allied to Q. filipes, Baker PI. Maur. Seych. 46, a Mauri- tian species, but is sufficiently distinguished by the short, thick pedicels, the larger size of the flowers, and the heteromorphic leaves. Plate XIX. Pig. 1. Twig from adult jilant. 2. From a younger plant showing the pinnatiiid juvenile leaves. 3. From a plant younger than that of 2. 4. Flower 331 IJOTANY OF K()l)KI(ii:i:Z. bud. 5. Expanded llo\v(>r. (5. Corolla and androocium spr(>ad out. 7. Uotachod petal. S. Four staiurns do1acli(Hl. 0. (lyna^cium enclosed in calyx. 30. Transverse section of nearly mature ovary. 13. A'ertieal section of nearly mature ovary. Figs. 1, 2, 3, about nat. size. Best mac:nined. An introduced species ol" ]\[elia, called by the inhabitants Lilas, which furnishes a ■nood, termed J3ois de Siiicjapore is lound jdanted in a lew places on the island. CELASTRACE^. Eloeodendron orientale, Jacq. Ic. t. 48; DC. Frod. ii. 10. Nom. Vulp;. Bois d'olive. This is the n\ost frequently met Avith of all the trees in the island, and many specimens, where far removed from dwellings, have attained a largo size. The leaves are heteromorphic. The young ones are linear and acute, about 8 in. long by J in. broad, ■n'ith no petiole, or nn exceedingly short one, the midrib and veins being of a bright red colour, and the margins faintly undulate. From this we trace a suc- cession of forms to the most mature, which are obovate or elliptico -oblong, obtuse or retuse, 3 to 3.\ in. long by an inch or more broad, and with a distinct petiole I to ,}- in. long, the midrib and veins usually green and the margins distinctly crenate. The inhabitants distinguish two varieties of this tree. One thcj call " rouge " the other " blanc " according as the wood of the tree is red or not red, and the " blanc " is the harder wood of the two. But it is impossible to recognise these varieties. The wood of the yoving plant is usually reddish, and as the plant grows the new wood tends to become paler, untU in matm'e trees the new wood may be hardly red, or it may have a distinct red tinge. And so it happens that in every tree of this species the wood in some part of the diameter of the trimk has a reddish or pink colour. The wood is tough, and is used, more than any other tree, in car- pentry and for making " pirogues." I think this must be the tree to which Leguat refers when he says, " The Tree bore a fruit something like an olive, and the parrots " lov'd the nuts of it mightily." From this tree exudes an enormous quantity of gum in the form of tears, which soon harden and form large masses in crevices of the stem or on the ground around. RHAMNACE.ffi. Zizyphus Jujuba, Lam. ; DC. Prod. ii. 21 ; Wt. Ic. t. 99. Nom. vulg. Masson. Is occasionally met with on the island. The infusion of the leaves is frequently used for cough and cold. Scutia Commersonii, Brong. in Arm. Sc. Nat. x. 368, 1. 15,/. 1 ; Sc. indica, JBrong. I.e. Isom. vtdg. Bois senti. Frequently met with in the less frequented valleys, where it is very annoying IS. 15AYLEY liALFOUK. 335 from its recurved spiaes. It also occurs on some of the most elevated hill slopes. The hark has great repute in Mauritius as an astringent. Gouania Retinaria, DC. 'Prod. ii. 40; Retinaria scandens (volubilis in icon.), Gdrlii. Fnict. ii. 187. t. 120, f. 4. This endemic Mascarene twiner is not common. I only found it in two localities, hotli in the higher parts of tlie island. Baker (Fl. Maur. Seych. 52), following Bojer (Hort. Mam-. 77), makes this species a synonym of G. tiliccfolla. Lam., a Bourbon plant according to De Candolle, Prod. I.e. ; hut I camiot identify them. The description of G. Bef inaria " huctihwi^ alato-triquetris " seems to me irrecon- cilable with that of G. tilicefolia "fructibus subovatis apteris." SAPINDACE.ffi. Cardiospermum HaHcacabum, Linn., var. microcarpum ; C. microcarpum, JS.B.K. ; DC. Frod. i. GOl. T}-pe figured Bat. Mar/, t. 1049; Trt. Ic. t. 508. Norn. vulg. Bonnet des pr6tres. Is a common twiner. Allophyllus Cobbe, Blume ; Hiern in Hook. fil. Fl. Brit. Ind. i. 673 ; Schmidelia Cobbe, Wt. III. t. 141. This is a glabrous form of the species and the leaves are almost entire. It is not common, only found in one or two of the valleys. Sapindus trifoliatus, Linn. ; Hiern in Hook. fU. FL Brit. Ind. i. 682 ; S. emargi- natus, Vahl; Wt. III. t. 51; DC. Frod. i. 608. Norn. Vulg. Bois savon. A few trees are found on the shore a little east of Venus Point. Dodoncea viscosa, Linn. ; DC Frod. i. 616 ; D. Burmanniana, DC. Frod. i. 616; fVt. III. t. 52. Nom. vulg. Bois goiu-nable. Very commmon on the hill slopes. ANACARDIACEiE. Sclerocarya castanea, Baker Fl. Maur. Seych. 63. (Plate XX.) Arbor ramulis validissimis, terctibus ; foliis imparipinnatis 7-11-foliolatis, ad apicem ramulorum confertis, petiolatis ; foliolis oppositis, sessilibus, rarius brevissime petiolulatis, oblougis v. ovatis, acutis v. breviter acuminatis, basi inajqualiter rotuu- datis, obscure crenulatis, submembranaceis, supra nitidis glabrisquc, siiljtus medio nervo subhirsutis ; floribus dioicis ?, breviter pedicellatis, in racemis brcvibus ses- sUibus V. breviter pedunculatis sohtariis in axillis foliorum terminalium dense confertis; ? sepalis 5 minutis, rotimdatis, ciliatis ; petalis 6 oblongis, obtusis, reflexis, imbricatis ; disco crenato-lobato ; staminibus 10, partim anantheris ; ovario oblongo, glabro, biloculari (?) ; stylis 5 validis, distantibus, brcvibus, erect is, divari- ."."(! HOTANY OK KOOKlGrKZ. catis, sub apicem ovarii sitis ; stigmatibus capitatis, spongiosis ; oviilis solitaviis. l)oiululis. Ai'bur parva, glabra, us(|iu- m\ ;>() pedes altn. Folia O-i) iKtII. longa ; foliola 2^-3i poll, loiiga, i— la poll. lata, iulcriora minora, latioraquo ; potiolus communis bispitliilus, \\-'l poll, longus. liKcrmiiN potiolum subo^quans. Pela/a albida, ,'o i)oll. longa, sopalis triplo lougiora. S/amiiia basi disci inscrta, altcrne brcviora. Sii/ll .Ji poll, lougi. Fnictns ignotus. In the valley of the Rivirrc Pabnistc and at the top of the valley of the llivi6rc Mourue. This is a rare tree 20-30 ft. high, and I found it only in the two localities mentioned. I have followed Baker in describing it as a species of Sclerocarya ; but its live-symmetrical flowers and crenated disk are marked points of difference from the description of that genus. Of the many points of agi'cement I specially would note the mode of attachment (if tlie ovule to the placenta, which is very peculiar. It answers mucii more closely the description of Sarpejjhyllum, an imperfectly known monotypic Cape genus ; the male flowers of whicii are alone describ(>d in Bentham and Ilookc^r's Genera Plantarum (sec also llarv. and Sond. PI. Cap. f. 525), but of the female flowers I have seen a MS. description at Kew. Unfortu- nately the male ilowers of my plant are unknown ; but the female flowers correspond so closely with the description that I should have been inclined rather to place it in this genus had Baker not previously described it as Sclerocarya ; and as our information regarding the genera is imperfect I prefer to leave it so. May not this species bring Sclerocarya and Uarpepkyllum into one genus ? It seems to me to point in that direction, but with such fragmentary material as we possess it is impossible to decide. Plate XX. Pig. 1. Terminal portion of twig with leaves and inflorescence, 2- Plowcr bud. 3. Expanded flower, 4. Vertical section of flower. 5. Detached stamen. 6. Style detached with stigma. 7. Branehlet bearing fruit. 8. Transverse section of imripe fruit. Pig. 1 nat. size. Rest magnified. Mangifera indica, Linn.; DC. Trod. ii. 63; Bot. Mag. t. 4510. Nom. vulg. Le manguc. Many forms of this tree occur on the island. The inlial)itants told me that the seed of the Mango in Kodriguez never contains a small grub which is always, or nearly always, present in the Mango in Mamitius. The trees not being in fruit during my stay I had no opportunity of verifying the statement. But supposing it true, it seems to point to the probable absence from Rodriguez of some insect which is present in Mauritius and visits the Mango flower. May this have anything to do with the fertilisation of the Mango ? I may mention in this connexion that a great numb(,'r of Mango trees which flowered most profusely showed no signs of producing fruit. IS. BAYLEY BALFOUK. 337 LEGUMINOSJE. Crotalaria vetnsa,, Zinu. ; BC. Frod. ii. 126; Bot. Mag. t. 2561. Nom. vulg. Casse-cavclle. This is found under two forms on the island. One, an erect branching plant with bright green leaves nearly two inches long, grows in the valleys, and places where the scrub has not been shortened. The second form is a dwarfed and stunted irregu- larly spreading plant with smaller leaves, which are more silky ; this grows on the barren plains and open ground, where the vegetation is kept short by the cattle. Arachis hypogaea, Linn. ; DC. Prod. ii. 174) ; Mart. Fl. Bras. xv. pt. 1, t. 23, f. 1. Nom. vulg. Pistaehc. Cultivated as an article of diet. Indigofera argeutea, Linn. ; DC. Prod. ii. 224. Nom. vulg. Indigo batat. Is found in many places. Indigofera tinctoria, Linn.; DC. Prod. ii. 224; Wt. Ic. t. 365. Formerly was cultivated largely on the island, especially on the higher parts, and it has escaped in several places and is now naturalised. Tephrosia pui-purea, Pers. ; DC. Prod. ii. 251. Nom. vulg. Indigo sauvage. A common weed on the island. Desmodium incanum, DC. Prod. ii. 332. Nom. vulg. Gros treff. Grows everywhere. Desmodium mauritiauum, DC Prod. ii. 334. Nom. vulg. Petit treff. Everywhere on the grass slopes. Desmodium triflorum, DC Prod. ii. 334; Wt. Ic t. 292. Nom. vulg. Petite oseille marron. Common. Desmodium, sp. I got a single specimen of a Desmodium, but not in flower or fruit, which resembles a specimen in like condition gathered by Home in Seychelles, and which Baker (Fl. Mam*. Seych. 75), considers as most likely D. adsceudens, DC. Prod. ii. 332 ; Bot. Reg. t. 815. Abrus precatorius, Linn.; DC Prod. ii. 381. Nom. vulg. Reglise. A few plants only near Mathurin. Clitoria Ternatea, im«. ; DC. Prod. ii. 233; Bot. Mag. t. 1542. Nom. vulg. Ambrevade marron. Not common. Only near the shore at English Bay Point. Teramnus labialis, Spreng. Syst. Veg. iii. 235. Common on the hill slopes. Erythrina indica, Xam.; DC. P/'0(/. ii. 412; Wt.Ic.i.hS. Nom. vulg. Moiu-uc. u u 338 BOTANY ()!• UODKIGUEZ. This tree is occasionally found in most parts of tlio island, Init is s])(H'ially abundivnt at the mouth of the valley Eivicro Mouruc, on the south side of the island, to which it ijivi^s the name. Atylosia scai'abaeoides, Bent It. Fl. JidkjIi, 2'i2. Common near the shore. Rhynchosia luiuima, DC. Prod. ii. 385. Common everyAvherc. This plant is not re])orted from Mauritius nor from SeycheUes. Caesalpiuia Bouducella, Flem. in Asiat. Res. xi. 159 ; Guilandina Bondue, Boj. Ilort. Mitiir. 117- Nom. ^'^llg. Cadoe. Common ever}T\here. CsBsalpiuia sepiai'ia, Bo.vb. Fl. Iiid. ii. 360 ; jrt. Ic, i. 37. Nom. vulg. Cassic. Is not common, but is met with in a few valleys. Hseinatoxylou campecliiamiin, Linn.; DC. Prod. ii. 485; Benll. and. Trim. 3fed. PL t. 8. Nom. vulg. iJois campcche. This tree is planted as a hedge around gardens in the vicinity of Mathurin. Poiuciana regia, DoJ. Mort. 3fanr. 119 ; £oi. Mag. t. 2884-. Nom. vulg. Flam- boyant. A few trees of this are found planted on the links at Port Mathurin, in front of Government House, where they flower and fruit very freely. Cassia occidentalis, Linn.; DC. Prod. ii. 497; Pot. Reg. t. 83. Nom. vulg. Casse puante. A few plants of this arc found upon the island. It is often used medicinally. Tamarindus indica, Linn. ; DC. Prod. ii. 488 ; Bentl. and Trim. lied. PI. t. 92 ; . T. officinalis, Ilook. Bot. Mag. t. 4563. . Nom, vulg. Tamariu. Is found scattered over the island. It is said that the early Dutch settlers introduced this tree to Mauritius. It is used very extensively by the Creoles in treating disease. The bark is said to be astringent and tonic. The pulp of the fruit is well known as a mild laxative. Mucima gigantea, DC. Prod. ii. 405; Wt. Illustr. in Hook. Bot. Misc. ii. 351. Suppl. t. 14. Nom. vulg. Mort aux Eats. Common in many places. The popular name indicates the power ascribed to it by the natives. Mucuna, sp. I have the leaves of another Mticuna which I have not been able to identify with any species. The leaves are thick and coriaceous, and are more oblong-oval than those of M. gigantea. The flower and fruit I have not seen. It is a common climber in the valleys, forming very thick festoons from tree to tree. IS. BAYLEY HALFOUK. 339 Canavalia obtusifolia, DC. Prod. ii. 404. Grows on the sliorc at Englisli Bay, interlacing with Ipomcea pes-caprce. Roth. Canavalia ensiformis, DC. Prod. ii. 404; Bot. Mag. t. 4027. Nom. vulg. Coco- rico. Only found in the upper part of Rivitire Cascade valley. Phaseolus lunatus, Linn.; DC. Prod. ii. 393; Wt. Ic. t. 755. Nom. vulg. Haricot vert. Is cultivated on the island, and is occasionally foimd as an escape. Many varieties of Haricot were formerly cultivated, hut lately a caterpillar has appeared and so damaged the crops that they are now less commonly cultivated. Cajanus indicus, Spreng Sysf. iii. 248; Cajanus hicolor, DC. Prod. ii. 406; Bot. Beg. t. 31. Nom. vulg. Ambrevade. This is cultivated occasionally, and the seeds are used as DhoU. In some places it has escaped and grows spontaneously. This plant is reputed most efficacious medicinally as diuretic. Desmanthus virgatus, Willd. ; DC. Prod. ii. 445 ; Bot. Mag. t. 2454. Grows abundantly in the neighbourhood of Port Mathurin. Leucaena glauca, Benth. in Rook. Bond. Joiirn. Bot. (1842) iv. 416. Nom. vulg. Acacie. This plant was introduced into the island about 30 years ago, and now has spread everywhere, filling up completely many of the valleys, and destroying the indigenous vegetation. The young twigs are a favourite food for the goats, and the straight stems of the young trees are used as poles for propelling " piro- gues." Acacia farnesiana, Willd. ; DC. Prod. ii. 461 ; Bchb. Fl. Germ. Ic. t. 2052. A few plants of this occur planted as hedges along with Scematoxylon cam- pechia}mm L., near Mathurin. Albizzia Lebbek, Benth. in Hook. Bond. Journ. Bot. (1844) iii. 87. Nom. vulg. Bois noir. Is found abundantly on the island. This tree is said to have been introduced into Mauritius about 1767, from Bengal, by Cossigny, but I have no record of its reaching Rodriguez. ROSACE.^:. Prunus communis, Benth. et Hook.f. Gen. Plant, i. 610. ; Amygdalus commimis, Linn.; DC Prod. ii. 530. Nom. vulg. La pcche. Is naturalised. Rubus roiMo^ViS, Smith ; DC. Prod. ii. 556; Hook. Ic. Pi. iii. t. 349. Nom. Tulg. Framboise. Everywhere on the island. u u 2 310 r.oTANY ov KonijicrEZ. CRASSULACE^. Bryophyllum calycimuu, *S'(///67a ; DC. Prod. iii. 390 ; Bot. Mag. i. liOQ. Nom. i;. Soutu t'al'an. Not coinmon on \\\o island. Ts used as an a]i])lioation to bruises. COMBRETACE^. Termiualia Benzoin, Lhin.fil Sitppl. l^l {e.vcl. syn. et loc.) ; T. ruauiitiana, l/flw. ; J)C. Prod. iii. 11. Nom. vulg. Bois charron. This eiulemic Mascarone species is one of the licterophyllous trees of the island. It occurs abundantly. The contrast between the leaves of younjij plants and the adult form is so great that it was some time ere I could convince myself they belonged to the same species. The young leaves are linear and about 2 inches long, and | in. broad, very shortly petiolate, clustered at the ends of the branchlets, densely pubescent with undulated and recurved margins. The adult leaves are quite glabrous, with long petioles almost equalling the lamina, which is over 2 inches long and nearly an inch or more broad ; oval-oblong with a crenate margin and coriaceous. T. (ntgustifolia, Jacq. Hort. Vind. iii. t. 100, is a form of this with leaves narroAver than usual. The wood of the tree is very hard, and is the best for the purpose of wheelwrights, hence the common name. The bark is sujiposed to be a good astringent. Terminalia Catappa, Linn. ; DC. Prod. iii. 11; Bat. Mag. t. 3001. Nom. vulg. Badamier. Common on the island. MYRTACEiE. Psidium pomiferum, Linn.; DC. Prod. iii. 234; P. pyriferum, DC. iii. 233. Nom. vulg. Goyave. Frequent. The fruits are frequently preserved by the inhabitants. Psidium Cattleianum, Sabine ; Bot. Beg. t. 622. Nom. vulg. Goyave de Chine. I doubt A'cry much if this is really my plant. The fruit in the figure is purple, and is quite globular. Mine has a pyriform fruit which is bright yellow. It may be, however, a variety such as we have in the case of Ps. pomiferum Jj. Eugenia uniflora, Linn. Sp. PL 073; E. Michelii, Lam.; DC. Prod. iii. 263; Plinia pedunculata, Bot. Mag. t. 473. Nom. vulg. Roussaille. Often met near habitations. Eugenia Jambos, Linn. Sp. PL 672 ; Bot. Mag. t. 1696 ; Jambosa vulgaris, DC. Prod. iii. 286. Nom. vulg. Jamrosa. Jamrose. Very common in the valleys. I am told that in the seed of tliis fruit in Mauritius a grub is always found, just as in the Mango, but it is absent in the fruit as grown in Rodriguez. IS. 13AYLEY BALFOUR. 341 Eugenia Jambolana, Lam. Encyc. iii. 198 ; Wt. Ic. t. 535 ; Syzygium Jambo- lauum, DC. Prod. iii. 259. Norn. vulg. Jamlongue. A few trcos near habitations at Oyster Eay. Eugenia Balfourii, Baker FI. 3Iaiir. Sei/ch. 11(5. Ai'borea, ramosissima, ramulis glabris tctragonis ; ibliis brcviter pctiolatis, ob- longis V. ovali-oblongis v. ol)lanceolatis acutis, basi cuncatis, glabris, rigidc subcori- aceis, pellueido-punctatis, pcnninerviis, ncrvis tenuibus plurimis arctc positis ; paniciilis longe pedunculatis, paucifloris, axillaribiis, folia tequantibus, glabris, ramis late patentibus ; floribus scssilibus v. brevissime pediccllatis, paucis; calycc obscui'e dentato, dentibus deltoidcis. Arbor parva, glabra, usque ad 15 pedes alta, cortice albido corrugate. Folia opposita, 2i-3 poll, longa ; petiolus 1^ poll, longus. Fanicnla 1^2 poll. lata. Calyx ^ poll, longus. Fructus \ poll, diam., globosus, ruber, lobis calycis i)ersisten- tibus coronatus. Nom. TTilg. Bois clou. " Near E. Jambolana, from which it differs by its smaller leaves, with much closer veining and fewer larger flowers, Baker, 1. c." This is a small tree which I found growing in the higher part of the island, which has a remarkably white bark, and Baker has considered it a new species. It agrees very well, however, with the descriptions of Syzyg'mm paniculatum, DC. Prod. iii. 259, collected by Com- merson, in Bourbon, where it is known as Bois a ecorce blanche, which is Eugenia paniculafa, Lam. Diet. iii. 199. Unfortunately my specimens are very imperfect, wanting flower and perfect fruit, and I have found no specimens of Commerson's plant in the Kew herbarium with which to compare it, and it is therefore very difficult to determine the point, but I do not think it improbable that my plant is this species. In the meantime I have followed Baker. Eugenia sp. Growing on the slopes of the Grande Montague I found another species of Eugenia, only in leaf, which is not far removed from the foregoing, but it is impossible to determine it. Eugenia cotinifolia, Jacq. Obs. iii. t, 53. Nom. vulg. Bois de fer. Specimens of a small shrub, about 12 feet high, with a habit very Like a Holly, growing very abundantly on the island, but of which I neither got flowers or fruit, has been refen-ed by Baker 1. c. to the above species. The wood of the tree is vexy hard and heavy, hence its popular name. I think this species is probably heterophyllous, for I have found bushes with leaves much less rounded than the adult, but have no positive evidence. Foetidia mauritiana, Zam. III. t. 419; DC. Frod. iii. 295. Nom. vulg. Bois puant. 3.j,2 BOTANY OF IJODRiniTEZ. This peculiar Mascvurno plant is very common. The tree is heterophyllous, but the speeimous of hoteromorphism I !i;ot show by no means so extensive a variation as do many other trees. The h(>t(M-opliylly, so fas as I observed it, is most mai-lced Avhen the the ]dant is cjrowing on the seasliore, the young- leaves then being more elongated and approaching a linear form. The plant receives its name Bois puant on account of the sickening and disgusting odour exhahd t'ldiu tlie lcav(>s wlicu the sim shines on it. Tlic -wood is very good, and is often used for making " pirogues." LYTHRACEiE. Pemphis acidula, Forsi.; DC. J? rod. iii. 89. Norn. yulg. Bois matelot. Gro-ws abundantly on the shore where there is coralline limestone. Punica Granatum, L'uin. ; DC. Frod. iii. 3; But. Mar/. 1832, A and B. Noni. vulg. Grenade. ■ Is sometimes found natiu-alized. Of it the inhabitants distinguish two varieties. Grenade rouge and Grenade hlanc, the difference lying in the colour of the cndo- carp of the mature fruit, which in one case is tinged with red. These Avould corres- pond respectively to the varieties a, ruhrum, and h, albescens, distinguished by De Candolle, 1. c. I cannot say that I satisfied myself of the validity of the dis- tinction. Used as a powerful astringent. TURNERACE^. Mathurina, Balf.jil. in Linn. Soc. Journ. xv. 159. (Plate XXI.) Sepcla 5, ovato-lanceolata v. elliptico-oblonga aciiminata, costa prominente, "•landula mas-na bilobata intus basi adnata. Fetala 5, subhypogyna, obcuneata v. ohovata acuta, nuda, subungiiiculata, basi sepalis leviter adnata, reclinata et corrugata. Stamina 5, subhypogyna, exserta ; filamentis subulatis calycis glandulis in fundo imo vix adhcerentibus ; antheris lineari-oblongis introrsis. Ovarium sessile, unUo- culare, libcrum, oblongum, glabrum ; owila adsccndentia ; styli 3, filiformes, termi- nales stigmatibus dilatatis subfimbriatis. C(i2)snla oblonga, triquctra, glabra, 3-valvis, poly spcrma. Semina obovoideo-cylindracea, lente curvata, funiculo brevi, arillo lono-e piloso-sericeo basim seminis circumdante, testa Crustacea extus foveo- lata, albumme carnosa ; embryo axilis, rectus, cotyledonibus ovatis plano-convexis, radicula tereti. Arbor parva. Folia alterna, petiolata, lanceolata, saejie obovata, acuta, crenato- serrata ; petiolus 2-glandulosus ; stipulae 'glanduliformes, deciduse. Flares magni, pedunculati, vulgo solitarii, axillares, nonniinquam in cymas trifloras dispositi, albi ; pedunculi infra medium articulati, 2-bracteolati ; bracteola) subfoliacea3, scrratse V. crenatse, lineares. M. pendiiliflora (species unica). Arbor parva, usque ad 20 pedes alta, ramis erectis, foliorum cicatricibus magnls. Folia- lanceolate v. obovata v. obcuneata, acuta, 3-4 poll, ^onga, 1 poll, lata, penni- IS. BAYLKY HALFOUR. 313 ncnia, norvis teniubiis ncnailis iutramarginalibiis conjunctis, seel per adolcscentiam lincaria tcI ligulata, ^ poll, lata, iu formam adultaiii gradatim transeuntia ; petiolo breri, marginc utroque versus medium glaudula instructo, lamina decurrcntc. Flores albidi ; pedunculi 1-2 poll, lougi ; bracteolae lineares. Glanduke sepalorum intus sulcatDD, pubescentes, apice emarginatte. Sepalu petalaque 1 poll, louga. Stamina periantliium dimidio excedentia. Ovarium glabrum, oblongum, 1 poll, longum ; sty lis iucurvatis. Nom. Tulg. Bois gandine. Usually found on the higher parts of the island. The stem is usually thickly clad with lichens, and the wood is light coloured and fine grained. The tree is heterophyllous. The young leaves are quite linear, about ^ inch broad, with slight widely seimrated serrations, but the adult leaves are usually obovate or obcuneate, almost an inch broad, and with very marked crenaturcs. The nearest affinity is with the monotypic genus Erblichia of Seemann, a native of Panama, from which, however, it is distinguished by the sepaline gland, absence of petaline fringes, stigmas, and the arillate seeds. Plate XXI. Pig. 1. Prom a photograph, 2. Twig from advilt with typical leaves. 3. Leaf from a young tree, more linear and elongated. 4. Twig with leaves from a very young plant or adventitious shoot. 5. Plower spread out. 6. Detached sepal. 7. Detached petal. 8. Gynajcium. 9. Ajdcx of style. 10. Pruit dehiscing. 11. Ti'ansverse section of fruit. 12. Seed (nat. size). 13. Seed magnified. 14. Embryo, PASSIFLORACE^. Carica Papaya, Linn. Sp. Fl. 1466 ; Bot. Mag. tt. 2898, 2899 ; Papaya vulgaris, DC. Frod. XV. 1,414. Kom. vulg. Papaye. Grows now spontaneously in several places. The juice is used most extensively, specially as an anthelmintic, and also for several other diseases. The inhabitants hold the common idea that fresh killed meat if hung up under this tree for an hour or two becomes quite tender. CUCURBITACE^. Lagenaria vulgaris, Ser. ; DC. Frod. iii. 299; Wt. III. t. 1057. Is cultivated. Momordica bakamina, Linn. ; DC. Frod. iii. 311. Nom. vulg. Margose. Cultivated and occasionally an escape. Citrullus vulgaris, Schrad. in Hckl. et Zeyh. Ennm. 279. Nom. vulg. Melon d'eau. Cultivated and sometimes found as an escape. The Water melon was one of the plants introduced by Leguat. lie says they 3-i-J. I'.OTANY OK KODlilC'.UKZ. brouijlit five sihhIs tVoin Ihr C';i))(' of Goodllopo, and the plants sjiriii^in^' from ihose he doscribos tliiis : — " Among oni- iiv(> plants of walor luolons there wcvv two sorts, ' red and white; the first Averc the l)est. 'riii> riiul Avas grecMi and the inside red ; * they are very refreshing, and never do any Innt, no more than the others (i.e. ' ordinary nudons) : they are so lull u\' water that one may easily go -without drink ' when they are ealiMi ; sonietimes they were so big liiat all eight of us (;onld • hardly i>at np one of then). These several kinds of melons grew Avithout taking ' i)ains about them, as 1 have said already, and produecd fruit in great abundance. ' AVhen we mingled a little ashes Avith the earth in the ])laee Avherc they AA^cro soAvn ' it maile 'em grow and fructify extraordinarily, and the fruit AA^as more than ' ordinarily delicate." Citrulliis Colocyuthis, Sclirad.; Naiul. in Ann. Sc. Nat. Ith scr. xii. 09; TFt. Ic. t. 198; Cucumis Colocynthis, Liidi. ; DC. Prod. ili. J502. A plant Avhich is provisionally referred to this species is found in several places on the island, usually on coral or on sandy soil. The specimens are too imperfect for absolute determination. Seeds are, hoAvcver, soAvn at Kcav, CACTACE^. Opimtia Tima, Mill. ; DC. Frod. iii. 172. Nom. vulg. Raquctte. Occurs near habitations, often planted as a hedge. FICOIDEiE. Sesuvium Portulacastrum, Linn. ; DC Prod. iii. 453 ; Bot. Mag. t. 1701. EvcryAvherc on the shore about high Avater mark. UMBELLIFERiE. Hydrocotyle bonariensis, Lam.; DC. Prod. iv. 60. GroAVs in many valleys. Daucus Carota, Linn.; DC. Prod. iv. 211; Eng. Bot. t. 515. Nom. vulg. Carotte sauvage. Common on the hills. ARALIACE.ffi. Gastonia ciitispongia, Zc/^m. ; DC. Prod. iv. 256; Polyscias cuih^ow^m. Baker Fl. 3Ianr. Seych.lil ; Polyscias repanda, ^«A;er, pars quo ad habitat Rodriguez. Nom. viilg. Bois blanc. A scarce tree only grovdng on coralline limestone. I have referred my plant to this, the solitary species of an endemic Bourbon genus, but the Rodriguez plant is not typical. The leaves are more rounded at the base, are less coriaceous and have distinct petioles ; the calyx is smaller ; the fruit also is more globular, the IS. BAYLEY BALFOUR. 315 style disk louprcr and style branches shorter and more recurved, and the whole more deeply umhilicatod than iu IIk; type form. These variations might almost be con- sidered specific, but as my specimens are not perfect, I think it is better to include it under this species, which seems to be very variable, until more complete specimens are obtained. Baker, Fl. Maur. Seych. 126, unites Gastonia with the genus Foli/scias and refers my plant to Foli/scias repamlo, Baker, to which he also refers Gilihertia repanda, D.C. I cannot agree with him. Folyscias and Gastonia are very closely allied, but the articulated pedicels and the calyculus of the former are very characteristic, as also the few-celled ovary, and keep them sufficiently distinct. P. repanda, Baker, so far as the description applies to the Mauritian plant is a true Foli/scias ; but the Bodriguez plant, which he also includes, is a Gastonia, and if not a variety of, is very nearly allied to, Gastonia outisponrjia, Lam., as above mentioned. RUBIACE^. Danais corymbosa, Pr///". y?/. Herba scandens volubilisve, ramulis tenuibus, glabris, tetragonis ; foliis oppositis, breviter petiolatis, oblongis v. lanceolatis acutis v. acuminatis, basi cuneatis, iutegris, glabris, subcoriaceis, reticulato-venulosis, subtus paUidis ; stipulis minutis ; cymis corymbosis axillaribus in axlllis foliorum terminalium, densifloris, breviter peduncu- latis, pedicellis erectis, tenuibus, brevibus, bracteolis mimitissimis ; calyce 5-dentato, tubo campanulato, deutibus lanceolatis ; corolla hypocrateriformi, segmentis oblongo-spathulatis acutis, patentibus, tubo dimidio brevioribus, fauce dense villosa ; flonun brevistylium staminibus longe exsertis ; stylo fiu'cato tubum corollte tequante, ramulis teretibus clavatis; capsula globosa, glabra. Serba lignosa, late scandens. Folia pallide-virescentia, ad extremitatem utramque attenuata, 2-1 poll, longa, ^-,1 poll, lata, marginibus siccitate revolutis sub-repandis ; petiolus J—^ poll, longus ; stipulse deltoidea3,Vo poll. longa>. Corym- bus l|-2 poll, diam ; pedunculus ^ poll, longus ; pedicelli sub-pubcruli, j'.,-J poll, longi. Corolla ^\y T^oU. longa, caljce ti'ijAo-longior. ^«^/f- fragrans, Comm., D. rotundifolia, Poir., and D. laxijlora, DC, are merely forms of one species, and they have been all reduced to one D. fragrans, Comm., by Cordemoy in Adansonia x. 356, whom Baker follows, Fl. Mam-. Seych. 137. D. sulcata, Pers., the fourth species mentioned by De CandoUe, is probably also a form of D. fragrans, Comm. The Rodriguez plant is not unlike some of the forms of J>. fragrans, Comm., but differs conspicuously in the form and long petiolatiou of its X X 316 BOTANY OF RODRIGUEZ. leaves and its longer paniculate infloreseenco, the racliis exceeding considerably the petiole. The flowers in this genus formerly considered dioecious have been shown by Cordenioy I.e. to be really dimorphic. I only collected the short-styled form in Rodriguez. Oldeulaudia Sieberi, Baker, var cougesta. Herba perennis dense caespitosa, caule brevi, rauiis coufcrtis stcUatim pa- tentibus, tetragonis, subalatis; foliis oppositis -^ poll, longis, ovatis v. oblongo- ovatis V. obovatis obtusis, iufcrnc in petiolum brevem attenuatis, glabris incras- satis, coriaccis, nitidis. This plant gi"ows only in tufts on the coralline limestones along with a small species of ^nia. Baker, El. Maur. Seych. 138, considers it a distinct species, but I do not think the characters arc sufficient to separate it from the Mauritian 0. Sieberi, Baker. Its congested habit, w'hich is the only marked point of distinction, is quite accounted for by its habitat on dry limestone soil ; the type O. Sieberi, Baker, being a plant of roadsides and damp ground. 0. callipcs, Griseb. of Coll. Wright, PI. Cub. n. 2G7S in Kew herbarium seems also a very close ally. Raudia heteropliylla, Balf.Jil. Plate XXII. Suffrutex glaber, ramulis tetragonis ; foliis oppositis, breviter petiolatis, rigidc coriaccis, glabris, supra nitidis, adultis oblongis v. cllipticis, obtusis mucronatis v. emarginatis, ad cxtrcniitatem utramque rotuudatis, vcl sa:!pe lanccolatis et versus extremitates attenuatis, juvenilibus liucari-lanccolatis, clongatis, acutis, hispidulis, gradatim informam adultam transcuntibus ; stipulis brcvibus, connatis, subtruncatis ; cymis solitariis, extra-axillaribus, patentibus, 1-5-floris, peduuculis glabris petiolum longe excedcntibus, bracteolis fere obsoletis ; floribus sessilibus v. brevissimc pedi- cellatis, ereetis ; calyce angnste-infundibuliformi, minute 5-dentato ; corolla hypo- crateriformi, fauce breviter villosa, segmentis lanccolatis ; anthcris pai'tim exsertis, ligulatis, acutis ; ovario 5-gono ; fructu ovoidco-oblongo 5-angulato. Suffrntex inermis. Folia heteromorpha, opposita, adulta 2\~Q poll, longa, 1^-2^ poll, lata, juniora pedem excedentia vixque poll, lata; pctiolus J^ poll. longus ; stipulse \ jioll. longse, extus glabrge, intus piloso-sericeae. Pedwictdus glaber £-1 poll, longus. Calyx angulatus, dentibus obscure deltoidcis, ciliatis. CoroUce tubus \ poll, longus, calycem sextuple excedens, segmentis tubo longioribus. Discus pulvinaris. Anthei'a ^ poll, longa. Fruclus costatus, coriaceus, l|-2 poll, longus. Nom. vulg. Cafe marron. This is one of the prettiest and most interesting plants from Rodriguez, be- longing, as it does, to a genus hitherto unknown in the Mascarene Islands, though abundantly represented in Africa. It is one of the few relics of the old Flora of the island, and is only found ia most unfrequented spots at the heads of the valleys. It is heterophyllous as is represented in the plate. I am inclined to think there are two species on the island, one of which in the adult has eUiptical leaves or leaves IS. BAYLEY BALFOUR. 347 roiinded at both cuds, and the other with leaves narrowing to both ends; but my sjiecimens are not sufficient to determine the point, and where we find, as we do in this flora, such variations in individual characters amongst so many species, we must allow a very wide range of specific variation. The parts of this plant are not put to any use by the natives, and indeed it is so scarce many of them are imawai-e of its existence. The wood is hard and white. Plate XXII. Fig. 1. Twig from a young plant, with narrow but not very long leaves. 2. Leaf from a younger plant. 3. Leaf from an adult, typical form. 4). Flower bud. 5. Flower expanded. 6. Stamens detached. 7. Vertical section of ovary, with style attached. 8. Transverse section of ovary. 9. Fruit not mature. 1, 2, 3, 4, 9, nat. size. Rest magnified. FerneUa buxifoUa, i«?«. ; 2) C. P>'0(^. iv. 398. Plate XXIII. Nom. vulg. Bois bouteille. This small tree or shrub is very common on the island, and exhibits a very marked heterophylly ; the leaves or young plants and on adventitious shoots being very minute, oval, and rigid, but not spiny. In the adult they become almost orbicular and lose much of then* rigidity. This is one form of what I have referred to in my introductory remarks as the fii'st type of heterophylly. To illustrate this a figure of the plant is given, but the heterophylly is not so clearly marked in the plate as I could wish for. This variation accounts for the multiplication of species in this gemis. In DC. Prod. iv. 398, there are two described in addition to F. buxifoUa, Lam., viz., F. obooata, Lam., and F. pedunculata, Gartu., but these have rightly, I think, been reduced by Baker, Fl. Maur. Seych. 142, to the type species, F. buxifolia, Lam. So that we have in Fernelia a Mascarene endemic monotypic genus. Some confusion as to the popular name of this tree has arisen, and it is often referred to as Bois de ronde. This is the name of Psiadia rodriguesiana, Balf. fi.1., a Composite plant. But the name Bois de ronde is often erroneously given to Carissa Xylopicron, Pet. Th., an Apocynaceous plant, of which the leaves, and specially the young leaves, are very hke those of Fernelia biixifoUa, Lam., and hence the name Bois de ronde has been sometimes associated with F. huxifolia. Lam. The wood is hard, but not used for any special purpose. Plate XXIII. Fig. 1. Twig from an adult plant. 2. Twig from a young plant. 3 and 4. Leaves of a different form from an adult plant. 5. Flower expanded. 6. Flower in vertical section. 7. Fruit. 8. Fruit in vertical, 9. in transverse section. 10. Seed in vertical section. 11. Embryo. Figs. 3-6 from Mam*itius specimens. Antirrhcea frangulacea, DC. Prod. iv. 460. Nora. vulg. Bois goudron. A shrub 9-10 feet high I found growing on the coralline limestone at the south- west end of the island, unfortunately not in flower or fruit, resembles in foliage this endemic Mascarene species. The wood is bright yellow. XX 2 318 l^OTANY OF IJODKIGUEZ, Vangueria ediilis, J'ahl ; DC. Prod. iv. l,")!. Nom. vulg. Vavanguo, Voavang. A common plant on the islaiid. Pyrostria trilocularis, Jiaff. Jll. ; rioctronia ? trili)c\il;nis, Baker Fl. Maur. Sei/cli. 117. Plato XXIV. Piutox ramosus glabcv, ramnlis totragonis ; fuliis rigitlc coriacois, marginihus lovitov rovolutis, glabris, nitidis, ponnincrviis, subtus eosta proniincntcs adult is brcvitcr petiolatis, oblongo-ovalibus v. oblongis v. fere obovatis acutis, juvcnilihiis sub-scssilibus, liuearibus, nmcronatis ; stipnlis connatis glabris, deciduis, longe cus- pidal is ; floribus axilla vibus ; fructu 3-raiius 2-loculavi, globose, subtrigono, brcvitcr sti])itato. Folio bcteromorpba, 2-5 poll, longa, 1-1.! poll, lata, pctiolus } poll, longus ; juvenilia 1-1 o poll, louga, ,', poll, lata; stipultc cuspide lamiuani icquantc, { poll, longa'^. Dr/qxi rubra, {'.^ poll, diam., pyrcnis }. poll, longis ; seminibus infra medium afiB.xis ; embryone axili, curve ; cotylcdonibus cordatis, compressis ; radicula tcrcti, inforiere. Nom. Tulg. Bois cliauvc souris. Tbis is a small lietcropbyllous tree or slirub about 15-25 ft. liigb, of A'cry erect babit, tbc young leaves being narrow and linear and acute, becoming in the adult broader and more obtuse and usually shorter. It is very abundant and bats feed greatly on its fruit, hence its common name. I have had considerable difficulty in determining to Avhat genus to refer this plant, and its position in the genus Pyrostria must only be regarded at present as provisional, as my specimens are imperfect, I have no flowers. With Pyrostria it agrees in most points ; the fruit is, however, fewer-celled than is common in that gen;is. The embryo of the Rodriguez plant is slightly curved, and has an inferior radicle. . Of Pyrostria the embryo is at present unknown, but the genus is located amongst the Vangueria;, Benth. et Hook. Gen. Plant, ii. 22, a group characterised by a superior radicle. An important result then of the confirmation of this plant as a Pyrostria will be the removal of that genus from its present position to one in the tribe Lvorece, where the radicle is inferior when it would come near 3Iyonima and Coffea. With Ilyonima the plant has many points of resemblance, more especially with a species, M. heteroplujlla, Bej. Hert. Maur. 169, but it differs therefrom in its axillary clustered infloi'cscenee, not terminal panicles. With Coffea also it has certain resemblances, but there are sufficient characters to keep it out of that genus. Baker refers it doubtfully to Plectronia, but it is impossible to put it there, as in that genus the radicle is superior. Plate XXIV. Fig. 1. Fruit bearing twig from an adult tree with type leaves. 2. Twig from a younger tree with narrower leaves. 3. From a still younger plant, or from an adventitious shoot. 4. Fruit. 5. Fruit in vertical section. C. Fruit in transverse section. 7. Pvrene in vertical, 8. in transverse section. 0. Seed in IS. BAYLEY BALFOUR. 349 profile. 10. Seed from the side of attachment. 11. Emhryo. Figures of leaves natural size. Rest magnified. Scyphochlamys, Balf.fil. Calycis tubus campauulatus ; limbus truucatus v. obscure 4-lobatus, coriaceus, persisteus. Corolla infuudil)ularis, tubo brevi, fauce villosa ; limbi lobi 1-5(6 ?) lanceolato-triquetri, acuti, iucrassati, valvati. Stamina 1-5(6?) ore corollte inserta, fiiameutis brcvibus, autlieris dorso infra medium affixis, vix exsertis, lanceolato- acutis. Discus annularis v. pulvinaris. Ovarium 4-5(6 ?) loculare ; stylus validus, profunde 1-fidus ; ovula in loculis solitaria, infra apieem loculi pendula. Frudus pyriformis, costatus .... (maturum non vidi). Scmbta pendula, non compressa. Arboi-es parva?, glaberrima}, ramuhs subtetragonis. Folia opposita, breviter petio- lata, rigide-coriacea, penninervia ; stipultc intcrpetiolares, latne, coiinatse, coriaceae, persistentes. Flores in capitulum 6-12-florum densum pedunculatum congest!. Capltula intra bracteas duas ojipositas conniA-entes crasse coriaceas persistentes in involucrum cyathiforme connatas inclusa. S. revoluta, i?«//.^/. Plate XXV. Arbor parva, ramosa ; foliis hetcromorphis, adultis breviter petiolatis, ellipticis V. late ovato-oblongis, nitidis, venulosis, margiuibus revolutis, 3i-4 poll, longis, 1^-2J poll, latis (petiolo excluso |-i poll, longo) ; juvenilibus linearibvis 4-6 poll, longis ^ poll, latis ; stipulis ^ poll, longis ; capitulis solitariis, axillaribus, erecto patentibus ; pcdunculis petiolum excedentibus, -^-xV poU- longis ; involucre ^ poll, diam., basi intus tuberculis ajipressis brevibus teretibus linearibus dense vestito ; corolla \ poll. Ibuga, calyce triplo lougiore ; fructu in singulo involucro solitario. Nom. vulg. Bois mangiie. This small tree grows commonly along with Randla heterophylla, Balf. fil. in secluded parts of the island, and is not common. I am unable to refer it to any known geniis, and have taken it as the type of a new one. Its nearest affinity is with Pi/rostria, but it is distinguished by the sessile anthers, which are included or sub-included ; the style, which is thick and is deeply 4-clef t ; but the most lorominent feature is the large bracts, which form a complete involucre to the flowers. These bracts are clothed at the base within by many short conical adpressed processes, which may represent abortive flowers. The embryo in the single immature fruit obtained is imfortunatcly too decayed to determine its relations. This tree is heteropliyllous. The leaves of yoimg plants being linear pass through a gradation of forms to the adlilt type, which in many cases is almost orbicular. Plate XXV. Fig. 1. Twig from an adult tree, with type form of leaf. 2. From a younger plant. 3. From a very young plant or adventitious shoot. 4. One half of involucre. 5. Flower bud removed from involucre. 6. Flower expanded. 7. Corolla and androccium opened out. 8. Stamen detached. 9. Gynajcium with portion of calyx attached. 10. Transverse section of ovary. 11. Vertical section of ovarv, 350 liO'l^VAY Uh' KUDKlCiUEZ. Avitli portion of calyx attarhcd. 1:2. Siiiylo not ripe fruit enclosed in involucre. 13. Vortical section of unripe fruit. 1, 2, 3, I, 12, 13, nat. size. Rest magnified. Coffea arabica, Li int.; DC. Froil. iv. ii)0 ; Bot. May. t. 1303. Nom. vulg. Caf6. lias Ix'cn largely cultivated on the island, but now seldom ; but many good- sized trees arc found in the vicinity of old plantations. Psychotria r lauceolata, Balf. fil. Frutex glaber, ramulis tetragonis ; foliis oppositis, lanccolatis, acutis, inferne in petioluni brevem gradatim attcnuatis, coriaccis ; stipulis deciduis ; floribus niinutis, breviter pedicellatis in paniculam corymbosam pedunculatam ebracteatam v. minu- tissimc bracteatam dispositis, paniculis in axillis foliorum superiorum congcstis ; ealyce late cupuliformi, trimcato, obscure 5-dentato ; corolla subeampanulata, lobis lanccolatis incrassatis tubo brevioribus, fauce dense pilosa ; antheris dorso aiSxis, filamentis brevibus ; disco magno ; ovario 2-loeulari ; ovulo oblongo-ovato, com- presso, in singulo loculo soUtario, erecto, basilari ; stylo f urcato, ramis brevibus. Fruct. ign. Frv.tcx erectus. Folia opposita, 3-4* poll, longa, ^.-l poll, lata, internodiis brevibus ; petiolus \ poll, longus. Pcduncidns petiolo triple longior, pedicelli petiolo breviores. Corolla \ poU. longa, calycem triplo exccdens, tubo ^ poll, lougo. Anthers iV poll, longse. Nom. vulg. Bois lubinc. Tliis small shrub is exceedingly rare. I only know of one plant in the island on the slopes of the Grande Montague. My specimens arc very imperfect, as I have only flower-buds, and no fruit. It is difficult therefore to be certain of the genus, but the 2-celled ovary, with a solitary basilar erect ovule in each loculus, places it distinctly in the tribe Fsychotrice, and it seems to fall most naturally into the genus Psychotria. COMPOSITE. Ageratum conyzoides, Linn. ; DC. Prod. v. 108 ; Soak. Exot. Flor. i. 15. Nom, vulg. Abgrat. A common weed. Eupatorium triplinerve, Vahl. Symb. iii. 97 ; E. Ayapana, Vent.; DC. Prod. v. 169. Nom. vulg. Ayapana. Frequently met with. A reputed panacea. Was introdviced into Mauritius from Rio Janeii-o in 1797 by Captain Augustin Baudin. The story is, that he having heard at that j)ort of its medicinal virtues, and having endeavoured in vain to obtain living plants, the night before leaving stole a plant from the window of one of the houses, which he carried to Mauritius. Psiadia rodi'iguesiana, Balf. fil. Suffrutex, ramulis teretibus pubescentibus ; foliis lanccolatis, acutis, superne pro- IS. BAYLEY BALFOUR. 351 funde-scrratis, inferne cuneatis integrisque in pctiolum pilosum brevcm attcnuatis, dense pilosis, uervo ccutrali subtus promiiientc ncrvos 7-9 adscendentcs parallclos subtus proniinulos emittente ; capitulis breviter peduuculatis, in corymbos laxos axillares patentes dispositis; bracteis involucri glabri interioribiis lauceolatis v. oblongis acutis margiuc submemliranaceis, exterioribus brevioribus post antbesin patcntibus ; acbrerdis radii compressis angulatis, rugis calvis, inter rugas strigosis, pappo rigido setoso-scabriusculo. Folia 3—1 poll, longa, f-1 poll, lata ; petiolus \-\ poll, lougus. Cymce oligo- cephalae ; peduuculi breves, puberuli ; bractese interiores ^ poll, longse. Nom. vulg. Bois de roude. This is now a very rare plant in Rodriguez. It must have existed foi-merly in great abundance, as the limestone plains are in many places thickly strewn with fragments of branches and stems. The wood is very hard, and has a dark ochry colour. It differs from most species of Fsiadia in Maiu'itius in being velutino- pubescent, not glutinous, and resembles many of the Bourbon species, originally described as species of Conyza, but now referred to Psiadia. Psiadia Coronopus, Benth. et Sook. Gen. Plant, ii. 285 ; P. trinervia, Willd. var. macrodon, Baker Fl. Maur. Seych. 172; Sarcanthemum Coronopus, Cass.; DC. Prod. v. 367. Plate XXVI. SufTrutex ramulis teretibus glutinosis ; foHis anguste oblongo-lauceolatis, basi attcnuatis, apice serratis, triplinerviis, coriaceis ; capitulis breviter peduuculatis in corymbos terminales dispositis ; involucris hemisphgericis fioribus paulo brevibus ; receptaculo paleaceo fimbrilHfero ; florima tubo crasso ampliato ; achceniis radii glabris compresso-obovatis pappo setoso paleaceo. This is a very interesting plant. It is very rare on the island. I only found it at one point, on the shore, near the mouth of the Riviere Pou.rsuite. It differs from all other species of Psiadia in the presence of scales on the receptacle, and in the dilatation at the base of the corolla tube. The plant was originally described by Lamarck, Encyc. ii. 89, as Conyza Coi'o- twjius, fi'om specimens brought by Commerson from Rodriguez. I have compared my plants with the original specimens of Commerson in the Museum at Paris, and have confii'med their identity. Cassiui in Bull. PhUom. 1818, p. 74 {see also Diet. Sc. Nat. xlvii. 349, and DC. Prod. v. 367), created a new genus, Sarcanthemum, for the plant, distinguishing it by the receptacle and dilated corolla tube. But Bentham and Hooker, Gen. Plant. I.e., place Sarcanthemttni under Psiadia, remarking " planta Mascarensis a nobis non visa ex charactere dato a Psiadia non dififert nisi " corollis basi crassis ampliatis." The specimens brought now from Rodi'iguez admit of a full examination of the plant, with the result that it is confirmed as a species of Psiadia, although it differs slightly from that genus in the corolla, and also in the fimbrilKfcrous centre of the receptacle. 352 BOTANY OF KODRTrj'F.Z. Baker, I'l. Alaur. Soycli. 173, rot'ors it as a varirly fo P. (nnevv'ia, Willd., but this it is certainly not. Cordemoy in Adansouia x. 21, supposes Comniorson's plant may he I*, rctusa, Lam., hut this also is erroneous. As to the (list rihut ion of the plant, I helieve it to he endemic in Rodrij^'uez. l)e Candollc I.e. puts ^NTauri- tius as a locality on the authority of Sprengel, hut querrics it. Sprengel himself, Syst. A'eg:. iii. 510, only says, '' Ins. Mascarcn," which docs not necessarily refer to Mauritius. And neither Bojer, Ilort. ^laur. T79, nov Eouton,knew the plant there. Bourhon is also given hy De Candolle I.e., on the authority of Bory, from whom he received specimens, hut I am inclined to think there has heen some confusion as to the locality. Cordemoy knows nothing of the plant in J5ourhon. Plate XXYI. Pig. 1. Capitulum isolated. 2. Capitulum in vertical section. 3. Flower of ray vritli an involucral scale. 4. Flower of disk with hasal scale. 5. Stamens and style from flower of disk. Partheuiiim Hysterophorus, Linn.; DC. Prod. v. 532; JBot. Hag. t. 2275. Nom. vulg. Ilerhe blanche. A comnion weed in waste ground near habitations. An infusion of this plant is a favourite tisane. Siegesbeckia orientalis, Linn. ; DC Prod. v. d05 ; TFt. Ic. t. 1103. Nom. vulg Herbe de flacq. A common weed near habitations. Bidens pilosa, Linn. ; DC. Prod. v. 597. Nom. vulg. Villc bague. Common near habitations. Tridax procumtens, Linn. ; DC Prod. v. 079. Abundant on the coralline limestone of Ptodrigucz and on the coral islets around. Abrotanella rhyncbocarpa, Balf.fd. Plate XXVIIa. Herba pusilla, pulvinata, glabra, caulibus dense confertis ; foliis imbricatis, stel- latim patentibus recurvatis, late amplexicaulibus, lyrato-pinnatifidis v. piunatipar- tilis, rarius et solum versus apicem ramulorum simplicibus spathulatisque, acutis, coriaceis, unmerviis, nervo prominente ; capitulis solitariis sessilibus, singulo G-10- floro; involucro 6-pbyllo, bracteis ciliatis, exterioribus plurinerviis latioribus, in- terioribus 2-3-ncrviis angustioribus ; receptaculo foveolato ; floribus 3-4 exterioribus femineis, interioribus hermapbroditis] fcrtilibus ; feminea corolla basi globosa, 3-dentata, stylo exserto brevitcr bifido ; hcrmaphrodita corolla 4-dentata ; antberis inclusis, ovoideis acutis, inappendiculatis ; stylo valido tubo corolla; subocquilongo, alte bifido, ramis ciliatis ; achccniis obovatis, compressis, calvis, lente curvis, obscure 4-5 angulatis, apice in rostrum primum tortum demum rectum elongato. Ccndes csespitosi, 1-1|^ poll, longi, inferno foliis vetustioribus dense obsiti. Folia \-l poll, longa, -jJg poll, lata, lobis erectis, rotundatis, nervo prominente. Cctpilula \ poll. lata. Bractece }, poll, longai. Flores flavi. Corollxe dentibus tubus quad- ruplo longior. Achcenia \ poll, longa. IS. BAYLEY B.\XFOUR. 353 This iutei-csting littlo plant grows only ou the coralliuc limestone, and there in no great ahimdaiice. The young leaves are undivided and gradually become piu- natilid as they get older. This is hardly to be considered a form of heterophylly comparable to what is so common iu many of the plants from the island . The genus Abro fa neUa, to whieli it is referred, originally founded 1)yCassini for a species A. emargmata, brought by Gaudichand from the Falkland Islands, now eon- tains about nine species, and these are all Australian, New Zealand, or Antarctic forms. The Rodriguez plant, while it closely roscml)lcs the other s])ocies of the "-enus, its nearest ally apparently being A. {Ccratella) rosulata, Hook. fil. Fl. Antarct. i. 25, a species from Campbell's islands, possesses a very remarkable peculiarity in the ovary and fruit. The apex of the ovary and of the young fruit is prolonged into a beak, which coils on itself upon the top of the ovary, making usually one complete turn, and to its extremity is jointed the corolla tube. In the mature fruit this beak becomes straightened out into a long process. The genus is unknown in the otlier Mascarene islands. Plate XXVIIa. Fig. 1. Capitulum enclosed in leaves. 2. Flower of rav with outer involucral scale. 3. Flower of disk with inner involucral scale. 4. Flower of disk in vertical section. 5. Achenc with beak still coiled on the summit. 0. Achenc with beak straightened. Senecio linearis, DC. Prod. vi. 375; S. Lingua, Polr. ; DC. 1. c. ; S. salicifolius, Pers. ; DC. I.e. ; S. Boutoni, Baker Fl. Manr. Scych. 181. This plant is common on the island, but under two forms. When growino- near the sea the leaves and stems usually become more succulent and fleshy than they are when the plant grows inland. The species was founded by Do Candollc on specimens from Mauritius in the herbarium of the ^luseum at Paris. The type specimens I could not find for comparison, nor did I find named specimens of ^S*. Lingua, Poir., and »S'. salicifolius, Pers. There are, however, speciniens unnamed, both in the general herbarium and in Jussieu's herbarium from Mauritius and Bourbon, collected by Commerson, which are identifiable with the descriptions of the above-mentioned species, ^S*. linearis, DC, S. Lingua, Poir., and S. salicifolius, Pers., and which are manifestly mere insular variations of one species, to Avhich the Eodriguez plant may also bo referred. There is also a specimen at Paris from Bourbon, collected bv Abbe Pourrct and named in MS. S. borbonicus, Avliich is clearly this plant; and a specimen of Commerson's from Madagascar is probably the same. I have no hesitation in vmiting the four species in one. Baker, I.e., from want of sufficient information as to ^S*. linearis, DC, describes the llodriguez plant as S. Boutoni. The species is exclusively Mascarene. Sonchus oleraceus, Linn. sp. Plant. 1110 ; Eng. Bol. t. 810. Nom. vul". Lastron. Y Y 351 BOTANY OF KODKIGUKZ, Commou. Two forms of tliis occur. When grown on coralliuo limestone it becomes much stunted, and the leaves are exceedingly narrow and congested. The leaves of this plant are frequently oaten as salad, and as a hrMe. In a letter to me at Eoilriguez 'Slv. Ilorue mentioned that a collector in 1861 brought to Mauritius a yellow flowered Achillea from llodrigucz, but I do not know it. CAMPANULACE^. Lobelia vag-aois, Balf.fil. Amiua ; caulibus repentibus, tenuibus, glabris ; foliis membranaccis, inferioribus ovato-lauceolatis v. clliptico-ovato-acutis v. obovatis, basi cuneatis integrisquc, apice creuatis, breviter petiolatis, superioribus liuearibus acutis v. obtusis retusis V. submucronatis subsessilibus crenato-seri-atis ; tioribus solitariis in axillis foliorum superiorum, pedicellatis, pedicellis l)racteis brevioribus ; calycis tvibo obconico laciniis subulatis sub-a^quilongo v. breviter excedente ; corollso tubo calycis lo])is quadruplo lougiore, lobo ini'eriorc obovato-oblongo acuto, superiore liucari ; antheris 2 inferioribus apice barbatis, superioribus pilosiusculis ; stylo longo, stigmate bilobato ; capsula obovoidea. CauUs late patens, gracilis. Folia 1-2 poll, longa (pctiolo incluso J—^V poU-). ^_2 poll lata, superiora ,V poll. lata. Fedicclli adscendentes, 7}-4 poll, longi. Corolla alba, \ poll, longa, labio inferiore prof undo inciso. CajmUa tubo calycis triplo- longior, i poll, longa. Grows abundantly in the upper and shady parts of the valleys. Is very closely allied to L. serpens, Lam. ; DC. Prod. vii. 368, and which I take to be the same as L.jiliforinis,lmn.; DC. Prod. vii. 368. It is distinguished, however, from these species by the larger size of the flow^ers and its distinctly two-lobed stigma and the much broader leaves. Lobelia Cliffortiana, Willd. ; DC. Prod. vii. 372. Only a few plants on the shore at the mouth of the Eivi^re Grande. PLUMBAGINE.ffi. Plumbago zeylanica, Linn. ; DC Prod. xii. 692 ; Dot. Reg. t. 23. Very common everywhere. MYKSINACE^. Ardisia, sp. Nom. vulg. Bois de boeuf. There are several forms of Ardisia growing in Rodriguez. Unfortunately they were only in fruit during my visit, and my material is not sufficient for the deter- mination of species. But there is certainly more than one. IS. BAYLET BALFOUR. 355 SAPOTACE^. Sideroxylon, sp. Nom. vulg. Bois des pouimcs. This tree I obtained in fruit only. It is undoubtedly a Sideroxylon, and is not improbably a new species. It has some resemblance to the Mauritian ^S*. Bojermium, DC. Prod. viii. 179, but differs from all the Mauritian species by its very large fruit and seed, the former being as large as a pigeon's egg. EBENACE.ffi. Diospyros diversifolia, Jllern in Trim. Jonrn. Hot. iv. 353. t. 172. Norn. vulg. Bois d'ebdne. A not uncommon endemic heterophyllous tree, usually about 16-20 feet high. The young leaves being narrow and linear, the older much broader. This hetero- phyUy in Diospyros is quite a novelty. Hiern 1. c. remarks, " I am acquainted " with no parallel to this diversity of foliage in the case of any other species " throughout the order." One seldom meets with a large tree. In all the laro-e ones I met with the dark heart wood was quite decayed, leaving a shell of newer wood outside. I was told by inhabitants that this rotting of the centre always takes place in this species after it attains a certain height. OLEACEiE. Olea lancea, Lam. ; DC. Frod. viii. 286. Nom. vulg. Bois tambalacoq. This Mascarene sj)ecies is a common tree in Eodriguez. In addition to the fore-mentioned, I have leaves of another very abundant small tree, which is probably an Olea, and Baker, PI. Maur. Seych., makes it a variety latifoUa of Olea lancea. Lam. I doubt its being so, as the tree difPers in habit and general facies from Olea lancea, Lam., l)ut it is impossible to decide what it is from my materials. It is known on the island as Bois d'olive petites feuilles. APOCYNACEiE. Carissa Xylopicron, Pet. Th. ; DC. Frod. viii. 333. Plate XXVIIb. Xom. viUo-. Bois sandal. This Mascarene tree is remarkably heterophyllous. The young leaves are small, about \—k, in. long, spinose dentate on the margins, and with the apex obtuse or slightly retusc ; also the brauchlets bear a number of interpetiolar spines, one between each pail- of leaves. In the adult the leaves are 1-2 in. long by |-1 in. broad, with entire recm-ved margins and the apex deeply emarginate or obtuse, and there are no spines or rarely on the branchlets. Between these two forms all stages are found. This belongs to the first type of heterophylly referred to in my introductory remarks, and is well represented in the plate. The tree is very common on the island. The wood is very hard and dense, quite like boxwood, and is much used for making netting needles and such like articles. TY 2 35G BOTANY OF KODRlOrKZ. The wood is not uiilikc thai of tlio IJois do i\)n(l(> (Psiodia rudrif/iiesiaiKi, Ball', ill.), but the heart wood is of a much brii^htor yellow. TIic plant is abundant still on the island, but I'orniorly existed in ononnous ciuanliiy, for tlu* limestone ])lains arc in many plaees strewn with dead l)ranelit's and stems of it mixed witii liiose ol" Buis de ronde. riale XXVIIu. Fiii'. 1. Twiii' I'rom an ndiill plant witlityiu" lorm oF leal'. 2. Twiu: i'rom a young ])Iant with small and spinose leaves. ',',. 'l\\\'j; Irom ;i young shoot with smaller leaves. 1. Plower bud. o. Flower expanded, (i. Corolla spread out with stamens included. 7. Sfyli\ S. Ovary in transverse section. ], 2, :?, nat. size. Best magnified. Fig. 3 and the analyses are of ^rauritian specimens. Vinca rosea, L'niii.; DC Prod. viii. 3rc. Of this there are the two varieties rosea and (dha, knotvn to the natives as Chdjionli're rouge and C. hlaiic rcsjicctively — both abundant. The leaves arc said to make an excellent infusion for bronchitis and other chest affectious. ASCLEPIADACE^. Tanulepis, Balf. fil. Cohjx miniitus, 5-partitus, basi iutus 5-glandulosas. Corolhi subrotata, profunda o-fida, subvalvata. Coronce sqnanux; 5, lincari-lanceolatic, clongatie, petalis sub- fcquilonga?, antheras breviter exccdentcs, subincrassatse, basi conuatyc, gynostegioque adhterentes. Sfamiiia filamentis latis brevibusque basi annulo coronac affixa; antheras apicc euspidata^, connivcntes ; pollen granulosnm, in quoquc locnlo in massas duas siibcohoerens, appendicibus latis orbicidaribus corpusculornm longe stipitatornm applicitum. Stigma 5-gonum. FoUicidi divaricati, tenues, subteretes, subcostati. Semlna comosa. — Siiffrntex volubilis, glaber. Folia opposita, nitidula. Cymce laxc ramosae, ad apices raniorum breviter pedunculataj, Florcs parvi, pedicellati. Bractece minutissiraiie. T. Sphenophylla, Balf.fil. Plate XXVIII. Caulibus tenuibus; foliis mcmbranaceis, breviter pctiolatis, lanceolatis acumiuatis V. oblaneeolatis, basi cuneatis, 1^-2^ poll, longis (petiolo excluso ~ poll.), ^-J poll, latis, integris, nitidis, subtus pallidioribus, pcnniucrviis ; cymis paucifloris, peduuculis fLj poll, longis, bracteis deciduis, pcdicellis ^-^ poll, longis ; corolla? lobis |^ poll. longis ; stipite corpuscidorum appcndicem duplo excedente ; foiliculis f usco-nigris, glabris, 2-3 poll, longis, V^^i po^l- l^tis ; seminibus compressis, bisulcatis, ^ poll, longis. This twiner is fairly abundant, and is also found on Frigate Island. Its nearest congener is Brachylepls, W. and A., a monotypic East Indian genus, from which the elongated corona scales, its inflorescence, and the arrangements of the stamens exclude it. The Madagascar monotypic Sarpanema, Dene, is also not far removed, IS. BAYLEY BALFOUR. 357 l)ut the same cliaractevs separate tliem. I ouly obtained the plant in fruit, and the flowers are deseriljed from specimens sent home by Eouton. Plate XXVIII. Fig. 1. Elower bud opened. 2. Flower expanded with a portion of one of the corona scales removed. 3. Stamen seen from the inner side. 4<. Flower with corolla, corona, and stamens removed. 5. Corpuscles. G. Follicles. Sarcostemma viminale, li. Br.; DC. Prod. viii. 538. Xom. viilg. Liane cal6. Plate XXIX. Figs. 1-3. Very abundant. The Rodriguez plant differs slightly from the type in having puberulous not glabrous receptacles. Plate XXIX. Fig. 1. Portion of a stem with inflorescence. 2. Flower expanded. 3. Flower still more niagnifled and with corolla and portion of one of the scales of the inner corona removed. Sarcostemma Odontolepis, Balf.Jil. Plate XXIX. Figs. 4-10. Planta aphylla scandens v. decumbens, caulibus teretibus, gla1)ris, carnosis, ad nodos articulatis ibique squamis minutis ojipositis in loco folioruni instructis ; umbcllis solitariis ad nodos scssilibus breviterve pedunculatis v. terminalibus ; receptaculo jialeis vestito ; floribus breviter pedicellatis ; calycis segmentis ovatis V. suborbicularibus, concavis, niargine ciliatis, extuspuberulis; corolla profunde lobata, laciniis oblongo-ellipticis glabris ; corona exteriore corollae adnata profunde 5-partita, lobis tridentatis, singulo squamse interiori adhterente, dente centrali maximo ; corona interiore exteriorem triple excedente, segmentis superne saccatis dolabri- formibus gyuostegio subcequilongis ; stigmate apiculato, obscure bilobato, umbilicato. FediccUi subpubcruli, /.-g poll, longi; Corollcc laciniaj J poll, longte, calyce quadruple longiores. Nom. vulg. Liane cale. This plant resembles the foregoing species in almost every point save in the flower, and therein it differs in the corona. In place of the sinuose-dentate cyathi- form outer corona, characteristic of Sarcostemma, and well seen in ^S*. cuninale, R. Br., the outer corona consists of 5 scales, distinct except at the very base, each tridentate, the central tooth being largest. These are connected with the inner corona scales, which are twice as long as the outer scales, but shorter than the gynostegium, whereas in S. viminale, R. Br., the inner scales are thrice the length of the outer, and are longer than the gynostegium. This is a very imjjortant diffe- rence in this family. But on examining a large series of flowers, I discovered one in which there was an ari-angcment of the coroiia exactly half-way between the two forms I have described. The cyathiform outer corona of S. viminale, R. Br., Avas divided about half-way down into five tridentate lobes, and the relations of length ])ctween outer and inner scales was intermediate. This then connects the two forms. It is then a question, have we to deal with two or one species ? I rather incline to the former view, that there are two distinct species, for we find certain other 358 - liOTANY OF KODKlGrKZ. minute differences between them, and at Rodriguez I Avas led to suspect there might be two species, from a certain dillereuce in facios of the plant in dilTerent sitvia- tious. Ilowevor we consider them, the cliaracter oi" the genus must he slightly emended. Shoidd they turn out to be one species, we have here a species A'arying beyond the generic limits. Acting on the opinion expressed, I have described a new species of Sarcostemma. Plate XXIX. Fig. 1. Portion of a stem bearing inflorescence. 5. Unopened llower bud. G. Flower expanded and magnilled. 7. Flower still more magnified and with corolla and portion of the scales of inner corona removed. 8. Pollen-masses. 9. Gvnostegium. 10. Ovarv in transverse section. Two species of Asclepiadaceous twiners are occasionally met with, l)ut as I only obtained them in leaf, and they are unlike any known Mascarene or Seychelles species, it is impossible to determine them. LOGANIACE^. Buddleia madagascarieusis. Lam. ; DC. Prod. x. 417 ; Bot. Hag. t. 2821. The leaves of a plant, Avhich seems to be this Mascarene species, occurs near the dwellings of the early settlers. BORRAGINACE^. Tom-nefortia argentea, Llan.f. ; DC. Prod. ix. 514. Nom. vulg. Voultie. Common on the coralline limestone, on the shore of Rodriguez, and on most of the coral islets. Hehotropiiun (Heliophytum) indicum, Linn.; DC Prod. ix. 556; Bot. Hag. t. 1837. Nom. vulg. Herbe a papillons. An occasional weed. The juice of this is said to be an excellent vulnerary, and also the infusion as a wash for ulcers. Trichodesma zeylaniciun, P. Br.; DC Prod. x. 172; Bot. Hag. t. 4820. Nom. vulg. 3Iadame Tombe. A common weed. CONVOLVULACE^. Argyreia tihaefoha, Benth. et Hook. Gen. PI. ii. 869. Is very common. Ipomoea paniculata, P. Br. Prod. 486; Bot. Peg. t. 62; Batatas panieulata, Choisy in DC Prod. ix. 339. A plant I obtained in leaf, which seems to be this, is common. IpomcEa purpm-ea. Lam.; Pharbitis hispida, Choisy in DC Prod. ix. 341. Is not common. IS. BAYLEY BALFOUR. 359 Ipomoea fragraiis, Boj. MSS. ; Pharljitis fragrans, Boj. ; Chohy in DC. Trod. ix. A plant in leaf resembling this occurs ou Gombrani and some of the other coral islets. I never found this on the main island. Ipomcea Nil, Both.; Pharbitis Nil, Choisy in DC. Trod. ix. 343. On the coast in many places, and on the coral islets. Ipomoea (Calonyctiou) miiricata, iio.rft. Tl. Ind. 499 {non. Cav.) ; Choisy in DC. Prod. ix. 353. Very common. Ipomoea Batatas, Lam. Nom. vulg. Patate or Batat. Many varieties arc cultivated, and it is the chief staple of food on the island, being easily grown, and not affected by hurricanes. Ipomoea Pes-Caprae, Sw. ; Choisy in DC. Prod. ix. 319. Nom. vulg. Batatran. Very common on the shores. In English Bay it grows mixed with Canavcdia ohtnsifolia, DC, and it is difficult at sight to discover the one from the other. A cm'ious example of mimicry. Ipomoea peltata, Choisy in DC Prod. ix. 359. Only found at the top of valley of Ri^dere Coco, where it grows in great profusion, covering the trees. Ipomcea Tm-pethum, B. Br. ; Choisy in DC. Prod. ix. 360; Bot. Beg. t. 279. Occurs pretty frequently in the valleys. Ipomoea tuherosa, Linn. ; Choisy in DC. Prod. ix. 362; Bot. Beg. t. 768. I have leaves of a twining plant very probably this species. It grows frequently in the vaUeys. Ipomoea obscura, Ker ; Choisy in DC Prod. ix. 370; Bot. Beg. t. 239. Very common on the shore. Ipomoea leucantha, Jacq. ; Choisy in DC. Prod. ix. 382. Very common on the coralline limestone. Ipomoea palmata, Forsk. ; Choisy in D C Prod. ix. 386. Common in the valleys. Dichondra repens, Forst. ; Choisy in DC. Prod. ix. 4-51. Onlv on coralline limestone towards the south-west of the island. Usually along with Rypoestes inconsjncna, Balf. fil. and Selaginella Balfouri, Baker. SOLANACEiE. Lycopersicum cerasiforme, Z)«ttrt/ wi DC. Prod, xiii, 1. 26. Nom. vulg. Pomme d'amour. Is cultivated and is seen near habitations occasionally. 3G0 r.OTAXY OF KODninrr.z. Solaimm nigrum, L'niii. ; Diatal in DC. Prod. xiii. 1. 50. Nom. vult^. Bi'i^lo martin. A vory common plant, mucli used l\v tho iiiluibitauts as a br^do. Solanum indicum, Liini.; Ditiial in DC. Proil. xiii. 1.30!) ; 7/7. fi\ t. 3K>. Xom. villi;. IVtito anghive. Is not common. Solanum macrocaj-pou, Linn.; Dnnal in DC. I'ruJ. xiii. 1. 353. Nom. vulg. Bringollc. A specially IMascarcnc form not very common on the island. Solanum Melongena, Linn. ; 7/7. III. t. \(jG. Nom. vulg. Gros brlngcllc. In the vicinity of dwellings. Solanum sanctum, Linn. ; Dunal in DC. Prod. xiii. 1. 369. Nom. vulg. Bringclle marron. Very con^nion in the vicinity of dwellings. Pliysalis Peruviana, Linn.; Dunal in DC. Prod. xiii. 1. IIO. Nom. vulg. Pocke-pocke. Common. Capsicum frutescens, Linn. ; Dunal in DC. Prod. xiii. 1. 413. Nora. vulg. Petit piment. Abundant everywhere. Capsicum cordiforme, Jlill. ; Dunal in DC. Prod. xiii. 1. 427. Nom. vulg. Gros piment. Not common. Leguat speaks of certain trees " that bear a sort of pcjjper, and are not a little " like plum trees of a moderate size ; their leaves arc much like that of the jessa- " mine ; they bear their fruit iu little bunches, and it did very well in our sauces." I suppose it is to one of these species of Capsicum he refers. Lycium tenue, TTilhl., var. Sieberi, Dun(d in DC. Prod. xiii. 1. 515. Usually on the coralline limestone near the sea. Specially abundant on the coral islets. In one spot at the top of the valley Ptivicre dc I'Est, I found this plant inland, and it there had lost its stunted, short-branched, rigid character, and formed a widely-spreading, long and slenderly -branched plant. The leaves also were larger and less fleshy. The description of this species answers my specimens, but they differ rather from the specimens named L. tenue, Willd., in the herbarium at Kew. It is a Cape species, but the variety Sieheri is Mauritian. I have seen no type specimen of the variety. The inland form of the Ptodriguez plant most resembles the Kew specimens of L. tenue, "Willd. Datm-a alba, Xees ; Dunal in DC. Prod. xiii. 1.511; Wt. Ic. t. 852. Nom. "viilg. Herbe du diable. A frequent weed. IS. BAYLEY BALFOUR. 3G1 Nicotiana Tabacum, X/ww. ; Dnnol in DC. Prod. xiii. 1. 557; Uaijne Geiodchse 12. t. il. Xora. Tulg. Tabac. Cultivated and iu some places now grows spontaneously. scrophulariace.se. Herpestis Monuieria, H.B.K. ; Benth. in DC. Prod. x. 400 ; Dot. Mag. t. 2557. Common in the streams towards the west end of the island Avhere the woods hare been destroyed and the streams pass over bai*ren plain. ACANTHACE^. Barleria cristata, Linn.; Nees in DC. Prod. xi. 229; Wt. Ic. t. 153; Bol. Mag. t. 1615. Not at all uncommon near habitations and also in some of the more frequented valleys. Barleria Prionitis, Linn. ; Nees in DC. Prod. xi. 237 ; JVt. Ic. t. 452. Only found in the Champ de Eoi near Port Mathiu-in, but there abundant. Hypoestes rodrigiiesiana, BaJf.fil. Plate XXX. Herlm parva, ramosissima, glabra, ramis crassis lignosis ; foliis longc petiolatis ovato-lanceolatis, inferne deltoidcis, integris, subcoriaceis, substrigulosis ; racemis breviter pedunculatis in axillis foliorum confertis, bracteis obovatis v. oblanceolatis capitulis imifloris oppositis arete dispositis ; involucro tubuloso, subventricoso, tetra- phyllo, ad medium discreto, lobis inrequalibus, exterioribus longioribus ovato-acutis, extus j)laga lineari lateraliter notatis, interioribus lanceolatis brcvioribus ; calyce brevi, extus hispidnlo, laciuiis tube l)revioribus, ciliatis ; corolla profunde bifida, pilosa, labiis tubo suhrequilongis, superiore oblongo, inferiore tridentato v. trifido ; staminibus 2, filamentis exsertis ; ovario oblongo, compresso, glabro ; stylo filiformi, exserto. Herba hnmilis, corticc albido. Folia 1-1-i poll, longa ; petiolus \-\ poll, longus. Capitula uniflora, opposita decussataque, brevissime pcdunculata ; involucrum f poll, longum, lobis integris hispidulis. Cahjx i poll, longus. Corolla uncia longior, churneaj labio .superiore integro ohtuso v. acuto. Filamenta glandulosa; anthera oblonga. A rare plant and not far removed from the Madagascar H. Bojeriana, DC. Prod. xi. 507, hut distingviished by its densely clustered racemes, unequally lobed involucre, and large flowers. I only found one plant on the top of Malartic, one of the highest points on the island, growing in a very exposed situation, which may account for the dwarfed growth, for it did not rise more than G inches from the ground. Plate XXX. Fig. 1. FloAver bud enclosed in involucre. 2. Involucre opened out \A-ith calyx included. 3. Calyx spread out. 4. Flower expanded. 5. Corolla z z 3G2 BOTANY OF RODRIGUEZ. opened out with androociiuu aucl gyiuociiun. (5. Stameu detached. 7. Transvorso section of ovary. 8. Vertical section of ovaiy. Hypoestes iucouspiciia, Balf.fil. Uci-ba (oiuiis, caulo ropcntc valde articuhxtOj ad uodos radicaute; foliis oppositis distantibus, ovalibus v. obovatis, obtusis v. siibacutis, strigosis, iutcgi-is, inlcrne in pctiohun subtequilouyum longiorcmvc attonuatis ; capitulis solitariis axillaribus siibsessilibus, bracteis nnllis ; involucro nnifloro, strigoso tctraphyllo, laciniis lineari- lanceolatis imcqualibus nsquc ad nicdinm connatis, cxtoriovibus dimidio longi- oribus; calycc brcvissimo, lobis lanceolatis acuniinatis, involncro multo In-cviore ; corolla inA'olncro longiorc ; ciet. ign. Cdidis raniosissimus, glaber scd extremitatc strigosus. Folia 1-| poll, longa ; Fetiolus :]— fW poll, longus. Involucrum ;[ poll, longum. Calyx .}-^ poll, longns, A very small plant discovered on one small patch of coralline limestone at tlio western end of the island growing along with Dlchoiidra repens, Forst., and Selayi- iiella, Balfoun, Baker. It has some affinity with M. serj^ens, R. Br. ; Nees in DC. Prod. xi. 501, and U. Alsine, Nees in DC. Prod. xi. 502, but is distinguished by its nou-pubcsceut involucre and very minute calyx. MYOPORINE-ffi. Myoporum maimtiamim, A. DC. Frod. xi. 711. A very cm-ions Mascarcnc plant. Rare in Rodriguez, ouly a few plants having been seen on the coralline limestone on the shore in Ansc Colon at the east end of the island. A. De Candolle described the species from Mauritian specimens in Sieber's herbarium, including it doubtfully in Myoporum. Por Avhilst most characters place it there it is peculiar in having a distinctly 4-lobed stigma. The Rodriguez plant differs from De Candolle's descrij)tion in its 5, not 4-lobed corolla. VERBENACE-ai. Nesogenes decumbeus, ^f?//".^^. Plate XXXI. Herba diffusa perennis, ramis laxis, oppositis v. suboppositis, teretibus, tenuibus, divaricatis, huniifusis, glabris sed versus extremitatem biseriatim pilosis ; foliis oppo- sitis, lanceolatis \. rhomboideis acutis, integris, marginc ciliatis, basi in petiolum brevem attennatis, siccitatc non nigrescentibus ; floribus in cymas 1-2-floras axillares dispositis, pedicellis brcvissimis pubescentibus ; calyce amplo, \ poll, longo, alte 5-dentato, dentibus acutissimis crenatis, extus intusque glandulosis ; corollse lobis svibgequalibus, rotundatis, tubo quintuple v. sextuple brevioribus; antheris oblongis ; fructu Isevi compresso, apice hispido, basi glabro, stylo mucronato calyce cincto. Canlis ab basi ramosus. Folia breviora J-f poll, longa, \-\ poll, lata: petiolus ^— ^ poll, longus. Calyx \ poll, longus, laciniis tubo cequilougis. Anlhera oblonga. Ovarium ovoideum, stylo liliformi incluso. IS. BAYLEY BALFOUR. 363 Only fouucl in one place, on a patch of coralline limestone about f of a mile from the sea, at the west end of the island, close to where Hij poesies inconspictia, Balf. fil., and Dichondra repens, Forst., were found. A most interesting plant belonging to a hitherto Polynesian monotypic genus. The other species, iV. Euphrasioides, Hook, and Arn., was fli'st described from Whitsunday Island in Botany of Becchey's Voyage G7, and there doubtfully referred to Myoporum. De Candollc, Prod. si. 703, constituted the genus Nesogenes for that species including it in Myoporacecc. But it is now referred by Bentham and Hooker, Gen. Plant iii. 1141, to Verhenacece and placed near to Sparto- thamnus. In Kew herbarium are specimens from Sow Island, Chain Island, and one or two islands of the Dangerous Archipelago, so that the species is probably common there. But it is cm-ious that another species should occur so sparingly in Bodriguez, an island almost antipodal. Plate XXXII. Fig. 1. Expanded flower. 2. Corolla and andrcecium spread out. 3. Detached stamens. 4. Gynsecium. 5. Apex of style. 6. Fruit enclosed in calyx. 7. Fruit removed from calyx. 8. Transverse section of fruit. 9. Ver- tical section of fruit. All magnified. A species of Lantona was shown to me in Mauritius by Mr. Home, who informed me it came from Bodriguez. I found no such plant there, which is strange, as species of Lantcma usually grow so freely and spread so widely it is not likely it should be overlooked. Stachytarpheta indica, Vahl ; Schemer in DC. Prod. xi. 564. Nom. vulg. Qiieue des rats. A very common weed. The juice is said to have a styptic action, and is often used for woimds. Premna serratifolia, Linn.; Schemer in DC. Prod. xi. 632. Nom. vulg. Bois siu'eau. Not at all uncommon. Clerodendron laciniatum, Balf. fil. Plate XXXII. Frutex v. arbor parva ; foUis oppositis, petiolatis, coriaceo-membranaceis, poroso- punctatis, adultis ovatis v. ovato-oblongis, acutis, basi subdeltoideis, integris, glabris, subtus pallidis, juvenilibus filiforme-tripinnatipartitis segmentis distantibus pubcr- uHs ligulatis oljtusis, per formas varias in adultam transcuntibus ; cymis axillaribus, paucifloris, bis trifidis, patentibus, breviter pedunculatis, bracteolis muuitissimis ; calyce cupuliformi truncato, ore integro v. obscm-e lobato, glabro, demimi patulo ; corolla infundibuliformi calycem sextuplo excedente, tubo intus resino-papillato, laciniis obtusis tubo brevioribus ; genitalibus breviter exsertis. Rami teretes, cinerei, apice puberuli. Folia heteromorpha gravcolentia, 2-3 poll, longa ; petiolus ^1 poll, longus. Cymi corymbosi, 2-3 poll. diam. ; pcdimculi -c—{'-2 poll- longi; pedicclli breves i\,— » po^l- loiigi- Calyx j'^- poll, longus. z z 2 301 BOTANY OF rxonmouEz. Nom. vulg'. Bois cabri. A small oiulomio fairly abundant tree cxbibiting an oxlrcmc form of hctoropliylly well I'oprosonto.l in the plalo. It is very closely allied to the Mauritian and Bourbon species, C. hctd'opIuiUuDi, 11. J5r., but its larger pinnalipartite leaves on the young plant aud relative size of the parts of the liower sulliciently distinguish them. This small tree is easily recognised by its disagreeable odour, -n-hicb has occasioned its popular name. The -wood is very Avhite and close-grained, but is not put to great use, save for bvu-ning, probably on account of the odour. I have no doubt this is the tree to ■niiieh Leguat refers when he says, " There is a tree we call the " Nasty tree because it stunk. 'Tis the best wood of all for carpenter's use, but " 'tAvas of no service to for us it slinks so, that it makes all ilie places about " it smell of it, and the smell is very offensive." Some have thought he refers to the Bois puant {Fa'tkUa inauvitlana. Lam.), but the odour of that tree though exceedingly objectionable is only apparent "when the sun shines upon it, and is evanescent, and not at all like the persistent odour Avhich the Bois cabri emits. Plate XXXII. Fig. 1. Twig with flowers. 2. Leaves from adult, typical form. 3. Leaf from a young tree. 4. Leaf from a still younger tree. 5. Twig with leaves from a very young tree, or from an adventitious shoot. 6. Flower bud. 7. Ex- panded flower. 8. Corolla and audroeeium opened out. 9. Anthers detached. 10. Gynascium enclosed in calyx. 11. Transverse section of immature fruit. 12. Vertical section of immature fruit. 1, 2, 3, 4, 5, natural size. Uest magnified. LABIATE. Ocimiim caniim, Sims ; Benlh. in DC. Trod. xii. 32 ; Bot. 3Iag. t. 2152. A few plants in the vicinity of Port Mathurin. Salvia coccinea, Linn.; Lentli. in DC. Prod. xii. 313. Is very common in the valleys. Stachys, sp. A plant which is apparently a species of Stachys was found in leaf only in the valley of Ptiviere Grande des Bamboux (Cascade). Leouiirus Sibiricus, Linn.; Benth. in DC Frod. xii. 501. Frequently found near dwellings. PLANTAGINE.ffi. Plantago major, i«72». ; Dene in DC Prod. xiii. 1. 091; Lug. Bot. 2nd edit. t. 1162. This occurs abundantly near the top of Valley Bivi^re Baleine on the south side of the island along with a species of Burner. IS. r.AYLEY BALFOUK. 365 NYCTAGINACE^. Mirabilis Jalapa, Lam.; Choisy in DC Prod. xiii. 2. 1-27 ; Bol. Marj. t. 371. 2som. vulg-. Belle dc nuit. Occasionally fouud uear habitations. Pisouia viscosa, Balf.fU. Plate XXXIII. Arbor iuermis ; foliis petiolatis, ovatis v. obovatis, obtiisis v. acutis v. breviter cuspidatis, basi deltoideis, uiembranaceis, subrepandis, primum subpiiberulis dcmum scaberulis, siccitate nigrescentibus ; petiolo tenui ; $ panicula corymbosa in axillis folioriun tcrminalium orieute, brunnco-pubosccutc, dcnsa, pedunculo compresso, bractcolis uiiuutis decidiiis ; pcriautliio iniuudibuliformi, extus pilose, breviter 5 dentato, dentibus reflcxis ; stamiuibus 10 exscrtis ; ? panicula laxe latcquc patenti, pediceUis erecto-patentibus ; j^erigonio clongato, 5-augulato, angulo siugulo tuber- culis brevibus uncinatis glandulosis seriatim instructo. Arbor parva, ramis validis. Folia 3-4 poll, longa, l|-2 poll, lata ; petiolus ^\ poll, longus. S Panicula 1^24 poll, diam., folia superans ; pedunculus \ poll, longus ; bracteola? subtilatte glauduloso-pubescentes ; flores subsessiles. Perian- th i urn \ poll, longum, tubo extus papillis subrigidis erectis 5-seriatis instructo. Stamina basi coalita ; anthera rotundata. Ovarium breviter stipitatum ; stylus crassus ; stigma capitatum subfimbriatvim. ? Panicula 6-8 poll. diam. ; pedicelli ■g— ^ poll, longi. Periffoiiiiim i poll, longum. Fructiis loculus semine impletus. Nom. vulg. Bois mapou. This tree is very abundant on a small ledge of coralline limestone on the west side of Frigate Island, where it is the favourite nesting place of the Fou, but it is not so abundant on the main island. The wood is very soft, and of no practical value. The plant has some affinity with P. ovo.Ufolia, DC. Prod. xiii. 2. ill, a Mauritian species of which the Mauritian P. lanceolata, DC. Prod. xiii. 2. 442, is also probably a form, but is distinguished by the structure of the flowers, and the bristly fruit with the seed filling up the whole cavity. Plate XXXIII. Fig. 1. Male flowers. 2. The same cut open. 3. Portion of female inflorescence. 4. Single female flower. 5. The same more enlarged in transverse section. 6. Embryo. Fig. 3 nat. size. Eest enlarged. Bcerhaavia diffusa, Linn.; Choisy in DC. Prod. xiii. 2. 452. Widely spread in the island ; when grooving on the coralline limestone it has a more stunted and woody habit. POLYGONACE.ffi. Eumex crispus, Linn. ; Jleissn. in DC. Prod. xv. 44. A plant exceedingly closely allied to this, if it be not identical, grows abundantly with Plantarjo major, Linn., in the valley of Riviere Baleine. 3G6 I50TAXY OF UODElGUnZ. AMARANTHACE^. Aiiiarautliiis tristis, Ziini. ; Moq. in DC. Prod. xlii. 2. 2G0; JFi.'Ic. tt. 511, 713. JNom. vulg', BrMc malabar. Very common noav habitations, ami is very much nsed as a hrC'dc by tlic inhabitants. MxwA congesta, Balf.fil. Ilorba pcrennis, dense cfespitosa, ramosissima, pulvinata, vamis lirmis lignosis humifusis ; foliis altcrnis pctiolatis, obovato-spathulatis, obtusis v. aeut is, subcori- aceis, glabris, junioribns subtus pilosis ; spiels oblongis v. subrotundatis, in axillis villosis foliorum supoviovum scssilil)us solltariis crectis, bracteolis deltoidois pcrlan- tbio brcnoribus, glabris ; perianthii scgmentis oblongo-lanccolatis unincrviis, extcrioribus duobus latioribus pilosiusculis, intcrioi'il)us tribus angustioribus dcnseque pilosis ; autheris rotundis ; staminodiis minntissimis ; iitriculo comprcsso sub- orbiciilari glabro ; semine insequaliter reniformi, minute tuberculato, margine obtuso. Ilerba pusilla, ramis pateutibus usque ad 2-3 poll. Folia |-i- poll, longa. Spicce I poll, longa?. Pcrianlhil scgmenta ^ poll, louga bracteolas albidas dimidio exeedentia. Semen nitidum nigrum. A small tufted plant growing only on coralline limestone, along with such plants as Abrotonella rhynohocarpa, Balf. fil., and Oldenlandia Sieheri, Baker var. congesta, and frequent on the coral islets of the reef. Achyrantlies aspera, Lmn, ; Moq. in DC. Frod. xiii. 2. 314 ; Wt. Ic. 1777. Nom. Tulg. Hcrbc I'argcnt. Very common weed. Achyi-anthes argentea. Lam.; Moq. in DC. Frod. xiii. 2. 315; Sibih. Fl. Grcsc. t. 211. jS'om. Tulg. Herbe I'argent. Occiu-s on the coralline limestone near the shore and on the coral islets. P erhaps is a mere form of the foregoing. An infusion of the root of Herbe I'argent is said to be of great service as a cure for cough. The juice of the leaf is a vulnerary of great repute. Alternanthera sessilis, B. Br. ; Moq. in DC. Frod. "xiii. 2. 367 ; Wt. Ic. t, 727. A common -weed. CHENOPODIACEiE. Chenopodium ambrosioides, Linn.; Moq. m DC. Frod. xiii. 2. 72. Wt. Ic. t. 1786. Is an occasional weed in waste ground in the vicinity of Port Matliurin. BASELLACE.ffi. Basella rubra, Linn. ; 3Ioq. in DC. Frod. xiii. 2. 222. Nom. vulg. Brede d'angole. Is cultivated and used much as a brfede. IS. BAYLEY BALFOUR. 367 LAUEACEJE. Persea gratissima, Gartn. ; 21eissn. m DC. Frod. xv. 1. 52; Wt. Ic. t. 1823 ; Bot. May. t. 45S0. Nom. viilg. Avoca. A few trees at tlie top of valley Rividre Palmiste, near the site of some old dwellings. Tetrauthera laurifoUa, Jacq. Mori. Sclmn. t. 113; Melssn. in DC. J? rod. xv, 1. 178 ; T. apetala, Bo.vh. Cor. t. 147. Nom. valg. Bois Zozo. Planted near Port Matliui'in. Cassytha filiformis, i/»«. ; Meissn. in DC. Prof^. xv. 1. 255 ; Wt. Ic. t. IB*?. Nom. vxilg. Liane sans fin. Very abundant, covering the ground and trees in many places. URTICACE^. Obetia ficifolia. Gaud. Atl. Bon. t. 82. Nom. vulg. Figne marron. Occasionally found in the upper parts of the valleys. Pilea Balfoui-i, Baker Fl. Mattr. Seych. 276. Plate XXXIV. Herba par\T.ila monoica perennis glaberrima, caule brevi 4-gono basi csespitose ramoso, ramis patentibus oppositis, exsiccatis cystoHthiferis ; stipulis deltoideis minutissiniis ; foliis oblongis rhomboideis, acutis v. acuminatis, basi cuneatis inte- grisque, apice profundo iuciso-crenatis, tripHnerviis, nervis usque ad apicem pro- ductis, laminis exsiccatis membranaceis, paginis utrisque cystolithis linearibus sufful- tis; cymis patentibus, multifloris, breviter pedunculatis, axillaribus, petiolo brevi- oribus ; floribus breviter pedicellatis ; i perianthio bilobato, lobis concavis subcucullatis acutis; ? achceniis Isevibus, acute marginatis, compressis, ovoideis, vix segmeutuni intermedium perigonii supcrantibvis. Bami late patentes. Folia 1-4 poll, longa, |— |- poll, lata, opposita, intemodiis longis usque ad 2 poll. ; petiolus f-lj poll, longus. 5 Alabastrus -^ poll, longus, exsiccatus cystolithiferus. Calycis lobi ^ poll, longi. Common in the shady and moist spots in the upper part of valleys. This is a near ally of P, cuneiformis, Wedd. in DC. Prod. xvi. 1. 133, a Maiu'itian species, but is distinguished by its habit, longly petiolate leaves, and spreading inflorescence. Plate XXXIV. Fig. 1, Portion of male inflorescence. 2. Male flower oj)ened. 3. Perianth of female flower. 4. Female flower. MORE^. Artocarpus integrifoha, Linn. Suppl. 412 ; Boxh. Cor. t. 250 ; Bot. Mag. tt. 2833, 2834. Nom. vulg. Jacque. Frequent near habitations. There are said to be two varieties, Jacque labou and Jacque hlanc, but I never found any plant which could be considered a variety. 368 BOTANY OF KODKIOUEZ. Ficiis cousimilis, Bid-rv Fl. Urdiir. Sei/cli. 280. Noin. vulg. La fouchc. Is very common. The bark is said to be astriui^cnt and llie juice t!,'ood for wart>. The bast layers make excellent cordage Avliieli is much used. The fruit is not edibU'. The Tvood is very liard and tough. Ficus rubra. J'o///. var. amblyphylla, Baker Fl. Maiir. Sei/ch. 285. Nom. \ ulg. La fouche rouge or La fouche petite feuille. Common. The fruit of this tree is said to be edible. I have followed Baker, 1. c, regarding this ])lant and the preceding. My specimens are not sufTicient to allow of a very satisfactory determination. Leguat speaks of " a wonderful tree whose branches are so ro\ii\d and so thick " 'tis impossible for the sunbeams to penetrate through it. Some of these trees " are so big that two or three huiulred people may stand under them and be sheltered " from the sun or the weather. The vast extent of it is occasioned thus. Some " of the great branches naturally tend downwards, and reaching the ground, take " root and become new trunks themselves which make a sort of little forest." He refers evidently to a species of Ficus, and gives a figure of it, but I do not know to what species, certainly neither of those above mentioned, for he describes it more particularly : — "The Bodriffo Kasfas (for I sought to keep i\\c Indian name at least in the Indies) bear leaves as broad as one's hand, pretty thick and somewhat like that of a lilach or heart in shape, they are softer than satin to touch. Their ilowcr is Avhite and smells well. Their fruit is red and round, and as big as a black damask plum. Theii* skin is hard and ^vithin it is a thin seed, a little like that ()r a fig. The fruit is not prejudicial to health, but 'tis insipid. The batts commonly feed on it, and multitudes of them nest in the tufted branches of the tree." I found no species which would answer this description, which if accurate indicates a species formerly extant, now extinct, or at least of which all the large individuals are destroyed, only young and inconspicuous ones left. Mr. Home showed me in the Botanic Gardens at Pamplemousses, Mauritius, a third species of Ficus which he said Mr. Duncan got from Rodriguez, but I do not know what this is. EUPHORBIACE^. Euphorbia pilulifera, Linn.; Boiss. in DC. Brod. xv. 2. 21. A common weed. Euphorbia thymifoUa, Biirm.; Boiss. in DC. Brod. xv. 2. 17. Common specially on coralline limestone. Euphorbia daphnoides, Balf.fil. Suffrutex caule tenui, ramulis lignosis teretibus ; fohis ad apicem. ramulorum confertis, breviter petiolatis, oblanceolatis v. anguste oblongis, obtusis, mucronatis, IS. BAYLEY BALFOUR. 3G9 tcnuiter coriaceis, glabris, cito deciduis, exstipulatis ; cymis tcrminalilms, peduncu- atis, bracteis 2 magnis coriaceis ovalilius v. siiborbicularibus mucronatis v. rctusis v. eraarginatis ; involucris campaniilatis, breviter pedicellatis, glabris, lobis brevibus, glandulis 5 soepe rubris rotundatis v. lunatis poroso-punctatis intcgris ; stylo brevi ; capsula glabra sub-depressa trisulcata transvcrso diamctro rnajore, coccis pauUum compressis ; seruine glabro, irregiilariter papillate, ovoideo. Siiffi'tdex glabcr, liabitu Daphnes. FoUa 2-1 poU. longa, i-|poll. lata. Bvactea; ultima? 5— I poll, longa; ; pedicelll glabri, ^ poll, longi. Involucvum \ poll, longum. Copsulo \ poll. diam. Stylus -^.^ poll, longus. Semen yV poll, longum. Not a common plant. Only found in tbe valley Riviere de I'Est, and on tlio flanks of the Mount au Sel. It forms a very bandsomc underslirub and of it there are two varieties, one with red glands in the flower, whilst in the other they are uncoloured. Euphorbia peploides, Gouan ; Boiss. in DC. Trod. xv. 2. 111. In waste ground near dwellings. Securinegu durissima, Gmel.; 3Iv.U. Arc/, in DC. Frod. xv. 2. 117. Norn. vulg. Bois dur. A tree, apparently referable to this Mascarene species is very abundant on the island. I only obtained specimens in leaf, which are therefore not fvilly determin- able. Phyllantlins Niruri, Linn.; 3IuU. Arg. in DC. Prod. xv. 2. 106. IS'om. vulg. Ananelle. Common around Port Mathurin. Phyllantliiis dumetosus, Toir.; 3Iv.lI. Arc/, in DC. Drod. xv. 2. 398. Abundant in Anse Baleine. This is one of the plants Commerson brought from Rodriguez, where it is endemic. Phyllanthus Casticum, TFillem. ; Midi. Arg. in DC. Prod. xv. 2. 318. Nom. vulg. Castique. Xot uncommon in many places. Many forms of this Mascarene tree occiu-, varying slightly in the leaves, but aU referable to the one type. Is said to be a powerful astringent. Manihot utilissima, Polil. ; 3IuU. Arg. in DC. Prod. xv. 2. lOGl ; Jatropha Manihot, Bat. JIag. 3071. Is^om. vulg. Manioc. Cassava. Commonly cultivated. There are several varieties which grow well when planted in a dry soil, otherwise the roots rot. It is always planted in sheltered situations on the hill slopes. Some varieties ripen in 3 months, others not for 14. The roots are boiled whole, or ground down and made into small round cakes known as Coup de poing manioc, or large flat ones called Gullet manioc. 3 a 370 BOTANY OF KODKIOUEZ. Riciuus communis, Linii.; Midi. uiry. in DC Prod. xv. 2. 1017; J3o/. Mag. t. 220i). Nom. vulg. Tang-Tang. Forms donso thickets in many places, and in some places produces a spiny fruit; in other places the I'mit is quite smooth. Claoxylou parviflornm, A. Jnss. ; Midi. Avg. in DC. Prod. xv. 2. 7S5. A tree greatly resemhling this Mascarcuc species occurs on the island, but as I only obtained it iu leaf, it is not fully determinable. PIPERACE^. Peperomia liirta, Balf.fd. Ilcrba rcpens pilosa, caule simplici v. ramoso, c nodis radicante ; foliis oppositis, pctiolatisj cllipticis v. oblongo-cllipticis v. obovatis, 5-ncrviis, utrinquc villosis, ner- vulo obscure juxta marginem currcnte, petiolo villoso. Ctet. ign. Caules pedali niinores iV poll, crassi. Folia ^} poll, longa ; petiolus ^-^ poll, longus. Discovered within a few yards of the summit of INIoimt Limon, and nowhere else seen. Its nearest affinity is with P. eUiptico, Dietr. ; CDC. in DC. Prod. xvi. 1. liO, a Maimtian sj)ecies from which its pilose character sufficiently separates it. Of this plant I have only leaves, and I am indebted to M. Casimir dc CandoUe for the specific determination. Peperomia reticulata, Balf.fil. Plate XXXV. Herba carnosa repens, caiile simplici v. uniramoso, in parte inferiore e nodis radi- cante, versus apicem adscendente ; foliis oppositis petiolatis, summis ternis, elliptico- rhomboideis, apice emarginulatis v. acutis, utrinqtie glabris, 5-nerviis, et reticulato- venulosis, nervulo obscure juxta marginem currentc ; amentis solitariis, axUlarilnis, folia terminalia superantibus, breviter pcdunculatis, bracteis subrotimdato-peltatis, breviter stipitatis ; ovario globoso immerso, stigmate umbilicato, prominente, glabro. Caules ^1 ped. longi, \-\ poll, crassi. Folia 1-2 poll, longa ; petiolus l-^^ po^- longus. Amenta 2-J; poll, longa, ^ poU. crassa ; pedunculus glabrus petiolum superans, Fructus ^ poll. diam. A small trailing plant not imcommon in the shady parts of the valleys. It is nearly allied to an Indian species, P. dindygidensis, Miq. ; CDC. in DC. Prod, xvi. 1. 412, but dift'ering in the glabrous leaves, shorter petioles, and non-puberulous stigma. Plate XXXV. Pig. 1. Portion of spike. 2. Bract, 3. Flower. 1. Stamen detached. 5. Fruit in vertical section. All magnified. Peperomia Rodriguezi, BoJf.Jil. Herba ramosissima, carnosa, repens, caulibus ad nodes radicantibus subpilosis ; IS. BAYLEY BALFOUR. 371 foliis oppositis, petiolatis, obovato-ellipticis, basi ciiueatis, obtusis iitrinquc glabris, ciliolatis, subtus albitlo-pallidis subtiliter 3-5 ncrviis cnei'vulosis, petiolo piloso. Citt, ign. Caules subpcdales, t^'^- poll, crassi. Folia ^Z\ poll, longa; petiolus .1-4 poll, longus. This small species I found only on the southern slopes of Mount Piton creeping over the surface of large boulders. Its nearest ally is P. Ventenoti, Miq.; CDC. in DC. Prod. xvi. 1. 116, an East Indian species. I only obtained the plant in leaf, and should not myself have ventured on a specific determination ; but M. Casimir de Candolle has kindly examined the specimens, and pronounced it a novelty. LORANTHACE^, Viscviin taenioides, Comm.; DC. Prod. iv. 283. Only in the valley of Eivi^re Baleine, and abundant there on the branches of Fernelia huxlfoUa, Lam. This plant is peciiliar to the Mascarene Islands. MUSACE.ffi. Musa paradisiaca, Linn. ; Trew Bhret. tt. 18-20 ; IR^ecl. Lil. t. 443. Nom. vulg. Bananc. Cultivated. Musa sapientum, Linn. ; Trew Ehret. tt. 21-23. Nom. vulg. Banane. Cultivated. Ravenala madagascariensis, >S'o««erfli^ Fb^. ii. 223. tt. 124-6; Jacq. JEort. Schosn. t. 93 ; Urania speciosa, Willd. ; Boj. Sort. Mmir. 333. Nom. vulg. Eaveual. A few trees near dwellings of old settlers. ORCHIDACE^. Mr. S. Le M. Moore has kindly determined the Orchids. Oberonia brevifoUa, Lindl. Gen. and Sp. Orch. 16; Fol. Orcli. No. 36. Not very common. Only found in valley Riviere Baleine on Fernelia huxifoUa, Lam., along with Viscuni tmnioides, Comm. Bulbophyllum incurvum, Thotiars Orch. Afr. t. 95. A Mascarene plant not uncommon on the branches of trees. Aeranthus arachnites, Lindl. Bot. Beg. sub. i. 817; Bot. Mag. t. 6034. Var. Balfoiu^ii. Leaves 10 in. long ; lateral sepals 1^ in. long. A distinct variety of tliis Mascarene species. It is very common on the branches of trees. Listrostachys Aphrodite, Balf. jil. and S. Moore in Baker Fl. Maw. Seyoh. 354. Plate XXXVI. Caidibus erectis, validis, -^--1 pcd. altis ; foliis subimbricatis, carnosis, lineari- oblongis, oblique-emarginatis, 2^-3 poll, longis, 3- poll, latis v. latioribus ; racemis 3 A 2 372 BOTANY OF RODRIGUEZ. atlsconilentibus foliis oppositis, fcrc 5 poll, longis, sqiiaiiiis nounullis lagis loutis scariosis vaginautibus iul'ra lloros vestitis; bractcis rotunclatis, ,1 poll, longis; sopalis lanceolatis, aciimiuatis, lero \ poll, longis ; pctalis brovioribus, linoaribus, labcllo ovato-votundato, 3-lobalo, lobis latcralilnis crcnulatis, inlVa circiun t-ohunnam oon- volutis, lobo L'cntrali augusto-linoari, iiitcgro vix ,'; poll, longo ; calcarc /^ poll, longo I'orc recto; polliniis oblongo-ovoidois, candiculis 1incaril)us Icviter in glandulan\ ovatam atteuuatis ; lobis rostcUi vcrticalitor rcsupinatis. Nom. viilg. Faliame. An eudoniic^ plant only occasionally inct with on stems and branches of trees. Plate XXXA'I. Pig. 1. Plowcr bud partly open. 2. Pront view of cxpandcnl iknver. 3. Side view of column. Augrsecum, sp. near A. cauJescens, Thoiutrs. Too far advanced for description. A very common plant. AMARYLLIDACE-ffi. Crimxm asiaticiim, Llitn.; Kitnih. Eiuhih. v. 51-7; Bol. 21 wood is very hard and durable, of a rieh mottled blaek api^earauee, and used for building huts, though now this is interdietcd as the trees arc becoming more scarce. The leaves arc chiefly used for thatching huts and are also made into baskets. The fibres of the petiole form a very excellent material for cordage and the reticulum is also put to various uses. Male and female trees exist iu about equal numbers. This is the Lalania aui'ca of horticulturists, and has been known in Europe for some years. The original description of the plant in 111. Ilort. I.e. errs in the dcsciiption of the pyrcna;, which arc represented as inverted, the apex being described as the base and vice versa. Leguat speaks of tliis tree as the " plautanc," and thus quaintly describes it, — " Tlic plantane is a sort of palm tree, and the arborists place it in the same class. Our plantanes have a straight trunk, Avhich seems to be formed of large rings at an equal distance. They have no such prickly scales as I have talked of in the palm tree. At the top of the trunlc is a cabbage, very like to that of the palm tree. At the foot of this cabbage instead of palm boughs are broad leaves, witli stalks about six or seven foot long ; these leaves are strong and thick, and like a fan when it is open ; the sticks of which come a little out of the circumference, and arc pointed at the end. Some of these leaves are eight foot diameter, insomuch that they serv'd to make rare coverings for om* cabbins. "We cut 'em out into little pieces, and made hatts and umbrcllo's of them. The stalk is four inches broad, an inch thick, and a little roundish at the sides ; at the bottom where it joyns to the tree, it widens and grows like a flat shell which sticks to the trunk, and in jiart embraces it. This wide and hollow plate is sometimes above a foot diameter, and of the thickness of a crown piece. We made use of it for dishes, plates, and sawccrs. The first rind of the stalk served us instead of ropes, and the fibres of the second made good thread to sow with. One might have wove stuffs with it, had it been prepar'd. " We cou'd not perceive any difference in the tast, or in any other qualities of the palm tree or plantane. This liquor is whitish like white whey, and so sweet, that no other sweetness, if I may judge of it, can compare to it : the newer it is, the more agreeable. In three or fuor days it begins to turn sowre, and in seven or eight, 'tis as sharp as the strongest vinegar, without changing its colom*. " The dates of the plantane are bigger than those of the palm tree. Having abundance of better things to feed on, fish and flesh, fruits, &e., we left the dates for the tiu-tles and other bu'ds, particularly the Solitaires, of ■which we shall here- after make mention. '•' About the cabbage of the plantane, near the bottom of it, and between the stalks of its broad leaves, is a sort of cotton of a limon colour, which all thro' IS. BAYLEY BALFOUR. 375 India is known by the name of capoc. We made very good quilts oT it. It may be wove and mannfactiir'd for all the uses that cotton is put to. Perhaps we might have thought of making a sort of stuff, both of the capoc and the fibres of our plantane leaves ; but wc had stuff enough of our own to serve a long time, and the air is so mild, so sweet, that we did not make much use of our cloaths." The genus Latania is Mascarene and is represented by three species. Of these one here mentioned is endemic in Kodriguez. Another, Latania Loddigesl, Mart. Hist. Palm. iii. 221. t. 161. f. ii. 10-11, the L. gknicophylla of horticulturists, is found only on Hound Island near Mauritius, while the third species, L. Commersoni, Gmel. Syst. viii. 1035, and which iuchides L. rubra, Jacq. Frag. Bot. 13. t. 8, and L. borbonica, Lam. Encyc. iii. 427, is distributed in both Mam-itius and Bom-bon. The three species are easily distmguished specially by the pyrense. It is cm-ious to note the similarity of constitution and distribution of this genus with another Masca- rene one, Si/ophorbe. Phoenix dactyhfera, Linn, Nom. vulg. Le Datte. Is not abundant and occurs occasionally on some of the coral islets, Hyophorbe VerschaflFelti, Wendl. in III Sort. tt. 462, 463. Norn. vulg. Palmiste marron. A palm spread over the whole island, but never occurring on the coralline limestone. It is of a very striking appearance by reason of the bulging which takes place in the stem towards the middle, the stem on both sides of the swelling decreasing in size. If the tree be lofty, there may be a second ventricosity. But the tree seems rarely to attain an altitude above 20-25 feet. The external hard part of the stem is very thin, not more than an inch thick, and within it is a soft succulent mass of cellular tissue and fibro-vascular bundles. The juice from this tree is said by the inhabitants not only to be unwholesome, but even poisonous, causing, if taken in small quantities, severe emesis. The leaves have an exceed- ingly plumose appearance, and with theu- yellow stripe beneath are very pictui-esque. The parts of the tree are put to no use. Hyophorbe, to which the species belongs, is a Mascarene genus represented by only three species. Of these the one here mentioned is endemic in Rodriguez. JT. amaricanlis, Mart. Hist. Palm. iii. 309, formerly cultivated in Europe imder the name Areea speciosa, is a second species endemic in Roimd Island, about 6 miles from Mam-itius. This, from its habit, is not unfrequently termed in Mauritius the " bottle palm," and hence it has for long been confounded with the Chilian bottle palm, Jubcca spectabilis, with which, however, it has no connexion. The Rodri- guez palm I should have said has also been confounded with Jubcsa. The Round Island palm is very distinct from the Rodriguez plant. Tlie third species is the most delicate. Originally described as Hyophorbe indica by Giirtn. de Fruct. ii. 186, the name R. Commersoniana was substituted by Martins Hist. Palm, iii, 164. o70 IJOTANY OF IJODKIOUKZ. There seems, however, 110 sulllciiMil ground t'ov ilir alicMviiioii. ]Vir_v St. \ ineont, Voy. ii. 2!)(), mentions and deserihes this palm as Jrcca Intesccus, under wliicli name it is fiviiuently and most commonly met Avith in gardiMis. The species has a Avider distrihution than \\\v (^thers, oceuiTing-in hoth Mauritius and JJDurhon. It is in these ishmds conliued to tht> shady jKirts ol" the Avoods and Aalh^vs and is now extremely rare. It dillers in hahit from t lie other species in having a sleniUu- tapering stem not dilated and with no Acntrieosities. Dictyosperma album, If'ciiill. in JjiiiiKca xxxix. 181. Aar. am*eiim ; Arcca alha, J)Oi\>/ T'oi/. i. 300. Palma typo minor tenuiorque usque ad vigiuta jiedes alta ; l\)lia 1-S ])e(l. loiiga ; petiolus eonimunis curtus, S ])oll. longus, intus paullo-eoiiA'exis squamulis adspersis munitis; A'agina l-'2 ped. longa ; pinme lineari-lanccolattie, angustc aenminatoo, lo-2 ])ed. longoe, 1 poll. lat:e, A'eniilis secundariis ohscuris ; spadix i'astigiatus ramulosiis, ramulis rigidis (n-eetis 9-11 poll. longis, omnino rcctis v. ad hasim torti- lihus ; tlores illis typi dimidio minores ; fructus A'iolaceus, -rl-.j poll, longus, ,'~iV poll, latus, cylindro-eonicus; semina i poll, longa. Planta) juA'eniles aurautiaccse, pinuis fere Unearihus, spinalis vestitse. Nom. vulg. Palmist c bon. This palm is very abundant in Hodriguez, growing freely both on the volcanic soil and on the coralline limestone. It bas for many years been cultivated in the gardens of Europe as Areca ain'ea. The genus Areca lias long been a receptacle for many species of doubtful affinity, but Wendland bas recently revised the genus and removed tbcrefrom many of its hitherto contained species. As a result of bis revision the genus is unrepresented in the Mascarenc islands. Some Mascarenc sjiccics are referred, as already noted, to Syophorhe ; U\o others, Areca riihra, Bory, and A. crinita, Bory, constitute the endemic Mascarenc genus AcanthoplioenLv, and the old Areca alba, Bory, is tbe type of a ncAV genus, Dlcti/osperma. This is a very variable plant, and bv reason of this several crarden names bave been given to its forms under cultivation. Thus we bave Diclyosperina furfuraccum, D. ruhriim, and D. aurenm. These arc, however, all varieties of the one palm, Dictyosperma album, Wendl., and tbc last mentioned is that form wbicb occurs in Ptodriguez. Leguat mentions only one species of " palm tree " in Rodriguez, and it is difTi- cult to determine whether he refers to Hijopliorbe or Dictyosperma. He tbus describes it . — " Our palm trees are commonly 30-40 ft. higb, their trunk is straight and with- out leaves, but 'tis cover'd Avith a sort of prickly scales, whose prickles stand out a little. Some have a smoother bark than others. On the top of the trunks groAV those boughs of palm, of Avliich no man ever saw a liA'cly picture. These boughs form a great knot and fall doAvn all about it in plumes. BeloAv these boughs, or rather IS. BAYLEY BALFOUR, 377 below the trunk from which they grow, ai-e produc'd long hunches, each fruit or grain as big as a hen's egg, and of the same shape, known by the name of dates. " In the center of this great knot, and at the summet of the trunk, is what we call the cabbidge. One cannot see it, being hidden by the boughs that rise a little all about it. This summet consists of tender leaves Avhich closely embrace each other, joyn together, and form a mass something like that of a cabbage lettice, or common cabbage ; 'tis about two foot high if the tree is large, and of the same bigness with the trimk. The large outside leaves of this mass are white, soft, pliable, and as strong as buff, Avhich it resembles. They will serve also for linen, satin, for napkins, table-cloths, and any thing what you please. The membrana, or inside leaves, are tender and brittle, like the heart of a lettice. They are good to eat raw, and last like a filberd ; but we made an admirable ragout of them when we fricasi'd them with the fat and liver of a turtle. We put them also in our scops. " We come now to the liquor, or rather the nectar of the Isle of Rodrigo. 'Tis call'd palm-wine all over the Indies. There are two ways of drawing out the juice. We make a hole in the trimk of the tree at about a man's height, as big as one's two fists. We presently put a pot or other vessel there to receive the delicious liquor, which runs out fast enough, otherwise we dig the cabbage, and make a little cistern at its head. We need only go twice or thrice a day and draw this rare wine at the fountain head, and we may be abundantly supply'd with it. The wine of the trunk, and that of the cabbage, are in my opinion of an equal goodness. But those who would be good husbands of their trees (for as for us we were lavish enough of them), the first way is the best, because after the cabbage has yielded its liquor for about a month, it withers, and the tree decays and dies. 'Tis the same thing if you tear off the cabbage, when its head and brain are gone it dies almost suddenly. " Whereas if you only pierce its side the tree do's not die, provided the wound is not too deep ; but the liquor will not run out at that hole above four days. The wounded tree must afterward have time to recover. I do not know what is done elsewhere, but I can tell by experience what I have said here, we having made trial of it daily for two years together. The bark of this tree is very hard, 'tis an inch thick, porous and tender in the inside. If one make the hole on the side .of the trunk too wide, there is reason to fear 'twill weaken the tree tlicre, and that then the next Inu'ricane Avill break it." " The fruit or grain as big as a hen's egg " makes it diificult to refer his palm to either Jlijophorbe or Dicti/osperma, and points rather to the Latanier. And then the trunk "covered with a sort of prickly scales whose prickles stand out a little," does not agree well with either, and least with Ilyophovbe. Jkit then he says afterwards, " The bark of this tree is very hard, 'tis an inch thick, porous and 3b 378 r.OTANY OF KODUIOrF.Z. '• tender in llic inside," whidi morc nearly corresponds with JT^opZior&f' ilian with Dicti/ospcrma. But, on the other hand, the inliabitants say the juice of Hifophorbe Vcrschtiffi'Ui, Ch. Lorn., is poisonous, Avhilst L(>guat and his companions seoin to have used it much. It is then a diilicult matter to decide, but ii' we discard the statement as to the size ol" the fruit, I think avo may most safely consider that Loguat failed to diagnose two species of Palm tree, and in his description includes both species. And as pointing in this direction I think we may take his remark " Some have a smoother bark than others ; " for the bark of Kyopliorhe Verschaf- fcUi, Cli. Leni. is much smoother than that o'^ Dicli/osporma album, Wondl. Cocos nucifera, Liun. \ few trees of this occur in the compound of the Government llousc at Port ]\Iatluu-in, and one or two trees are found on the south side of the island. The introduction of this plant was on this Avise as described by Leguat : — " The sea " having thrown us up some Cocos which began to bud, we planted some of that " fruit some months after our arrival, and wdien we left the place, the trees were " four foot hii^h." 'o' PANDANACEiE. Paudanus heterocarpus, Balf.fil. Arbor ramosa, caudice Itcvi, radicibus aereis plurimis, ramis patentibus ; foliis lanceolatis acumiuatis, basi amplexieaulibus dilatatis, crecto-jmtentibus, coriaceis, strictis, vii'idesccntibus basi sgepe glaucescentibus, subplanis, marginibus per totura costaque subtus a medio distanter rubrosi^inosis, spinulis subincurvatis ; c? sj)adicibus laxe racemosis odoris 15-spicatis, spathis subito acuminatis, carinatis, basi latis, sj)icas trigono-oblongas excedentibus, marginibus costaque subtus per totum spinescen- tibus ; staminibus lase dispositis, filamento commuui brevi, 5-10 ramoso, ramulis patentibus, antheris linoari-oblongis mucronatis ; ? capitulis solitariis globosis sub- depressis v. elliptico-oblongis nimc longe nunc breviter pedunculatis, pendulis v. inclinatis, spatliis pluribus brevibus deciduis ; di-ujiis 5-locularibus, obpyramidalibus, non compressis, 5-6-gonis, fere per totum coadunatis, apice humiliter pyramidalc rarissime convexo sa^pius applauato v. umbilicato, distincte 5-6 angulato, areola centrali 2-5 stigmata plana sessilia reniformia suberosa cingente instructo. Arbor umbraculiformis, usque ad 20 pedes alta. Candlces fuscobrunnei, 5-7 poll, diam., Iseves, erecti, oblique annulati, ramosissimi, ramulis termiualibus non adscendentibus, radices aereas plures seriatim spinesceute-tuberculatas emittentes. Folia ensiformia, recta v. rarius recurvata v, dependentia, 1^-3 ped. longa, 1^-2^ l^oll. lata, supra non sulcata, non reduplicata, pallide v. atro-virentia, supra nitida, subtus opaca venulis lateralibus subtiliter striata, marginibus incarnatis a basi spinosis, spiuis ssepe in medio deficientibus acutis rubris subincurvatis adscendentibus, costa subtus pallidiori subrosea a medio spinescente, spiuis in margiue majoribus intervallo- que longiore dispositis. Spadices masculi 1 ped. longi 15-spicati peuduli, spicse basales IS, BAYLEY BALFOUR. 379 3 poll, lougoe, spathoe sursiim sonsim minores oymbiformcs cuspidata; abrupte angustatcque aciiuiinatoe basi submcmbranacete latoe, spicas amplcetentcs polli- bicus pJui'ibus suj)erantes, marginibus caruaaque subtus per totum spinulescentibus. Staniinum fQamentum commune ^ poll, longum, ramiili singuli breves patentes ^ poll, lougi ; antheroe albidno -^ poll, longge. Capitula 60-70-(U'upata, nunc 35, rarius 100, 4^-62- poll, longa, -i^-S^- poll, lata ; peclunculus trigonus plcrumque 8-12 poll, rarius 3 v. 15 longus, -iV~i"2 pol^- crassus. Dnip(C rJ— IJ- poll. long£e, 1-li poll, latiu borizontalitcr, f-1 poll, in diam. vcrticalitcr, usqvie ad |— | a basi coadimata?, parte conjuncta matui'itatc rubra v. flava, apicis margine nonnunquam ruguloso ; areola centrali distincte marginata ; stigmatibus -^-j poll. diam. reni- formibus, pelvis centrum versus spectante; putamine rotundato lateribus liBvibus apicc applanato processibus 2-5 instructo ; mesocarpio spongioso, amplo ; semine \ poll, longo. Nom. vulg. Vacoa calc rouge, V. cale blanc, V. sac, V. poteau, V. parasol, V. male. It is a very variable tree, and the popular names indicate this. It occui'S very abundantly everywhere on the island from the seashore to the highest points. And according to its situation, its habit and appearance vary. Thus when on sites ex- posed to the wind it has a stimted habit. The branches are few, tliick, and short, and the leaves are also short and are erect. In such situations it is known as the Vacoa cole. The inhabitants make a distinction between two varieties of this according as the head, peduncle, and united parts of the drupes are red, or are greenish yellow or yellow. The former they style roiige the latter blanc. The first of the characters on which they base theii* varieties by the colour of the head and pedimcle is worthless, and the difference in colour of the bases of the drupes is found in all the forms of the species, but is not sufiicient to characterise varieties. If the tree grows in siiitable soil, and in a sheltered position where it has room to develop its branches properly, then it forms a dense and compact dome and the branches may droop downwards, so far as almost to conceal the stem, and is then known as Vacoa parasol. When in any situation the tree develops a trunk of good size, and is allowed to grow until the wood is hard and firm to the centre, and is capable of being used as a post for a house, then the tree is called V. poteau. The name V. sac is given to the young plants when the leaves are long and broad, and may be made into bags or sacs. V. male is of course the male tree, known by the inhabitants as not bearing frait. It is not only in habit that the tree varies, but few plants exhibit such an amount of variation in the fruit and it well deserves its specific designation. It is impossible here to enter into a consideration of these which will form the subject of a separate memoii*. This species belongs to the TJtilis group of the species of Pandamis. It has a close aflinity with P. utilis, but is sufficiently distinguished in habit and in fruit. 3b 2 380 r.OTAXY OF HODRICI'KZ. Lt'guat gives a curious description oL" the Screw pines ol" llodrig-uoz as folloAvs : — " Among the great nunil)cr and variety of trees in (liis island ])lanted by nature, " there is one ^vhiell is wonderful and wortliy our ])ai'iieular observation for its '•■ beauty, bigness, roundness, and llie rare symmetry of its exaef brandies, tlie ends " of which are A'cry much tufted, and its leaves so great and lliiek tliat they fall " do\\ n almost to the ground all about it, so that come which way you will at this " tree, you can perceive but a small part of its trunk, and that at Ihc bollcnn of it, '• and sometimes you can see nothing at all of it. It being as one may iuiagine, all '• shady in the middle ; the branches are within-side like dry poles, Avhich seem to " be the work of a carpenter, and set there to bear up the i)lumes or branches a\ hich '• ai'e quite about it, and thus make a sort of cage or tent of the tree, 'Tis true, the '• greatest beiiuty of t bis tent, is in its charming outside, though the coolness and " shelter of the inside have also tiieir charms : 'twas unhapi)y that its fruit was not " good to cat. Those of us Avho had the curiosity to eat it, found it sour, and knew " by experience that was all the hurt that was in it. It had the smell ol' a very " line quince. 'Tis a sort of a grape, the seeds of which are close and altogether; " it looked at a distance like the fruit of the ananas, for Avhich reason we us'd to " call these trees Ananas, tho' there's a great difference between the (wo plants: " as for me I was calling it Wie jjavilion or tent. The leaves are of an admirable " green, and the stalks of them are so short, that one wou'd think they grcAv imrae- " diately to the Avood. The greatest are four or five inches broad, sharp at the end, " and about five inches long. They form a great bunch, and here and there one " may see the grapes, which are of diAers colours, according as they are more or less " ripe. I have often taken pleasure to survey these natural palaces, and was " equally ravish'd with its largeness and singular beauty." Pandauus teuuifolius, Bclf. JiL ^Vrbor parva, caudiee kevi, radicibus aereis paucis, ramis ercctis crassis ; foliis angustate lincari-lanccolatis acuminatissimis, apicc subflagcllatis, basi dilatatis lateque amplexicaulibus, coriaceis reciu'vatis pendulisque, subjjlanis atro-viridibus ssepe basi dealbatis, marginibus subincrassatis per totum costaque a medio rubro- spinosis, spinis brevibus acutis incurvatis adscendentibus ; capitu.lis soKtariis pauci- drupatis subglobosis, pedunciilo curvato pendulosis ; spathis pluribus deciduis ; drupis subpyriformibus 2-5-locularibns, non compressis, usque ad f-f a basi coad- unatis, parte conjuncta 5-6-gona obpyramidali, apice libero rubiginoso rotuudato umbraculiformi obscure 5-6-angulato, vertice applanato v. depresso, areola margi- nata 2-5 stigmata plana sessilia rcniformia cingente instructo. Arbor usque ad 15 pedes alta pleramque magnum spatium instar arbustum occupans, rare solitaris. Cuiidices fusci, laeves, 3 poll, diam., ramos erectos validos abrupte terminantes augulo acuto emittentes. Folia 8 poll. — 2\ ped. longa, :}— fV poll, lata, supra non sulcata, subplana, non reduplicata, supra nitida, infra opaca, late- ralibus nervulis subtiliter striata, marginibus subincrassatis a basi s2)inosis, spinis IS. BAY LEY BALFOUR. 381 ssepe in medio deficicntibus, rubric, arcto dispositis, iucuvvatis v. subapprcssis, adscendoutibus, costa prouiiiiente rubra, a medio spinis distantibus acutis lougi- oribus instructa. Capifnla 20-40-drupata, 3-5} poll. diam. ; peduuculus trigonus, ciirvatus, 5-11 poll, longiis, l—{'.r crassus. Drupct breviter pyriformes, 1^ poll, longa?, parte superiore ^\ libera, horizon taliter \\-\\ poll, latne, verticaliter If-lji poll. latte, parte [conjuneta matnritate rubra ; stigmatibus ^g-\ poll, diam., reniformibus, pelvi centrum A'crsus spectante ; putaminc rotuudato latcribus Itevibus apice de- presso 2-6 jn'ocessus gereutc; mesocarpio crasso spongioso; seminibus 'j poll, longis. Nom. viilg. Vacoa cbevron. A small and very graceful tree, usually growing in clusters and forming frequently a very regular dome. It is confined to the higher parts of the island and the upper parts of valleys. Though nearly allied to, it is easily distinguished from the foregoing by its arching narrow delicate leaves, and its few large druped fruits. It is also a very much smaller plant. But certain trees on the island seem to possess characters intermediate between these two species and may be hybrids. This is, however, a difficult matter to settle. Its local name is derived from its common use as a rafter for huts, for which its close-grained and hard wood when matiu'e well fits it. These are the only two species I found on the island, but Sir Henry Barkly has sent home figures of two (and of one a specimen) other species from the island. These are P. vfilis, Bory, and P. odoraUssimus, Linn. f. His figures and specimen of the former are most distinctly those of P. utilis, Bory, but whilst I do not dispute the fact that this plant may have existed or may exist on the island, I may notice that Sir Henry Barkly's drawing and specimen are of a tree grown in the Botanic Garden in Mauritius from seeds which Mr. Duncan is said to have brought from Rodriguez. I think this takes a^vay considerably from the value of the observation, especially as the trees Mr. Home pointed out to me as from Uodriguez are growing closely siu'rounded by Mauritius ones. At the same time I may remark that in the yovmg state, before many brandies have been given off, P. utilis, Bory, and P. heteroearpus, Balf. fil., are so much alike as to be almost undistinguishable, and I may have in that way missed P. utilis, Bory, in Rodriguez. As to P. odoratissimns, Linn, f., I can only say I never saw it, and it must be very scarce, if there at aU ; and it seems to me very strange that such a very marked form should be unknown by the inhabitants, especially as several of them were employed by Mr. Jenner, the magistrate of the island, at the request of Sir Heiuy Barkly, to collect Pandani for him. Mr. Hornc showed me trees of this in tlie Botam't; Gardens at Mauritius, said to have grown from seeds brought from Rodriguez by Mr. Duncan, and as this may have been the source of Sir H. Barkly's figures the same objection may be urged as in the case of P. utilis, Bory. It seems to me veiy strange that no mention is made by Sir H. Barkly of P. tenu/foliiis, Balf. fil., from Rodriguez, which is very abundant, whilst P. fitilis, Bory, and P. odora- 382 liOTANY OF KODIUGUEZ. fisshiiiis, Linn. f. nuist be very scarce, if there. Bojcr in his Ilortus Mauritiauus says of P. »i;iric(ifi(s, Pet. Thouars, that it grows naturally at Rodriguez. 1 believe Bojer visited Eodriguez, and he onght to have known the Pandiuu, but he says nothing about JP. hcterocarpus, Balf. ill. or P. tcnnifollus, Ball", fil., which are so common on the island. He appends to his notice of P. muricatus. Pet. Thenars, that some trunks are to be found at Grand Port. To recognise a species of PdiichoiKS by leaves, fruit, and stem is in many cases difTicnlt enough, but to dis- criminate species by stems only is very fallacious, if not impossible. I am convinced the species does not now exist on the island; and, indeed. Petit Tliouars only re- corded it from Madagascar. In the Botanic Gardens in Maiu'itius a species was shoAATi mc by jMr. Home which he supposed might be P. muricatus, Pet. Thouars, but this is merely a form of P. iitilis, Bory, with drupes much divided at the apex. AROIDE^. Colocasia autiquorum, Scliott Trod. 138 ; Arum colocasia, JJlnn. ; Wt. Ic. t. 78G ; Calatlium esculentum and nymphetefolium. Vent. Nom. vulg. Songc rouge. Grows very abimdantly in the streams. Is eaten in times of scarcity by the inhabitants, but is said to produce a severe form of skin eruption if much eaten. It is chiefly used to feed pigs when they can get nothing else. Alocasia macrorhiza, Schott Frodr. 146 ; Arum macrorhizum, Li)iii. ; Boj. Hort. 31(1. ur. 356. Xom. vulg. Songe blanc. Grows abundantly with the former, and is the better to eat of the two. NAIADACEiE. Ruppia maritima, Linn. ; Ktinth Enum. iii. 123; R. rostellata, Koch; Beich. Ic. ii. 66. 1. 174. Abundant at the mouth of English Bay River. Zannichellia pahistris, Linn. ; Kuntli Enum. iii. 124. Abundant, especially in Rividrcs Bouteille and De I'Est. This plant is not recorded from Mam*itius. Halophila ovalis, Rook. fil. FL Tasman. ii. 45 ; H. ovata. Gaud, in Freyc. Voy. 429. t. 40. f. 1. Grows abundantly on the sandy flats on the reefs, and is left uncovered at the faU of the tide. Halophila stipalacea, Aschers. in lAnncEa xxxv. 172 ; Thalassia stipulacea, Konig ; Kuntli Enum. iv. 120. Abundant on the sandy flats on the reefs, specially where there is a current, and does not seem to be left uncovered by the tide. IS. BAYLFA' BALFOUR. 383 CYPERACEiE. Cyperus Isevigatus, Ziuu. ; vav. albidus, Beklr. in Linncea, xxxv. 487. Very coiiunou sj)ecies, occurring under several forms, according to situation, CjT)enis polystachyus, Jiottb. ; K/aif/i Eimm. ii. 13; Beklr. in Linncea, xxxr. 477. Not uncommon on the islaiul. Cyperus dubius, Bottb. ; Beklr. in Ziiincea, xsxvi. 336. Nom. Tulg. Herbe a oignon. Common everywhere, and is a great pest. Cyperus distans, Linn.Jil. ; Kunth Eniim. ii. 93 ; Beklr. in Linncea, xxxv. 612. Frequently found. Cyperus, sp. A species without flower I found close to Gabrielle village. Kyllingia monocephala, Botth. ; Beklr. in Linncea, xxxv. 427. Very common. FimbiistyUs gloinerata, Kees ; Beklr. in Linncea, xxxvii. 47. Common, especially on coralline limestone. Fimbristylis communis, Kunth Enum. ii. 235. Common. Carex gracilis, B. Br. ; Kunth LJnum. ii. 513 ; Boott Car. i. 59. t. 154-156. Xom. vulg. Herbe Madame Pitelle. Very common. GRAMINE^. Coix Lachryma, Linn. ; Kunth Emim. i. 20 ; Bot. Mag. t. 2479. Nom. vulg. Larme de Job. Very abundant, especially near habitations on the banks of streams. Zea Mays, Linn. ; Kunth Enum. i. 19; Mart. Flor. Bras. ii. 2. t. 11. Cultivated, but not so much as it might be. Is exported to Mauritius. A good head is supposed to yield 640 grains, and each plant bears usually two heads. In planting thi'ee seeds are put in each hole, and five heads are expected from the three plants. Paspalum scrobiculatum, Zw«. ; Kunth Env/ni. i. 53; Trin. Sp. Gram. ii. t. 143. Isom. vulg. Herbe file. Very common. Panicum brizoides, Linn. ; Kttnth Enum. i. 78 ; Trin. Sp. Gram. ii. tt. 158-9. Common. Panicum fluitans, Betz. ; Kunth Ennm. i. 78. 384 BOTANY OF KODRUIUKZ. Pauicum maximum, Jaoq. Ic. t. 13. Paiiicum plicatum, Lam.; var. P. costatum, Ro.vb. FL [lul. i. 311; Knnlh Emmi. i. 91. Commonest grass on the island. Panicum serpens, Kit nth En inn. i. 117. Panicum Balfom-ii, B((kcr Fl. Jfaiu: Seych. ISS. Frequently met with in \\\c valleys. Closely allied to P. midnlatiformm, Ard. Steuotaplirum complauatum, Schnmk. ; Kwith Emini. i. ]37. Common. Steuotaplirum subulatum, Triii. Sp. Gram. iii. t. 3G0. Only growini;' on Gomljrani and Pierrot islets. Ceucliriis echiuatus, Ijinn. ; Knnth Emm/, i. IGG. Nom. vulg. Ilerbe cateaux. Is widely dispersed. Zoysia pungeus, 7/7/A/. ; Kiinfh Ennm. i. 171. Very common on tlie sandy and coral islets on tlie reef. Saccliai-um officiuarum, Liiiii. ; Knnth Enum. i. 471 ; Ilook. Bat. Jlisc. t. xxvi. Formerly much cultivated. It grows well, hut is not much cultivated now, on account of the difficulties in manipidation from want of Avater. Andropogon contortus, I/iiin. ; Knnth Ennm. i. 186. Audropogou mm'icatus, Bel:. ; Kunth Enum. i. 505. Nom. vulg. Vetivert or Cuscus. Gro^As in abundance on the slopes of Movuat Malartic, and the roots used much hy the natives for scenting their cabinets and apparel. Audi'opogon Schoenanthiis, Linn.; Kunth Enum. i. 193. Nom. vulg. Citi'onellc. An infusion of this is a very favourite tisane. It grows abundantly on the slopes of Mount Limon. Andropogon foliatus, Stcnd. Si/n. Gram. 389. Probably a variety of A. Schosuaii- thus. Andropogon finitimus, Hochst. ; Steucl. Syn. Gram. 385. Andi-opogon halepensis, Sibth. ; Kunth Enum. 502. Nom. vixlg. Petit mille. Cultivated. Sporoholus virginicus, Knnth Enum. i. 210. Aristida adsceusionis, Linn. ; Knnth Enum. i. 190. Cynodon Dactylon, Pers. ; Kunth Enum. i. 259. Nom. vulg. Chiendent. On the shores where sandy, and on most of the coral islets. Eleusine indica, Gdrtn. ; Kunth Enum. i. 272 ; Trin. Sp. Gram. i. t. 71. Everywhere on the island. IS. BAYLEY BALFOUR. 385 Chloris barbata, »S'/c. ; Kunth Emm. i. 2^1; Tri/i. Sj). Gram. iii. t. 306. Norn. Tulg. Esquinc. Very common. Phragmites communis, Trin.; Einith Enum. i. 251. On the banks of the stream in valley Eivi5re Grande des Bamhoux. Bambusa, sp. Abundant on the banks of the streams. LYCOPODIACE^. Lycopodimn Phlegmaria, Lhtn. ; Spring. Mono(j. Lycop. i. 63. Kot common. Only at the top of the valley of the EiviOre aux Huitres. Psilotum triquetrum, Sw. ; Boj. Sort. Maur. 423. Very common. Selaginella Balfouri, Baher Fl. Maur. Seych. 522. Canlibus decumbentibus dense caespitosis ad 3-4 poll, procumbentibns, ramulis compositis plurunis curtis adscendontibus instrnctis ; foliis majoribus obliquis, oblongis, obtusis, patentibus, yV poU- longis, albo-viridil)us, mombranaceis, ciliatis, siunmis marginibus comiiventibus ; foliis minoribns obovatis, mucrone longa armatis ; spicis |— ^ poll, longis, tetragonis ; bracteis iiniformilius, deltoideis, cuspidatis. The specimens from which this species is described are from two localities. On the coralline limestone growing along with DicJionclra repens, Forst. and Sypoestes hiconspicua, Balf. fil., a plant of which the leaves have a reddish and dried appearance was found. Other specimens of a pale green and fresher look were obtained in the valley Riviere Grande des Bamhoux (Cascade). Baker 1. c. considers the plants from both localities one sj)ecies as above described. Professor MacXab informs me that he believes the two forms belong to different species, and he thus describes them : — Form 1. Growing on coralline limestone. Large leaves ovate, rather obtuse, looft long, -j-^^nj broad, margin serrated, midrib central, base loide, loioer side rounded not produced, zipper side prodiiced not so short, denticulate auricle; small leaves yffg- loiigj tw<7 broad, contracted below, widest above middle, then narrowed into an acute point, margin of leaf denticulate, base with lower side produced into a long denticulate auricle, inner margin with a very minute aimcle. Form 2. Growing in moist localities in valley Eivi^rc Grande des Bamhoux (Cascade). Large leaves elongate, widest in middle, narrowed towards each end, apex rather blimt yet pointed, y^J^^ long, yf|,j- broad, up^jer side of leaf tcith a roxmded outline, loioer side nearly straight and less serrated, vein nearer lower than upper margin, base auricled, upper auricle long curved and almost ciliate, lower ciliate and denticulate; small leaves y^^^ long and y^^ broad, contracted below, widest above middle, then narrowed into an acute point, margin denticulate, base 3c 386 IJOTANY OK KonuTorr.z. with outer marcrin proilaccd iuto a large denticulate auricle, inner margin with a very minute auricle. Selagiuella rodri^nesiana, Baker Fl. JIdiir. Sei/ch. 523. Caulibus (Iccuiubeu(il)us, non-articulatis, ad podi>iu ])rocuml)entibus, ranmlis compositis pluriinis distant ibus adsceudeutibus iiistructis ; i'oliis majoribus obliquis, oblongis, ohtusis, falcatis, -,'.t poll, longis, yiridibus, mcmbranaceis, integris obscure ciliatis, summis marginibus conniA'entibus ; foliis miuoribus obovatis, longc mucro- natis ; spicis :}— 3- poll, longis, totragonis ; bracteis uniformibus, dcltoideis, valide cuspidatis. Very abundant. A species not Car removed from S. concinna, Spring, Monog. Lycop. ii. 190. Professor MacXab says of this plant, " the form of the large leaves " is like ^S*. concinna, Spring., bu.t the expression ' foliis rigidis longe biauriculatis ' " can hardly apply, the leaves being soft, thin, and with very small auricles. Then " the small leaves are much more oblique, and by no means ' aristato-mucronatis.' " The species comes near one I have from Kew Gardens cultivated as S. concinna, " but locality unknoA\'n to me." OPHIOGLOSSACE^. Ophioglossiim reticulatum, Linn. ; Hook. S^ Baker Syn. Fil. 446. Amongst the grass on the slopes in a few of the valleys. FILICES. Trichomaues cuspidatiim, TTiUd. ; Hook. <§• Baker Syn. Fil. 73. On the rocks at the sides of streams, only near their source. A Mascarene plant. Not very abundant. Davallia mam-itiana, Hook. Sf Baker Syn. Fil. 96. A Mascarene species common on the dry slopes of the valleys. Adiantiim caudatum, Linn. ; Rook. Sf Baker Syn. Fil. 115. Nom. wilg. La capillau'e. Very common everywhere. An infusion of this is used as a tisane in fevers. Adiantum Capillus-Veneris, Linn. ; FLook. Sf Baker Syn. Fit. 123. Not common only in the valleys of Eiviere Cascade and Ilivi^re Poursite, a few plants growing on moist rocks. Adiantum hispidulum, Swartz ; FLook. 8j- Baker Syn. Fil. 126. Common in the upper parts of valleys. Pellsea (Platyloma) hastata. Link. ; Hook. 8f Baker Syn. Fil. 152. Very abujidant everywhere. Pteris flabellata, TJiunh.; Hook. ^ Baker Syn. Fil. 161. Not very common. Grows in shady places in the upper parts of valleys. IS. BAYLEY BALFOUR. 387 Asplenium (Thainnopteris) Nidus, Linn. ; Hook, cj- BaJcer Si/n. Fil. 190. Xom. Tiilg. Laiigue dc boeui', Commou on trees iu the valleys. Asplenium hirtiun, Kanlf. ; Soak. Sf Baker Syn. Fil. 205. Common on the banks of streams. Asplenium falcatum, Lam. ; Hook. Sf Baker Syn. Fil. 208. Not commou. GroAving on dry stony slopes of valleys. Asplenium fiuxatum, Thunh. ; Hook. Sf Baker Syn. Fil. 211. Common on banks of streams. Asplenium (Anisogonium) decussatum, Sw. ; Hook. S)- Baker Syn. Fil. 213. Only found at the foot of Cascade Victoire. Neplirodium (Lastrea) crinitum, Best. ; Hook. 8)- Baker Syn. Fil. 265. A Mascarenc species abundant in all the valleys near the bed of the stream. Neplirodium unitum, B. Br. (non Sieh.) ; Hook. ^- Baker Syn. Fil. 289. Grows in large patches on the di-y slopes of some valleys. Nephrodium molle, Besv. ; Hook. 8f Baker Syn. Fil. 293. Yery abundant in all valleys. Nephrodium elatum, Baker ; Hook. §)• Baker Syn. Fil. 502. Only at the foot of the Cascade Victoire with Aspleniiini decussatum. Nephrolepis exaltata, Scliott. ; Hook. 8)- Baker Syn. Fil. 301. Everywhere. Nephrolepis acuta, Presl. ; Hook. 8f Baker Syn. Fil. 301. EveryT\'here. This and the foregoing species in many places cover the ground for several acres, notably on the eastern side. Polypodium (Niphobolus) adnascens, Sw.; Hook. §' Baker Syn. Fil. 319. Abimdant on dry rocks near the mouths of the valleys. Polypodium (Phymatodes) phymatodes, Linn. ; Hook. <5' Baker Syn. Fil. 361. Nom. vulg. Polypod. Everywhere abundant. An infusion of the rhizome is an excellent tisane for cough. ]\lr. Home showed me in the Herbariimi of the Botanic Gardens, Mauritius, specimens of a crested form of this from Eodi'iguez. There are specimens of a Nephrodium {Lastrea) grown by Lady Barkly at E.cduit, which is said to have come originally from E-odriguez, in the Herbarium at the Botanic Gardens, Mamitius. I do not know what species, and I did not gather it. In a letter to me at Rodriguez, Mr. Home mentions that in 1861 a collector brought from Ilodi'iguez Nephrolepis splendens, but there must be some mistake as to the name, as there is no such species. lAndsaya acuiifolia, Desv., is also reported from Eodriguez, but I have not seen specimens. 3c 2 388 BOTANY OF RODRIGUEZ. MUSCI. ]5y AVlLLIAM MllTEN, A.L.S. The number of the mosses as yet known to occur in the island of Eodriguez is thii-ty-throe, of th(^se ten were gatliered in a fertile state, the remaining twenty- three being barren. With one exeeiitionall the genera are the same as those known from Africa or tlie Mascareuc Islands, and thirteen of the species are identical with those found in Ihose countries. Three species are found also in India and the Indian Archipelago, and seventeen are considered to be new, of tliese eleven are without fruit, and six fertile. The most interesting moss is the Orthotrichnm j)^icatiim, Beauv., before known only from the specimens gathered in Bourl)on by Du Petit Thouars, described and •well figured by Schwajgriclien ; it is undoubtedly a member ol' the family of the Orthofriclia, but its characters as a genus are intermediate, and its position in that family unique. A few species correspond with those from the western coast of tropical Africa, and it appears from Avhat litth; is yet known of the mosses of the eastern coast that some species liave an enormous range in the African continent. Cosmopolite mosses arc represented solely by Weisia controversa. Octoblepharum albidum, Lin. (Bryum). Not a common species. Found on decayed branches of trees and vegetable humus. Calymperes (Hyophilina) Isevifolium, Mitt. Caulis ascendens ramosus. Folia dense inserta c basi erecta dilatata late obovata margine minute crenulata intus cellulis hyalinis areolata exinde patentia lineali ligulata apice lato acuta subapiculatave margine incurvo canaliculata inte- gerrima cellulis minutis sublsevibus obscuris et ad margincs folii basis erections limbum formantibus descendentibus areolata, vitta intramarginalis nulla. Folia anomala longiora erecta attenuata apice propagulis glomerulatis coronata. Stems about half an inch high, with the leaves a line and a half wide. The uppermost leaves green, the lower all In'own, about a line long, when dry, curved ajid looselv contorted. Not uncommon. In its leaves this moss comes nearest the South American C. Bichardi, C. Mull, and C. parasiticiim, Sw., and differs from the African C. Afzelii, Sw. in the absence of the marginal vitta. Calymperes pallidum, 3Iitt. Cmdis brevis subsimplex. Folia a basi parum latiore erecta cellulis hyalinis areolata margine superiore crenulata exinde patentia oblongo-lanceolata, cellulis rotundatis vix obscuris papillis brevibus aspera, nervo crassiusculo concolori intus MUSCI.— WILLIAM MITTEN. 389 extusque scabro percursa, vitta angusta sub-intra-marginali lutescente circumdata. Folia aiiomala angustiora ex apice ucrvi glomerida corpuscularum vii-idiimi fusi- formium feroutia. Stems about two lines high. Leaves a line long, pale glaucous green when dry, loosely incurved and contorted. Frequent. In the structure of the leaf nearest to C. varmm from Ceylon and Moulmein, in the glaucous appearance of the upper cells like C Dozyanum, Mitt. ( C. Moluc- cense, Bryol. Jav. t. 37 ; but not of Schwsegrichen), so widely spread in the Indian Archipelago and Pacific Islands. Weisia (Euweisia) controversa, Heche. Fund. Muse. Frond, iii. 12 t. 5. Everywhere ou the banks of streams. Weisia (Hyophila) Roscheri, Zot'enfs. Found with other species of Weisia. Small barren stems, which agree pretty well with others from Zanzibar. The species is very similar to W. {S.) involuta, so common in India. Weisia (Tortularia) incerta, Mitt. Ccmlis brevis ramosus. Folia patentia patenti-recui'vaque lineaU ligulata apice obtusa, nervo crasso rufescente in mucronem brevem excurrente, marginibus superi- oribus inflexis integen-imis, cellulis superioribus rotundatis obscuris basalibus paucis oblongis parvis rectangulatis hyalinis. Stem a quarter of an inch high. Leaves one line long, dark green, when old of a rusty brown colour. Common on banks of streams. This appears to be a moss very nearly resembling W. navicularis, Mitt. Muse. Austr. Amer. 139, but it is a sj)ecies of which too little is known to render its position here well ascertained. Tortula (Planbelia) mutica, Mitt. HumiUs, subsimplex. Folia erecto-patentia ovato-oblonga obtusissima, nervo vaUdo concolori sub apice abrupto dorso supcrne aspero inferne loevi carinata, inte- gerrima, cellulis superioribus parvis fere obscuris inferioribus basin versus rotundo- quadratis parvis pellucidis. Stems two lines high, red. Leaves one third of a line long, glaucous green, when dry contorted. Along with species of Weisia, but more scarce. Very near to T. orientalis, Willd., and to Fottia {Hyophila) paj)illine7'vis, Lorentz, from Zanzibar, which is doubtless a species of this group, being very similar to T. orientalis in size, habit, and in the structvu^e of its leaves. 300 BOTANY OF RODRIGUEZ. Leiomitriuiu, genus nov. Cdiilis ropcns ramis fortilibus crectis, Theca cxscrta. Fcristoinium duplex extcmuui octodeutatum (dentibus bigemiuatis) iuternum oetociliatum. Caliiplra cain]vuiulata plicis nullis. L. plicatimi, Bcaxc. Frodr. 81; Sdnocccj)'. 1. 52 (OfiholiicUuui). Not uucommon. This eurious aud interesting species has been well ligurcd and described by Schwa^grichen, its habit is that of a Macromitrium, its leaves like tliosc of the species of Ziigodou which are allied to Z. viridissimus, and its capsule and peristome as in V^lota ; there is, however, no species of Orthotrichiim or TJlola to which it has any resemblance, except in its capsule. In the non-plicate calyptra this genus agrees witb Schlotheimia, without, however, having any other near aflinity or resemblance. Macromitrium astroideum, Milt. C«»//5repens prostratus, foliis e basi patente divergentibus apice iucurvis hastato- lanceolatis acuminatis apice angustis ccllulis iufernc oblongis nounullis curvulis supcrioribus rolimdatis obscuris areolata, rami dense disjoositi breves erecti dcnsifolii, foliis lincalibus o basi suboblonga erccta cellulis pcUucidis angustis oblongis panels curvatis jjapillis elevatis intus asperis, exinde patentibus apice obtuse acutis incurvis cellulis rottindis densis obsciuis vix papillosis areolatis costa superne concolori inf erne flavcscente percursa carinatis. Stems creeping in a centrifugal manner with closely set branches about a line high. Leaves one line long, when dry ciuied and contorted, forming globular tufts. Bai'c. Only at the top of Oyster Eivcr valley, trailing over boulders. This appears to be near M. fimhriatum, Schw., bu,t has narrower leaves. Macromitrium aciculare, Brid. ; Schwcegr. t. 111. Occurs more commonly than the foregoing. Stems all barren, bu.t agree with Mam'itian specimens. It appears to be one of the most abundant amongst a group of very closely allied species found in Mauritius, Bai-tramia (PMlonotis) pungens, llitt. Dioica. Cutdis erectus ascendensve late cajspitosus gracilis rufus infra jieri- chetium ramis pleriimque ad unam lateram decurvis ramosus. Folia laxe imbricata erecto-patentia lanceolata sensim angustata acutissima costa angusta dorso denti- culata percursa margine parum recurva serrulata e cellulis oblongis laxis pcllucide areolata, perichsetialia e basi dilatata ovata subulata nervis laevioribus. Theca in pedunculo rvibro sjepe sinuoso parva globoso-pyriformis inclinata sicca j^licata oper- ciilo convexo breviter mamillata peristomio normali. Flos mascidus gemmiformis rufus foliis perigonialibus e basi dilatata excavata vaginante sensim angustatis eloneatis remotius scrriilatis. MUSC I.— WILLIAM MITTEN, 393 SfcDis half an iuch or more Ligb, with the leaves half a line wide. Leaves when cliy appressed, rather soft in substance, the upper of a pale sulj-glaucous green, the lower pellucid, with the nerves and margin pale brown. Seta three-quarters of an inch high, shining. Capsule when dry and deoperculate curved and plicate, of a red brown colour. On moist rocks near the sources of all streams. Very nearly rescmliling the Ceylon 13. imb/'icatnla, Mitt. Muse. Ind. Or., but with narrower and more lax areolation. The male flower is like that of £. rnfijlora, Hornsch. in Mart. Flor. Brasil. Plate 37 A. 1. Plants of the natm'al size. 2. Leaves. 3. Portion of peri- chsetium. 4. Capsule mature with operculum. 5. Older collapsed capsule. 6. Male flower : all magnified. Bartramia (Pliilonotis) flaccidifolia, Mitt. Gracilis, subsimplex. Folia caulina laxe inserta apicalia subsecunda planiuscula late lanceolata acuta nervo dorso Irevi in acumen pungentem dissoluto margine pamm recurva serrulata, cellulis oblougis laxis pellucidis areolata. Polia in ramu- lorum inferiorum uonnulla parva obtusa. Found with the foregoing. Larger than B. pnngens, with leaves twice as wide, but in their substance less different, although more flaccid. It is possible that this may be only an altered state of B. ptmgens by some peculiarity of locality, but its appearance is diffe- rent. Brachymenium pulchrum, Hook. Bot. 3Iiscell. i. t. 38. On the stems of many trees. A small quantity of plants which appear to be a young state of this or some nearly allied species. When dry the foliage is much twisted, but the form of the leaf is as in the S. African moss. Brymn erythrocaulon, Schiocegr. t. Ixxx. Widely spread on the island. The specimens agree well with barren examples from Mauritius and Natal. The species appears to have softer and narrower leaves than is usual in B. leiitothecinm, Tayl., which in other respects is very closely allied. Bryum megalacrion, Schwcegr. t. 71. Common on trees. Specimens barren. Bryum pachypoma, 3Iont. in Ann. So. Nat. 3d ser. iv. 104. Eare. Only in a few places on rotten wood. Stems mostly barren, all without fruit. They agree with specimens from Ceylon and Java. 392 BOTANY OF KODKICUEZ. Rhacopilum Africanum, ^Lill. in Joiini. Linn. Sue. vii. I(i2. rouml iiUorniiiiiclcd wilh species ol" Eetropotlieciuiu on luoist rocks in the bed of streams. Specimens barren, but they a))pear to lie in no way (liU'i'icnt IVoni lliose from the western coast of tropical Africa. The species dill'ers i'rom 11. mno'onaium, Beauv. Prodr. GG, in the cells of its leaves being larger and easily defined with a lens that leaves thosf^ of 7i'. muo'ondfiini obscure. CaUicostella Iseviusciila, Mill. Synoica. Caiilis procumbens. Folia comjiressa, media late ovalij oblonga apicc obtusa late sub-apiculata, intermedia longiora apice truncato obtusa, lateralia apice late acuta, omnia costis divergentibus sub apice al)ruptis dorso parum crosis mar- ginibus versus apicem crenulatis cellulis parvis I'otiuulis kcvil)us areolata ; perichre- tialia erecta vaginulam vix togentia. Tlicca in pedunculo rubro lu3vi borizontalis dccurvave ovalis operculo longc subulato calyptra ad medium tliecai desccndente apice scabra, pcristomii deutibus crassiusculis solidis cajterum normalibus. Abundant. One of a large genus, most of the species of which to the naked eye present scarcely any obvious differences ; on a close examination, however, of the arcolation of the leaves considerable ditferences are apparent in the form of the cells, in the presence or absence of the papilla?, also in the position of the papilloc on the cells ; due consideration being given to this, C. Iceviuscula is allied in structure to C. papillata, Mont., so widely distributed in the Pacific islands, to C. Africana, Mitt. in Linn. Trans, xxiii. 63. t. 5, f. ix., from tropical Africa, to C. depressa, Swartz, from the West Indies, and to C Martiana and C. Merkeli and C. viicrocarpa, Hornsch., all from S. America, but appears to differ from all in its shorter leaves which are not evidently papillose. Plate 37 B. — 1, stem of the natural size ; 2, a leaf from the middle series of the upper side of the stem ; 3, a leaf of the intermediate series on the same side ; 4, a lateral leaf ; 5, perichsetium ; 6, capsvile with calyptra : all magnified. Meteorium involutifolium, Mitt, in Journ. Linn. Soc. vii. 156. IS'ot common. At the top of Oyster Ptiver Valley. Specimens showing the moss in a young or incomplete state, with simjile stems about an inch long, arising from a creeping rhizoma-like stem denuded of leaves. On the ascending branches the leaves are imbricated, and at their points are con- gested into a cuspidate bud, they are inserted in five twisted rows. This species is found also in Madagascar as well as on the Western Coast. Fertile stems are frond- like, and show the species to be nearly related to M. Mauineiise, SuU., and Crijpto- theco cochlea rif olio, Hornsch. Pterogonium curvifoUum, Mitt. Monoicum. Caulis primarius repens radicellis purpureis apice ascendens arcu- atus. Eami laxe dispositi ascendentes simplici vel interdum ramosi. Folia ramea MUSCI. — WILLIAM MITTEN. 393 pateutiu iucm-va parum comprcssa sicca torcti imbvicata, late ovato acuminata con- cava mai-gine recto ob prominentia cellulamm parictum minute crcuulata nervo fur- cato brevi subobsoleto, cellulis superioribus anguste elongate oblongis apicibus in papillas clcTatis basin versus ad lateras densis minutis transverse oblongo-rotundis obscurioribus, pcricbiictialia duplo majora patentia ovato-lanceolata Iseviora. Flos masculus cxilissimus. Stems about an inch long. Branches three to four lines long, when wet straight, when dry curved, with the leaves one fourth of a line Avide. Found sparingly on trees in Oyster Bay Valley. This pretty moss forms extensive rather loosely interwoven patches. In minia- ture it has the habit of P. gracUe ; it differs from Clasmaioclon, Ilahrodon, and Anisodon in the almost obsolete nerve and papillose leaves. Plate 37 C. — 1, stem of the natural size ; 2, a portion -vvif h leaves ; 3 and 4, leaves detached : all magnified. Anomodon exilis, Mitt, in Journ. Linn. Soc. xiii. 309. Rare. A single stem amongst Pterogoninm ciirvifolium agrees well with the South African specimen. The species is allied to A. triste, Cesati, but has the foliage of its branches compressed. Neckera Lepineana, Ilont. in Ann. Sc. Nat. 3'' ser. x. 107. Only in Valley Riviere des Huitres, trailing over the boulders. Agrees with specimens from Madagascar and from Johanna, whei-c it was gathered by Speke. The species is found in Ceylon, the Indian Archipelago, ai^d the Pacific islands. Its fruit is rare. Sematophyllum fulvifolium, Jlitt. Monoicum. Caiilis decumbens ramosus. Folia compressa interdum sccuuda laxe imbricata ovata ovaliaque parum acuminata acuta concava, margine late subplana integerrima breviter o])soleteque binervia cellulis oblongis distinctis obscuriusculis alariljus utrinque tribus majoribus peUucidis areolata, pcricha3tialia erecta longiora acuminata. Tlieca in pedunculo gracili rubro ovalis suberecta inclinatave basi callosa, operculo convexo tenui curvirostro, peristomii dentibus luteis iuferne latis crassis. Stems an inch long, with the leaves scarcely a line wide. Leaves green, almost glossy. Seta foiu lines long. Capsule greenish brown, after the fall of the red operculum, much contracted below the mouth and urceolatc, at its base where con- tracted into the seta is a callous ring. Not uncommon. The specimens appear to be specifically identical with a moss gathered in Mauritius by Telfair, Bewsher, and Pike, in Avhich the leaves are usually fulvous. S. brachycarpum, Hampe Ic. xi,, has narrower leaves with a different areolation. 3d 391 IJOTANY OF KODUIGUEZ. In the oLscurc cells of the loaf S.fiih'ifoliitm np;rces with the South American S. KcgcJianum, C. Mullcr. Plate 37 B. — 1, stem of the natural size ; 2, leaves from the same ; 3, pcrichoo- tium and male flower ; -t, capsule ^A•ith operculum : nil magnified. Seuiatophylhnu iucnrvnm, Mitt. Monoicura. Caiilis repeus pinnatus ramis assuvgcntibus incurvis. Folia cauliua ovato-lanceolata acuminata coneava nervis binis brcvibus uno A'alidiorc tcI obsoletis, niavg-inc obsolete serrulata ramea vindique patcntia sicca sursum secimda elliptico-lanccolata coneava margine rccurva iuteg-errima enervia c ccllulis clongatis augustis basi ad angulos rotundatis incrassatis alaribus oblongis hyalinis : pericha}- tialia crecta caulinis similia duplo longiora. Theca in pcdunculo brcviusculo rubro ovalis liorizontalis pendulavc, basi apophysc calloso, opcrculo convexo longc oblique rostrato. The prostrate stems are about an inch long, tlie branches from two to three lines long, and with the leaves about two-thii*ds of a line Avide. The leaves glossy green and straw coloured. Seta three lines long. Common. This differs greatly in the substance of its leaves from S. fiilvifolinm, and agrees more nearly with S. hraeln/cavpum, From which it may be distinguished by the form of its leaves, which are all du'ccted upwards on the incmwed branches. Plate 38 A. — 1, stem of the natural size ; 2, leaves detached ; 3, perichsetium and male flower ; 1 and 5, capsules, one with calyptra. Ectropotliecium scaturagineum, Br id. ii. 418 (Hypnum). Perhaps the commonest species on the island. Specimens all without fruit, vaiying greatly in appearance, but not much in the form of the leaves. Ectropotheciiun subenerve, Mitt. Cctulis procumbens pinnatus ramis assm'gentibus in csespitem intertextus. Folia cauliua compressa subsecunda ovata acuminata integerrima, nervis obsoletis, cavius- cula ad margines vix implana, e ccllulis clongatis angustis firmis areolata. Branches from three to six lines long, with the leaves three-fourths of a line wide. Along with the foregoing. Ectropotheciuin doliare. Mitt. Monoicum. Caulis procumbens intricatus foliis a symmetricis ovato-lanceolatis. Mami ascendentes in csespitem laxum congesti foliis distichaceis compressis apicibus decui*vis subfalcatis fulcatisve anguste lanceolatis omnibus margine apicem versus minute seiTulatis, nervis binis brevibus subobsoletis, ccllulis clongatis augustis Ifcvibus areolatis; perichpetialibus longioribus subulato-angustatis. Theca in pedunculo gracili breviter ovalis pendula operculo convexo brevirostro, peristomio e dentibus lIEPATICiE.— WILLIAJM :MITTEN. 395 latis crassis interno pi-ocessibiis ct ciliis in uno coalitis in membranam alte exsertam insidentibus. Branches mtk the leaves about two-thirds of a line wide. Leaces pale green, glossy. Seta half an inch long. Capsule small and short. Very common on clay rocks on the banks of streams. Forming extensive soft loose patches, in which the ramification, so frequently elegantly pinnate in allied species, is in the present scarcely perceptible. Plate 38 B. — 1, a stem of the natural size ; 2, a leaf from the middle series on the upper side of the stem ; 3, a leaf from the intermediate series of the same side ; 4, a lateral leaf : 5, perichcetium and male flower ; G, capsule. Ectropothecium subulosum. Mitt. Caul is procumbens radicans raniis brevibus pinnatim ramosus. Folia subcom- pressa lateralia patentia apicaHa subfalcata, omnia lanceolato-subulata, caulina integerrima, ramea apicem versus sub-scrrulata, costis birds brevibus iuconspicuis, subobsoletis, cellulis angustis firmis areolata. Branches about two lines long ascending, with the leaves two-thirds of a line wide. Leaves green, shining, scarcely altering when dry. Frequent along with the preceding. This moss nearly resembles a Mauritian species in its narrow leaves, but seems a little larger and of a firmer substance. Hypnuin (Rhynchostegium) pectinatum, Milt. Caulis procumbens ramis assurgentibus. Folia laxe iuserta patula compressa, sicca pectinatim patula, ovata acuta vel breviter acuminata argute serrulata, costa tenui ultra medium evanida, cellulis ubique elongatis teneris mollibus areo- lata. Stems with branches an inch long, slender, with leaves one line long, subpellucid, thin and glossy. Not common : on diy places. Nearly resembling a species from Natal, and also the S. African S. rhaphido- rhynchum, C. Muller, but not exactly agreeing with any, Fissidens procumbens. Mitt. Caulis elongatus inferne procumbens crassiusculus simplex. Folia alterna patentia lineari-lanceolata acuta, costa vix pcllucida in apice evanida, lamina vera ultra mediiim producta acuta, dorsali angusta baud decmi-ente, ubique integerrima e cellulis minutis obscuris densis areolata. Stems an inch or more long with the leaves a line wide. Occurs in tufts occasionally. A species to all appearance different from any of the large group to which it belongs by its immarginate obscure leaves, Avhich are not crenulatc from the promi- 3 D 2 39(5 r.or.wv of kohuiguez. ncnt marginal cells, as is so frcqiicutly observable among the species destitute of llio hyaline limb. » • Fissideus brevifrons, Miff. C'ni;u;r !•:/.. apitr rtrtanu'uJato atl inediuin in:iri;iuis A'ontralis i)rodiicta v coUulis idlundo-hoxa- gonis pilhuidis toqualibus arcolat-ii. Amjj/iigasti'ia votmuhxia caulcm triplo supev- autia, apifo fissa, rariiis indivisa. Branches hall' a line wido. Foliage wliitisli iirecn. Not common on live stems. This appears to be not dilTercnt from a specimen, also barren, from Mauritius. The length the lobule continued along the ventral margin renders this immediately distinguishable from the numerous species resembling L. serpi/llifolia. Lejeuuia aiigiilifolia, Jli/l. Coiilia repens raniosus. Folia alterna pat(Miti divergentia sul)ovata a]iice angu- lata subacuta integerrima, lobulo sul)rotun(lo saccato, cellulis prominulis pa|)uI()so apiee acuto, oelluUv Iblioruni rotundata,> pellucidyc ltcvi;c parietibiis erassis obscuris viridibus. Anqjh'igastria miuiita bifida caulem vix latiora. Folia involucralia acutiora lobulo ovalo acuto. Amphigastrium ovale profundc fissum laciniis acutis. Stems with the leaves about one third of a hue wide. Leaves all angled at the apices, and occasionally with one or more angles on the dorsal margin. On tree stems. Not uncommon. A small species, less than L serpyllifolia, of a pale green colour. Allied to L. Nilgiriana, ^Nlout., but more tender and with a more lax areolation. Lejeunia miuutissima, Sm. Gathered sparingly on stems of trees. Lejeiuiia subciliata, 3Iitt. Caulis repens appressus ramosus. Folia imbricata, dorso planiuscula, late ovata apicc rotundata, margine dorsali celhilis elongatis hyalinis tcnerrimis scarioso mar- ffinata, cellulis reliquis rotundo-bexagonis parietibus crassiusculis arcolata, lobulo tumido ovale apice apiculo spiniformi terminato. AmpJiigastria nulla. Nearly related to L. planissiraa, Mitt., from Ceylon, which has the same kind of scarious margin, but is much more closely adherent to bark. Stem with the leaves scarcely more than half a line wide. Leaves whitish green, closely appressed to the bark on which the specimens were growing. The scarious margin is composed of a single row of elongate cells placed side by side, and an irregularity in their length causes the edge to be erose ; this hyaline limb of diverse cells is here and there broken off, leaving the margin of the leaf itself entire. Eare on stems. Plate 40 B. — 1. Stem of the natural size. 2. A portion of the same as seen on the ventral side. 3. Portion of the edge of the leaf on the dorsal edge : all mag- nified. FruUania obscurifolia, 3rdt. Cdulis procumbeus repensve ramis vagis subpinnatim ramosus. Folia imbricata orbiculata e cellulis angulato rotundis parietibus angustis teneris sed chlorophylloso IIEPATICiE.— WILLIAM MITTEN. 401 obsciiris, lobulo ovali dimidiuiu iblii loiigitudinis toycnte coiuprosso apice paululuma cauli clivergente interdum explanato lacinia ininuta interjecta. Ampkigastria caulem diiplo latiora rotxiuclata apice breviter bifida laciniis extus unideiitatis. Folia invo- lucralia ovalia lobiilo lato ovato-lanceolato acuto infenie lacinia dentiformi instructa ampbigastrioqiie profundo bifido laciniis aciitis integerrimis. Ferianthium obo- vatiuu obtusum, apiculo brevi, compressnm, doi'so uni ventre bieavinatnm. Common on trees. Similar to F. trinervis, Lelmi. et Liudenb., from S. Africa, but differing in its obscure areolation and the longer lobule of its leaves. . Plate 40 C. — 1. Plant of natural size. 2. A portion of the stem with leaves and stipules on the venti-al side. 3. Perianth with involucral leaves and stipule. 4. Transverse section of perianth : all magnified. Finillania squaiTosa, Kees ah E. ^'ery common on trees. Fnillania apicalis, Mitt. C'aulis gracilis pinnatim ramosus. Folia dorso convexa divergentia orbiculata integerrima cellulis parvis rotundatis distinctis sed utriculo collapso obscuratis areolata, lobvdo ovaH-cylindraceo erecto cauli parallello lacinia parva interjecta, in ramvilorum apicibus lobulo apice ad caulem incumbente, amphigastria obovata bifida laciniis extus unidentatis, folia involucralia acnta lobulo ovato acutiore amphigastrioque margiuibus dentatis. Ferianthiiim oblongo-obovatum, dorso piano ventre unicarinato. Common on trees, specially on Pandanus. In size and colour, as well as in the form and imbrication of its leaves, nearly resembling F. Cai)ensis, Gottsche Syn. Hep. 449, but with more the habit of F. gracilis, and the leaves of the involucre dentate. Plate 40 D. — 1. Stems of the natural size. 2. Portion with leaves and stipules as seen from the ventral side. 3. Perianth with involucral leaves and stipule : all magnified. Anthoceros Isevis, Linn. Common on moist clay rocks on the banks of streams. Anthoceros fucifoiinis, 3Iont. in Ann. Sc. Nat. 2d ser. xx. 296. Growing along with the foregoing. This appears to be the same as the specimen from Bourbon. The spores are opaque and echinulate. CHARACEiE. Chara Commersoni, A. Braiin. In manv streams. 3 £ ■1.02 LICHENES. By Ecv. J. :\1. CuoHBiK, F.L.S. Family COLLEMACEI. Tribe Collemei. Collema, sp. ? Several specimens of a Collema occur "whicli externally bears some resemblance to youno: states of C.Jlacckhim, but sterile and Milb the Hialhis discoloured so as to be quite indeterminable. On trunks of trees sparingly. Leptogiiuu tremelloides, Ach, SufTieiently typical, and also occasionally somewhat atypical. Common, on the trunks of trees ; fertile. Family LICHENACEI. Tribe Cladoniei. Cladouia Ealfouri, Cromh. Tballus minor, basi squamulosus, squamulis parvis, albo-glau.eesccntibus, sv\b- fii'mulis, crenatis, podetiis simplicibus, brevissimis, apice raro divisis, albo-pulvc- rulentis (vol pulvere subvirescenti-albo obductis), K — , contra squamis K+ (flaventibus) ; apothecia fusca (ut Tidetur), rite evoluta non visa. The general aspect of this plant is suificiently peculiar to warrant its being regarded as a distinct species, notwithstanding that the specimens seen arc probably not fully developed as to thallus. Only a few faint traces of young apothecia arc visible, which would refer it to the Phseocarpte, where it seems to have its affinity in the section of C. pi/xidata. On dead stumps of trees ; not common. Tribe Koccellei. Roccella Montagnei, J3el., f. teretior, Cromh. This form is distinguished by the narrow rounded thallus, which gives it more the aspect of JR. lihyeopsis than of B. fuciformis, which typical specimens more closely resemble. On rocks ; infertile. Various and evidently young states, which seem referable to this form, are also present. LICIIENES.— REV. J. M. CEOMBIE. 403 Tribe Usneei. Usuea florida (X.), Ach. On the branches of trees ; sparingly and infertile. Usnea das3rpogoides, Nyl. " Subsimilis Z^snecc dasypogce, sed thallo subflavido, loevi, glabro, nee nisi in ramiUis vcl fibrillis s?epius Icviter sparsim setuloso (quibus setulis ctiam distinguitur ab Z7. trichodea, Ach.). Apothecia ignota." — Nyl. in Utt. These specimens vary considerably in length, the largest being upwards of a foot long. Unfortunately none of them are fertUe, though scattered " cephalodia " not unfi'equently occiu". Abimdant on the branches of trees in the higher parts of the island. Tribe Ramalinei. Ramalina gracilenta, Ach. On rocks and trees in the higher parts of the island ; fertile. Ramalina gracUenta, Ach., f. nodulosa, Cromh. Thallus minute, densely branched and fastigiate, forming more or less rounded nodules. This is evidently a yoimg or rather stunted condition of the plant, presenting a somewhat singular appearance. In one specimen a single minute apothecium was visible. Rare ; on rocks with the type. Ramalina farinacea, Ach., Nyl. On the branchlets of trees ; not common. In higher parts of island. Also found in Mam'itius. Ramalina farinacea, Ach., Nyl., f. pendula (Schrad.). On the branchlets of trees, with the type. Ramahna subfraxinea, Nyl. Common on the branches of trees in the higher parts, along with Usnea dasy- pogoides ; fertile. Ramalina canaliculata, Tayl. On the branches of trees ; fertile. Ramalina canahculata, Tayl., f. brevior, Cromh. Thallus shorter, ^-1 inch ; lacinia; rather broader. On the branches of trees ; fertile, but like the type. Ramalina intermedia, Del., Nyl. On rocks : not uncommon, but sterile. 3e 2 iOl BOTANY OF KODKIGUEZ. Tribe Parmehei. Parmelia latissima, Del. Vi'rv I'oiuinou on tiros and rocks : sterile. The thallus is often discoloured, probably from the action of salt water. Parmelia subcouspersa, Nyl. Common on rocks ; fertile. Parmelia atriclioides, Nyl. Ou rocks ; sparingly and infertile. Tribe Stictei. Stictiua dissimilis, Xi/I. On trees; sparingly and infertile. Stictina agyracea, f. insidiata, Njil. Thallus somewhat densely covered with isidia. On trees ; sparingly and infertile. Sticta aiirata, Ach. On trunks of trees and dry places, with Parmelia latissima; infertile. Tribe Physciei. Physcia flavicans, f. crocea [Ach.). Rare ; on the branches of trees. Only on Mount Piton, near south end ; sterile. Physcia speciosa {Wtilf.). Abundant, on trees and on boulders ; fertile. Physcia segialita {Ach. Syn. p. 179, sub Lecanora). Ou rocks ; fertile, but very sj)aringly gathered, only a single small specimen being present. Physcia picta {Sw.). On bark of Pandanus, not uncommon. ■ Tribe Pyxinei. Pjrxine petricola, Nyl. " Tliallus glaucescens, laciniis subintricato-contiguis, adnatis (latit. 1 millim. vel angustioribus), late diffusus ; apothecia nigra, planiuscula, lecideina, marginata (latit. 1 millim. vel minora), vel margine extus epithallino obducta, intus obscura ; sporce 8nse, nigrescentes, oblongae, biloculares, longit. 0'011-0"015 millim., crassit. 0-0015 millim. ; epithecium cseruleo-nigricans ; paraphyses graciles ; hypothecium sub- incolor. lodo gelatina liymenialis bene cserulescens. " Species certe bene distincta notis allatis. Thallus K=."—Nyl. in lift. On rocks ; sparingly. Tribe Lecanorei. Pannaria rubiginosa, f. cinerascens, Nyl. Thallus greyish ; otherwise similar to the type. LICHENES.— REV. J. M. CROMBIE. 405 On rocks. Tliis species occurs only in the liiglier parts of island. Only two apothocia visible on the specimen. Two other sterile specimens may belong also to this form. Pamiaria rubiginosa, var. dispartita, Ni/l. " Forsan propria species, thallo tenuius cliviso, effuso. Apothecia non visa." — Ni/l. in lift. On rotten stumps on the ground ; sparingly. Pannaria luridiUa, Ni/l. " Thallus luridus, squamulosus, squamiilis creuatis vel crenato-lobatulis, sub- intricatis ; apothecia testaceo-rufa vel tcstaceo-nifescentia (latit. circiter 1 millim. vel minora), bene lecanorina, margine thallino crassiusculo crenato ; sporse ellip- soideae, longit. 0-016-0'025 millim., crassit. 0009-0-011 millim. lodo gelatina hymenialis cternlescens. " Affinis P. immixtie, Nyl., sed color thalli luridus et hypotliallus niger ullus vix visibilis etc." — Nyl. in Utt. On the ground ; rare. Coccocarpia molybdaea, Fers. Not uncommon on bark of trees ; infertile. Heppia Rodriguezi, Cromb. Thallus cervino-liu'idus, adnato-squamulosus, squamulis planiusculis vel alibi convexiusculis, subanguloso-rotundatis (latit. 1-2 millim. vel saepius minoribus, crassit. circiter 0*25 millim.), contiguis aut subdispersis ; apothecia rubello-carnea, concaviuscula, innata (latit. 0"2-0-3 millim.); thecse polysporae; sporge breviter ellipsoideae vel oblongo-ellipsoideae, longit. 0*004-0"006 millim., crassit. 0'0025-0'0035 millim., paraplayses gracilescentes. lodo gelatina hymenialis cserulescens, dein thecae prsesertim violacee rubescentes. This is a fine species, which is at once distinguished by the colour of the apothecia. The syngonimia are usually sufficiently small, and then contain but few gonimia. With the exception of the colour of the apothecia, it presents almost the appearance oi Lecauora cervina, Pers. {squamulosa , Schrad.). On rocks ; A'ery rare in the Cascade Valley ; only a few apothecia visible. Lecanora murorum, var. lobulata, (Snirf.). As the thallu^s is not sufficiently developed in the specimen seen, the plant is somewhat uncertain, although the apothecia and spores agree with this variety. Spores 0-010-0-013 millim. long, 0-006-0-007 millim. thick. On rocks ; not common. Lecanora cinnabarina, (Sic). Frequent on rocks ; in the specimens seen associated with other lichens, and not very characteristic. lOG r.OTANY OF KODHIOUEZ. Lecanora aurautiaca, {Lightft.). On the bark of trees ; siiflu'ienfly iypieal, but very sparingly gallicrctl. Spores 0-011-()-01G millim. long;. O-OOO-O-OIO millini. tliielc. Lecauora aurautiaca, var. isidiosella. A somewhat peeuliar variety, ^liieh is well tlistiuguislied by the thallus being more or less covered "with subconcolorous isidia. The apothccia arc also smaller and only sparingly in-esent in the few fragments seen. On the bark of trees, associated with other lichens. Lecauora am-autiella, Nyl. " Siibsiniilis L. aura lit iactc ct cn/flirelhc, Ach., scd apothcciis minoribus ct iunatis, sporis minoribus (iongit. 0-010-0-012 millim., crassit. O'OOU-O'OOS millim.)." — Ki/l. ill litt. On rocks in the Cascade Valley, associated with other lichens. Lecauora leucoxautha, {Sj)r.). On the bark of Mathiirina pendulijlora ; found in fine fruit. Lecauora glaucoftiscula, Nyl. " Thallus glauco-ciuevascens (vel humidus cincrco-vircscens), tenuis, continnus, passim subrimosus ; apothccia fusca, su.binuata, plana (latit. circiter O'Oo millim.), margine thallino tenui vel parum distincto cincta ; sporaj 8na3, incolorcs, placodince, Iongit. 0009-0-011 millim., crassit. 0006-0-007 millim. ; epithecium (in lamina tenui) lutescenti-inspersum (K purj)u.rascens) ; hypothecium incolor. lodo gelatina hymenialis intensive cDcrulesccns. NuUi cognitse afi6.nis, vcrgcns autem versus L. enccphaJarti (Kphb)." — Nyl. in lilt. On rocks ; sparingly. Lecauora glaucofiiscula, f . biatoroidea. This differs from the type only in the apothccia having a more distinctly biatorine appearance. On rocks ; like the type, very sparingly, Lscanora apostatica, Nyl. " Tliallus albus vel albidus, tenuis, areolato-rimosus, hypothallo cajrulescenti- nigricantc passim visibili ; apothecia nigra, lecideina, plana, marginata (latit. 0"05-0'07 millim.), intus albida ; sporre 8nse, incolores, placodinaj, Iongit. O'Oll-O'OlS millim., crassit. 0'005-0'008 millim. ; paraphyses mediocrcs ; epithecium et peri- thecium cserulescenti-nigrescentia; hypothecium incolor. lodo gelatina hymenialis intensive ccerulescens. " Est species c sth-pe Lecanorce diphyodis, proxima L. Kurzii (Ki^hb.). Affinitas videtur inter Lecanoros, ctiamsi apothecia lecideina. Thallus K flavens. Sper- mogonia arthrosterigmatibus munita ; spermatia minutula, oblongo-cyliudrica, Iongit. 0-002 millim., crassit. 0-0006 millim."— A^y/. in litt. Kot uncommon on cindery rocks. LICHEMES. — REV. J. .AI. CROMBIE. 407 Lecaiiora obliquaus, JS^i/l. " Diffcrt thallo ciucrasceutc tcauissimo continuo. Occurrit quoqiic cpitliccio albo-suffuso. Forsan propria species." — ]Vi/l. in lUf. Very common on rocks, giving them a white appearance. Lecauora carneofusca, JS'yl. " Thallus glauco-cinerascens vel glaiico-albidus, tennis, intequalis, areolato- rimulosus; apothecia carneo-fusca vel rufescenti-fusca, Iccanorina (latit. 1 millim. vel minora), marginc thallino iutegro cincta ; sporte 8na?, incolores, placodiniic, longit. 0'011-0"018 millim., crassit. 0'007-0"009 millim. ; paraphyscs fere mediocics ; epithecium sordide sublutescens. lodo gelatina liymenialis intensive ccerulescens. '•' Species accedens ad Lecanoram cmnptidiam, Tnck., a qua thallo areolato rimnloso et sporis nonnihil majoribus prassertim differt." — Nyl. in litt. On rocks. Common. Lecanora subfosca, f. pumicicola, Nijl. " Differt a f. campesM, Schger. prtesertim thallo subdisperso, c granulis sat teunibus constans vel variantibiis convcxinscnlis, ambitu subcrenatis (latit. 1 millim. vel miuoribus), passim contignis. Apothecia fusca (latit. 1 millim. vel minora) ; margine thallino integro aut obsolete subcrenulato ; sporse longit. 0-011-0"012 millim., crassit. 0'006-0-007 millim. ; paraphyses tenues, molles ; epithecium rufes- centi-fuscum. lodo gelatina liymenialis cseriilescens, dein violaceo tincta. — Xyl. in litt. This very distinct form may easily be recognised by the above characters of the thallus, &c. Common on cindery rocks. Lecanora subflavicans, Nyl. " Thallus albidus vel albido-flavescens, tenuis, granuloso-insequalis (K+, flavens) ; apothecia testaceo-lutescentia vel testaceo-pallida (latit. (0-5-0-8 millim.), margine crassiusculo-ruguloso-ingequali vel demum integro cincta ; sporce Suae, incolores, eUipsoidte, longit. 0-010-0-012 millim., crassit. 0-006-0-008 millim. ; para- physes gracilesccntes ; epithecium iuspersum. lodo gelatina hymenialis cteridcscens, dein thecal violaceo tinctse. "Est species quasi intermedia inter Lecanoram subgramilatam et L. albellam." Nyl. in litt. On the bark of trees ; common. Lecanora achroa, Nyl, "Thallus glaucescens, tenuis, rugulosus, demum rimosus, nigricanti-limitatus (K flavens) ; apothecia lutescenti-paihda, plana (latit. OS-O'G miUim.), superficialia, marginc thallino subcrenato vel subintcgro cincta ; spornc Sna?, cllipsoidiv, longit. 0-010-0-016 millim., crassit. 0-006-0-008 millim. ; paraphyscs graciles ; epithecium 408 nor ANY of uouRiorKZ. !:>;ranuloso-inspprsimi. lodo ^clatiiir. liviucni.-ilis ranailesccns, dciii tlioca* obscure nonuihil violacoc tiiictas. " Ex at1iiiit:ito vidotuv Leca)>o)'(P chlanDhP. Spormntia viilc^o lovilor nrcviata, loiigit. (>018-U02;> iiiillim., rrassit. O'OOOS millim. (minus arcuata (|uaiu in compa- randa L. ochroeUo, qutv simul aliis not is dilfert." — Nyl. in Hit. On bark of trees, specially on Pandamis. Lecanora acliroella, yyL "Thalhis glaucesccns, tenuis, subltevigatus, rimulosus, uij^ricanti-limitatus (K flavescens) ; apotbecia lutescenti-pallida, plana, subinnata, minuta (latit. circitcr 0'3 nuUini.), margiue tballino tcnui integ'ro cincta ; spora) 8nse, cllipsoidcoo, longit. 0009-001 1 millim., crassit. 0006-0007 millim.; parapbysos graciles ; epitbccium vix vol parum inspcrsimi. lodo gelatina bymcnialis cairulesccns, deiu tbccae sub- violaceo tinetiu vel violaceo-fulvescentes. " Ex affiuitate videtiu- Lecanorce chlaronw. Spermatia arcuata, longit. 0015- 0-018 millim., crassit. 0-0005 millim."— J\^^/. /;/ ////. On \ho trunks of trees, associated Avitli tlic preceding. Lecanora perlutescens, Ni/l. "Tballus Havidus vel albido-flavescens, tenius, sublcprosus, determinatus ; apotbecia cerino-lutea vel luteo-pallescentia (latit. 0-4-0-6 millim.), plana, margine tballino tenui intcgro vel evanescente cincta ; sporrc Sniie, oblonga? vel ollipsoideoe, longit. 0-009-0013 millim. ; crassit. 00035 millim. ; parapbyses gracilescentes ; epitbecium inspersum. lodo gelatina bymenialis cserulescens, dein lutescens. " Afiinis Lecanorce lutescenti (DC) sed minor, magis determinata, sporaj minores, etc. Tballus et nonnibil apotbecia CaCl erytbrinose vel aurantiace tincta. Spermatia arcuata, longit. circiter 0016 millim., crassit. O'OOOo millim." — Nyl. in Utt. On tbe bark of trees. Lecanora conizaea {Ach.), Nyl. Entirely similar to om- European plant. Tballus CaCl — . Spores 0-010-0-014 millim. long, 0-003-0-004 millim. tbick. On decaved wood, not common. Lecanora conizopta, Nyl. " Tballus albidus, tenuis, opacus, passim rimulosus (K flavens) ; apotbecia carueo-lutea vel luteo-rubella, subinnata, plana (latit. 0-2-0-4 millim.), margine tballino depresso insequali cincta ; sporae Suee, ellipsoidete, simplices, longit. 0010- 0-012 milUm., crassit. 0-005-0-006 millim. : parapbyses gracilescentes. lodo gelatina bymenialis cserulescens. " Yidetur e stirpe Lecanorce conizcece, et notis datis facile agnoscitur." — Nyl. in lUt. Common on rocks, and very abundant in many places. LICIIENES.— REV. J. M. CKOMHIE. 409 Lecauora pai-ella * phloeoleuca, Ni/l. " Thallo teuulssiiuo albo pricsertim differens a typo, et facie apothecioriim sub- pertusariodea. Sporae longit. 0038-0050 millim., crassit. 0023-0025 inillim. Fortasse propria species."— ^yZ. in litt. On the bark of trees. Lecanora atra, f. succedanea, Nyl. " DilVort a L. atra ouropwa saltcMu spermatiis uonniliil brcAdoribus (longit. 0-007-0011 millim., crassit. fere O'OOl millim.)." On many rocks in the higher parts of the island ; more rarely on dead Avood and bark of trees. Lecanora punicea, Ach. On the bark of trees ; common, as it is in Mauritius. Pertusaria velata {Turn.) On the bark of trees, specially Mathurina ipenduliflora ; quite similar to our British plant. Pertusaria impallescens, Nyl. " Thallus albidus,opacus, continuus, tenuis (crassit. circiter 0'2 millim.), sublaevis, passim rimosus vel rimulosus (K — ) ; apothecia incoloria in protuberantiis thalhnis depresso-convexulis innata, 1-4 in quavis tali protuberantia, ostlolis nigris, puucti- formibus, vix j)rominulis ; sporae Snse, incolores, solitae generis, longit. 0-070-0-082 millim., crassit. 0-032-0-040 millim. lodo gelatina hymenialis intensive cinerulescens. " Notis datis satis distincta species, nee ulli cognitse affinis. Spermatia peculiaria, aciculari-fusiformia (vel fusiformia utroque apice longe aciculari-producto), longit. 0-007-0009 millim., crassitie medio vix O'OOl millim." — Nyl. in litt. On rocks. One of the commonest lichens on the island, giving a white appear- ance to the rocks. Tribe Lecideei. Lecidea achroopholis, Nyl. "Thallus albidus vel pallesce nti-alhidus, squamulosus, squamulis (latit. ch'citer 1-2 millim.) adnatis, plauis, contiguis vel subdispei"sis, ambitu sublobatulis vel interdum subcrenatis, subtus nigris; apothecia nigra, planiuscula, marginata (latit. 0"7-l'5 millim.), intus obscura ; spora3 Snse, incolores, oblongae, simplices (vel obsolete vel spurie 1-septatae), longit. 0-009-0"014 millim., crassit. 0'0035-0'0045 millim. ; paraphyses mcdiocres apice subpurpurascenti-fusco ; epithecium ethypothe- cium suhpurpurasceuti-infuscata (vel hoc parte supera rufescente). lodo gelatina hymenialis cserulescens, dein subfulvescens. " Species iusignis in stirpe Lecidece dec/jiientis, ut videtur. Thallus firmus K non reagens. Thalamium totixin dilute subpurpurascens et K simul cum cpithecio et perithecio et hypothecio inferiore pm'purascens. Spermogonia supra nigella, intus 3 F •110 BOTANY OF KODKIGIEZ. incoloria ; spcrmatia tcniiitci" bacillaria iiouiiihil vol ubsoletc cuvvula (ciuarc siibrocta apparent, licot vix iiuquam siut vcrc recta), longil. 0005-0'007 ruillim., crassit. 0-000() millim."— 37//. In litt. On rocks, usually oi" a vesicular eliaracter. Lecidea coccocai'pioides, Ni/l. " Thallus albido-glaucesceus, squamaccus, subadnatus, squamis (latit. 2-5 millim.) iuciso-lobatis, ineisuris clausis (ncc biantibus) ct uigricautibus, ambitu subcrcuulato, patjiua iufcra nigra ; apotliceia nigra, innata, plauiuscula, iinmarginata (latit. 1-2 niillini.), interdum 2-3 confluoutia, intus obscura ; spora) Siia^, cllipsoidctc vol obloagiv, l-septat;\> (scpto saltern spurio), longit. O'OUG-0"U12 millim., crassit. circitcr 0'0035 millim. ; parapbyscs mediocrcs ; cpitbeciimi ctcrulcscens ; bypotlic- cium obscure rufesceus. lodo gclatina bymenialis bene casrulcscens. '*' Species omnino pcculiaris ct proprite stirpis. Facie fere CoccocarpicB molyhdcecs, sed tballus subnitidiusculus, niliil superficie couceutrici babet, ct cctcroquin est gonidicus." — Nyl. in Jitt. On I'ocks. Lecidea inelopta, Nyl. " Tballus albus, tenuis, areolatus, in liypotballo nigro subprscdominante instratus ; apotbecia nigra, in arcolis vulgo demum singulatim innata, plana (latit. 0"5-0"7 millim.), immargiuata ; spora^ 8nce, incolores, cllipsoidca), longit. O'OOS-0'010 millim., crassit. 0"004 millim. ; parapbyscs non distincttc ; epitbccium cierulescens ; bypotbccium incolor. lodo gclatina bymenialis bene ctcrulescens. " Species bene distincta, et fere urceolariiformis, inter species stirpis Lecidece plance. Tballus K flavens (I demum uoanibil subroseo tinctus, lisec reactio vero passim obsoleta). Variat tballus crassior (crassit. 0*5 millim.) et pluribus apotheciis in quavis areola (arcolis circumcirca lateribus nigris),*cipotlieciis impressiusculis." — ■ Nyl. in litt. On rocks ; common. Lecidea disciformis {Fr.), Nyl. On tbe bark of trees specially Fcetidia mauritiana, Lam. Lecidea triphragmia, Nyl. ' On tbe bark of trees, specially of Ficus. Lecidea spui'ia {Schcer.) On decomposed rocks. Lecidea immutans, Nyl. Tballus ocbraceo-rosellus (verisimiliter autem typice albidus), areolato-diiVractus (crassit. cii'citer 0*2 milbm.), determinatus (K +, flavens) ; apotbecia nigra, supcr- ficialia, plana, obtuse marginata (latit. 0-5-0-9 millim.), intus coucoloria ; sporse 8njB, fuscfe, ellipsoidea?, 1-septatge, longit. 0-016-0-023 millim., crassit. O-OOS-O'OIO millim., leviter superficie granuloso-inrequales ; parapbyscs gracilcscentes (non LICHENES. — REV. J. :\r. CRO.AIBIE. 411 bene (liscretco) ; epithoeiiim ct hypotliccium luteo-fuscesccntia (aut hoc raagis in- fuscatum). lodo gelatiua hymenialis intensive ccerulcscens. " Species bene distincta in stirpe LecidecR spurice. Tliallus I-. Spermatia breviter baciUai-ia, longit. 0-00-t millini., crassit. 0*0006 millim." — Nyl. in litt. On the rocks Avitli the fore2:oin2:. Lecidea continens, Nyl. " Thalliis lactcus rel sublutescenti-albidus, sat tenuis, continuus, superficie arco- lato-riunilosus (K flavens, medulla I Cceruleo-obscurata) ; apothecia nigra, plana marginata (latit. 0-5-0-6 millim.), margine fuscescente, intus obscura ; spora? 8uce, fuscte, cllipsoideje, 1-septatae, longit. 0-011-0'014 millim., crassit. 0-006-0-007 millim. ; paraphyses fere mediocres ; epitheciimi et hypotheciimi fuscescentia. lodo gelatina hymenialis intensive casrulescens. " Ad Lecideam spuriam quodammodo accedens et reactionibus conveniens, at tballus alius. Variat perithecium fusco-pallesccns. Spermogonia non visa.'' Not uncommon on rocks. Lecidea configurans, Nyl. "Thallus kitescenti-glauco-albidus, Isevis (crassit. 0*5 roillim. vel tenuior), sub- rimulosus, dcterminatus ; apothecia fusco-uigra yel f usca, innata, plana (latit. 0"3-0"5 millim.), immarginata, circumscissa, intus albido-incoloria, demum 2-G submoniliformi-confluentia (turn quasi lomentaceo-liuearia et varie flesxiosa) ; sporae Sure, incolores, oblongse, simplices, longit. 0*011-0'014! millim., crassit. 0*004<-0'005 millim. ; paraphyses graciles ; epithecium sublutescens aut dilute glauco-fuscescens ; hypothecium incolor. lodo gelatina hymenialis cserulescens, dein violascens. "E stii'pe Lecidece ^iUw. Geog. Dist.— European shores. Leptothrix lutea, Kiz. Geog. Dist. — "111 tlicrmis Engauearum " (Menegliini). NOS lOCHINEiB. Nostoc commune, T'aiich. Geog. Dist. — Europe ; America ; Java. 3 G 2 120 z 0 0 L o c; Y. EXTINCT FAUNA. Observations on the Bone Caves of Kodrigitez. — By Henry M. Slater, B.A., Natitralist to the " Transit of Venus " Expedition. I. — The Cave Region of Rodriguez. The cave tract in Rodriguez is situated about tlie S.W. side of the island, and is of a very curious natiu'c. We find there 10 or 12 patches of limestone scattered upon the basalt which forms the island ; these patches are of irregular form, and usually terminated by an escarpment of various height, from 3 to 10 or 12 feet, which marks their junctm'e Avith the basalt. On examination, these patches are found to consist of marme coral upheaved with the basalt. Sometimes the limestone preserves its original structure or grain which it had in life ; sometimes this grain is brought out by weathering, the outer walls or thccse being harder than the rest, and standing out in relief when the rest is denuded away ; sometimes, but more rarely, the septa arc visible. But, as a rule, the; lime- stone is much altered : whether by pressure dm'ing or previous to uphcavnl, or whether by an infiltration of acidulous water with carbonate of lime in solution, I must leave to more able geologists than myself to settle upon an investigation of my colleague, Mr. Balfour's, collection of rocks ; but the stone is usually rendered very hard and solid, though always very porous. Sometimes weathering produces a quasi-oolitic grain on the sm'face. It is in these coralline limestone patches solely that the caves are situated ; in Bom'bon, on the contrary, the only caves were in the igneous rock itself, and appeared to be the effect of rapid cooling. The caves have been, if not originated, at least much enlarged by water, of which many bear abundant traces, and in the rainy season some are evidently the coiu'ses of subterranean streams. Another proof of the marine origin of the caves is found in the presence therein of several marine shells, as Fleurotoma sp., 150 yards above the sea, and 2,500 yards from the coast line; and Terebellum sp. In places where the coral is at its minimum of alteration, perforations were seen in it, which from their slightly oval shape, and ffradiiallv increasina: diameter, seemed identical with those of the Lithodomi, at present common on the reefs in Bodriguez. I quote in allusion from Sir 0. Lyell's "Principles:" "Perforations of Litliodomi in limestone cliffs, and in the BONE CAVES.— HENRY H. SLATER. 421 " Temple of Serapis at Puteoli, afford conclusive evidence of change in the sea level " of coasts in modern times." There exists near my camp at the caves a sort of ravine, terminated at each end by a cavern, and having others opening into it. The terminal caves and precipitous sides at once determined me that this has been a vast cavern, the roof of the greater part of which has fallen in ; this fall has left a sort of ravine with a level bottom siuToundcd on aU sides by precipices, nearly, if not quite, perpendicular, and having a height of from 30 to 90 feet ; the bottom is now covered with earth and full of large trees, the tops of which rise to the level of the cliiTs. Descent can only be effected with ease in two places, where two heaps of limestone blocks rear themselves against the precipices. There is no reason to believe that water ever acciunulates in the caverns opening into this " Gorge," and in these caves most of my specimens of any value were found, whilst the trees in the Gorge, being the abode of guinea- fowl, supplied my table. There were, judging from the hollow sound of the coral in various places under my boots, other caves with no mouths at all, but I did not judge it needful to open them. Capt. Wharton, of H.M.S. " Shearwater," opened one about a mile and a half "W.S.W. of the Gorge, and descended by means of a rope, but nothing was to be seen but stalactites, nothing like earth on the floor.* II. — Bone Earths and Stalagmites. The cave-earth was always similar in colour to the reddish brown earth foimd on the basaltic parts of the island, but differed from it in being generally mixed with a greater or smaller per-centage of coralline sand, formed from the crumbling of the interior of the cave, or washed in. The earth itself was probably washed into such caves as were liable to the influx of water ; into the dry caves it seemed to have been blown in. in the form of dust, which form it often kept. Mixed with the cave-earth was a large admixture of land shells, including — 1. Cyclostoma articiilatiim, now li'sing in the island abundantly. 2. Cyclostoma hcemastoma, much smaller than above, and also living in the island. 3. Helix heiosheriana. If inches in diameter across the whorls. 4. Fupa sp., 1 inch in length, and ^ inch in diameter. The two latter are now extinct, and only found in the caves ; there are also a small Fiipa, and two small Bitllmi now living in the island, and an Acliatina (extiuct), which occvir in the cave-earth, with remains of a huge Julus, much larger than those now in the island, cciualling the species in the Seychelles. I never met with any human remains, either bones or implements, in the cave- earth. * I was surprised when I made my preliminary survey of the cave tract in Rodriguez lo discover the amount of work already done. Out of the first 13 caves, I found 12 to bear uuniistakeablo signs of previous research. Nevertheless, I deemed it best to dig them over again, a proceeding which elicited a certain number of bones, though it gave no information whatever as to their deposition. 122 ZOOLOGY OF IIODIJIGUEZ. . The depth of the bouc-earth is very variable ; in some caves wc find it with a depth of from six inches to three feet; in others, liowever, it varies from four to nine feet in depth. 13elow about two fed T n(>ver found many bones, wliieh makes me believe that the agency Avliich deposited I he bones in ilie caverns, never operated until the later days of the existence of tlu- Solilaiic 'flic bones mighi certainly have dceaycd, but yet I usually found thai the hones whii-h were well coA'ered with earth were in much better preservation than those near or upon the surface, which were usually much decayed. This makes me think that the Solitaire resorted to the caverns in case of lire in the island, which has been known to have denuded it several times of its trees ; more so, as in several cases I found nearly perfect skeletons, which lay evidently as they died ; this precludes the idea that they were carried there by wild cats. Again, in the bottom of a cleft near the moid li of a cave, I found the greater part of the skeletons of a male and fenialc Solitaire ; they had clearly fallen in and were luiable to extricate themselves, but the bones being but imperfectly covered, many bones Avere so much decayed as to prevent their removal. But I could not doubt that they arrived there alive, and if not cbiven into the neighbouring hole as a shelter under some alarm, what could have allured them there ? Nor were any bones found in the caves at any distance from the mouths, at least, if there were, they were single, and looked as if they had been washed there. Land crabs inhabit some of the caves, and may have conveyed the dead Solitaires into the inmost recesses, when decay enabled them to remove portions of the carcases at a time. Stalagmite of any thickness is uncommon in the caves ; in three caves only did it occur in anything like extensive sheets, but it often occurred in small patches of a yard or so in diameter under some long stalactite. But in one long cavern, near a fishing station called Palates, there was evidence of there having been rather an extensive system of stalagmite layers. This cave was nearly a mile in length, and ran downhill, having an elevation of about 200 feet above the sea level at one mouth, and not more than 25 at the other ; moreover, it was clearly occasionally the bed of a stream. In one place in this cave there was an interesting relic in the shape of a sort of column composed of alternate floors of stalagmite and cave-earth. It occmTed in an angle of the cave Avhere the force of a stream coming doAvn from the higher mouth would just have missed it. It was about 8 feet in height, and there were three or four layers of stalagmite in it. At one period, before the streams began to flow through the cave, this system of alternate layers had existed all over the cavern, but had been broken up by the stream, which would wash out the earth, when the stalagmite would fall in. Evidences of the stream were to be found in the trunks of trees and beds of leaves, with smooth ripple-marked sand in the floor of the caves. 1-23 The Extinct Birds of Rodriguez.*— i?// Br. A. Gi'mther, F.B.S., and Edward Newton, C.M.G., M.A., F.L.S. (Plates XLI.-XLIII.) The following notes arc the results of our examination of two sets of materials, one collected by Mr. H. H. Slater, one of the naturalists of the Transit of Venus Expedition, the other obtained previously by Mr. George Jenner when magistrate of Hodriguez. The majority of the remains brought home by Mr. Slater have been deposited in the British Museuu], whilst the bulk of the latter set has been deposited in the Museum of Zoology of the University of Cambridge. As far as practicable, and without detracting from the value of each set, the two sets have been mutually supplemented by an exchange of duplicate specimens, but originally they consisted of the following remains : — Carine murivora. Cranium Mandible Pelvis Humerus Ulna Cranium Mandible Sternum Coracoid Humerus Sterna Humerus Upper part of cranium Mandible Coracoid Humerus Humerus Ulna T. of V. Jenner. — 1 1 1 2 3 4 5 . 9 Metacarpus Coracoid Femur Tibia Metatarsus Necropsar rodericanus. Ulna Metacarpus Femur Tibia Metatarsus Turtur jncturaftis. Tibia Metatarsus Necropsittacus rodericanus. 1 2 6 Ulna Femur Tibia Tarsus Po.lc^omis exsul. Femur Tibia T of v. Jenner 1 2 1 9 3 2 10 3 6 1 2 15 11 2 23 5 4 1 — 2 2 2 1 I 1 2 — 1 5 * With the exception of the Solitaire, which is the subject of a subsequent separate paper. 121 zooT.nny of t^odrtgfez. Crnnitini Bill - AIniulible Pi-lvis SliTiium Iliiinonis Erijthromach us Icguati. T. of V. Jciincr. 1 2 Uliin o 2 Ratlins — T) Mclticnrpii a () Foiiiur 2 — Tibia 1 — Tiirsus 3 12 T. of V. Jcmicr 8 4 2 — 2 — 13 15 16 18 10 17 Ardea megaccphala. C'rnnium with bill T_vmi>niiic Maiulilile Pelvis "V'ertcbFft Stcnumi Corncoid Scapula Hum or us 1 Ulna 1 Kmlins 1 IShliicaipus 1 I'halanx 3 Femur 1 Tibia 2 Fibula 2 Metatarsus 4 Fbalaux — 2 — 1 1 2 - 1 2 1 2 2i — 1 4 4 — 3 Carine murivora [A. 3£.-Edw.) (Plate XLI., Figs. A.-F.) M. X. Milne-Edwards (Ann. Sc. Nat. Zool. t. xix., art. 3, p. 12) recognized two tibiaj and one metatarsus as belonging to two species of Owls, one of which he named StrLv {Athene) murivora, the other (represented by one tibia only) he left uimamed. This last was G mm. longer than that of C. murivora, and said to be distinguished by a shorter and less projecting peroneal crest. Having this individual bone before us, together with a considerable niiml)er of tibia? of the shorter and longer kind, we have been able to ascertain that a great portion of the peroneal crest is broken away, that in perfect specimens the crest of the longer bones is as much developed as in the shorter ones, and that, in fact, the former are evidently those of females and the latter of males, according to the difference in size which obtains between the sexes of nearly all the nocturnal birds of prey. After having satisfied ourselves that all the Strigine remains from Rodriguez are referable to one species only, Carine murivora, we hoped to be able, with the aid of so much more material than was at the disposal of M. Milne-Edwards, to throw more light on its generic affinities. In this we were not successfid, chiefly for the reason that the genera of Owls adopted by some ornithologists do not by any means coincide with osteological modifications, and that the Rodriguez Owl shows a combination of osteological characters which we have not fou.nd in any of the numerous skeletons of Striges with which wc have compared it {Bubo, ScojJS, Surnia, Carine, Ninox, Glaucidium, Asio, §"0.). Of the craniiim (Fig. A.), the frontal region, with a portion of the parietal region, has been preserved. The frontal bone is not swollen at the root of the bill, as, for EXTINCT BIRDS.— A. GUNTHER and E. NEWTON. 425 instance, in Cariiie boobook or C. hirsuta, but rather concave, as in C. hifmcula and C. noctua. The orhit appears to have been of an unusually short diameter, and the eye is not protected by the long supraorbital process of C. noctua and C. oifitscata, this process being reduced to a very short prominence, as in C. novce- zeland'ue, C. boobook, and C. Jiirsuta. The margin of the orl)it in front of the supra- orbital protuberance is longitudinally grooved. Width of anterior part of Frontal region, 13 mm. ; least width of interorbital space, 15} mm. The mandible (Fig. B.) does not show any peculiarity ; one of the rami is 40 mm. long. The form of i)xQ pelvis (Eig. E.), particularly of the postacetabular portion, differs from that of all other Owls with which we are acquainted. The anterior iliac blades are of moderate width, presenting a deeply excavated outer surface, the lower portion of which is nearly horizontiil, a small pointed tubercle projecting on its outer margin (as in Scops riitilus). The convergent dorsal edges of this part of the ilium do not coalesce into a crest, but are connected by a flattened lamina, which forms the roof over the large longitudinal vacuities on each side of the sacral crest. In its middle the pelvis is but moderately constricted. The postacetabular portion is broad, chiefly owing to the great width of the sacrum, which is of rhombic shape, dorsally slightly depressed in the median line. The post-iliac crest is unusually prominent and more strongly curved, in an S-shaped line, than in any other Owl with which we are acquainted. There exist considerable modifications of the configuration and arrangement of the foramina for the sacral nerves in the various species of Owls, even in such as are referred to the same genus. The Rodriguez Owl resembles in this respect Carine noctua, and still more Scops riitilus, the three anterior foramina being equidistant from each other, of nearly the same width, and separated by the strong pleurapo- physes, all of which are perpendicular to the axis of the vertebral column. The fourth foramen is much constricted by pleurapophyses, which are much stronger than the preceding, and situated at a difi'erent level, opposite to the point of the greatest constriction of the pelvis. The deep anterior and posterior renal excavations are separated from each other by a pair of slender pleurapophyses. On the whole, the pelvis of this bird is of an unusually strong build, and well adapted to form the base of the powerful muscles of the trunk and hind limbs by which this small-bodied bird was enabled to keep down the rats vrith which the Island was infested, as Leguat informs us, and later to overpower " les petits oiseau et les petits lezards " as the author of an old manuscript says it did.* * Proftssor Newton has published the abstract of this M.S. " Relation do I'llo Rodrigue," in Proc. Zool. Soc. 1875, p. 39 ; and the entire MS. has been piiblislied by Professor A. Milne-Edwards in Ann. Sei, Nat. (6) ii. art. 4, pp. 20. The passage runs as follows : — " On voit une oiseau qui est a peu pres commc la chouettc, et (jui mange " les petits oiseau et les petits lezards." 3 n 426 ZOOLOGY OF RODKIGUEZ. The pelvis does not appear to vary much in size. Lonjrth of pelvis (to oiul of siicrum) Givntosi width of niitcrior pnil. Least wiilili of anterior part Greatest wiilili of posterior part Asio iiMpitrinus. 36 nun. 21 „ 2!) „ ;!■} mm. "-' j» H „ 23 „ •10 mm. 21 „ 13,4 V 25 „ The smg'lc Umb-boues do not show any characteristic featm-es, and very little need he said ahont them heyond the general conclusion at Avhich we have arrived from comparative measurements of those of other Owls. The coracoid (Fig. C.) is a little weaker than in C. boohook, and 30 mm. long", its sternal extremity being 10 mm. broad. The foramen niitritium of the shaft of the humerus (Fig. D.) is generally (but not constantly) nearer to the distal than to the proximal extremity. The ulna shows elevations for the secondaries. In the femur (Fig. F.) all the lines and grooves for muscular insertions are strongly marked ; the outer inferior condyle is very prominent, and provided with a deep groove for the tendon behind. The tibia and metatarsus have been described by M. Milne-Edwards. The discovery of a part of the cranium and the pelvis enabled us to obtain a jiretty accm-ate idea of the size of the Rodriguez Owl. Its body was rather larger than that of C. boobook, but considerably smaller than that of Asio accijntrimis. The wing was clearly somewhat shorter and a little less developed than in either of those Owls, though the power of flight of this bird cannot have been much imj)aired. On the other hand, the length and strength of its legs, as compared to the size of the bodv, and especially to the length of the wing, was much more than in any of tlie species of Owls, the skeletons of which we have been able to compare, as will be seen from some of the measm'cments which we have taken, and which we tabulate thus : — Humerus. Ulna. Metacarpus. Femur. Tibia. Metatarsus. longest, shortest. longest, shortest. long, short. long, short. Carine murivora 69 X 64 74 36 54 X 50 76x69 46x41 „ boobook 69 76 36 48 74 41 „ infumata 49 28 58 39 58 36 „ uoctua 51 28 56 41 59 37 „ hirsuta 56 30 64 39 60 32 „ castaueoptcra 48 OO 24 33 49 23 „ conuiveus 115 137 60 65 109 — . Spiloirlaus novse-zelandise 58 30 65 41 64 35 Asio accipitrinus 81 90 42 54 77 40 Scops rutilns - 47 54 23 32 50 28 Bubo virginianus 163 178 85 101 146 75 Ketupa zeylonensis - 76 121 68 120 129 54 EXTINCT BIRDS.— A. GUNTHER and E. NEWTON. 427 Necroi^sar rodericamis. Slater. (PL XLII., Figs. A.-G.) The discoveiy of an extinct Starling in Eoclriguoz, allied to the Fregilujms of Keunion, which appears to have held out a little longer in struggling for its existence within so narrow limits, is undoubtedly one of the most interesting results of Mr. Slater's labours. He recognised and determined the affinities of the remaias found by him, and says, in prehminary manuscript notes appended to his general report : " Professor Newton prophesied, as it were, the existence of this bird (Proc. " Zool. Soc. 1875, p. -il) when, alluding to a bird mentioned in a manuscript " ' Relation de I'lle Rodrigue,' he said, ' I am at a loss to conjecture what these bu-ds " were, unless possibly of some form allied to Fregilupus.^ Necrojpsar is altogether a " smaller bii'd than Fregiliipus, to which it is most closely allied. If it were a " mammal, I should not make a separate genus of it ; but as it is a bird, I think I " cannot do less." After a close scrutiny of all circumstances, we quite agree with the discoverer of this bird in questioning the propriety of genei'ically separating two species, on what appear to us very slight modifications of the osteological frame ; and we retain the name Kecropsar merely from the wish of conforming with the present ornitho- logical practice. In the following notes more especially those points will be mentioned in which the Rodriguez Starling differs from that of Reunion, the latter having been fully described in Dr. Miu'ie's excellent paper in Proc. Zool. Soc. 1874, p. 474, to which we have to refer for other details. The skeleton of Fregilupus itself, which is imique, has been lent to us by Professor Newton. Of the skull of Necropsar (Fig. A.) the cranium with the maxilla and mandible has been preserved ; but all the bones of the palatal region and the tympanic have been lost. It is extremely similar to the skull of Fregilupus, and more especially possesses a postorbital muscular impression above the tempoi-al fossa, which is very conspicuous in FregiUqms and Acridotheres {cristatellus), and faintly visible also in the common Starling {Siin'iins). But it differs from Fregiliqms in having not only the supraoccipital ridge conspicuously more developed, but also in showing, besides, an additional line of muscular insertion (biventer *) running across the parietal region, subconcentric with the supraoccipital ridge. The development of these ridges indicates a corresponding increase of strength in the muscles of the neck and articular end of mandible. The maxilla is a little shorter, and decidedly less curved, and less slender at the extremity than ui Fregi- liipus, the nasal aperture being longer and generally larger. The mandible f is on * See Owen, Anat. Vert. II., p. 85, fig. 35. 18. f The articular facet of the mandible is figured on I'l. XLII., Figs, a and w', half as large again as the natural size. Fig. a represents its configuration as it appears in two of the specimens before us ; but in 3 H 2 428 ZOOLOCY OF RODRIGUEZ. tlio whole of a stouter build, ami al)out the luiddUMil' its longtli (at tlie angle of the mouth) more bent, in the fashion of the corrcspondin;:;' ])art of the Starling. Peenliaritics all of whii-h indieate that Nccropsar had to use greater foree than Fi'cgiliquis in scarehing and perhaps digging for its I'ood, and that it possessed in an eminent degree the capability found in several species of Starlings and Crows, of removing objects or forcing an entrance in their search for food, by- first inserting their wedge-shaped bill and then opening the mandibles. The greater strength of the bill would also be fully accounted for, if Necropsar bo tin- liird described in tlie old manuscript jireviously mentioned as feeding on eggs and (lead tortoises' " qu'ils savent assez bien dechircr." * The Ibnestva in the interorbital se])tum is wider tliaii in Frrff/l/'/nt.^. Fregilupiis. Necropsar. Length of cranium (moiisuroil from occipitiil cnndyk) - Greatest width of cranium . . - - Wiiltli of interorbital space .... Length of bill ..... ,, nasal aperture .... „ mandible - - - - - Greatest depth of mandible - - - - A. B. C. D. 26 mm. — — 28 mm. oi 29 mm. 23 „ — — SI „ 21 „ 7i„ — — 8 „ 8 „ 41 „ — — 39 „ 36 „ 11 „ — — 13 „ 12 „ 58 „ 59 5.3 53 „ — 5 „ 5 5 4i „ — Of a siernnni only the fore part has been presei-ved, which does not show any difference from the corresponding bone of FregUiqms, except in the foremost part of the keel being 1 mm. lower. The coracohl (Fig. B.), 27^ mm. long, is otherwise identical with that of Fregihtpus (29 mm.). Such of the wing and leg bones as have been preserved differ nowdse from those of Fregilupiis, Acridotheres, or Sturinis. Figm-es, half as large again as the natural size, are given of the htouerns (C), metacarpns (D.), femur (E.), tibia (F.), and metatarsus (G.). The ulna is distinctly marked with elevations for the secondary c[uills. All these bones vary in size ; but as far as we can judge (with any amount of the third specimen (Fig. a) the angular developes a transversely extended process directed inwards, which is entirely absent in the two other specimens. One of the authors (E. N.) is inclined to consider this process as an individual malformation. * The passage runs as follows: — " On trouve un petit oiseau qui n'est pas fort commun, car il ne se trouve pas sur la grande terre ; on en voit su I'ilo an Mat, qui est au sud de la grande terre, ct je crois qu'il se tient sur cettc ile a cause des oiseaux de proie qui sont a la grande terre, comme aussi pour vivre avec plus de facilite des oeufs de ces oiseaus de pechent qui y pondent, car ils ne mangent autre chose que les oeufs ou quelques tortues mortes de faira qu'ils savent assez bien dechirer. Ces oiseaux sont un peu plus gros qu'un merle et ont le plumage blanc, une partio des ailes et de la queue noire, le bee jaune aussi bein que les pattes, et ont un ramage merveiUeux ; je dis un ramage quoiqu'ils en aient plusieurs, et tous differents, et chacun des plus jolis. Nous en avons nourri quelques uns avec de la viande cuite hachee bien menu, qu'ils mangeaient preferablo- meut aux graines de bois." — Proc. Zool. Soc, 1875, p. 41. EXTINCT BIRDS.— A. GtNTHER and E. XEWTON. 429 safety) by comparing their average length with that of Fregiliqms, the length of the forearm of the latter was relatively to that of the humerus soniewhat greater than in Necropsar, while the proportionate lengths of tihia and melatarsiis were exactly the same, as may be ascertained from the following measurements : — Lens til of humerus ulna metacarpus femur - tibia metatarsus Necropsar. Fregilupus. Sturnus vulgaris. longest shortest 35 32 ! 40 37 091 --2 34 32 59 52 41 36 38 47 24 37 65 45 27 34 26 46 Necropsittacns rodericanus, M.-E. (PI. XLII., Figs. H.— X.) As more parts of the skeleton of this extinct bird are discovered, its aflinity to Falaornis (which has been maintained by M. Milne-Edwards, in his second treatise on this bird, p. 134,) becomes more and more apparent. The most peculiar portion of the skeleton (as far as we know it at present) is the craniimi (Fig. H.), which is singularly depressed, the occipitals being low, the occipital foramen wider than deep, the parietals but slightly convex, and gently sloping behind towards the occi- pital region. This depressed form of skull we have not observed in any other genus of parrots except in Ara, but the birds of that genus have a complete suborbital arch, while in Necvopsittacus, as in Falccornis, the uncinate bone is not anchylosed to the squamosal process. To judge from the figure of Lophopsittacus mauritianiis (Encyl. Brit. Ed. 9, iii., p. 732), that bird may have had a similarly flat and depressed cranium. Length of base of cranium (without bUl). Width of cranium - - - Depth of „ - Xecropsittacus. 50 mm. 38 „ 24 „ Palffiomis (alexandr). Ara militaris. 35 mm. 28 „ 21 .. 59 mm. 48 „ 34 „ The several limb-bones vary considerably in size and length, in a similar manner as those of the other extinct birds of Rodriguez ; but this variation does not seem to indicate a sexual difference, as there are intermediate specimens between the longest and shortest. Such of the bones as have been preserved do not show any noteworthy feature; Ave have given figures of the natural size of the coracoid (J.), humerus (K.), femur (L.), tibia (M.), and metatarsus (N.) A comparison of 430 ZOOLOGY OF KODKIGI'KZ. their relative lengths with those oi" Am and FaUcoDiis appears to be of direct interest. Necropsittiiciis. ! ," "-'"•''"^ Ar:i militaiis. ' ' ( iiluxaudn). loiiyc'st sliortfsi Length ol ' coracoid - - - — 3.5 31 53 liumortis - . . 53 49 43 79 ulim - . — 57 52 99 j» femur . - 49 46 37 60 tibia - - r 63 59 50 83 » motntiu-sus " ■ ^^ 22 18 31 If we considered the skull as a sate guide to obtain an estimate as to tlie size of the bird, we should infer that the length of the wing as well as of the leg had been diminished in this bird, as compared with Falceornis and Ara. But we believe that the pelvis aifords a mucli surer basis for such calculation, and are, therefore, contented to leave this question undecided for the present, merely maintaining that in the extinct E.odi'iguez Parrot all the limbs were smaller relatively to the size of the head than in Ara and JPala'oniis, and that the fore-limbs were relatively to the hind-limbs not less developed in the Rodriguez bird than in those two genera. Falceornis exsul, A. Newton. The fragments of the skeleton collected by Messrs. Jeuner and Slater have been mentioned at the beginning of this paper. As, however, a nearly complete skeleton of a freshly killed bii'd has been obtained, a description of the isolated bones in their collection would be obviously superfluous. Colmnha rodericana, M.-E. M. Milne-Edwards has described this Pigeon from a single sternum so different from that of other members of this family (as far ^as they are known), that we cannot refer it, at present, to any particular genus. A metatarsus seemed to him to afford evidence of the existence of a second species, which he believed to be Turtur picturatiis of Madagascar, B.6union, and Mauritius. "We have before us two meta- tarsi, not differing from that examined by M. Milne-Edwards, and do not see the reason why they should be separated from C. rodericana, to which, evidently, all the other bones of oiu- series belong. All the limb-bones {Jiumertis, femur, tibia, and metatarsus) vary not incon- siderably in size, but there is a gradual passage from the shortest specimen to the longest, so that this difference is not likely to be sexual. The only evidence we have found of the possible existence of a second larger species of Pigeon is a femur in Mr. Jenner's collection which is 42 mm. long, and comparatively more slender than that of C. rodericana. The Hne of muscle-insertion on the front of the shaft starts from the very margin of the bone below the trochanter, as in a EXTINCT BIRDS.— A. GUNTHER and E. NEWTON, 431 Pigeon, and not inwavds of the margin as in a Parrot. But no definite conclusion can be safely arrived at from this single bone. The limb-bones of the Rodriguez Pigeon do not show any perceptible noteworthy peculiarities. The humerus has a deep semi-globular pneumatic cavity, and a small tubercle on the ulnar margin just above the condyle. Thefenmr is slightly bent forwards, with the line of muscle-insertion starting from the margin of the bone, and running inwards along the middle line of the front side of the shaft. The following are the measiu-cments : — Shortest. Longest. Length of humerus - - - 34 mm. 37 mm. femur - - - - 36 ,, 37 ,, ,, tibia - - - - 44 „ 48 „ ., metatarsus - - - 25 ,, 26 ,, Aphanapteryx and Erythromachus. (Plate XLIII.) The classical pajier " Sur les aflfinites zoologiques de V Aphanapteryx''' in which M. A. !Milne-Edwards, with the most convincing precision, demonstrates the Ralline characters of a few ornithic remains sent from Mauritius, recognizing their connexion with the picture so fortunately and ojjportunely discovered by Von Prauenfeld, offers such a sure base for all succeeding workers that scarcely more remains to be done than to supplement his memoir by description of those parts of the skeleton with which lue was not acquainted. At a somewhat later period M. Milne-Edwards received a metatarsus and sternum from Rodriguez, in which he recognized the representative of the Mauritian Aphan- apteryx, but misled (as we think) by Leguat's description, he believed that the Rodri- guez bird differed by having a straight bill and shorter legs, referring it to a distinct genus which he named Erythromachus (Ann. des Sc. Nat. Zool. t. xix. art. 3, p. 6). These differences do not in reality exist, the bill of the Rodriguez bird is as much, curved as in the Mauritian, and their legs have the same proportion. The wings were clearly more developed in the species inhabiting the smaller island, and certainly lack the power of flight, but this can scarcely be relied upon as a generic distinction, and, therefore, we prefer looking upon the Rodriguez bird as a smaller species of Aphanapteryx with somewhat longer wings, and to treat of it as Aphanapteryx legnati. The search for further evidence of Aphanapteryx in Mamitius, subsequent to M. Milne-Edwards' s memoir, has resulted only in the discovery of some more or less broken metatarsals, besides fragments of other portions of the leg which do not admit of determination. From the small size of the former bones, it is clear that Aphanapteryx hroeclci varied considerably in size, or that two races existed, which is not very probable. 132 ZOOLOGY OF KoniilGUEZ. T!n< craninm of A. loc/iiatl (Fig. A.) is, in general appcarauce, exti'cmcly similar to that of Bolliis {iiqualicits) and Ocydromna, being slightly elongate, equally eonvox in every dix'eetion, and with tlic proeesses and ridges but slightly projeeting. The orbital edge of Oci/ilro»iiis is truncate nearly in its whole circunirercnce (perhaps l\)r the attachment of a glandular band), but in A. legiiati it is rounded off, as in BaUits. The frontal region is narrow, long, with the laieral margins slightly diverging in fi'ont. All the bones belonging to the palate and the lachrymals arc lost, but so much of the jaws has been preserved as to give a perfect idea of the form of the bill, which is curved in exactly the same manner and to the same degree as in A. broecki. It is very feeble, and varies extraordinarily in length, being in some specimens one third shorter than in others. The extremity of the longer beaks (Fig. A.) is more conspicuously curved than that of the shorter ones (Fig. B.) Besides, we may state at once that throughout the series of bones at our disposal a marked difference in size may be traced ; but whether the short bills belong to the smaller individuals, or whether the difference in length of Ijcak and in size generally is attributable to sex, we arc unable to say. The culmen of the bill above the nostril is almost straight; the nasal aperture of extraordinary length, slightly longer than the cranium. The extremity of the bill is provided with a deep and narrow groove along its under side. The mandible (Figs. A., C, C) is identical with that of the Mauritian species in every respect, except in size, more especially it has the three postarticulary apophyses (C.) arranged in exactly the same manner, but less projecting, as may be expected in the smaller species : — Length of cranium measured from occipital condyle - 38 mm. Greatest width of cranium - - - - 20 ,, Least width of frontals - - . . . (3-5 ^^ Length of bill - - - - - . 77 „ „ nasal aperture - - - - 43 ,, „ lower jaw - - - - - 98 „ Greatest depth of lower jaw - - - - 8 ,, The 2ielvis (Fig. D.) is large, very strongly built relatively to the size of the bii'd, and has thoroughly Ralline characters ; its prseacetabular portion is narrow, with the iliac blades scarcely divergent, and dorsally confluent into a high crest, the profile of which is much more strongly curved than in Ocydromiis ; a strong crest runs along the median line of the back of the sacrum. The postacetabular portion is dilated, much broader than in Ocydromus or Rallus, with deep and capacious renal excavations, and ^nth the outer descending surface of the os iscliium formed into a deep broad short hollow, bordered by an upper (jjostiliac) projecting crest, and by the lower reverted margin of the bone. The postiliac crest is not continuous, but interrupted above the ischiatic foramen as in Rallus and Ocydromus. EXTINCT BIRDS.— A. GtlNTHER and E. NEWTON. 433 The iscliiatic foramen is circular, about equal in extent to the acetabulum. The slender pubic rods are not dilated posteriorly or attached to the ischium, so far as we can judge from a single specimen, in wliich the pubic hone is preserved, with only a very small portion of its extremity broken off. Length of pelvis to end of sacrum - - - 57 mm. Greatest width of anterior part - - - - 20 „ Least width of anterior part - - - - 13 „ Greatest width of posterior part - - - - 36 „ A single scapula (Fig. E.), niiich we are inclined to refer to this bird, is im- perceptibly dilated behind the middle of its length, and tapering in its posterior third. The articular surface for the shoulder-joint is very small. The whole bone is 45 mm. long, with a greatest width of 4 mm. The humerus (Fig. F.) resembles much that of Ocydronms, its shaft being straight, somewhat compressed, and moderately slender. The projecting process behind the head is almost hamate ; and there are no pneumatic foramina. Notable is the great difference in the size of this bone, the largest specimens being 50 mm., the shortest 45 mm. long. The larger and smaller specimens are represented in about equal numbers in the series before us ; and several are intermediate between the extremes mentioned. The ulna is considerably bent, in some specimens rather more so than others ; relatively to its length it is stout, with a conspicuous ridge running from the inner side of the elbow joint along the greater part of its length ; some faint elevations for the attachments of the secondary quills are visible. As regards length, it varies to a similar extent as the humerus, the longest specimen being 42 mm., the shortest 37 mm., long. The radius is much more slender and weaker than in Ocydromus. In the metacarpus (Fig. G.), the imperfect development of the wing is stiU more apparent than in the proximal bones ; its constituent parts being very slender, and the middle and annular bones anchylosed at the extremities only, leaving a large vacuity between them. A minute tubercle is situated on the side of the middle bone which is opposed to the annular. The two specimens before us measure 16 and 14 mm. in length, the larger being as long as the metacarpal of Ocydromus, and not much longer than that of Rallus aquaticus. The femur (Fig. H.) is, relatively to its length, stoiiter than in Ocydromus; the trochartcrian ridge well marked. The shaft is slightly bent, with a faint ridge for the insertion of muscles along its anterior and posterior side. As regards length, the specimens vary from 56 to 63 mill. The tibia (Fig. I.) and metatarsus (Fig. K.) are, in every detail of their con- formation, identical with those of A. hroecid, so that any description would he merely a repetition of that given by M. Milne-Edwards. The shaft of the tibia 3i 131 ZOOLOCY OF nODHIGUKZ. seems to be a little more elliptical in a cross section in (he Rodriguez bird; ami its middle troi'hlea i)rojeets somewhat less beyond the outer ones. The great extent of variation in size ol" the Rodriguez Rail is clearly shown by our scries of these two bojies ; specimen A (from Mr. Slater's coUecliou) must have belonged to an iu- diviilual of unusually large size ; but the passage from specimen J3 to C is very gradual, as is also the case between the extreme sizes of the metatarsus. Length of tibia "Width of middle of tibia Length of metatarsus "Width of middle of metatarsus "Widtli of trochlear extremity It is not by any means certain that the til)ia and metatarsus of A. broecki described by M. ^Milne-Edwards belonged to the same ijidividual, although the state of their preservation, the same deep black colour Avith which they arc stained, the cii'cumstance that they were found in the same locality, and the corresponding size of their condyle sm-faces, would seem to lead to that conclusion. Neither have the fenim- and metatarsus of the same individual of A. leguati been fovmd, or at least preserved. It is therefore impossible to state with precision the relative lengths of these bones in cither species. But taking such examples of bones of A. leguati as appear to be of an average size, and comparing them with those of ^. broecki, avc come to tbe conclusion that in all probability the Rodriguez species did not differ from the Mauritian as regards the relative lengths of the long leg bones.* S|jc(.'. A. 1$. c. 101 mm. 96 mm. 84 mm (i „ ■li» Longest. Smallest. - 60 52 - 5i 5 . lU 11 Nycticorax megucephulus, {Ardea megacephala, M.-E.).t (PI. XLL, Fig. G.) Researches made subsequently to M. Milne-Edwards's description of this bird, have brought to light several bones with which that author was not acquainted, viz., tlu> penultimate and antepenultimate cervical vertebrge, the fifth dorsal vertebra, the pelvis, the scapula, iilna, and radius, the second phalanx of the inner toe, and the first of the hind toe. These bones need scarcely any description, as they are * Note on Fulica newtoni, M.-E. Like the other birds indigenous to the Mascarene Islands, tlie Coot of Mauritius varied considerably in size. The tibia described and figured by M. Milne-Edwards is 144 mm. long. Two others which are now before us, and which like the former were found in the Mare aux Songes, and are now in possession of Mr. Lucas, of Upper Tooting, measure 140 and 124 mm., the shortest specimen being that of a full-grown bird. t Ann. des sc. nat. Zool. t. xix. art. 3, p. 10. EXTINCT BIRDS.-A. GUNTHER a.nd E. NEWTON. 435 of the same form as those of the hirds of this family, and especially of Ni/cticorax, hut they are important additions, assisting us essentially in amving at safe con- clusions as to the relative development of the various parts of the skeletou, and the real aflinities of this hird. M. Milne-Edwards had considered it to he a time Ardea ; finding the fore part of its frontal region fiat as in a Heron, whilst this part is concave in Nycticorax, he dismissed the question of the affinity to the latter which to us appears to he undeni- able. The distinctive features of the skull common to the European and Rodriguez Night-Herons are : — first the great width of the occipital region ; then the mastoid processes, which are as distant from each other as the temporal : the relative distances between the mastoid, temporal, and postorhital processes are the same in Iwth birds. The temporal fossa is nearly of the same -width ; the foramen occipitale is broader than deep, more as in the Rodriguez species than in N. griseus ; the arch of the supra- orbital margin is in both much more open than in the Heron ; the prajnasal groove is equally deep, and extending equally far forwards; the bill of the Rodriguez species is not less curved downwards than in N. grisetis, though the bill is equally slight in either. The bill of N. megacephalus is much stronger than in the European bird, but not more so than in other species of this genus. The prin- cipal difference between these skulls is that that of the Rodriguez species is much more depressed, with scarcely any transverse and longitudinal depression near the base of the bill ; it is also a little longer. The evidence gathered from a comparison of the pelces leads to the same con- clusion. The priBacetabu.lar portion is nearly equally narrow and constricted; the anterior iliac blades coalesce for a short distance only, leaving a great part of the sacral crest uncovered. The width of the postacetabular half, and the arrangement of the foramina, pleurapophyses, &c. are nearly the same. Length of pelvis Width in its narrowest part Greatest width above acetabulum N. megacephalus. 63 mm. 15 „ 35 „ N. griseus. 61 mm. 14 „ 33 „ The pelvis of Ardea cinerea has a length of 82 mm., and a greatest width of 39 mm., and is therefore in general sliape (as well as in other details of configiu-a- tion) widely different from that of Nycticorax. Taking i\\Q pelvis as guide, the body of the Rodriguez Night-Heron was of nearly the same size as that of N. griseus ; the cranium also was nearly of the same size, whilst the bill and mandil)le were much stronger, and in accordauce with this powerful development of the maxiUai'y apparatus, tlie cervical portion of the ver- tebral column was proportionally stouter than in N. griseifs. The reduction of the power of flight has been already demonstrated by M. Milne- 3 I 2 N. mo gnceiili 1. N. gi •i.sous. Gil mm. 09 mm. 34 )) 37 j> 14 )j 20 >5 72 )) 72 5) 55 jj 60 >> 114 >) 126 )J 121 J) 139 )J 117 )j 132 J) 62 >» 70 »> t all tlie bones of the wiug 430 ZOOLOGY OF KODKIGUEZ. Edwards, and wo are able to corroborate his opinion by completing the measure- ments of the sternum and wing in comparison wth the European Night-Heron: — Length of sternum - - . Breadth of sternum ... Givatest depth of keel Length of scapula - . . „ coracoid „ humerus „ ulna . - . - „ radius - - - - „ metacarpus - . - It will be observed from this table that wl have been reduced in length (and strength), the scapula has not been affected by the diminished amount of exercise taken by this bii'd. As regards form, it is exactly the same as in other Herons, viz., tapering behind, without dilatation. "With regard to the leg, M. Milne-EdAvards, guided in his estimate of the general size of the bird by the length of the skull and femur, inferred that this part of the osseous frame was much reduced in length. Having shown from the pelvis, with which M, Milne-Edwards was not acquainted, that the body of this bird was considerably less in size than he supposed, in fact equal to that of the Eiu-opean Night-Heron, we arrive at the opposite conclusion, viz., that the leg is proj)ortionally much more developed in length and strength. And this will be readily perceived from the following table : — N. megaceph. Length of femur - - 86 mm. „ tibia - - 136 „ „ metatarsus - - 93 „ Width of metatarsus - - 6|^ ,, Length of 2nd phalanx of inner toe 20 „ „ 1st „ hind toe 30 „ In this table we have added also the corresponding measurements of the Heron, in order to show that the Rodriguez bird agrees with Ni/cticorax, and differs from Arclea in the length of the femur as compared with that of tibia and metatarsus. The metatarsus and the phalanges are the parts in which the greatest development has taken place, the thickness of these bones being nearly twice as great as in N. grisens, the bird having been clearly of much more cursorial habits than its congeners, chasing rather terrestrial animals (Hzards) than aquatic. Thus the effect of the prolonged isolation on the two vertebrate-hunting birds of Rodriguez, the Owl and the Night-Heron, was precisely the same. Without N. griseus. Ardea cinerea, 82 mm. 89: mm. 136 „ 185 j> 89 „ 139 5> 4 „ — 19 „ 26 JJ 28 „ 35 JJ EXTINCT BIRDS— A. Gt'NTHER jcsd E. NEWTON. 4,37 losing the power of flight, they became brevipennate ; but the increased development of the legs compensated for the reduction of this power, and enabled the one to destroy animals of larger size when the smaller kinds became scarcer, and the other to chase its swift-running prey. In the Night Heron the increase of development was confined to the legs in conformity Avith its acquu-ed habit ; and it was principally the metatarsus Avhich became enlarged to receive and form a base for the tendons of the foot. But the Owl requii-ed additional strength for the purpose of mastering and tearing its prey, not only in the muscles of the lower leg, but also in those attached to the trimk, and hence we find hi this bml the greatest development in the femur and pelvis. Explanation of the Plates. Plate XLI. :— A. — F. Carine murivora (nat. size). A. Ci-aniuru. B. Mandible. C. Coracoid. D. Humerus. E. Pelvis. F. Femur. G. Pehds of Kycticorax megaceplwlus (nat. size). H. Pelvis of Nycticorax gi-iseus (nat. size). Plate XLII. :— A. — G. Xecropsar ivdericanus. All these figui-es are half as large again as the natui-al size, with exception of A. (skull), which is of the natural size. A. Cranium. a, i'}. Ai-ticidar surfaces of mandible. B. Coracoid. C. Humerus. D. Metacarpus. E. Femur. F. Tibia. G. Metatarsus. H. — N. Kecropsittacus rodericanus (nat. size). H. Cranium. I. Coracoid. K. HumeiTis. L. Femur. ]\I. Tibia. N. Metatai'sus. Plate XLIII. :— Aphanapteryx leguati. All the figures are of the natural size, with exception of C. and C, which are tliree times enlarged. A. and B. Cranium of long and short billed forms. C. Articular surface of mandible. C Posterior aspect of end of mandible. D. Pelvis. E. Scapula. F. Humerus. G. Metacarpus. H. Femur. I. Tibia. K. Metatarsus. ■138 ZOOLOGY OF KODIJTOT'KZ. On the Osteology of the Solitaire {Pczophaps solHar'ta, GmcL). — By Edwakd Newton, C.M.G., M.A., E.L.S., and John Willis Clark, M.A., Supcrin- tondont of the ^Museumof Zoology aud Comparative Auatomy in the University of Cambridge. (Plates XLIV.-L.) In the iMemoir on the Osteology of the Solitaire in the Philosophical Trans- actions for ISGO, by Messrs. Alfred and Edward Newton, the authors entered fully into the history of the Bird, and recounted the circumstances under which a large collection of its remains had been discovered in the caverns of the island of Rodriguez. That island having been selected as one of the stations from which the Transit of Ycnus in 187-t shoidd he observed, it was suggested that a thorough examination of the caves should be instituted in the hope of obtaining those portions of the skeleton which the previous researches had failed to discover. The naturalist ai^pointed to that station, Mr. H. H. Slater, sent home a large series of bones, out of which several male and female skeletons, almost complete, have been sorted. The major part of this collection is now in the British Museum. Previous to this expedition Mr. Edward Newton had requested Mr. George Jeuner, resident magisf ' of Rodriguez, who had before been good enough to interest himself in the t, ■ make a fresh examination of the caves. The result was a large collection oi Doucs, supplementing many deficiencies ; but unfortunately, neither in Mr. Slater's collection (when it arrived in England) nor in that of Mr. Jenner were there any remains which could be proved to be associated. Mr. Jenner's collection, found in Januai*y and February 1871, has been deposited in the Museum, of the University of Cambridge. He accompanied his collection by a most interesting report, in which he described the localities where the bones had been found. This it was our wish to have been allowed to print in connexion Avith this paper. ■ As, however, Mr. Slater had gone over the same ground, and composed a report of a similar character to that by Mr. Jenner, we were reluctantly compelled to acquiesce in the suppression of the latter. In the following Paper we have naturally drawn our descrij)tions mainly from specimens in the collection that was most accessible to us ; but the Council of the Royal Society having placed Mr. Slater's collection at our disposal for description, we have availed ourselves of that series of specimens when it appeared to us necessary to do so. ON THE SOLITAIRE.— E. NEWTON axu J. W, CLARK. 139 Vertebrce. la the former Memoir the difficulty of coming to any accurate conclusioa respecting the number of vertebrae was felt to be so great that the authors stated that " it is beyond oiu- power to determine precisely the number which the skeleton " contained." The assum])tion that nineteen vertebrae intervened between the skull and the last dorsal (that which is anchylosed to the pelvis) was made doubt- fully, and "wdth the full expectation that more extended study might reverse it. It Avas based on the probability that JPezophaps would possess as many vertebrae as Didus, and on the fact that Prof. Owen's artist had assigned that number to the latter.* The former collection contained one hundred and sixty-one vertebraj. Those received in the recent collection of Mr. Jenner alone augment this number to three hundred and thirty-two assignable to the cervical region, which we conceive should be distributed as follows : — I. IL III. IV. V. VI. VII. VITT. IX. X. XL XII. XIII. Total. ? 2 1 10 12 11 13 12 13 IG 17 14 16 13 18 13 22 13 14 14 18 15 12 11 10 10 12 154 178 The penultimate free dorsal, the eighteenth of the series according to our present view, is represented by ten female and eleven male examples. We are, therefore, now enabled to give a complete figure of this vertebra, Plate xliv. figs. 1, 2, of which the former collection possessed only more or less broken examples. The neural spine is developed so as to attain a height cqvial to that of the three anchylosed vertebrae that precede it, and as in them its summit is marked by a ridge. The other parts of the vertebra were sufficieutly described in the former paper (p. 332, Plate XV. figs. 56-59). The foui-teenth vertebra (figured in the former paper, Plate xv. figs. 48-50) has its neural spine prolonged into a thick broad ridge, which does not reach quite so high as the coalesced spines of the three succeeding vertebrae. It sends forwards a long blunted process extending rather beyond the anterior edge of the centrum. The thh'teeuth vertebra, Plate xliv. figs. 3, I, has its neural spine developed in a similar manner to the fourteenth, but not to so great an extent. The walls of the * Phil. Trans. 1869, p. 332. The number ninclcen is not given in the text, but nineteen vertebra) are figured (Phatc xv.), and in the " Description of the Plates," p. 359, the words occur, "Twelfth (and last?) " cervical vertebra," " Penultimate (seventh ?) dorsal vertebra." no zoolo(;y of RODiJiorKZ. neural canal slopo inwards, and tlio roof curves downwards in tho centre, so that the ontline becomes sub-cordate. The anterior zygapophyses are elliptical, and nearly flat ; their surfaces slope inwards and upwards. The posterior arc sli2:htly concave and slope upwards and outwards. The transverse processes arc broad and flat and inclined downwards and backwards. In some specimens of this vertebra in tlie collection the canal for the vertebral artery is developed, but in only one instance on both sides. The hyi)apophysis resembles that of the fourteenth vertebra, but is smaller. Iri the twelfth vertebra, Plate xliv. figs. 5, G, the neiu-al processes have not united into a spine, but are present as two thin plates of bone, including a canal. The ti'ansverse processes are extremely wide and strong, and develop broad processes which extend downwards and backwards to meet the process sent up from the anterior edge of the centrum, and so form a large arterial canal. Bibs. Six appears to be the normal number of dorsal ribs, though m some examples, as in the large male skeleton, of which the pelvis is here figured, Plate xlvi. fig. 1, a seventh appears. The fii'st of these articulates with the fourteenth vertebra : the sixth with tlie nineteenth vertebra, which coalesces with the pelvis. The fii-st and second are without any corresponding sternal ribs, but the third, fourth, fifth, and sixth, had corresponding sternal appendages. The seventh, when present, bad a similar appendage, which failed to reach the sternum, but was connected by ligament with the sterno-costal rib inmiediately preceding. Only three or four examples of these ribs, attached in this manner to each other, have been found, even in the combined collections, and then only on one side or the other. The figure, Plate xlv. fig. 1, from a specimen in the British Museum, represents the structure in a very old individual, where the two sternal ribs have become anchylosed. In the former paper it was said (p. 331), " There appears to have been eight " paii's of dorsal and foxu* of sternal ribs. The first j)robably articulated with the " thirteenth vertebra, the last or eighth with the twentieth." This statement we now hold to be partially incorrect. It was formerly estimated that nineteen vertebrae were interposed between the cranium and the last dorsal vertebra, that which is anchylosed to the joelvis. We now iucHne to consider that there are eighteen only, of which we assign thirteen to the cervical region, and five to the dorsal ; so that there are six dorsal vertebrae in all, for we cannot count as normally dorsal that which we have found iu a few examples only, as above stated, with a rib attached to it. In Didus the number of ribs appears to be eight. The first two, as in JPezo- phaps, have no sternal appendages ; the next five have sternal appendages ON THE SOLITAIRE. — E. NEWTON ahu J. W. CLARK. 141 articulating -n-ith the sternum, and tlie last an appendage Tvhicli fails to reacli the sternum, but articulates with the sterno-costal appendage preceding it. This description is derived from a study of Professor Owex's figure.* In Didnnculiis there are only six ribs ; the first is free ; the second has a much attenuated sternal appendage corresponding to it, wliich articulates with the sternum, but fails to meet tlie rib by a slight interval ; the next three have sternal appendages articulating both to them and to the stemum, and the last has a sternal appendage attached by ligament to the similar appendage of the preceding rib. In the following Pigeons of which we have been able to examine specimens, the dorsal rib, articulating with and anchylosed to the first pelvic vertebra, has no proper sterno-costal rib, but only an appendage which joins its extremity, and is attached to the sterno-costal rib immediately preceding by ligament. The species are Phaps chalcop/cra, Phaps eleyans, Phlogcoios crinigera, Phlogmnas cruentata, Carphophaga microcerca, Caloenas nicobarica, Gcotrygon violacea, Starnoenos cyano- cepliala, Treron calva, Colnmba oenas, Diduncuhis strigirosfi^is. In Goiira and Pezo2)haps, however, this rib has its own sterno-costal rib. Out of all these species Pezop)hops,\ Pliajjs chaIcop)tera, and Goura coronata, alone present any evidence of a rib articulating with the second vertebra of the pelvis. We figm'e (Plate xlv. figs. 2-6), the vertebral ribs of the right side of a female skeleton. The uncinate processes are somewhat more slender and recurved than those of Diclus, but iu other respects the ribs bear a close general resemblance tc those of that bird. We also figure three of the sternal ribs on the same plate. Of these, the first and second, figs. 7, 8, are of the left side ; the third, fig. 9, of the right. The surface articulating with the sternum is marked a; that with the rib is marked b. Pelvis. The present collection formed by Mr. Jenner contains eight examples, three of which were probably males, and five females, in which the sacro-caudal vertebrae are complete. We find the number of these, as stated before, to be eighteen. The anterior portion of the pelvis was so complete iu the former spccimeus, that any fiu'ther description is unnecessary. Wc can now, however, describe and figure the posterior extremities of the ilium, ischium, and pubes, which before were unknown. The last six sacro-caudal vertebrae are of uniform width. The first three of these are separated from the ilia by a depression on the superior surface in older specimens, • Transactions of the Zoological Society of LonJon, vol. vi. pi. xv. ■f It will be obsen-ed that the first pelvic vertebra is the eighteenth of the series in Goura and Phaps ehgans, the nineteenth in Calcenas nicobarica, Phlogcenas cruentata, Carpophaga violacea, Pfiaps chalcop- tera and Pezophaps, and the twentieth in Phloyienas crinigera. 3k ii=2 ZOOLOGY OF RODRIGUEZ. and by a slight intorval iu yoimgcr ones. The last three are quite free, and the inner odgo of the posterior portion of the ilium bends away from them, first out- wards, tlien iiiwai'ds, then outwards again, and forms a blunted process at the termination of the ridge which separates the ilium from the ischium, Plate xlv. fig. 2. The extremity of the ischium was figured before, (Plate xviii. fig. 70) ; and the perfect specimens now before us show that the extremity, as there drawn, was hardly, if at all, broken. The assertion there made (p. 336), as to the " lower margin " sloping downward and outward, as if to pass and avoid the pubic style " is fully borne out by extended observation. In one specimen only, a very large male,* (Plate xlvi. fig. 1) they might possil)ly have met, as there is a trace of an articular surface on each bone (Plate xlv. fig. 13). The extremity of the pubic bone, perfect in two specimens, a male and a female, would appear to vary considerably (Plate xlv. fig. 13, Plate xlvii. fig. 1). In the former (the specimen mentioned above) the extremity turns round the posterior process of the ischium, and terminates bluntly at about a quarter of an inch beyond it. In the latter it turns inwards and slightly upwards on reaching the same parts, making a hook-like process about half an inch in length. The collection contains six caudal vertebrce, including the coccyx. It is of com'se quite impossible to say to how many individuals these belong. Their most noticeable peculiarity is the feeble development of the neural ridge, as might have been expected from the similar conditions of the sacro-caudal vcrtebroe, and the ex- traordinary shape of the coccyx (Plate xlvi. fig. 2.) Tliis bone is nearly rectangular, and tapers very slightly from before backwards. Its length is nearly equal to its height. From its inferior sm-face it sends forward a small process to pass under the centrum of the preceding vertebra. The drawing of a restored pelvis (Plate xlvi.) will show how completely the tail must have fallen within a line joining the pelvic bones. This must have been the case in Diclus also ; but in that bird the ilium was suturally connected with the entire series of sacro-caudal vertebrae, so that the caudal alone were free.f Sternum. We are now able to figure nearly complete specimens of this bone. In the former paper the anterior portions of the ** plcm'osteon" were described and figiu'ed, J but the posterior portion was broken in all the specimens. Immediately behind the articulations for the ribs is a short broad " metosteon," subject to considerable indi- * This pelvis has beeu placed on the mounted skeleton in the Museum of the University of Cambridge. I Owen, ut sttprci, Plate xix. J Plate xviii. Figs. 71-74. ON THE SOLITAIRE.— E. NEWTON axd J. W. CLARK. 443 vidual variation, but generally projecting out-wards and backwards. Behind this, the edge of the bone ciu'vcs gradually inwards, until near the posterior extremity, when it curves outwards again. There is a short blunt process on cither side of the extremity of the posterior edge, and a more or less deep notch in the centre, between the two ridges that mark the origin of the keel. Between these pro- cesses and the central notch the edge curves outwards. The form of these parts wiU be best understood from the figiu-es of a male and a female sternum, here given, Plates xlviii., xlix., figs. 1, 2. Individual variations, probably depending to some extent on age and sex, seem to be very great, and the median notch is in some specimens wholly absent. The costal border is, moreover, perfect in a sufficient number of specimens to place beyond all doubt or question the existence of four articular surfaces only.* Scapular Arch. The present collection, made by Mr. Jenner, contains three complete furculse (Plate xlvii. fig. 3, Plate, 1. fig. 7) and the fragments of three others. "We are thus enabled to decide that this bone is most thoroughly Columbine in form. We have nothing to add to the description before given of the scapula, coracoid, and the bones of the wing, for unfortunately no phalanges have yet been found. Bones of the Leg. The femur, tibia, and metatarsal have been already so fully described by Dr. Melville, Mr. Strickland, and Messrs. Newton, that we need not enter upon that part of the subject in this place. As, however, it is on the difference of size in these bones that especial stress has been laid by Strickland, and more lately by Professor Owen,f in their attempt to prove that there were two species of Solitau-e, which they designate respectively Fesophaps solitaria, and P. minor, we have carefully measured a very considerable number of specimens, as will be seen by the following table : — Presumed Males Presumed Females Femur. Tibia. Metatarse. E. L. R. L. R. L. 41 28 42 41 36 34 31 38 40 42 29 33 * Compare Newton, p. 338, with Owen, Trans. Zool. Soc. vii., p. 514. ■j- P.S. — Aug. 1878. — Still more recently (Ann. and Mag. Nat. Hist., January 1878), Professor Owen has tacitly admitted the error he had espoused. 3k 2 Uk ZOOLOtJY OF UODIJIGUKZ. la this mimbor of bones wo have before us th<3 remains of at least 12 specimens of caeh sex. The lUmensions of these are as follows, iu inches : — Pi-esuiued JInlcs Presimioil Females - Feiiuir. Tibiu. Metatnrse. Largest. 7-28 «-10 Smallest. 6 •(50 5 ■ 55 Largest. 10-90 9-30 Smallest. 10-20 8-35 Largest. 7-45 (il4 Smalkst. ()-80 .'■)-40 Femur. Tibia. Metatnrse. Diftereuce in length between largest and siunllost presumed male. Ditloronce in length between largest and smallest presumed female. DilVeienee in length between smallest male and largest female •68 •55 •50 ■70 •95 •90 -fi5 -74 •66 From this it will be seen that tliough the differences in length between tbe largest and smallest of each supposed sex are generally greater than the diiTerence between the smallest male and the largest female, and though there are bones of length intermediate between the largest and smallest of one sex, there are no bones of a length intermediate between the smallest male and the largest female, as would doubtless have been the case had there been more than one species on the island. The present collection contains twenty-eight ungual phalanges which agree in character with the three figures referred to above, but wc consider that it is almost if not t|uitc impossible to determine their exact j^osition on the foot. We believe, however, that the present collection contains those phalanges which were noted in the former paper as wanting, and we have accordingly reconstructed the foot of Fezopluips (Plate i., fig. 7) after Strickland's figure of that of Dklus.* Skull. Of the cranial portion of the skull, the collection includes seventeen specimens, together with fragments of eight others. Moreover, we had nine submitted to us by the authorities of the British [Museum. Of the twenty-six sj)ecimens that are more or less complete, twelve may be assigned to females, and nine to males. In all these, the depression of the central tract noticed in the former Memoir * We have come to the conclusion that some of the ungual phalanges formerly figured (Plate xx., Figi>. 113, 114, 125, 127) were incorrectlv referred to Pezuphaps. They belong to some species of Chelotiia, On that plate, Figs. 108, 109, and 1 lo, alone refer to Pezophaps. ON THE SOLITAIRE. — E. NEWTON and J. W. CLARK. 445 (p. 246), is most markod, subject, of course, to individual vaviatious, being in some quite flat, in others slightly concave, or divided into two portions by a low trans- verse tumescence. The occipital and frontal ridges are well marked in all ; but the degree of their elevation, especially in I lie case of the former, varies with age and sex; being occasionally raised into warty protuberances separated by deep depres- sions : and the sevei'ance of the occij)ital from the ex-occipital portion is always observed. The frontal ridge, or boundary of the central tract, is less elevated than the occipital. It is always divided into a more or less distinct central portion, forming the forehead of the bird (Plate 1., fig. 1, a.), and two lateral protuberances marked by more or less prominent exostoses (Ibid., h.), directed forwards and inwards, and in some cases even meeting in the median lino. In old specimens, the masoteric ridge always exists as an elevated line, forming the lateral boundary of the central tract behind the orbits. This ridge is well shown in the specimen figured in the former Memoh (Plate xxi., fig. 142, Plate xxii., fig. ]49). On making a vertical and longitudinal section of the skull of the Solitau'e, the extent of cancellous structure disposed between the external and internal walls of the cranium is seen to be of considerable extent, but developed very differently from the same tissue in i)ic7? bone bends abruptly outwards, then forwards and inwards to join the maxillary ; the inner edge, which is more vertical, is produced at its anterior extremity into a very delicate and sharp point (Ibid. inl was very incomplete, and apparently 3'ouug. "J. Caldwell, G Dec. 1877. " P.S. — Tliero are no stones of similar composition to these in the neighbourhood of any of the caverns -where Solitaij-e bones have been found. I should think the nearest place where fragments of l^asalt could be found -would be at least t-wo if not three miles from the cavern whore I found them." One of these stones lie ■was so good as to give to Mr. Edward Newton. It is here liguiTd in three aspects (Plato xlvii. figs. 4, 5, 6). It weighed, hci'orc it was cut for the pm-posc of determining its inineralogical natiu'c, a little over IJ oz. It is hro\A-n, somewhat rough, heavy, and hard. It is hardly, however, to be called " flat on one side," as Leguat describes those he had the opportunity of obseiTing. In connexion with this fact, however, it may be remarked that the bii'd Avith whose remains it was associated appears to have been young.* We have to thank Professor Bouney, M.A., F.R.S., for the following report on the microscopic structure and mineralogical composition of the stone : — " This rock externally presents considerable resemblance to a dolcrite. Such it is proved to be on niicroscoinc examination ; the slide shewing a crystalline mixture of plagioclase felspar, augite, olivine, and a peroxide of iron (? hematite). The plagioclase is well preserved ; the sections are commonly about six times as long as mde, and exliibit the characteristic twinning ; probably it is labradorite. Enclosiu'cs of opacite, augite (?), and other microliths, with minute gas-cavities, are frequent in some crystals, rare in others ; colours with polarizing apparatus fairly bright. The augite in the sections is of a pale j)uce-brown colour, rather rough in textm-e, and with the nicols fau*ly rich coloured. The olivine also shews brilliant colom's ; with ordinary light it is nearly colourless, excej)t where stained a Avarm brown through incipient decomposition; some crystals are thus rendered almost opaque. Eor this reason, and the absence of serpentine, I conclude this to be a feiTiferous variety approaching hyalosiderite. The grains of iron peroxide are not very numerous and are rather irregular in form. It seems most probable that they are hematite. These minerals are enumerated in order of frequency. The felspar, as is not unfrequent in doleritic rocks, is pierced in places by long acicular microliths, nearly colom-less ; some of which may possibly be apatite." * P.S., August 1878. — The stone was exhibited to the Zoological Society, 5th March 1878, by Professor Ne-wton. The preceding description is borrowed from the remarks made by him on that occasion. (Proc. Zool. Soc. Lond., 1878, p. 291.) ON THE SOLITAIRE.— E. NEWTON and J. W. CLiVRK, 451 Concluding Observations. The minute examination of the enormous series of specimens at our disposal enables us confidently to affii-m the " bold " statement made in the former Memoir (p. 330) : " There does not seem to be a single bone in the skeleton of Pezophaps solitarm which is not liable to greater or less individual variation of some kind or other; the individual vai-iation is not at all confined to absolute size ; it extends to the relative proportion of divers paiis of the bone, to processes or depressions upon them such as are commonly held to be specifieall}^ characteristic, so that it is often utterly impossible to predicate any definite limits of individual modification." We are not aware that the osteology of any vertebrate, other than man, has been studied -with the same wealth of materials as that of the Solitaire. About the won- derful variability of this particular species no doubt can exist. The bearings of this fact on the theoiy of Evolution will not here be entered into ; but that they will of necessity prove to be most important in futiu'e discussions of that theory, and of many of the questions arising from it, can hardly fail to be admitted. We have deviated so far from the example set in the former Memoir as in certain cases to give precise measm-ements of certain bones ; but it must especially be borne in mind that variations in length, breadth, and thickness form but a comparatively immaterial portion of the variations which exist. These are often, nay almost always, of a kind which cannot possibly be expressed in words, and could only be represented by a series of figiu*es almost equal in number to the specimens. The age of these bones has not hitherto been taken into consideration. Some of those found on a former occasion were considered by Professor Steenstrup (Proc. Zool. Soc. 1855, p. 718) to bear traces of having been broken by man, or by some predatoiy animal, in order to extract the marrow. In the present collection, however, we cannot find any examples of this kind. There is no evidence by which we can determine the age of the bones ; and the collection may include the remains of bu'ds which lived at a very remote as well as at a very recent period. It is to this cause that we ought perhaps to ascribe some portion of the wonderful varia- bility observable in the remains of Fezophajjs solitaria. P.S., August 1878. — In his recently published paper (Ann. and Mag. Nat. Hist., Januai-y 1878, p. 87), Prof. Owen remarks of the prsesacral vertebrte of ]?ezop1wps : — " So much of the vertebral formula thus accords with that of Bidunculns." Tliis is a tribute to the fidelity with which the late Mr. James Flower obeyed the directions of Professor Flower, which were that the skeleton of Fezoplwps, at fii'st deposited and mounted at the Royal College of Sm-geons, and subsequently transferred to the British Museum, where it was examined by Professor Owen, should be articulated in accordance \n\\\ the skeleton of Didnnculus, in which of course the number of vertebraj is known. 3l 2 452 ZOOLOGY OF ROnHlOUEZ. The Extinct Hevttles op Eodkiguez. i)'y Dv. A. Gilalhcr, F.ILS. Tho oarliost notice of the Tortoises and Lizards of Eodrii^uoz we find in Lkguat (Voyag:es ct Avantures). He says "that tliere are such pk>nty of Land-Tiirllcs in " Rodriguez, that sometimes you see two or three thousand of them in a flock, so " that you may go above a hundred paces on their backs." According to Admiral Kempinfelt, who visited the island in 1761 (sec Grant's Maurit. p. 100), small vessels Avere constantly employed in transporting these animals by thousands to ^Mauritius for the service of the hospital. But early in the present century the work of extermination appeal's to have been accomplished, and there is, at present, of the Rodriguez Tortoise not a single living example in the island, or in any other locality. Remains of this Tortoise had been discovered and had reached Europe many years ago, but no particular attention was paid to them. M. J. Dksjakdins, one of the first explorers of the fauna of ^lauritius, sent a bone of a Tortoise found in 178U in a cave in Rodi'iguez, with some remains of the Solitaire, to'Paris,* where they were examined by Cuvier and Blainville, who erroneously stated them to have been recently found under a bed of lava in Mauritiu.s.f Another Mauritian naturalist, C. Telfair, in searching, in 1832, for bones of the Solitaire in Rodriguez, succeeded in obtaining " numerous bones of the extremities of one or more large species of Tortoise," which were presented to the Zoological Society of London, and cxliibited at one of the meetings. J These bones were still in the possession of the Society three or four years before the publication of Strickland and Melville's memoir on the Dodo (1848) ; but no further attention being paid to them they were lost. Another portion of Telfair's collection was presented by him to the Andersonian Museum at Glasgow, where they are still preserved. Some well-preserved bones, kindly sent to the writer by M. Bouton, of Port Louis, in 1872, satisfactorily proved that the Tortoise of Rodriguez is distinguished from all its congeners by well-marked characters (Ann. & Mag. Nat. Hist., 1873, xi. p. 397) ; but it was only when these remains were supplemented by those jireserved in the Andersonian Museum at Glasgow, and entrusted to me by the curators of that institu- tion for examination, and when, finally, the extensive series collected during the Transit- of- Venus Expedition arrived, that our knowledge of its specific characters became tolerably complete. No further important additions can be expected from Rodriguez, with the exception of the small bones of the foot and caudal vertebrae ; * Proe. Comm. Zool. Soc. ii. p. 111. Strickland and Melville, " The Dodo," pp. 51-53. t Edinb. Journ. Nat. Sci. iii. p. 30. t Proc. Zool. Soc. 1833, p. 31. EXTINCT REPTILES. — A. GtrNTMEU. 453 and these will be but of small value, unless they be found in their natural con- nexion. With the aid of the carapaces brought home by Mr. Slater we are now enabled to recognize the Rodriguez Tortoise in some carapaces which reached Europe in the last ccntiuy, probably during the lifetime of the species, and which we find noticed by the following herpetologists : — 1. ScHOEPFP (Ilistor. Testud. 1792, p. 103, pi. 22, fig. B.) has reproduced a sketch of a Tortoise 2f feet long, which was communicated to him by Vosmaer, who examined the specimen which then was in " Museo Principis Arausionensis " in the Hague. This seems to have been a male, with a carapace very similar in form to that of the male described below ; its front and hind margins, being still provided with the epidermoid scutes, have an undulated outline. Schoepff was informed l)y Vosmaer that the carapace had been brought from the Cape of Good Hope ; and expressing himself uncertain whether it should be considered a distinct species, or a sexual, local, or individual variety of the Tortoise described by Perrault, he named it " Testudo Liclica, Vosmaer/." 2. Di'MEUiL and Bibron recognized Schoepff's Tortoise in a skeleton with comi:)lete carapace in the Paris Museum, The description of the specimen, whose .shell measured 75 centims. over the curvatui-e, again perfectly agrees with om- male specimen, and supplies a detailed account of the outer epidermoid covering. The authors adopt the binominal term, " Testudo vosmceri" which, of com-se, supersedes that proposed by myself {Testudo rodericeusis, Ann. & Mag. Nat. Hist., 1873, xi. p. 397). By the singular resemblance of the general form of the male of this species to that of some of the Galapagos Toi-toises, they were led into the error of supposing that T. vosmcBri came from the Galapagos Islands (Erpetol. Gen^r. ii. p. 140). 3. A second specimen, probably a young female, likewise in the Paris Museum, and without known history, was considered by the French herpetologists a distinct species, Testudo j^eltastes (ibid. p. 138). This description agrees in every respect with oiu' young carapaces from Rodriguez. The Rodriguez Tortoise* differs from the Mau.ritius and Galapagos Tortoises by the more slender build of all the various parts of its skeleton ; its neck must have been capable of still greater flexion, as is evidenced by the deep postapophysial impressions or actual perforations of the cervical vertebrae. Although careful com- parative measurement show beyond doubt that this Tortoise had longer limbs and a longer neck than even some of the Galapagos Tortoises, yet, taking also into consi- deration the extreme thinness and fragility of its carapace, we must infer that this general slendemess of the bones must have been partially due to the same cause, probably a diminished supply of the calcareous salts, or a diminished power of assimilation of them. * For the detailed illustrated description of these remains, I refer to " Gigantic Land-Tortoises (Liviug and Extinct). Lond. 1B77. 4lo." d54 ZOOLOGY OF KODRIGUEZ. - Tlic bones collected by tbc Naturalist of the Trausit-of-A''enus Expedition belonged to several hundred individuals ; and there are in some eases as nianyas 10 specimens of one and the same hone in the collection ; yet no variation in structure equivalent to that observed among the Galapagos, Aldabra, and Mauritius Tortoises could bo detected, so that evidently in this small island there was room for one species only. The only variation which is worth recording is one which can be explained as a sexual difference, the female having been of a smaller size and somewhat stouter form than the male, as is the case in the other Gigantic Tortoises. Bones far exceeding in size the majority of their kind arc not rare, and prove that the Eodi-igucz Tortoise was quite equal in bulk to Testudo clephantina, many (proba- bly male) individuals having had a carapace i^ feet in length. From Dum^ril and Bibron's descriptions we learn that the scutes were perfectly smooth or nearly so, and that the shell of the adult was black, whilst the young were of a lighter brown colom-, the sternum being dotted with yellow. Two kinds of Lizards appear to have been known to Leguat. He says (p. 108), " Les Palmiers et les Lataniers sent tons chargez de L6zards de la longuem- d'un " pied : on ue s^-aiu-oit se lasser d'en considerer la beaut(5. II y en a de noirs, de " bleus, de verds, de rouges, de gris, et detout cela du plus rif, et du plus 6clatant. " Lem- nourriture la plus ordinau-e est le fruit du Palmier. lis ne sent nullement " malsaisans, et sent si familiers qu'ils venoient souvent manger nos melons sur la " table en notre presence, et memes entre nos mains. lis servent souvent de proye " aux oiseaux, sur tout aux butors. Quand nous les faisions tomber des arbres, " avec une perche, ces oiseaux accouroient et venoient les engloutir devant nous, " quoi que nous pussions faire povir les en empccher ; et lors que nous en faisions " seulement le semblant, ils venoient de la memo mani^re, et nous suivoicnt " toiijours." " II y a une autre espece de Lezards nocturnes, de couleur grisatre, dont la figure " est fort vilaine : ils sent gros et longs eomme le bras, et la chair n'est pas mauvaise. " lis aiment beaucoup les Lataniers." The former of these Lizards may have possibly a species of Fhelsuma, a genus which is well represented in the Mascarene region, and the species of which are subject to great variation in colour. The latter is probably the species of which Mr. Slater collected, with remains of the SoKtaire and Tortoise, several bones. He recog- nized them as the remains of a Lizard, possibly belonging to the family of Skinks. In my opinion it is a Geckoid Lizard, which, as far as the evidence before us goes, cannot be separated from the genus Geclco ; but the species from Eodriguez appears to have attained a much larger size than Gecko veriis (to which it is very similar), or than any other Geckoid known. Hef erring, then, this Lizard to the genus mentioned, I concur in Mr. Slater's proposal of naming it after Mr. E. Newton, — Gecko neiotonli. EXTINCT REPTILES. — A. GONTHER. 455 The bones collected consist of two pariotals, posterior half of right ramus of lower jaw, right humerus, right half of pelvis, live left and two right femurs, and therefore must have belonged to at least five individuals, of which the one indicated by the pelvis was the largest. In the following description these bones have been compared with the skeleton of a Gecko vems, the vertebral column of which is 100 milluns. long (exclusive of the caudal vertebraj), and the skull 45 millims. Thejyarietal agrees in size and shape entirely M^ith that of G. vents, in which the two long processes into which this bone bifurcates behind are separated by a large vaciuty from the paroecipital. In Fhelsnma (which genus is so well represented in these islands and on the coasts of tliis I'l • ii'-i •111 •■ Parietal bone of geographical region, and which might have been expected to occur in Gecko newtonn. Rodriguez) the parietal has quite a different shape (P. seychcllense), and and'of dmS its posterior processes are addressed to the paroecipital. The articulary piece of the mandible differs nowise from that of G. verm ; like the latter it is produced behind the condyle into a hamate grooved process, which, however, is much more concave on its upper siu'face in the Rodriguez species than in G. verus. Humerus of Gecko newtonii, nat. size. A, anterior, B, posterior, C, ulnar, and I), radial aspects of bone. The humerus offers a more striking difference from G. verus than the preceding bones ; it is much stronger, and especially its extremities are comparatively much more dilated. Its head is transversly elongated, passing into a curved and project- ing prominence, which answers to the ulnar tuberosity. The radial crest is strongly developed, and does not extend beyond the proximal third of the length of the bone. The transverse diameter of the distal extremity is nearly rectangular to that of the proximal. The whole of this part of the bone is much dilated, particularly by a broad trenchant crest running along the radial border of the bone. Of the two condyles the radial one is much the more prominent one and projects towards the anterior side of the bone. In all these particulars G. neiotonii resembles G. verus, aU the ridges and prominences being, however, much more developed. Length of the humciiis Least Avidtli of the shaft Transverse diameter of proximal end Transverse diameter of distal end - G. verus. millims. 2 6i G. newtonii. millims. 25 3 9 10 •i5G ZOOLOGY OF KODIUGUKZ. A. Inferior surfncc of the pelvis of Gcclio verus. B. The outer aspect of the right os iiinoniinatiini of the same iiiiiinal. C. Lower surface of the right half of the pelvis of (!. nciftoiiii, aud D, its exterior surface. All the CgureB are of iinturul size. The ^Ji'/r/*" and y*:'^?;^ are SO similar to those of G. verus that the accompanying li-'m-es aud statements of measm-emeuts will suffice to give a perfect idea of those bones. Length of the os ilium (from acetahulnm) - Greatest ■uidth of os ilium (from acetabulum) Length of os pul)is (from acetabulum) Width of os pubis at its base Length of os ischii (from acetabukim) Least width of the os ischii - - - Length of femur . . . . Width in the middle Width of its lower extremity G. verus. G. newtonii millims. millims. 1] 13 i 5i 11 (injure 3 5i 6i 11 2^ 11 26 30 li 3 54 71 ' 3 0 C B A The femur of Gecko newtonii iu four different aspects and of natural dimeusion. Tee Extinct Mollusca of E,obe,iguez. The only extinct form as yet discovered is a land-shell {Helix hewsheriana), which was obtained in a sub-fossil condition in the bone caves by Mr. Slater. This species wUl be ao-ain referred to in the account of the recent mollusca of the Island. 457 EECENT FAUNA. Mammalia.— jB// G. E. Dobson, M.A., M.B., etc. The only iudigcuous mammal consists of a species of frugivorons Bat, described below, -which appears to have hitherto escaped the notice of zoologists.* Bones of the rabbit, rat {Miis decumamis), mouse, and pig (introduced by the settlers), Avere broxight home by the collectors, but they possess no special interest. Pteropus rodericensis. Dobson, Catal. Chiropt. Brit. Mus. p. 36. Ears slightly longer than the muzzle, but projecting by theu' upper third only from the long fur siu'rounding them ; ear-conch moderately broad in lower three- fourths, very abruptly narrowed above by flattening of the upper fom-th of the inner margin and a deej) concavity of the corresponding part of the outer margin, termi- nating in a narrow acute tip ; more than two-thirds of the concave sm-face of the conch is well clothed with moderately long hairs, the upper foui'th and the back of the ear naked or with a few short hairs only. Fur long and dense, in quality intermediate between that of Pt. eclwarclsii and Ft. vulgaris ; that covering the back of the head, neck, and shoulders very long, on the back shorter but not appressed, the hairs directed backwards, extending thickly upon the wings almost as far outwards as a line drawn from the proximal thml of the humerus to the knee, upon the femur, knee, and adjoining wing-membrane along the proximal third of the tibige, about two-thirds of which are thickly covered with straight hairs directed backwards, theii* distal thirds being nearly naked as well as the feet ; and, although the fur of the back extends upon the interfemoral mem- brane, it does not conceal its posterior margin, and its lower third, supported by the calcanea, is naked. Beneath, the antebrachial membrane is covered with long thinly-spread hairs, and similar hairs clothe the wing-membrane between the humerus and femur, and extend outwards in a broad band behind the forearm. Face reddish brown, with a few greyish or shining hairs ; chin and throat darker brown ; top of the head and nape reddish or yellowish brown, jDassing into a band of bright yellow, which extends across the back of the neck from shoulder to shoulder, and downwards on the sides of the neck and thorax, limited by a longitudinal band of * Mr. Gulliver, who obtained some of tlie specimens of this Bat, remarks : — " Insectivorous Bats are " entirely absent in Rodriguez, though they exist in Mauritius. At the latter place these Bats may be seen " in the caves, but in the caves at Rodriguez I never observed a trace of them, nor did Mr. Shiter, who, " from the constancy of his occupation there, could hardly have failed to see them had they existed," 3m -JiOS ZOOLOGY OF RODRIGUEZ. dai'k fiu" passing baclc wards from the chiu along the thorax to the ahclomen, whieh is elothod with (hirk brown fur, ol' which some hairs have greyish or shining extremities; hw of the back behind the shoidders dark In-own, the extremities of the hairs more or less yellowish. In the male the extremities of the haii's on the head and back are much brighter colom-ed than in the female. Fii-st upper premolar deciduous ; the second upper premolar and the second and tliii'd lower, also the first and second ixppcr molars and the first lower, have each a small but distinct posterior basal cusp ; even the small first lower premolar has an indication of a posterior basal cusp ; last upper molar very small and eircidar, not as large as the fii'st lower premolar, last lower molar slightly larger than the first lower premolar. Length (of an adult c? preserved in alcohol) : head and body 7"; head 2"-2; ear 0"-9 ; ear from tip of nostril 2" ; eye from tip of nostril, 0"-8 ; forearm l<"-9 ; thimib-metacarp, 0 "-35 ; ph. and claw, 1"'6 ; third finger-metacarp, 3"-3 ; 1'^ ph. 2"-4, 2"^'ph. 3"-55; fom-th finger-metacarp, 3"- 25; T' pli. 2", 2'«' ph. l"-85; fifth finger-metacarp, 3"-35, T' ph. l"-5, 2'«i ph. 1"-15 ; tibia, 2"- 25; calcaneum, 0"-6; foot, 1"-15. This small species resembles Pterojms ruhricollis, GeofEr., from the island of Mauritius, in size only. The ears project by their extremities beyond the fur of the head, and their margins are quite naked. In JPt. ruhricollis they are covered by long hairs and quite concealed by the fvir ; the fur of the body is also quite diiferent in textm-e, being much coarser and not in the least degree woolly, and the lower third of the tibia is naked. 459 BiEDS. — JBi/ E. Bowdler Sharpe, F.L.S., F.Z.S., etc.. Assistant in the Zoological Department, British Museum. Our knowledge of the recent Avifauna of the island of Rodriguez has hitherto been confined to a few memoirs pubHslied by ]\Ii'. Edward Newton, and Professor Alfred Newton of Cambridge. The former has given us an accoimt of a hvirried trip of a few days diu-ation made by him in 1864 (Ibis, 1865, pp. 146-154.), and it says much for his zealous collecting that the collections now received add very little to his account of the ornithology of the island, and confirm in nearly every instance the correctness of his identifications. Two indigenous species were then discovered by him, both of them pro-sing new to science, and these were described by Professor Newton as Foudia flavicans and DrymcBca ? rodericana. More recently he has described a Falceornis from this island as Palceornis exsul (Ibis, 1872, p. 33), and in 1875 a figiu*e of this interesting bml was published (Ibis, 1875, pi. vii.). The naturalists attached to the present expedition did not prociu*e a specimen, but Mr. Henry Slater saw one on one occasion only : — " This was on the 30th of September, " towards the south-western end of the island, where there is a good deal of wood, " and he could have shot the bird if he had had a gun with him ; but neither of his " companions was favoured with a sight of this expiring species, and no further " information about it could be obtained from the Creoles." (Newton, Ibis, 1875, p. 343.) Of the collections brought home by the naturalists to the Expedition, that of Mr. Slater is the largest, as he procured not only several specimens of the Foudia and Brymceca, but also a number of waders and sea-birds which are found on the island. Mr. Gulliver's specimens of the two indigenous species were very well preserved, and he also succeeded in obtaining their eggs. Among Mr. Slater's series were examples of the following introduced species, Psittacula cana, Acridotheres tristis, Francolimis ponticeriamis, and Numida coro- nata. 1. Bradypterus rodericanus, Drymseca (?) rodericana, Newton, P. Z. S. 1865, p. 47, pi. 1, fig. 3. Drymoica rodericana. Gray, Handl. B. i. p. 199, No. 2771. Adult male. General colour above olivaceous brown, very little paler on the upper taU-coverts ; lores pale yellow, extending above the eye but not forming a distinct eyebrow ; round the eye a ring of pale yellow feathers ; cheeks and sides of face pale yellow, the ear-coverts olivaceous ; under surface of body pale yellow from chin to vent, the sides of the neck and sides of breast olivaceous brown ; luider wing- coverts light yellow ; wing-coverts above brown washed with olivaceous Ukc the 3m 2 tr.o ZOOLOGY OF KODHUiUKZ. liac'k ; quills dark l)ro\vu, the secondaries odgccl \\i(li olivaiToiis, inclining to yellow- ish white on the outer margin of the primaries; tail-leathers hrown, narrowly margined with olivaeeous, th(> outer ones slightly tii)])ed with while; hill honi- hrown in skin, the lower mandihle yellow ; feet fleshy hrowu in skin. The foHowing are the measurements of the specimens sent, seven in number: — 'I'Dtlll LiMigth. Culmen. Wing. Tail. Tuisiis. 1 a. cf lul. Rodriguez, Nov. 1874 (Gullivoi) - (i-a 0 • (i.j 2-45 3-0 o-y-j b. ? ad. ,, i« »♦ • ' 6-3 0-6 2-35 2-95 0-85 C. (j 11(1. » It (Slater) 6-2 0-(55 2-45 2-85 0-95 ih cJ ad. )i >» >) " " 5-8 0-6.5 2-45 2-9 0-95 e. cf nd. 11 n i» " ■" 6-2 0-65 2-45 2-95 0-9 /. ? ad. >9 » » ' ' 6-0 — 2-35 2-9 0-8.5 1/. ail. » »> n ' ' 6-0 0-6,5 2-2;) 2-8 0-85 As is usual among the Warblers the female is smaller than the male. Tlie difference in coloration is iiot very great, ranging from a dark olive-brown to a clear olive or greyish brown ; llic shade of yellow also varies in individuals. The i^resent bird is a difTicult one to place in the natural system. In general plumage it is luilike a Dvyuurca, but, as Professor Newton observes {I.e.), it cannot be associated with the Cisticohc, which have ten tail-feathers. Its affinities must, therefore, be with Drymmca, Orthotomns, Frinia, &c., or Avith the Reed- Warblers, Acrocephalits, Bradypterus, &c. It cannot be referred to Orthotonms, as all the members of the latter genus have the tail shorter than the wing, or about equal to it in length. In true Drijnupca the tail is always longer than the wing, and so it is in Frinia, but in Apalis the wings and tail are about equal, so that I can hardly understand Mr. G. R. Gray's suggestion that D. rotlencana resembled Apalis. The shorter and broader bill, the minuteness of the rictal bristles, and the fact of the first primary being equal to the inner secondaries in length, all show that Apalis is not the genus to which the present bird could be referred, though in the " Handlist " a clean sweep is made of all the Grass-Warblers and theii- allies, which are sunk into one enormous geiuis Drymoica. It remains therefore clear that the Rodriguez Warbler must be referred either to Frinia, to Drynueca, or to the immediate vicinity of the Reed-Warblers. After comparing them I have come to the con- clusion that it should be referred to the last-mentioned group of Warblers, as its abnormally long bill and strong rictal bristles separate it off from Brymceca and Frinia. In the two latter genera the bill, though long, never exceeds in length the hind toe and claw, and is only equal to about half the tarsus. The well developed first primary prevents the admission of D. rodericana into Acrocephalns, and it therefore must be referred to the genus Bradypterus, of which it exhibits the generic characters, but of which genus it is rather an aberrant member, on account of its long bill and vellow coloration. BIRDS. — K. BOWULER SHARPE. 4G1 The uests of this species, obtaiucd by both naturalists, are very similar in structure, but one is rather larger than the others. They are both cup-shaped and neatly constructed, but firmly built, and the walls of the nest decidedly thick. The principal material used in their construction appears to Ijc fibre with a few grasses interwoven ; some leaves and shreds of cotton are also added. Mr. Gulliver brought an egg of this Warbler. It is something of a White- throat type, being white, thickly spotted, more particularly at the obtuse end. All over the surface of the egg are small underlying spots of pale purple, and toAvards the thicker end of the egg are larger clay-brown spots, and also good-sized spots of reddish brown. Length 0-7 ; diam. 0-5. 2. Foudia flavicans. Foudia flavicans, Newton, P.Z.S., 18G5, p. 47, pi. 1, figs. 1, 2. Ploceus flavicans, Gray, Handl. B. ii., p. 45, No. 6624. Adult Male. Above brown, slightly washed with olivaceous, the back broadly streaked with dark brown, the lower back uniform, the rump yellow, forming a band across it ; upper tail-coverts light olivaceous brown ; wing-coverts blackish brown, with greyish olive edgings to the lesser series, the greater and median coverts broadly-edged with whitish : quills dark brown, the primaries narrowly edged with yellowish, the secondaries more broadly margined with whity brown, lighter on the innermost ; tail-feathers ashy brown, all edged with olive yellow ; head and neck all round bright yellow, as also the breast ; the forehead, lores, sides of face, and cheeks orange scarlet, the throat also waslied with the latter colour ; feathers in front of and round the eye black ; rest of under surface of body light yellow, the flanks brownish ; under wing-coverts light ashy-brown. Total length 5-2 incbes ; culmen 0-6 ; wing 2-7 ; tail 2-1 ; tarsus 0-8. The series of male specimens is very complete, showing gradual transitional changes from the time when the back is like that of the female without any trace of the yellow band on the rump, the orange colouring on the head and neck bein» apparently gained by a moult. The yellow band across the rumjj appears to be assumed after the yellow head and neck are fully coloiu'ed. Of two males killed by Mr. Gulliver in November 1874, one has the band tolerably well defined, the other only shows its approach by the presence of a single yellow feather. Adult Female. General colour above brown, the head washed with olive, all the feathers mesially centred with dark brown, moi'e distinct on the back and scapulars ; rump uniform brown ; wing-coverts blackish brown, with grevish edgings to the least series, the median and greater coverts rather broadly edged with white ; quills dark brown, narrowly margined with olive yellow, the secondaries with lighter bro\vn ; tail-feathers dark brown Avith narrow olive-yellow margins ; loi-es yellowish ; car-coverts olivaceous brown with paler shaft-streaks ; cheeks and 1G2 • ZOOLOGY OF RODRKiUEZ. imtlcr sm-Aace of body light yellow, the sides of the neck ashy, the sides of the body light bro^^'n ; under wing-coverts light ashy brown ; bill brown. Total length, 5 inches ; enlmon, 0"55 ; wing, 2'G5 ; tail, 2'1 ; tarsus, 0-85. The size of tlie bill in this species varies very much in length and also in stout- ness. In some examples of the brown pluniaged birds the beak is black, and these I take to be yoimg males, as the black bill is accompanied by a certain bright yellow shade about the face and throat, and the under surface is rather brighter yellow. Tlu'ee nests collected are very interesting. That obtained by Mr. Slater is a lai'ge domed structure, composed entirely of fibre, with a few leaves and a little moss. The two obtained by Mr. Gulliver arc firmly fixed in the small twigs of a tree, and in addition to the domed structure which is exemplified in Mr. Slater's specimen, these two have a doorway Icadmg into the nest, and the probability is, therefore, that the nest which Mr. Slater found is not finished. With the exception of a few shreds of cotton and scraps of lichens the nest is entirely constructed of fibres. Tvro eggs taken by Mr. Gulliver in December 1871', were of a clear pale blue. Length, 0"75 inch ; diam. 0"55. 3. .ffigiahtis geofiroyi. .fflgialitis geofiroyi (Wagl.) ; Harting, Ibis, 1870, p. 378, pi. xi. Mr. Slater obtained a single female of this Plover, in winter plumage. It measures as follows : — Total length, 8"3 inches ; culmeu, 0-9 ; wing, 5'55 ; tail, 22 ; tarsus, 1-35. 4. Strepsilas interpres. Strepsilas intei-pres (L.) ; Finsch. & Hartl. Vog. Ostafr. p. 662. Six specimens collected by Mr. Slater, They are all in winter plumage, with here and there an appearance of the siunmer dress being donned. 5. Numenius phseopus. Numeuius phseopus (L.) ; Finsch. & Hartl. Yog. Ostafr. p. 739. The vernacular name of this bird in Rodriguez is " Corbiseau," according to Mr. Slater, who obtained one specimen in full winter plumage. 6. Butorides atricapilla. Ardea atricapilla (ASzel.) ; Finsch. & Hartl. Vog. Ostafr. p. 701. Two specimens were collected by Mr. Slater, an old female and a young biitl, the latter in the usual mottled plumage peculiar to these Herons. The old bird agrees with the African JS. atricapilla, and I must say that it surprises me to see that BIRDS.— R. BOWDLER SHARPE. 463 some authors would unite this African sijecies to the Indian B.javcmica. The latter is much darker, and is ashy-brownish where the African species is light grey, while the green shade of the upper plumage is always more hronzy. Mr. Slater also procm-ed an egg of this Heron. It is light gi-eenish blue, and is not quite so brightly coloured as one procured by ^Ir. Monteii'o from iVngola. Length, 1-7 in. ; diam. 1'2. 7. Sterna bernsteini. Sterna bernsteini, Schl. Mus. P. B. Sterna, p. 9 ; Gray, Handl. B. iii. p. 120, No. 11,060 ; Saunders, P. Z. S. 1876, p. 657. Thalasseus bernsteini, Blasius, J. f. 0. 1860, p. 81. Adult in icinter plumage. General colour above bright silvery grey, the upper tail-coverts wliite ; forehead, sides of face, sides of neck, and hinder neck piu-e white, forming an indication of a collar on the latter ; crown of head mixed black and white, the plumes of the nape long and pointed, forming a crest ; the ground coloui* of the feathers of the crown white, with a greater or less amoimt of black on them, sometimes forming only a dusky black streak, sometimes occupying nearly the entire feather, evidently indicating a change of plumage ; wing-coverts light silvery grey like the back ; quills slightly darker, silvery grey, white along the inner web, this colom' more extended on the secondaries, occupying nearly the whole of the inner web of the innermost, and extending round the outer web of the shorter secondaries ; primai'ies with the shafts pm-e white, the first quill blackish grey on the Older web near the base, shading gradually off into lighter grey towards the end of the feather, the other long primaries with a broad line of dark grey extending alongside of the shaft on the inner web, and widening out towards the apex of the same web ; edge of the wing pure white, as are also the outer webs and tips of the thumb-coverts (bastard wing) ; taU light silvery grey, rather whiter on the inner webs of the feathers, the centre rectrices whitish ; entire under sm'face of body, in- cluding the thighs and under wing and tail-coverts and axiUaries piu'e white ; bill (in skin) yellow, lead coloiu'ed towards the base of both mandibles ; feet blackish brown in skin. Total length, 16'8 inches ; culmen, 2-3 ; wing, 13'5 ; tail, 3'5, to outermost feathers, 5-5; tarsus, I'O. The description is taken from a bird obtained by ]Mr. Slater. Anotlier specimen obtained by the same gentleman is not quite in such good condition. It is marked a male, and has the following measm'ements : — Total length, 17"5 inches ; culmen, 2'^ ; wing, 13'0 ; tail, 4'0, to tip of outermost feathers, 6'0 ; tarsus, 1'05. Probably the dimensions of the tail-feathers may be a shade wrong, as both the bii'ds are moulting. A misprint has evidently occurred in the dimensions of the bill (I'S inch) given by Mr, Howard Saunders {I.e.). 1G4 ZOOLOGY OF IJODRIiiUKZ. The species seems to be a very distiuet one and apparently linds its nearest ally in S/enia hcryii. Prom this, howmor, it dilViMw in maiiy iin]ior(aiil particulars : — Sterna bergii. Sterna bernsteini. Tppcr snrjhci- dark slaty grey. Z'ppt'i' tail-cocerts and tail uniform with back. Plumes of the bastard wing- uniform ■fl-ith the other coverts. Vpper surface light silvery grey. Z^jjper tail-corcr(s and tall mueh lighter than tlic l)ack, and nearly Avhite. Plumes of tlic ha-^tard wing tipi)ed wilh white. Friiiiaries externally blackish, but 7'/'m«/v>A' externally blackish or grey, Avith a broad grey line intervening be- perfectly uniform, Avithout any lighter tween the shaft and the black outer i shade intervening between the outer web web. ' I and the shaft. Mr. Slater says that the vernacular name for this species in llodriguez is " Goilon." 8. Sterna dougalli. Sterna dougalli pront.) ; Saunders, P. Z. S., 1876, p. 053. Two specin^ens obtaintnl by ~Mv. Slater have the bill ])erfectly black, and measure as follows : — Total Length. Culnien. Wing. Tail. Outer Tail-feathers. Tarsus. a. Roilriguez {Slater) - 15-0 15-0 1-5 1 "oo 8 -Go 89 3-0 2-9 6-0 6'5 0-7 0-7 9. Sterna fuliginosa. Sterna fuliginosa (Gm.) ; Saunders, P. Z. S., 1876, p. 666. A pair of the large Sooty Tern were collected by Mr. Slater, and agree with the other skins from West Africa and other parts of the Southern Ocean. The measure- ments are as follows : — Total Length. Culmen. Wing. Tail. Long Tail-feathers. Tarsus. a. Rodriguez {Slater) - 16-8 16-7 1-65 1-7 11-6 U-6 3-4 4-1 6-1 6-8 0-85 0-8 10. Gygis Candida. Gygis Candida (Gm.) ; Saunders, P. Z. S., 1876, p. 067. BIRDS.— R. BOWDLER SHARl'E. 165 A pair of birds were eollected by Mr. Slater, who j^'ivcs the vernaeuhir name as " Goilette blauehe." The following are the measurements of these birds : — Total Length. Culnien. Wing. Tail. Tarsus. a. (J Rodriguez {Slater) . . . - 6. ? ,, » ... - 12-2 12-3 1-75 1-6 9-2 4-3 0-45 9-25 4-3 0-45 It will be seen from the above that the sexes in this Tern are very similar in size, but the female has a slightly smaller bill. They do not, however, show auy api^roacli to the slender-billed species G. microrhyncha of Mr. Saunders, of which the Museum has a specimen presented by Admiral Sir E. Belcher, and collected somewhere in the eastern seas. The white shafts to the quills and tail feathers in addition to the small size of the bird seem to l)e adequate distinctions for G. micro- rhyncha. Four eggs of Gygis were obtained by Mr. Slater, but the dates of capture are not recorded, nor the circumstances of nesting. Mr. Saunders writes {I. c. p. 669) : — " The nesting of Gygis is peculiar, the single egg of clay-white mottled with brown " being placed on the cavity of the branch of a tree, or in a fork of two branches, " and on the points of the coral reefs — anywhere, in fact, where it will lie." The fact of there being but one egg laid by these birds will account partly for the differ- ence in type exhibited by the four specimens obtained by Mr. Slater. No. 1. Ground colour clay-white, sparsely spotted all over, the underlying spots and blotches being of a pale purple, the overlying marks being larger and of a dark brown colom*, deepening on some to blackish brown ; the larger blotches seem to be congregated more about the larger end and centre of the egg. Length, 1"55 ; diam. 1-2. Ko. 2. Very different from the foregoing, being of a dull clay colour, the under- lying purple blotches being very large, and distributed nearly equally over the whole e^^, the brown mottlings taking the form of scribblings and wavy lines, confluent and forming blotches near the top and centre of the egg. Length, 1'5 inch ; diam. 105. No. 3. The ground colom' of this egg is dull clay-colour like tlie preceding, but the whole surface is covered with small underlying spots of purple, vdi\\ surface scribblings of dark brown all over the e^<^ ; in no place are there any large confluent blotches. Length, 1'65 inch ; diam. 1'2. No. 4. The ground colour in this specimen is clay-white as in the first described, but the markings are most like those of the last mentioned, Ijeing a collection of innumerable small spots and blotches of pale pui'ple as the underlying markings, 3 X IGG ZOOLOGY OF UODIUGUEZ. with an equally confused mass of dark broAvn sjiots and scribbled lines on the upper surface. There are a few broad blackish scribblings near the upper end of the egg. Length, 1-55 inch ; diam. 1'2. 11. Auous tenuirostris. Auotts tenuirostris (Temm.) ; Saunders, T.Z.S., 1876, p. G70. Adult Female. General colour above sooty-brown, rather darker on the rump nnd upper tail-coverts; wing-coverts sooty brown like the back, the least series blacker, as also those near the edge of the wings ; quills black with an obscure gloss of sooty-brown on the outer webs, the inner secondaries sooty-brown like the back ; tail uniform dark sooty-brown ; crown of head light grey, gradually shading off into the ashy chocolate colour of the entire hinder part and sides of neck; feathers between the bill and the eye light grey ; in front of the eye a black spot ; above the hinder part of the eyes and below the same a narrow line of wliitish feathers on the evelid ; sides of face chocolate brown, shaded with grey as they approach the sides of the neck ; entire under sm-f ace of body sooty-brown ; lower surface of quills ashy chocolate brown. Total length, 13-1 inches ; culmen, 1-4 ; wing, 8-7 ; tail, 4-li ; tarsus, 0'75. The name of this species at Rodriguez is " Malannc " according to Mr. Slater, who collected a single specimen of this small Noddy. It bears out the distinctions of jSIi". Saunders' paper [I.e.), but as in the latter essay the characters do not seem very easily made out, I have described the present species in detail from Mr. Slater's Rodriguez example. 12. Anous stolidus. Auous stohdus (L.) ; Saunders, P. Z. S. 1876, p. 669. One specimen collected by Mr. Slater {vide infra). Like the A. tenuirostris it is called " Malanne." Mr. Slater has also sent six eggs of this species. No. 1 is white, with faint pm'j)le spots and blotches distributed over its surface, but only distinct at the larger end, which is further slightly clouded by large blotches and spots of chesnut, with a few spots of black intermixed. No. 2 is very similar, but the chesnut spots are not distributed so thickly round the thicker end of the egg, but there is a large blotch of dark chesnut at the apex. No. 3. Very like No. 1, but more yellowish white, and having the upper end thickly clouded with chesnut and black spots, the underlying pTU'plish blotches beinsf rather more distinct. No. 4. Creamy white, the blotches large but sparingly distributed round the larger end of the e^^ in a ring, leaving the apex unspotted. No. 5. White with the blotches extremely large and confluent round the larger end of the e^^, the chesnut colour being rather pale and blending with the under- lying piu'ple colour. BIRDS. — R. BOWDLER SHARPE. 467 No. 6. "Wliite with very few markings, tlic underlying purple spots as well as the larger chesnut marks being distributed all over the egg. 13. Puffinus chlororhynchus. Pufl5nus chlororhynchus (Less.) ; Gray, Ilandl. B. iii., p. 103, No. 10,844. ? . Total length, 17 ; culmen, 1-4 ; wing, 10-1 ; tail, 5-4 ; tarsus, 1*7. cf. „ 16; „ 1-45; „ 11-2; „ 5-2; „ 1-8. 14. Phaeton flavu-ostris, Brandt ; Gray, /. c. p. 124, No. 11,096. An adult female procured by Mr. Slater. 15. Sula piscator, (L.) ; Gray, t. c. p. 126, No. 11,109. An adult female prociu-ed by Mr. Slater. " Booby." Note on A nous. — The following would seem to be the differences between the members of the genus Anous, of which the British Museum has a good series. I give herewith a Hst of the specimens at present in the national collection, with the dimensions of the wings in each case : — Key to the Species. a. Crown of the head grey or whitish, not uniform with the back. «\ Lores and feathers in front of the eye sooty-brown or blackish like the sides of the face. c(}\ Forehead and crown grey, gradually shading off into ashy-brown or chocolate on the neck, a^ii. Very much larger; wing, 10'2-11'5 inches. a*. Crown of head rather clear grey, whiter on the forehead and over the eye ; lores and sides of face greyish chocolate . . stolidus. b\ Similar to A. stolidus, the crown greyish brown, with a narrow super- ciliary line of white in strong contrast ; lores and sides of face dull chocolate brown ....... swperciliosfUiS, h^^. Veiy much smaller ; feathers in front of the eye black ; wing not exceeding 9 • 5 inches, c*. Cheeks and throat all blackish chocolate, darker than the head and sides of the face melanogenya. d*. Cheeks and throat sooty -gi-ey, Uke the neck . . phmibeigularis. 6". Crown and nape pui-e white, forming a cap .... Icucocapillus. b\ Lores and feathers in front of the eye grey like the crown. c^*. Underneath sooty-brown, as also the sides of the face . . . tomiroslrw. {cceriUetts. cinerevs h. Crown of the head uniform brown like the rest of the upper surface . . galapagensia. 1. Anous stolidus, L. ; Saunders, P.Z.S., 1876, p, 069. a. ad. Africa. Lieut. Friend, R.N. Wing, 10-2. h. ad. Gold Coast. R. B. Sharpe, Esq. Wing, 107. c. ad. Cameroons. Capt. R. Burton. Wing, 10-9. 3n 2 .168 zool(X;y or noDRinuFZ. (/. <■. nd. Ciipe Seas. dipt. Thomas Row Hany. / ail Reunion. Hcrr. J. C. Van Dam. Wing, lOi). g. a.l. „ ,. ., Wing, 11-2. //. cJ ad. Roihiguez. H. H. Slater, Esq. Wing, 11-1.5. V. ad. S. Piuir.s Roek. Antarctic Expedition. Wing, H".") /,-. ad. Moity Lsland. A. R. Wnllaee, Esq. Wing, 1 1 -2. /. ad. Raines' Islet. J. B. Jukes, Esq. Wing, 11-2. m. ad. At sea between New Zealand and A. q. ad. Ellice Islands. Rev. S. J. Whitmee. r. ad. North America. H.R.H. The Duke of Sus.sex. Wing, 10-fi. s. ad. North America. Purchased. Wiiig, 10'2. t. ad. Western Coast of America. Sir W. Burnett and Aihuiral I<'itzroy. Wing, 11*3. 2. Auous superciliosus, .sp. n. A. similis ^1. atolido sed pileo summo cinerascenti-brunneo, linca .supcrciliari allia distincta distingueudus. Long, tot., 122 ; culm., 1'6; abw, 10'7 ; caudiv, 5'3 ; tarsi, 01)5. Hab. — Coast of Central America and the Antilles. a. ad. Coast of Jlexico. G. Barclay, Esq. Wing, 10-75. b. ad. S. Ncvi.s. W. Cottle, Esq. Wing, 97. c. ad. Jamaica. Purchased. Wing, 107. (1. ad. " Caught at sea in long. 91 E., lat. 8 N." R. B. Sharpc, Esq. Wing, 9-7. 3. Anous melauogenys, Gray ; Saunders, t. c, p. 670. a. Locality unknown. Captain Lord Byron. Wing, 89. b. ? . Moreton Bay. J. Macgillivray, Esq. Wing, 895. 4. Anous pluinbeigularis, sp. n. A. ubique cinerascenti-cbocolatinus ; capite sumrao canescente ; macula supra-et infraoculari albida ; plumis anteocularibus nigris ; loris, facie latei'ali tota clarfe cinerascenti-chocolatinis, corpore reliquo subtus chocolatino-brunnco ; subalaribus cinerascenti-chocolatinis ; tectricibus alarum superioribus cliocolatinis, remigibus rectricibusque nigricantibus. Long, tot., 14*4 ; culm., 1'55 ; alse, 9-8; caudse, 58; tar.si, 0-9. Hab.— B.i^d Sea. Ml-. Saunders hints that the Noddy of the Red Sea is probably incorrectly determined, and I have therefore examined a specimen which we have in the Museum from that locality. I find that the bird is apparently' distinct from A. melanogcnys, differing in its clearer grey face and throat, which are not so black ; the wing is also an inch longer, the back rather lighter brown, and the tail chocolate brown without any grey shade. The size of the bird is not nearly large enough to allow of its being confounded with A. stolidus. a. ad. Red Sea. J. Middleton, Esq. 5. Anous leucocapillus, Gould; Saunders, t. c. p. 670, pi. LXI„ fig. 3. a. Off Bristow Island, S. coast of New Guinea. J. Macgillivray. Wing, 8-5. h. Off Barnard Islands. J. Macgillivray. W^ing, 86. BIRDS.— K. BOWDLER SHARPE. 469 c. Locality uukiiown. ^Vi^g, 0"3. d. „ R. B. Sliarpe, Esq.* e. OS' Western coast of South America. R. B. Sharpe, Esq. /. Ellice Islands. Rev. S. J. Wliitmee. 6. Anous teuuirostris (Temm) ; Saunders, t. c. p. G70, pi. LXI., fig. 1. a. ad. Mauritius. R. B. Sharpe, Esq. Wing, 8-15. b. ad. Rodrig-iicz. H. H. Slater, Esq. Wing, cS-7. c. c? «id. South Island, Houtiuann's Abrolhos. J. Gould, Esq. Wing, 8-7. 7. Anous cinereiis, Gould; Sharpe, P.Z.S. 1878, p. 271. a. b. ad. Australia. Sir T. Mitchell. Wing, 8-4.5. 8. Auous csenileus (Bennett) ; Sharpe, P.Z.S. 1878, p. 272. «. b. ad. Ellice Islands. Rev. S. J. Whitmee. 9. Anous galapagensis, sp. n. A. similis A. stolido, sed rostro minors et capita suramo fuliginoso unicolori distinguendus Long. tot. 13-4; culmen, 1-4 ; aire, 10-6.5 ; caudpe, o-5 ; tarsi, 0'8.5. Hab. — Galapagos Archipelag(.i. rt. ad. Dalrymple Rock, Cliathaui Island. Mr. Saunders thinks that this may be a young A. stolidus. I believe the bird to be quite adult, and to constitute a distinct species. * Said to have been procured oft" the Gold Coast, but the locality is probably erroneous. 470 ZOOLOGY OF RODRIGUEZ. Reptiles.— i?i^ Dr. Albert Giinther, F.R.S. Numerous specimens of a Gecko, Feropns mufihifus, were collected by Mr. GiUliver. This small species is not only generally {listril)utc(l in the Mascarcne Islands, but occm-s also in numerous islands oi' the East Indian Archipelago. Mr. E. Newton has received quite recently a specimen of Phelsuma cepedianum, which was caught by ]Mi\ H. E. Desmarais on the 1st of July 1876, in the jungle of RocU'iguez. This species is common in Madagascar and almost all the Mascarene Islands, but seems to be very scarce in E/odriguez. Fishes. — By Dr. Albert Giinther, F.B.S. A. — Marine Species. The marine fish fauna of Rochiguez does not differ in any respect from the well- known fauna of Mauritius, the Seychelles, or Zanzibar ; and the species collected by the natm-alists of the Exiicdition being not uncommon species, it will- be sufficient to enumerate them in the following list : — Holocentrum diodema, Lacdp. „ spinifermn, Forsk. Serranus Jiexagonatiis, Forst. „ Jlavocccrnleus, Lac6p. Blesoprion fidcijlcnnma, Forsk. Dules marginatus, C. & V. Fentapus aurolineatus, Lac^p. Mulloides JlavoUneatus, Lacep. Lethrimis mahsena, Forsk. „ nebidosns, Forsk. Chri/sojyJiri/s hifasciato, Forsk. Chcetodon setifer, BI. „ lunula, Lac6p. „ vagabundus, L. ,, Uneolafus, C. & V. FLolacantlms alterncms, C. & V. Menioclnts monoceros, C. & V. Fsettns argenteus, L. Caranx crumenop)hthalmus, Bl. ,, melmnpxjgus, C. & V. Gobius albopunctatus, C. & V. Scdarias quadricornis, C. & V, Teuthis nehnlosa, Q. & G. Naseus unicornis, Forsk. Acanthurus desjardlnil, Benn. „ matoides {blochii), C, & V. „ strigosiis, Benn. „ triostegus, L. Atherina pinguis, Lacep. Foniacentrus trilineatns, C. & V. Glypihidodon sparoides, C. & V. „ ccelestimis, C. & V. Dascyllus aruanus, L. CMlinus punctatus, Benn. Anampses diademed us, Biipp. Fsendoscarus harid, Forsk. Flotosus anguillaris, Bl. Tetrodon nigropunctatus, BI. Balistes aculeatiis, L. „ rectangidus, Bl. Ostracion cubicus, L. Diodon hystrix, L. FISHES— A. Gt^NTHER. 471 B. Freshwater Species. Also the fresliwater fishes (wliich were collected by Mr. Giillivcr) are species well known to inhabit or to freely enter fresh waters of the Mascarenes and other parts of the western shores of the Indian Ocean. Two Grey Mullets, however, differ sufficiently from those of the neighbouring islands to be entitled to specific distinction. 1. List of the Species. Dules marginatus, C. & V. Mitgil rodevicensis, sp. n. „ rupestris, Lacep. Myxus ccBCutiens, sp. n. Chrysoplirys sarha, Forsk. Megcdops cyprinoides, Brouss. Fsetttis argenteiis, L. Chanos salmonoides, Bl. Gobms cdbopnnctatus, C. & V. Anguilla mauritiana, Bemi. „ ocelkrris, Brouss. „ virescens, Ptrs. Eleotris monteiri, O'Shaughn. 2. Description of the Grey Mullets. Mugil rodericensis. Allied to M. troscJielil, smithii and compressiis, but with a narrower and longer caudal peduncle. D. 4i A. I, L. lat. 30. L. transr. 11. The height of the body is a little more than the length of the head, and one fom-th of the total (without caudal); the head is two-thirds as high as long; the diameter of the eye is one-fourth of the length of the head, and contained once and three-fourths in the width of the interorbital space. Eyelids not deyeloj)ed. Upper profile of the head and nape nearly straight. Prseorbital scaly, emar'^inate, and finely denticulated. Snout a little shorter than the eye ; the maxillary extends beyond the prseorbital, and its extremity is uncovered ; upper lip thin. There are twenty-one scales between the snout and the dorsal fin. The least depth of the tail is less than one-half of the length of the head, and the caudal peduncle is consider- ably longer than deep. The spinous dorsal is higher than the soft ; the spines are strong, the length of the first exceeding that of the postorbital portion of the head • the base of the first spine a little nearer to the root of the caudal than to the end of the snout. The soft dorsal and anal nearly entii-ely scaleless ; the anterior third of the anal in advance of the origin of the soft dorsal. Caudal fin forked. Pectoral shorter than the head. Coloration uniform. One specimen 10 inches long, and several young ones were collected by Mr. Gulliver. 172 ZOOLOGY OF KODKltiUEZ. Myxus cceciitieus. D. ^, A. ^, L. lat. 43. L. tninsv. Ik Tcctb very small. mov(\a1)ii', l)onl, tlioso ol" lli<> upper jaw in a sini^'le scries; a notch in the micUUe of the upper jaw to receive the mandihular sympliysis. Lower jaw with a similar scries of horizontal teeth ; oilier smaller teeth hehind a])poar to be destined to replace those in runction. Lower sm-iace of the mandible without transverse folds. Two round, hard, apparently toothless, naked patches on each side of the palate. The maxillary does not quite extend to the front margin ol' the eye. Snout ol)tuse, shorter than the eye, wbicli is two-sevenths of the length ol" the head, and two-thirds of the width of tlic interorbital space. Eye with a broad anterior and posterior adipose eyelid. Tlie depth of the body is nearly equal to llie length of the head, which is two-sevenths of the total length (Avithout caudal). Pectoral extending to the commencement of the spinous dorsal, which corresponds to the tliirteenth scale of the lateral line. Dorsal spine moderately strong, the length of the first being more than half the length of the head. Caudal fin deeply emarginate. Coloration uniform. Two specimens, 5 inches long, were collected by Mr. Gulliver. 473 MoLLUSCA. — By Edgar A. Smith, F.Z.S., Assistant Zoological Department, British Museum. (Plate LI.) The following list comprises aU the species of mollusca collected at the Island of Rodriguez by Messrs. G. Gulliver and H. H. Slater. The marine shells, with one exception {Bhizocheilus monodonta), were obtained by the latter. They are for the most part very common forms, which arc found abundantly in the Indian and Pacific Oceans, on the shores of the Mauritius, Ceylon, Andaman, and Seychelles Islands, Philippines, North Australia, Samoa, Fiji, &c. The terrestrial and fluviatile molluscan fauna, as might be expected, and as shown by M. Crosse, assimilates most closely to that of the most adjacent land, the Mauritius, Boiu-bon, Madagascar, and the Seychelles, a few of the species being common to Rodriguez and these localities. The living terrestrial and fluviatile moUusca hitherto recorded from this island are 25 in number. To these are now added five others, and a species of Bit limits which in all probability never lived on Rodriguez, but has l)een conveyed there by chance, or may have been cast overboard from a ship and washed on shore, where it was picked up I believe, since it was packed up with marine shells collected on the beach. It "will be noticed how few of the species are quoted as Madagascarian ; this no doubt simply arises from the fact that this country has been so little investigated in regard to its molluscan fauna. The whole collection consists of 78 marine species, represented by 2C0 specimens ; 16 terrestrial forms, consisting of 1,149 examples ; and six fluviatile species, which include 574 specimens. The following species not obtained by the Transit of Venus Expedition are mentioned in M. Crosse's paper, entitled " Eaune malacologique terrestre et fluviatile de rile Rodriguez," published in the Journal de Conchyliologie, 1874, Vol. XXII., pp. 221-242: — 1. Pupa (Gonospira) metableta, Crosse; 2. Omphalotropis tce- 7«'«^rt, Crosse ; 3. O. littorinula, CvossG; 4. Truncatella teres, VMScr ; 5. Melania commersoni, Morelet ; G. 3Ielania tuberculata,Wyx\\Qv; 7. Ncvitina caffra, Gray; 8. N. mauritiana, Morelet ; 9. Strejdaxis distortns, Jonas ; 10. S. piriformis, PfeifEer. 3o 474 ZOOLOGY OF EODKICtUEZ, List of tee Species. I. MABms. Previously foimd elsewhere. Conim hcbrcBus, Linn, „ tessellatus, Born. „ catiis, Hwass - „ miles, Linn. „ pennaceus. Born. „ lividus, Hwass „ rattus, Hwass „ balteatus, Sowerby „ cei/laiieiisis, Hwass Sistrum arachnoides, Lamarck „ mortts, Lamarck „ mnticiim, Lamarck „ horrridum, Lamarck - Ehizochilus ( CoralUophila) monodoiifa,Qnoj & Gaimard Khizochilus ( CoralUophila) sqitawosissimus, sp. nov. Harpa impcrialis, Linn. „ iiiinor, Lamarck Fasciolariu trapezittm, Linu. Latirus pohjcjoims, Gmelin - „ nassatulns, Lamarck Strombus gibberulus, Linn. - floridiis, Lamarck - » elegans, Sowerby - samarensis, Chem- nitz. X A « c "3 Other Localities. a ■J. r. M S. = D 3 ■t ;^ n s < O Cape of Good Hope, Zanzibar, China Sea, N. Anstralia, Fiji, Sandwich Islands, Samoa, Society Islands, Aneiteum, Tonga Islands. Mozambique, Rod Sea, Philippines, N. Australia, Fiji. Tahiti, Hardy's Island. Pliilipi>incs, Solomon Islands, Fiji. Persian CUdf, Philippines, Moluccas, Sandwich Islands, Tahiti, Society Islands. Cape of Good Hope, Red Sea, Philip- pines, Sandwich Lsl.ands, Solomon Islands, Fiji, New Caledonia. N. Australia, Annaa Island. Philippines. Red Sea, Sandwich Islands, Lord Hood Island, Niue, Red Sea, Natal. Lord Hood's Island, Samoa, N. E. Aus- tralia, Sandwich Islands. Lord Hood's Island, Cape of Good Hope. Mozambique. Philippines, Port Essington, Society Islands. Fiji Islands, " Tonga Tabu " ( Q,uoy). Red Sea, Mozambique, Philippines, Society, and Navigator's Islands. Red Sea, Philippines, Solomon Islands. Moziimbiquo, Philippines, N. Australia, Tonga Islands. Cape of Good Hope, Philippines, Port Essington, Fiji Islands. Red Sea, Cape of Good Hope, Zanzi- bar, Philippines, New Guinea, Swan River, Fiji, New Zealand, Society Islands. Philippines, Society Islands, Sandwich Islands, Red Sea, Cape of Good Hope, Zanzibar. Red Sea, Philippines. Philippines. MOLLUSCA. — EDGAR A. SMITH, 475 •a o 1 o -§ 3 Previously found elsewhere. a •a n •| i Other Localities. Pterocera chiragra, Linu. Tritonium cynocephalum, La marck. „ tuberosum, Lamarck Cassis ribex, Linn. - Cerithium echinatum, Lamarck „ ( Vertagus) aspertim. Linn. Natica maroccana, var. Chem' nitz. N. {Rumd) melanostoma, La- marck, „ simicB, Chemnitz Janthina communis ? Lamarck Nerita plicatu, Linn. „ polita, Linn, „ ulbicilla, Linu. Fenella reticulata, A. Adams Cyprma tigris, Linn. „ arabica, Linn. „ mappa, Linn. „ mauritiana, Linn. „ vitellus, Linn. „ carncola, Linn. „ isabcUa, Linn. „ caput-scrpentis, Linn. „ nucleus, Linn. „ helvola, Linn. „ lamarckii. Gray „ moiieta, Linn. „ annulus, Linu. Dolium perdix, Linn. Amalthca acuta, Quoy & Gai- mard. Turbo margaritaceus, Linn. - Trochus (Cardinalia) virga- lus, Gmelin. Trochus {Jnfundibulum) deli- ratiihis, Philippi. Timor, Australia, Solomon Islands. Philippines. Sandwich Islands, Aunmi, Fiji. Philippines, Samoa, Pacific Islands, Annaa, Society Islands, Cape of Good Hope. Red Sea, Mozambique, N. Australia, Fiji, Louisiade Archipelago. Natal, Mozambique, Red Sea, N. Aus- tralia, Philippines, Fiji, Society Islands, W. Africa, W. Indies, Cah- fornia. Red Sea, Mozambique, Pliilippines, Darnley Island. Phihppines. Natal, Mozambique, Zanzibar, Round Island, N. Australia, New Guinea, Fiji. Red Sea, Cape of Good Hope, Zanzibar, Philippine Islands, Sandwich and Solomon Islands. Hardy's Island, Zanzibar, Cape of Good Hope, Philijjpines. Gulf of Suez, Japan, Mozambique, Samoa, Tongatabu, New Caledonia. Natal, Mozambique, Red Sea, N. Aus- tralia, Upolu. Annaa. Navigator's Island, Tongatabu. Fiji Islands, Cape of Good Hope. Natal, N. Australia, Fiji, Sandwich Islands. Sandwich Islands. Cape of Good Hope, Mozambique, New Zealand, Tahiti, Savage Island. Philippines, Borneo, Savage Island. Mozambique, Sandwich Islands, Cape of Good Hope. Mozambique. Fiji, Sandwich Islands. SaudOTcIi Islands, Cape of Good Hope, Samoa Island, " Florida " ( Green). " New Ireland " ( Qttoy), " Natal," Krauss. Philippines, Tongatabu. Kud Sea, Philippines. India. 3 0 2 -17G ZOOLOGY OF HODinr.UEZ. ^ Previous o .s> a 1 ilis, Kranss Pinna muricata (Linn.), Reeve. Pccten imbricattis, rar. Gmelin. Eastern Seas. Red Sen, Philippine Is., Tahiti, I'"iii, New Zealand. Philippine.^. ll|)olii, Aiifili'aliii, I'liilippine Islands, Red Sea. Indian Ocean, C'liina. Indian and Pacific Oceans, Philippines. Red Sea, Pliilippines, China Sea, Port Essington [West Columbia, AVt'.]. Zanzibar, Nicobar Islands, Australia. China " Cape of Good Hope " {Jai/). "New Caledonia" (Souvcrbie). rCape of Good Hope [Krauss). < N. Australia, Solomon Island.s, Persian L (iuir, Society and Sandwich Islands. Red Sea. Red Sea, East Africa, Zanzibar. Natal, Zanzibar, Red Sea. [W.Indies, /^fc.]? Red Sea. 11. FLm-IATILE. Melania scabra (Miiller), X X X Reeve. Melania rodericensis, sp. nov. X Planorbis rodericensis, Crosse Niritina gagates, Lamarck - X X Sumatra. „ lo7igispina, Recluz - X Navicella porccllana, var.. X X X Philippine Islands. Linn. MOLLUSCA EDGAR A. SMITH. 477 III. Tehrksteial. 1 N 3 1 o 1—* Previously found elsewhere. CD 3 c 1 1 CO 1 1 i 1 1 1 □ < d o 1 Other Localities. Helix bewsheriana, Moielet - X „ rodericensis, Crosse X „ sp. - - „ similaris, Ferussac X X Brazil, Cuba, Natal, Bengal, Java, Japan, Sanilwich Islan( China, 5. Bulimus Jihratus var., JIartyn New CaU'doiiia. „ i^Stcnogyra) yracUis, Hutton. India, Siam, Archipelago. Pupa lienardiana, Crosse X „ desmazurcsi, Crosse X „ {Gibbus) rodericensis. X Crosse. Pupa ( Gibbus), chloris, Crosse „ ( „ ), palanga, Lesson. X X Succinea nevilli, Crosse X X Vaginiila rodericensis,'3p. uov. X Cyclostoma articulatum. Gray '„ hamastoma var. X X? Round Islaud. Anton. Omphalotropis littorinula, Crosse. X Remarks on various Species. Conus lividus. Both forms of this species, the coronated {lividits) and uon- coronated (C Jlavidus Lamarck), were collected at Rodriguez. Of the former the specimens are strongly tubercled and broad-shouldered, clothed with a brownish olive epidermis and encircled around the lower half of the whorl by about eight nodulous, but only faintly elevated lii-ations. The latter variety also has similar lirae, and agrees precisely with certain specimens in the British Museum, from the Fiji and Solomon Islands and Port Essiugton. Conus ceylanensis. The single specimen of this species is in very bad coudititm, but is interesting on accoimt of its unusually broad form, very similar to that of C. sj)onsaUs. It has a purple aperture, and is painted exteriorly with spiral series of reddisli dots. Conus balteatus. Plate LI., fig. 12. The shells which I have referred to this species display considerable difference in form from the normal type. They are much shorter and consequently more briefly conical and even somewhat pyriform. The tubercles of the spire arc also more remote from one another. The mouth is deep 478 ZOOLOGY OF RODRIGUEZ. purple ■within, and wlutc at the upper and middle parts of the lip. The exterior surface is obscurely two-banded with brown, variegated with a few opake white dots here and there, and is clothed with a very thiu yellowisli-olive epidermis. Couiis rattus. Plate LI., llg. 11. Of this species, besides two fairly typical specimens, there is a third one iu the collection which is a very curious short shell apparently belonging to, although differing considerably from, this species. It is 30 millimeters long, and 18 in breadth and partakes much of the oiitline of C. spomalis. It also shows considerable affinity to C. haUeatus and, indeed, it is possible that it may be a variety of it. From this species it dilVers in having the spire blotched with brown and not wholly white, and it lacks the rosy apex. It is, however, coronated, which is very imusual in C. rattus. The body-whorl is broadly two-banded with brown and dotted with opake white and the aperture is deep pm-plc brown. Comis catus. Plate LI., fig. 13. A single specimen of this species is remarkable on account of its uniformly pink colour. Latii-us nassatnlus. The pretty example of this species agrees in all respects with the normal form, with the exception of colour being in every part perfectly white. Sistrum arackuoides. The only specimen of this species is a very fine large representative of it. It has fine large bold spines with the interrupted band on the labrum and columella of a rich orange. Sistrum horridum. This shell also, like the specimens of S. mortis, is thickly coated with lime, and is remarkable for the tliickncss of the labrum and the development of the teeth, which form an almost uninterrupted scries of eight, which are about equal in size, and the fom* on the coliunella are also very stout, and the aperture is consequently very narrow. Fenella reticulata. Plate LI., fig. 6. Besides the three lirations which encu'cle the upper whorls and are nodulous on crossing the longitudinal costse, which are about 14 in mmibcr in each whorl, there is a fourth fine thread-like lira around the upper part which borders the sutm'e and is only slightly nodose. The body- whorl below the fom- above-mentioned lirations is ornamented with four or five others (" liris numerosis !" Adams' Annals Nat. Hist. 1860, VI., p. 422) which generally are not nodulous like the upper ones, as the longitudinal ribs usually terminate at the fourth lira from the sutm-e. The largest example in the British Museum is from Japan, and has a length of 4 millim, and its last whorl is \ broad ; aperture 1 long and f wide. Specimens from Suez, the Persian Gulf, and Rocbiguez, have invariably rather smaller dimensions than the preceding. Janthina communis ? The specimens of tliis genus from Rodriguez have a rather elevated form, being well represented by the figure la in Reeve's Monograph in the Conchologia Iconica, to which he has given the name J. roseola, which, it is impossible to say definitely, may or may not be distinct from the common species. MOLLUSCA.— EDGAR A. SMITH. 479 Amalthea acuta. The single sliell from Rodriguez wliicli appears to be hut a variety of this species, differs slightly from the description and figui-e given by Quoy in the Voyage of the "Astrolabe." It is slightly more elevated and the margin is more or less stained with brown. It is a very stout, solid shell, and coarsely radiately ridged exteriorly, white and stained with reddish-brown, especially towards the apex. Natica maroccana. One specimen which is in a somewhat broken condition agrees almost precisely with others in the Museum from the Red Sea and Sandwich Islands. The sutm-al wrinkles are well marked. It is of a pale livid grey colour with a white zone beneath the sutm'e and another above the middle of the whorl, the base of which is also white. The aperture is brown banded. Strombus elegans. A curious example of this species was obtained by Mr. Slater. It is a small shell about an inch in length, whitish, with a few brown spottings ; the columella is deep red, and the upper and lower parts of the labrum are stained with the same colour. On the last whorl the longitudinal phcso become very strongly nodulous on the dorsal region, and are not prolonged downwards, and the transverse stiiation is particularly well defined. Strombus floridus. There are two dwarf specimens of this shell, only ten lines long, having a very thick labrum, and the thick callous of the columella is scarcely striated at all. They are white with only a few brownish markings. Pinna muricata, Reeve. The specimens of this species from Rodriguez do not appear to differ in any very material respects from that said to come from the West Indies, and described by Reeve as muricata of Linneus. However, there are one or two slight distinguishing characters in them, but not sufl6.cient to separate them specifically. The ribs on the ventral half of the valves which are scale-bearing, do not extend so close to the margin ; the dorsal edge is more arcuated and the base is also slightly sinuated and not convex, as in the shell figured by Reeve in the Conchologia Iconica, fig. 23. Helix rodericensis. The specimen of this species originally described was not in a perfect condition, as no mention is made of the beautiful epidermis which should adorn it. Two of the best preserved shells collected by Mr. Gulliver display a very pretty brown epidermis, wliich is tluck and almost lamellar, being elevated into close-set oblique shaggy laminae, which have here and there one or two somewhat pointed elevations, which produce the appearance of the whorl being spirally ridged. On the body- whorl there are about seven of these pseudo-ridges, three above and four beneath the periphery, and the oblique riblets or lamina? do not extend very far over the lower surface of the whorl. The height of the spire in this species is subject to variation, as some specimens are much more depressed than that depicted in the Journal de Conchyliologie, which also is rather a small one. The largest living shell at hand measures \2>\ mill, at its greatest diameter and some of the semi- ISO ZOOLOGY OF KODHIOLEZ. fossil specimens wliicli wer(> obtained iVoin i]\r l)on(>-cnves an> still larg'er, having a w idtli of 17 null. Helix bewsheriana. This sub-iossil species was cullecleil liy -Mr. Slater, in tiio hone-eaves. Helix sp. : This speeies is represented in the cnllcctidn 1i_v a single sj)ecimcn, obtained by Mr. Slater, it is very small, measuring only 1,1 niillini. at its greatest diameter, depressed, discoid, and consists of 3.\ Avhorls which are obliquely, finely lu-ate. The umbilicus is rather large; aperture lunat(> circular. As there is but a single specimen, and that possibly immature, I have not thought it advisable to designate it -with a specific name. Several closely allied species inhabit New Zealand and tlie islands of the I'aeihc. Helix similaris. This almost cosmopolitan shell was " very common " on the island according to the observations of Mr. Gulliver. Biiliinus fibratus, rar. The presence of this New Caledonian speeies in the island of l\odriguez is very surprising. Only one specimen was in the collection made by Mr. Slater. It seems impossible that it can be truly indigenous to the island, for, as is well-known, land-shells arc very local in their geographical distribution except in a few instances, such as in the case of the preceding ubiquitous species, JI. cel- laria and one or tAVO others. It appears to me therefore, that in all probability this specimen must have been conveyed to Rodriguez Island by human agency, and may have been picked up on the shore, having been thrown overboard from a vessel and washed on to the beach. The specimen varies somewhat from the typical fibratus and seems to be a connecting link between it and B. Souvillei, Morelet. Biilimus (Stenogyra) gracilis, Hutton. On comparing the numerous examples from Eodriguez, with Indian specimens of this species, some of them appear to be quite identical. It is a very variable species with regard to the thickness of the spire, and if the two extreme forms be separated from the others they appear to be distinct species ; but in a long series of specimens such as that before me, all the inter- mediate subtle gradations clearly show that they are but variations of one and the same species. B. {Stenogyra) subiila Pfr. and 02mranus Pfr. are scarcely separable from this species. Cyclostoma articvdatum, Gray. " This species is especially abundant on the Coral- " line limestone, the shells, chiefly dead, occurring in vast numbers. The living " animals arc to be found for the most part on trees " — ( Gulliver). The description of Cy. desmazuresi given by Crosse (Journal de Conch. 1874 XXII., p. 237 pi. VII., f. 12) agrees so exactly with the young stage of articulatum, that there can be no doubt that his shells are not a distinct species, but merely im- mature examples of the present one. Cyclostoma hseraastoma, var. The variety of this species which dwells on Rodriguez has been styled var. rodrigiiezensis byCrosse, who points out the following distinctions. It is " much smaller than the typical form, possessing only four and a half (instead MOLLUSCA.— EDGAR A. SMITH. 181 " of five) Avhorls and presenting a smaller number of spiral sulci (12 instead of 14 " on the peniiltimatc whorl). It occurs everywhere in abundance."— (6' »///(;c;-.) Omphalotropis hamiliana. The coloration of this species varies somewhat, the reddish brown colour differing in intensity, and the white band around the periphery sometimes being almost obsolete. The spiral striation will be found to be minutely punctured under the microscope. Omphalotropis littorimila. Crosse is very closely allied to, if not the same as, certain varieties of this species. Descriptions of Neav Species. Melania rodericensis. Plate LI., Figs. 9 and 10. Annals and Mag. Nat. Hist. 1876, May, p. 404. Testa subulata, pallidc olivacea, strigis longitudinalibus purpureo-fuscis irre- gulariter picta, et circa basim anfr. ultimi fascia lata colons saturatioris cincta, fre- quenter corio limoso rubro induta ; anfractus 11, convexiusculi, sutiu-a leviter obliqua discreli, liris transversis tenuibus plicisque longitudinalibus supra liras paululum granosis ornati ; apertura ovata, superne acuminata ; cokunella arcuata, alba. Long. 16 mill. diam. 5 ; apertm-a long. 5, diam. 2^. var. major, long. 23 mill, diam. 7. Shell small, subulate, thinnish, of a pale olive or yellowish olive colour, orna- mented with longitudinal little undulating purplish-brown or red stripes, which frequently stop short of, and do not flow into, the broad dark purplish-brown band encircling the base of the body- whorl ; frequently the whole surface of the shell is covered with a reddish earthy deposit ; whorls 11 moderately convex, encii-cled with fine subeequal sj)iral lirations (9 or 10 on the iipper whorls and about 20), which are crossed by slightly oblique longitudinal plications, which arc slightly granulous at the points of contact with the transverse ridges and scarcely extend to the suture beneath ; they are slightly flexuous, about 17 in number and on the last whorl they become obsolete about the midtUe. The aperture is ovate and acuminated at the upper extremity, and has much of the tint of the exterior of the shell, the longitudinal flames and the dark basal band being quite distinctlv visible ; the columella is oblique, not much arched sujjcriorly but more so below at its junction with the outer lip where it is white. Variety. The shell is rather larger, rather more ventricose and has the sculpture less pronounced. The dark band, encircling the base of the body-whorl, is very constant in all specimens. This species was collected by Mr. Gulliver, both in fresh and brackish water, but appeared most abundant in the latter. 3 p 482 ZOOLOGY OF RODRIGUEZ. There arc several species of JTelaniie, which partake very much of the forms of the present species. M.fasciolataoi Olivier appears to exhibit the greatest aflinity to it. The former is howovcr larger and the sculpture coarser, l)oth as regards the longitudinal ribs and lh(> transverse sulcatious. VagiuuJa rodericeusis. riale LI., Figs. 4 and !«. Annals ct Mag. Mat. Uist., 187G, May, p. 405. Corpus elongatum, utrinquc rotundatum, postice leviter angustatum ct acumi- natimi, supcrnc rotundatum, latcraliter carinatiuu, pallium supra ct infra minutis- sime granulatimi, testaccum, irregulariter confertim nigro tcsscllatum, vol punctatum infra paulo pallidius, marginibus lateralibus baud nigro-punctatis ; pes augustus latitudinis corporis } adequans, testaccus, usque ad extremitatem corporis I'ere productus ; tcntacula oculifcra nigrcsccntia ; caput tentaculaqvic buccalia flavo- testacea ; orificium fermincum paulo pone medium corporis situm. Long, 30 mill., diam. 10 (specim. in alcobol. serv.). In one or two specimens there is the faintest trace of a mesial dorsal narrow yellowish stripe. V. macula ta of Tcmplcton from Ceylon appears to be tlic nearest ally of this species, which is not at all common at Rodriguez according to the observations of Mr. Gulliver. Diplodonta lateralis. Plate LI., Figs. 7 and 7a. Annals and Mag. Nat. Hist. 1876, May, p. -iOS. Testa alba, transverse globoso-ovata, antice augustata, postice latissime rotundata, valde inaequilateralis ; margo dorsalis utrinque paululum dcclivis et rectiusculus ; valvse tenuiter concentrice striatte, hie illic fortius ; umbones baud prominentes, conspicue excentrici ; impressio muscularis antica elongata augustiuscida, possica aliquanto latior. Long. 18 mill., latit. 20|, crass. 11. Sbell pure white, transversely globosely ovate, a little narrowed anteriorly and posteriorly broadly rounded, very inequilateral, equivalved. The dorsal margin sloping on botb sides of the lunbones, tbe posterior slope is straight and rather short, the anterior, just a little excavated near the umbo ; tbe valves are concentrically striated, the striae being very fine with here and there the intervals of growth distinctly marked ; the umbone are small and not prominent and very eccentric, the anterior muscular scar is elongate and narrowish, the posterior broader. The linear fossa which holds the exterior hinge ligament is about 5 millim. in length. MOLLUSCA — EDGAR A. SMITH. 483 D. corecnsis Ad. and Bve. resembles the present si^ccies somewhat as regards outline. In that species, however, the iimboncs are deciiletUy more central and the anterior dorsal slope is consequently longer. The anterior end of the ventral margin is veiy faintly angled at its juncture with the anterior lateral margm ; but in I>. lateralis no trace of such an angulation is observable, and it is also a more compressed species than coreensis. Within the valves of the single specimen at hand parallel Avith the palial impression on the side towards the umbones, there is a second impression, which is punctured with small pits ; but probably this is merely an individual peculiarity. Avicula fiiscopurpui-ea. Plate LI., Figs. 5 and 5«. Annals and Mag. Nat. Hist. 1876, May, p. 405. Testa incequivalvis, perobliqua, postice pcroblique alata, striis concentricis confertis regulariter sculpta ; color externus f usco-piu*pureus, antice interdum pallescens ; rostrum valva3 dextrte medriocriter breve, paululimi declive, conf ertim rugose lamel- losimi, valvse sinistrae angustum, magnum ; cauda inconspicua ; ala perobliqua, imo reciu'va, in margine postico valde sinuata ; margo valvarum cardinalis rectus ; mar- garita pulcherrime iridescens, a Hmbo fusco-purpureo, antice pallidiore, posticeque latissimo cincta. Long. max. 55 mill., diam. 35. This species is more nearly allied to A. spadicea of Dunker (Kiister's Conchylien- Cabinet, Avicula p. 31, pi. 10 f. 1 and 8) than to any other, however it is more deeply sinuated on the posterior margin. The obliquity of the shell varies with age, for young specimens do not exhibit nearly so much of a wing as older ones. The colour of the exterior surface is a dark brownish or blackish purple which is of the darkest tint posteriorly, and beneath the rostrums, or on the anterior portion of the valves, it is generally paler. This is also the case with the inner surface, the anterior part of the Kmbus siu*- rounding the pearl being of a dirty yellow colour gradually blending into the dark blackish purple tint of the posterior margin. A. maUeoides, Reeve, agrees somewhat with the present species in form. It is however more obHque and produced, and lacks colouration. Rhizochilus (CoraUiophila) squainosissimus. Plate LI., Figs. 8 and 8«. Amaals and Mag. Nat. Hist. 187G, May, p. 404. Testa ovato-fusiformis, aliquanto umbiHcata, alba ; spira elevata, turrita ; anfrac- tus 7 convex!, costis spiralibus inoequalibus pulcherrime squamatis, irregulariter alternatim majoribus cincti, et plicis longitudinalibus obliquis clatlu-ati; anfi*. 3p 2 484 ZOOLOGY t)l' UODUKU'KZ. ultimus costis spiralibus circiter 2Goniatus, veiitricosiis, inl'cnn^ augustatus ; apci'tura ovato-pyrifonnis, all)a, loimitudinis lolius ;^ vix loqiiaus ; lahnnii luargiuo cronulato ; coluinoUa allia, I'alln la-vi induta ; canalis aui^'ustus, ohliciuus, lcvit(>r rocurvus. Long. 31 mill., iliaiu l(i ; a])ortura long. ID, diam. I). Shell ovately fusiform, narrowly umbilicatod, entirely white ; spire elevated, tun'ited, with the eonvergiug outlines almost reetilinear; Avhorls 7 convex, gi-adually increasing, encircled by numerous beautil'iillv and very closely imbricated lirations which are more or less regularly alternately stouter and liner and ai'c about 12 in number on the penultimate whorl and abcmt twice that number on the last. The whorls are also ornamented with obliquely longitudinal costations or ])li('ations, which are more conspicuous and closer together on the upper whorls than on the ultimate ; they number about 12 on the penultimate and the same on the last whorl ; the latter is ventricose above and narrowed inferiorly ; aperture white, ovately pear- shaped, scarcely equalling ;| of tiie whole length ol" the shell ; the outer lip is crcnulated on the edge ; columella covered with a thin white callosity, only slightly arcuate in the middle ; canal oblique, rather narrow and a little recurved. Fiisus inflattts, Dunkei*, in I'hilippi's Abbildungen neuer Conchylien, pi. iv., f. 2, must be allied to this species, but the sculpture and proportion of the aperture to the shell is very different. Perhaj)s R. exaratus, of Pease, which is found in the Sandwich Islands, has more the form of this species than any other in the genus. Tt is, however, at once distinguishable on account of its very distinct sculpture. Explanation of Plate LI. (Figs. 4 to 13— Molluscca.) Fig. 4. Vaffinula vodericensis, dorsal view. „ 4a. „ „ ventral surface. „ 46. „ „ orificium femineuin. „ 5. Avicula fusco-purpurea, upper side. „ 5a. „ „ interior of the same valve. „ 6. Fenella reticulata (much enlarged). „ 7-7rt. Diplodonta lateralis. „ 8. Rliizochilus (Coralliophila) squamosissimus. „ 8a. „ „ portion of tlie surface magnified. „ 9. Melania rodericensis. „ 10. „ „ var. „ 11. Conus rattu.s, var. „ 12. „ balteatus, var. „ 13. „ catus, var. 485 Crustacea.— ^y Bdward J. Iliers, F.L.S., F.Z.S., Assistant in the Zoological Depai-tment, BHtish Mmeum. The Ci'iistacea collected by Messrs. G. Gullivei" and H. H. Slater amount in all to 189 specimens, reiJi'csenting 35 species. All of these are forms that are ■nidely distributed throughout the Indo-Pacific or Oriental Region (which includes the eastern coast of Africa, the south and cast of Asia and islands adjacent, Avistralia, and the islands of Polynesia), witli the folloTving exceptions: — yltcrr/atojjsis signatiis (hitherto only known from the Mauritius), Caridina typus (original locality not known), Palcemon dispar (hitherto recorded only from the Malayan Archipelago), Falcemoti hirtimanus (from Mauritius, Rdimion, and the Indian Ocean), P. debilis (from Amboina and the Sandwich Islands), and the new species of Talitrus {T. gul- liveri), which is described below. With two exceptions all the species in the collection belong to the Podophthalmia. The following are the sub-tribes repre- sented, with the number of species belonging to each : — Sub-Tribe. No. of Species. Oxyrhyncha Cydometopa Catonietopa Oxysfamato Fagiiridea 2 8 9 1 5 Sub-Tribe. No. of Species. Caridea - Squill idea Oniscoidea Qammaridea 6 2 1 1 The Crustacea inhabiting the Red Sea have been made the subject of special study by Rtlppell and Heller, those of Madagascar and the islands adjacent by Hoffmann, of Mauritius and Reimion by Alphonse Milne-Edwards, and of the South African coast by M'Leay and Krauss. Valuable additions to our knowledge of the Crustacea of the East African coast have been published by Hilgendorf, in Van der Decken's " Reisen in Ost-Afrika," where will also be found a conspectus of all the known species of East African Crustacea by Von Martens. So far as I am aware, however, no species have hitherto been recorded as inhabiting the Island of Rodriguez. PODOPHTHALMIA. Decapoda. Mencethins nionoceros, Latr. A single specimen, an adult male, was collected by Mr. Slater. There is pro- bably no species of the Oxyrhyncha subject to greater variations, both in the form -186 ZOOLOGY OF KODKIGUEZ. and Icntjth of the rostrum aiul anterior legs and in tlio tiihoreiilaf ion of the cara- pace. Nearly a dozen supposed species of tlie geniis Ilohcthius have been described, all of which are united by M. A. Milne-EdAvards under the single name of M. moiiocci'os ; and ahhough one or two of these forms (as, for example, the M. tnhcrculatus. Ad. and AA'hite, from the Mauritius) appear at first sight to present marked distinctive peculiarities, it is probable that these are no more than individual variations. The specimen from Rodriguez is of tlic normal form, in which the carapace is indistinctly tubcrculated, the legs nearly smooth, the palm slender, and about twice as long as the fingers, Avhich are not strongly arcuate. This vai'iety is widely distributed throughout the Indo-Pacific region. Specimens ai-e in the British Museimi from the Gulf of Suez, Mauritius, Fiji Islands, Philip- pines, and Eastern Seas. Farthenope horrida, Linn. Two specimens, males, of this common species Avere collected by Mr. Slater. It is probably distributed throughout the Indo-Pacific region, and M. Milne- EdAvards records its occurrence in the Atlantic Ocean. Specimens arc in the British Museum fi'om the Bed Sea and Mauritius. This species is not to be con- founded with P. s2)iiiosissima of M. Alph. Milne-Edwards (Faune Carcinologique, in Maillard, Notes sur I'ile de la Reunion. Annexe F., p. 8, pi. xviii., fig. 1). Atergatopsis signatiis. Ad. & White. A single specimen, a male, was obtained by Mr. Slater. This specimen is a little smaller than the tyi^)ical example of this species in the British Museum Collection, from the Mam-itius, hitherto its only recorded habitat. Zozynms ceneus, Linn. One example, a male, of this beautiful but very common species is in the col- lection obtained by Mr. Slater. It is found in the Red Sea, at the Seychelles, the Mam-itius, Reunion, on the shores of the Indian Ocean, and in the Loo-Choo Islands, Australian Seas, and South Pacific (Samoa and Fiji Islands and Paumotu Archipelago). Aclceodes tomeiitosns, M.-Edw. Two specimens, male and female, of this species were collected. It is found at the Mauritius, and is distributed throughout the Indo-Pacific region ; its occurrence having been recorded from the Red Sea, Mozambique, the Indian Ocean, Nicobars, Simiatra, Sulu Archipelago, Samoa, and Fiji Islands ; and specimens are in the Collection of the British Museum from the Seychelles, Philippines, Sandwich Islands, and Minerva Reef. A. offinis, Dana (U. S. Expl. Exp. xiii.. Crust, i. p. 198, pi. xi. fig. 3), from the Pavimotu or Society Islands, differs only in its more CRUSTACEA.— EDWARD J. MIERS. 48? distinctly areolated and minutely gvauiilous carapace, and must be united with this species, as specimens in the British Museum Collection presented by the Smith- sonian Institution prove. Daira x>erlata, Herbst. Three examples of this species, a male and two females, were collected. It is probably distributed throughout the whole Indo-Pacific Region. Specimens are in the British Museum from the Maui'itius, Philippine Islands, Australian Seas, and Minerva reef. Milne-Edwards mentions its occurrence in the Atlantic Ocean and on the Coast of Brittany, and it has been recorded from the Samoa Islands, New Caledonia, Loo-Choo Islands, and Auckland. Cymo andreyossii, Audouin. A single specimen of this curious species was obtained by Mr. Slater. Its geo- graphical range is probably as extended as that of the preceding species. Specimens are in the British Museum Collection from the Bed Sea and Piji Islands, and it has been recorded from the Loo-Choo and Samoa Islands, Tahiti, and New Caledonia. The C. melanodactylm, of De Haan (Eaun. Japon. Crust, p. 22), from Java, as described and figm-ed by Dana (U.S. Expl. Exp. Crust, p. 225, pi. xiii. fig. 1) fi-om specimens collected at the Eiji Islands, is probably as M. A. Milne-Edwards has stated, a variety of this species. The colour of the fingers is not a constant character in C. andreyossii, but there appear to be two distinct forms ; in one of which the carapace and anterior legs are nearly smooth, and the front nearly cntu-e ; in the other the sides of the carapace and anterior legs are conspicuously granulated, and the fi-ont denticulated on its anterior margin. The form of the carapace is very variable ; it is ixsually nearly orbiculate, but sometimes longer and broadest in front. These characters, however, do not appear to be of specific value, as inter- mediate varieties occur, and forms in all respects identical are found at widely separated localities. Trujjezia riifo-punctata, Herbst., var. maculaia, M'Leay. One example is in the collection, a male, obtained by Mr. Slater. Specimens are in the Collection of the British Museum from the ^lauritius. Red Sea, and Ceylon, and its occvu-rence is recorded at Zanzibar, the Cape of Good Hope, Tahiti, the Sandwich Islands, New Caledonia. The specimen from Rodriguez is of the form figm'cd by Dana (U.S. Expl. Exp. Crust, i. p. 256, pi. xv. fig. 4), as T. maculata, of M'Leay {Grapsillns maculatns, M'Leay, in Smith Zool. S. Africa, Crust, p. 67). This variety is distinguished from the typical T. riifo-imnctata by the form of the frontal lobes, which arc obscurely defined and obtuse or rounded, not strongly prominent and acute. The two or three anterior teeth upon the anterior margin of the arm are truncate or slightly -188 znoi.ocY OF ROi>Ri(;rKZ. t'lnartjinafe at their apox, the red spots, with wliich lli.' wIkiIc ;niiinal is i-overed, arc sinalliM- and t'(MVor in niinih(M- ilinn in ihr tv|iical xariclv. Tetralia carimaiic, llilKr. Three specimens, two ni.ilcs and a IVmah*, wcn^ coUectcMl hy Mr. SlatiM-. Ex- aiii]des of this species are in t lie Collection of Iho Hritisli Museum I'roni the i'iji Islands and Australian Seas; and il occairs also at Taliiti and llic lied Soa and New Caledonia. Achelotis grauulatus, ^I.-Edw. One example, a male, was collected l)y Mr. Gulliver. This species is very commonly and widely distrilnitcd throug-hout the Indo-Pacitic Region. Spcciiiiens are in th(^ British Museum Collection IVom the lied Sea, .Mauritius, Ceylon, the Philippine islands, Piji and Sandwich Islands ; and it has hecn found also at Reunion, Zanzibax", Java, Samarang, and Japan, and New Caledonia. Gelasitmis vocans, Linn. Pom- specimens, males, Avere collected by Mr. Gulliver, who observed that both this and the following species were very common in the island, l)U]-rowing in t he sands by the sides of the estuaries rather than on the seashore. The prominent triangular lobes or teeth on the inner margin of the in\mobile finger A^ary considerably in size, and one or even both may be obsolete, as in the species or varieties described as G. cultnmonm by Adams and White (Voy. Samarang, Zool. Crust, p. 49), from the Philippines and ILong-Kong. Specimens of this S])ecies arc in the Britisli ]^[usenm Collection from Mauritius, Singapore, and the Fiji Islands ; and it lias been recorded from the Islands of Nossy-Falyand Nossy Be adjacent to Madagascar, Zanzibar, Java, Malabar, the Nicobars, and Loo-Choo. Gelcisimus aumiUpes, M.-Edw. Six specimens, including males, females, and young, were obtained by Mr. Gulliver, Avhich I refer to this very variable species. The larger hand and fingers are smooth externally, without indications of longitudinal grooves, or with one obscurely-indicated at the base of the palm, Avhich has a short transverse groove neai' its anterior margin. On the inner surface of the palm there is a high pro- minent oblique granulated crest, and two transverse lines of granules near the anterior margin. The fingers are granulated on their inner margins, the upper one arcuate, the lower nearly straight. There is always a tooth near the distal extremity, and usually one at about the middle of the inner margin of the lower finger, some- times also two or three other tubercles or teeth placed at unequal distances along the inner margins of both the upper and lower fingers ; often, hoAvever, these are absent. Specimens of G. (omulipes are in the collection of the British Museum CRUSTACEA. — EDWARD J. JUERS. 489 from Zanzibar, Ceylon, the Malayan Ai-chipelago, the Philippines, and tlie Fiji Islands ; and it has been recorded from the Islands of Nossy-Faly and Nossy-B^, Madras, and the Nicobars. Ocypode cordimana, Desm. Two specimens, a male and a female, were collected. I\Ir. Gulliver states that the female taken by him was foimd in a wood, and considers it i^robable that this species was hibernating in holes at the time of his sojonm on the island. It has been recorded from the Red Sea, Zanzibar, Mauritius, Reunion, Mozambique, Natal, and the Cape of Good Hope, and its range extends over the whole Indo-Pacific Region ; as specimens are in the British Museum fi-om Ceylon, Cliina, and the Australian Coast, the Samoa Islands, Fiji Islands, New Hebrides, and it has also been recorded from the Nicobars, Manilla, Japan, Hong Kong, Loo-Choo, and New Caledonia. The 0. arenaria. Say, which inhabits the Atlantic Coasts of the American Con- tinent, is very nearly allied to this species, young individuals are indeed hardly to be distinguished, but adult specimens may be recognized by the laterally-compressed ambulatory legs, which are smooth (not granulated as in 0. cordimana), and thickly clothed on the upper and lower margins with dense fulvous hair. Grapstis picttis, Latr. A female example of this very variable, common, and widely distributed species was taken by Mr. Slater. Specimens are in the British Museum from the Cape Verde Islands, St. Helena, S. Africa, the Mauritius, Ceylon, Fiji Islands, Galapagos, Peni, California, and New Zealand, and it also inhabits the West Indies and Coast of Florida. Geograpsm grayi, M. Edw. Three specimens, an adult male and two females, were collected. The examples obtained by Mr. Gulliver were found by him at some distance from the seashore. In the younger animal the hands are more granulous, and the series of teeth or spines upon the anterior margin of the arm more acute. In all the adult examples of this species that I have examined the frontal margin is straight or very nearly so, never bi-emarginatc as in the specimen figiu'ed by Hilgendorf as the adult of G. rubidtis, Stimpson (Crust, in van der Decken's Reisen in Ost-Afrika, pi. v. fig. Ic). This species has been recorded from the Red Sea, Mauritius, Reunion, Zanzibar, Madagascar, the Indian Ocean, Australia, Bonin, New Caledonia, and the Sandwich Islands ; and specimens from the Fiji Islands are in the Collection of the British Museum. Me(o]}oyra])8iis messor, Forskal. Four specimens, including males and females, were collected by Mr. Gulliver, who states that he found this species plentifully near freshwater streams in the 3q 4,90 ZOOLOGY OF KOnivlGUEZ. woods, but never actually on the seashore. It varies cousulorably in tlic marbled coloi-ation of the carapace. It has been recorded from tlir Eed Sea, Zanzibar, Natal, Persian Gulf, Coast of Malabar, ^Nfadras, Ceylon, Bunin, Sandwich Islands, Tahiti, and New Caledonia, and specimens arc in the British Museum from [Mauritius, the Eastern and "Western shores of Australia, and the Fiji Islands. The G. Ihiiknjar, of Owen (Beechcy's Voyage, Zool. Crust, p. SO, pi. xxiv. fig. 3), is, I believe, a variety of this species. Sesarma telragontim, Fabr. Three examples, male and female and young, were collected by Mr. Gulliver who found them in burrows near the estuaries. In the adult male the fingers are arcuate, and meet only at the corneous excavated tips ; in the young male the fiuLjers are straight. This species has been recorded from the Red Sea, Mauritius, Zanzibar, Natal, Madagascar, Iiulian Ocean, Chinese Seas, and New Caledonia. Specimens from the Fiji Islands are in the Collection of the British Museum. Sesarma qiiaclratian, Fabr. Seven examples, males, of a species of Sesarma were collected, which I refer with some hesitation to ^S". quadratum. The animal is of a blackish-brown colour ; the carapace has the lateral margins entire, and is slightly granulated on the post- frontal region, with the fom- truncate frontal lobes well-defined ; the arm has a prominent angulated lobe or tooth at the distal extremity of its anterior margin ; the An'ist is uniformly rugose Avithout spines; the exterior sm'faco of the palm smooth below, slightly granulated toward the superior margin, the upper sm'face has two small oblique very closely and minutely pectinated ridges, and behind these two or three oblique granulated lines. Sesarma quadratum has been I'ecorded from Zanzibar, Natal, Poudicherry, and New California ; and there are specimens in the British Museum Collection from Borneo and the Fiji Islands. The species figured by De Haan, (Faun. Japon, Crust, pi. xA-iii. fig. 5), under the name of Fachysoma affinis, is probably identical with this species ; also the Sesarma aspera, of Heller (Voy. Novara, Crust, p. 63, pi. vi. fig. 1), from Ceylon, Madras, and the Nicobars. Dr. Hilgendorf, basing his observations upon a large series of specimens from the Indian Ai'chijjelago, Malacca, Sumatra, Luzon, Flores, and Timor, states that *S'. quadratum varies very considerably in the form of the carapace, frontal lobes, and anterior legs, but it is possible that he has confounded two or more species under the above- mentioned designation, as he neither mentions nor figui-es the minutely-pectinated ridges on the upper margin of the hand, which are characteristic of S. quadratum. {See Van der Decken's Pieisen in Ost-^yrika, Crust, p. 90, pi. iii. fig. 3c. and pi. iv. fig. 3.) Cardisoma carnifex, Herbst. Two examples, a fine adult and a smaller male, were collected by 3Ir. Gulliver, who found this species bm-rowing in sandy ground by the estuaries of the streams. CRUSTACEA.— EDAVARD J. MIERS. 191 but (lid not observe that it penetrated far inland. There is usually a small tooth behind the external orbital tooth or spine, but this is sometimes rudimentary or entirely absent : il is present in the smaller, wanting in the larger specimen from Eodriguez. As the animal increases in age, the body becomes very much more convex over the branchial regions, and is expanded laterally beyond the marginal line, which is hardly distinguishable in the oldest specimens ; one or other of the anterior legs is very greatly developed, and the hair upon the penultimate and ante- penultimate joint of the ambulatory legs becomes very long and dense. Specimens of this species are in the British Museum Collection from the Mauritius, Pondicherry, Philippines, the Fiji, Sandwich, and Samoa Islands. It has been recorded from Zanzibar, the Xicobars, and Xew Caledonia, and a young example obtained at Madeira is referred by Dana to this species. Culappa hepatica, Linn. Three specimens, two males and a female, were collected. It is one of the commonest and most widely distributed of the Indo-Pacific Brachyiira. This species has been recorded from the Eed Sea, Zanzibar, Natal, Madagascar, the Mauritius, Indian Ocean, Xicobars, Auckland, the Piji and Sandwich Islands, and New Caledonia ; and there are specimens in the Collection of the British Museum from the Seychelles, Ceylon, the Coast of China, the North-east Coast of Australia, the Loyalty Islands, New Hebrides- and Samoa Islands. Paffurus jJiiuctitlatus, M. Edw. Three examples of this species, one of the largest and most beautifully-coloured of the genus Paguriis, were collected. One is an adult female, bearing a large quantity of ova. It has been recorded from the [Mauritius, Red Sea, Madagascar, Indian Ocean, Nicobars, the China Seas, Tahiti, and the Sandwich Islands; and specimens are in the British Museum from the Philippines, Australia, and the Fiji Islands. Aniculus typicus, Dana. Four specimens were collected. The recorded localities of this species are the Mauritius, Reunion, Batavia, ^Manilla, Japan, Auckland, the Paumotu Archipelago, Wakes Island. Specimens are in the Collection of the British Museum from the Seychelles, Australia, Samoa and Fiji Islands. Calcinus tibicen, Herbst. A single specimen inhabiting a shell of a species of Turbo was collected by Mr. Slater. This species has been recorded from Zanzibar and the Coast of Natal, the Nicobars, Straits of Balabac, Loo Choo, the Paumotu Ai'chipelago, the Society, Sandwich, Samoan Islands, and '^'akes Island. Specimens are in the Collection of the British Museum from the Pelew Islands and Mozambique. 3q 2 492 ZOOLOGY OF RODRIGUEZ. Cdlciniis elegaiis, M. E(bv. A iVmalo example, with ova, is in the Collcctiou, i)l)taiucd by jMr. Slater. Both the autcrior legs are unfortuuatcly >yautiiig, but the colouration agrees very well with the descriptions of this species ; the legs of the second and third pair being beautifully marked with l)road alternating bands of crimson and white, the tarsi with crimson sjiots upon a white ground. This species has been recorded from the Coast of Xatal, Loo Choo, the Faumotu Archipelago, IlaAvaiian Islands, Wakes Island, Tahiti, New Ireland, and "Western Patagonia. Ccenoblta rugosa, M. Edw. Foiu'teen specimens of this species were collected. The greater number of these inhabit shells of Ncriia polita, one specunen was foimd in a shell of the terrestrial Cyclostoma articulatuin. Mr. Gulliver observes of this species that it was often found in perfectly arid places a mile or so from the sea-shore. It is very commonly and widely distributed throughout the Indo-Pacilic llegion, and its range even extends to the "Western Coast of America. It has been recorded from the Hed Sea, Zanzibar, Mozambique, Reunion, Natal, Persian Gulf, Indian Ocean (Ceylon, Madras, Nicobars), Java, Cliina Seas (Loo Choo), Australia (Sidney), the Sooloo Sea, Panmotu ^\_rchipelago, Fiji and Samoan Islands, Tahiti, and Panama. To these localities I may add, on the authority of specimens in the Collection of the British Museum, the Mauritius, Coquimbo, and Vancouver Island. M. Milne-Edwards (Hist. Nat. Crust, ii. p. 211) states that a specimen from Mcssma in the Collection of the Paris Museum appears to belong to this species, but may have been wrongly labelled. Caridina typiis, M. Edw. To this species I refer a single specimen obtained by Mr. Gulliver, which only diflfers from Milne-Edwards' description and figure in having two instead of three minute teeth on the inferior margin of the rostrum, which is proportionately some- what shorter and deeper than that of the example figured by Milne-Edwards. In many of the Coridea the form and number of teeth of the rostrum are so variable that these differences can scarcely be considered as of specific importance ; and it may be that the C. exilirostris of Stimpson (Proc. Ac. Nat. Sci. Phil. p. 29, 1860) from Loo Choo, in which the inferior margin of the rostrum is obsoletely 2-3 dentate, represents an intermediate variety of this species. The habitat of the type-example is unknown = Alpheus laevis, Randall. A single example, a female with ova, is in the Collection obtained by Mr. Slater. The antennas are imperfect, the smaller hand is wanting ; the larger hand agrees well with the descriptions and Dana's figure of the species. It has been recorded CRUSTACEA.— EDAVARD J. MIERS. 493 from the Red Sea, Nicobars, Syclucy, Tahiti, and the Sandwich and Fiji or Friendly Islands. Falaemon ornatus, Olivier. A large number of specimens of this, one of the commonest, and most univer- sally distributed of the Pravrns inhabiting the freshAvaters of the ludo-Pacilic region, were collected, forty-one in all ; of these, however, but few are in a perfect condition, the greater number having lost one or both of the large legs of the second pair ; and but two or three examples are of full size. Many of the smaller examples are females bearing ova. The teeth on the rostrum vary between 6-11 on the upper and 2-3 on the lower margin ; there arc usually 8-9 upper and 3 lower mai-ginal teeth (-^i-). The males of this species are at once recognized by the strong greatly elongated second pair of legs, which are closely beset with numerous minute spines, the palm in the adult animal is much longer than the wrist, the fingers are inflexed at the tips, the upper often considerably arched, the immobile finger has two, the mobile finger a single strong conical tooth on the inner margin. In the females and young these characteristics are not nearly so strongly marked, and perhaps some of the smaller specimens ought to be referred to the following species, (P. dispar). P. reunionensis and P. longimanus of Hoffmann, (Eecherches Faune Madagascar, Crust, pp. 33 & 34, pi. ix., fig. 66 & 67, & 68, 69), from Reunion, appear to me to be founded respectively on younger and adult examples of this species. Falaemon- ornatus has been recorded from Amboina and the Islands of Waigiu and Adenare, the Celebes, Philippines, Tahiti, and Auckland. There are specimens in the Collection of the British Museum from the Mauritius, Australia (Moreton Bay), the Samoa and Fiji Islands, New Hebrides. Palacmoii dispa)', von Martens. To this species is certainly to be referred a male example in the Collection obtained by Mr. Gulliver. When adult, P. dispar is distinguished without difliculty from P. ornatus, by its smaller size ; the straighter rostrum, with more numerous marginal teeth, the very imequal legs of the second pair, and the form of the fingers, which, in the male, each have a series of conical acute spines on theii- inner margins. Hitherto P. dispar has been recorded only from Adenarcs Island, near Flores, in the Malayan Archipelago, but it may have easily been confounded with other species of this very difiicult genus. The P. alphonsianus of Hoffmann (Rech. Faune Madagascar, Crust, p. 35, pi. ix., fig. 63-65) from Reunion is in all probability synonymous with it. Falaemou hirtimanns, Olivier. One specimen, a male, is in the Collection, obtained by ]Mi-. Gulliver. In tliis example the large leg of the second paii- on the right side has Ijeen broken off 494 ZOOLOGY OF TJODIJIOrEZ. above the second joini, :iiul its place is supplied 1)\ ;mollier leg, evidently ncAvly- formed, and very sn\all and weak. Imt \V(>]l-])roportioned in all its joints. This species has been recorded IVoni ^lauritiiis and lu'iniion, and ilie Indian Ocean. Falacmon (Leai/dcr) dchilis, Dana. No fcAver than fifty-six spcciin(>ns were collected, bnt a considerable number of these are move or less imperfect. Like all llu> species ol" tins sub-u'enus, P. *2 5 )> {/ §2 1 „ <• t2 4 „ l> J2 2 „ 'I 42 2 .. 3 |2 1 The second pair of legs are very small and slender, resembling those of the first pail*. The specimens described by Dana were collected at the Sandwich Islands ; Von Martens records the occurrence of this species at Amboina, at the mouth of the Weynitu. River, in solt-toater ; Mr. Gulliver, on the other hand, found it in the freshwater streams of the Island of Rodriguez. The species of the sub-genus Leander are, I bcKeve, generally marine, those of Falaemou fluviatile. Dana has figm-ed a variety of P. debilis from the Hawaiian Islands, to which he has applied the name of P. attenuatus ; having six teeth on the upper and nine on the lower margin of the rostrum. Stomatopoda, Lysiosqiiilla mciculata, Lam. A single large male example (length nearly 13 in.) was obtained by Mr. Slater. This species has been recorded from the seas of Asia and Japan. Specimens are in the Collection of the British Museum from the Indian Ocean, Philippines, Samoa and Piji Islands. On account of its large size, and the colouration of the segments of the body, which are marked with broad alternating bands of black and yellow, this species is perhaps the most conspicuous of the Stomatopoda. CRUSTACEA. — EDNYARD J. MIERS. 495 GoHodactylus cliircf/ra, Fabi-. Two specimens of this species were collected. The largest of these, an adult male, has a length of nearly 3^ in. It is a very common and widely distributed species, and its range probably extends throughout the warmer latitudes of the globe. There are specimens in the Collection of the British IMuseum from the Mediterranean, Red Sea, Zanzibar, Indian Ocean, Ceylon, "Western, Southern, and North-eastern Australia, Sir C. ilardy Island, Pliilippiuc Islands, New Ilebrides, Fiji Islands, and Panama. It has been recorded from Natal and Reunion. EDRIOPHTHALMIA. Isopoda. Oniscus ? {sp.) Five specimens of a species of Terrestrial Isopod, probably belonging to the genus Oniscus or Forcellio, are in the Collection obtained by Mr. Gulliver, who states that this species is common under stones. In all these specimens botli antennae and m-opoda are wanting, and it is therefore impossible to ascertain even the genus to which they ought to be referred with certainty. The posterior margins of the first three segments of the body are straight, the terminal segment almost T-shaped, the posterior half being very narrow. From P. truncatiis, M. Edwards (Hist. Nat. Crust, iii., p. 171) inhabiting the Mauritius, this species differs in the form of the terminal segment ; from a species described under the same name by Brandt (Bull. Mosc. Soc. Nat. vi., p. 181), but placed in a different section of the genus by Milne Edwards (1. c. p. 173), from the Cape of Good Hope, in the form of the head, in which the front is scarcely at all prominent, the antero- lateral lobes large and rounded. AmpMpoda. Talitrns gulliveri. T. yulUveri, Miers, Ann. Mag. Nat. Hist. ser. 4, xvii., p. 406 (1876). Slender, smooth, with the buccal organs considerably projecting, as in T. locusta. Head small, with the anterior margin straight, eyes round, black. Second and third segments of the postabdomeu with the posterior margins straight, the infcroposterior angle somewhat prominent and acute. Superior antennae scarcely as long as the head and about reacliing to the extremity of the second joint of the inferior antenna?, with seven joints exposed, the basal very short, the two next longer, sub- equal, together about as long as the four last joints, of which three are equal and the last minute. Inferior anteunce slender, not as long as the body, with the basal joint of the peduncle very short, the second joint more than half the length of the third joint, the flagellum about eighteen-jointed and rather longer than the peduncle. First and second pairs of legs small and weak ; the first pair with the three last ■196 ZOOLOGY OF RODRIGUEZ. joints short, suhcqual, and not dilated ; the second jiair (in some specimens) longer, Avith the palm slightly dilatod and the finger quite rudimentary as in T. platychcJcs. Next tluve pairs of logs short, sixth and seventh pairs considerably longer, Avith the second joint moderately developed. Rami of the fourth and fifth pairs of tail- appendages suhequal, those of the fom-th shorter than those of the fifth pair, sixth pail- quite rudiraentai-y. The distal extremities of the joints of the autennse, the legs, and the rami of the postabdomiual appendages are fringed with short hairs. Leugtli about 5 lines. This species -nas found by Mr. Gulliver under stones in damp places, but never observed actually \i\ the water. I feel no hesitation in referring this species to the genus Talitrus, as in none of the specimens are either the first or second legs largely developed and subcheliform as in other genera of Oi'chcstUdcc. It diff'ors from the common European T. locusia in its much shorter and slenderer inferior antenna;, and in the form of the segments of the post-abdomen; the posterior margins not being sinuated as in that species. It seems to be more nearly allied to T. platycheles, Guerin. (Exped. Scient. Mor6e, Zool. Crust, p. 41, pi. xxvii., fig. 4), but differs in the proportionate length of the joints of the pedimcles of the antennae, in the larger second pair of legs, and sixth and seventh pairs of legs, and in the more obtuse infero-posterior angles of the post-abdominal segments, &c. i97 Myriopoda and AiiAciiNiDA.— Z>// Arlhu)' OonJiner Butler, F.L.S., F.Z.S., S,-c. (Plate LII.) Of the twelve species of Ilyriopoda ohtainocl by Messrs. George Gulliver and H. n. Slater, seven are referable to the Chilocjnatha and five to the Chilopoda : all, with the exception of the two species of Scolopendra (and possil)ly Splrostreptus soronms and the Mecistocephahis), appear to be new to science. So few naturalists have interested themselves in the study oC these animals, that it is impossible at present even to form an opinion as to the geographical distribution of the species of Eoflriguez ; all that can be said is that, in all prol)a- bility, the two forms of Scolojjcndra will be found abundantly in Madagascar. Of the Arachnida twenty-eight belong to the Armieidea, one to the Acaridea, and one to the Scorpionidea, eleven of them being new species. Of the remainder, three have been previously recorded from Mauritius, eight from Bom*bon, two from Madagascar, three from Mauritius and Bourbon, aud one from Mauritius, Boui-bou, and Madagascar. The last-mentioned species {Olios leucosius) is very common, and probably has a wide range. Specimens in the Collection of the British Museum are ticketed " Magdeleine." The species of the greatest interest is a new form of the curious genus Miagram- niopes of Cambridge, a group of delicate and extremely strange-looking four-eyed spiders, originally discovered by Mr. Thwaitos in Ceylon. There is also an abnormal form of Sjificsiis, which in coloiuiug, and general aspect much resembles Jliayraiu- mopes ; vinhappily both of these species seem to be very rare. MYEIOPODA. Previously obtained elsewhere. Name of Myriopoda found in Kodrigucz. to 1 . " t 't. B'T, u> Other Localities. ■2.S § 3 CS S 2 o 1 'a S K. S Strongj'losoma enucaria, Butler - X Cambalii nodulosa, Butler - - - X Spirostreptus avcrnus, Butler X Spirostrt'ptus sorornus, Butler X Spirostreptus gulliveii, Butler X Spirostipptiis simulans, Butler X Spii-obolus hecate, /?(///f/-- X Eiirylitholiiiis slateri, Butler X Scolopendra mossambica, Peters • Mozambique. Scolopeudra ferruginea, Koeh (Locality previously uuknowu). Rhoinbocc'pbalus smaragdiiuis, Butler X Mecistocephalus gulliveii, Butler • X Reunion ? 3 R -198 ZOOLOGY OF KODHIGUEZ. Polydesmidaa. Strougylosoma, Brandt. Sh'oiiffi/loso»iit errticaria. Aun. & Mag. Nat. Hist. s. 1, vol. xvii., p. 'Ill (1S7G). Doop chocolatp-brown ; under sixrface, legs, lateral Avings, and hind margins of the segments pale tcstaeeous ; segments glabrous, first dorsal segment with well- defined lateral carina; preaual segment terminally rostriform. Length 8 lines, width 1 line. (Coll. by Gulliver.) This speeies seems to be allied to S. Guer'mii of Gervais, but it dilTcrs in its well-marked lateral carina, smaller size, and deeper coloration ; young examples are, however, paler than the type. Ten specimens were brought home by Mr. Gulliver. Cambala, Gray. Cambala iiodulosa. Aim. & Mag. Nat. Hist. s. 1, vol. xvii., p. d41< (1876). Reddish testaceous ; dorsal segments with a lateral grey stripe ; first segment longitudinaUy miUtisulcatc ; remaining segments crossed transversely by two rows of tubercles, between which, on each segment, is a depressed line ; posterior mar- gins narrowly reborded. Length 8 lines, wddth 1 millim. (Coll. by' Gulliver.) Tliis species has about fifty dorsal segments, but they are difficult to count, owing to its small size and elaborate sculpturing. Nine examples were obtained. Julidse. Spirostreptus, Brandt. aS^; irostreptus avern ns. Ann. & M5ig. Nat. Hist. s. 4, vol. xvii., p. 445 (1S76). Black, legs and antennae castaneous ; forty-three dorsal segments ; first seg- ment smooth, with narrow marginal ridge; remaining segments rugulose and depressed in front, smooth behind, striated at the sides ; preanal segment j)roduced into a point behind. Length 13 lines, width 1 line. Cascade Valley. (Coll. by Gulliver.) Eight examples. Spirostreptus sorormis. Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. 445 (1876). Colom* and general structure of the preceding ; forty -nine dorsal segments, which are not rugulose in front. Length 1 inch 9 lines, width 3 niillims. (Coll. by Slater.) MYEIOPODA. — AETIIUR GARDINER BUTLER. 499 This may be the aS*. tiifferrinms of Newport, but the latter is only represented by the head, nuchal plate, and two or three of the anterior dorsal segments, in a dried state. Only one specimen was obtained. Sjiirostreptus gulliveri. Ann. & Mag. Nat. Hist. s. 4, yol. xvii., p. 445 (1876). Olivaceous ; dorsal region dai*k ; eyes black ; face, margins of first dorsal segment, and hind margins of other segments tawny ; legs reddish tawny ; fifty- seven dorsal segments, first segment with square anterior angles, above which are fom- depressed lines ; marginal ridge feebly developed ; each of the remaining segments crossed by a depressed line, behind which they are swollen ; sides striated as usual ; preanal segment slightly produced, with bisinuate hind margin. Length 2 inches 4 lines, width 5 milluns. (CoU. by GuUiver.) Nine examples were collected. It is said to be " common." Spirostreptus sinmlans. Ann. & Mag. Nat. Hist. s. 4, vol. xvii, p. 445 (1876). Olivaceous ; a dorsal series of dark spots in front and a continuovis lateral series of blackish spots ; first segment clouded with brown behind ; forty-three dorsal segments, first segment with marginal ridge feebly developed ; remaining segments rugvilose and depressed in front ; preanal segment regularly arched. Length 10|^ lines, width 2 millims. (Coll. by Gulliver.) This species looks like a small form of the preceding, but its structiire is quite distinct. Eight specimens were obtained. Spii'obolus, Brandt. Spirobohis hecate. Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. 445 (1876). Shining black ; front margins of the segments below castaneous ; fifty-nine dorsal segments ; lateral wings of first segment quadrisulcate in front ; remaining segments swollen behind, laterally striated; preanal segment deeply bisinuated behind, the centre of the hind margin being produced. Length, 4 inches 3 lines, width 7 millims. (Coll. by Gulliver.) Only one individual of this fine species was obtained ; it is "not common." Lithobiidae. Eurylithobms, Butler. -Lithobio affinis, multo latior, segmentis viginti ; antennarum articuHs septen- • 3b 2 500 ZOOLOGY OF RODKUaKZ. ilccim, quoniiu decern primoribus ct posircmo cliistiiK'tis, aliis ineonspicuis ; ocvilis utrobiquo quatuor; podiini paribus uno v\ Ai^iiili, lalis, brcviusculis. Gen. typ. ^. sla/eri. Fjo-jililltobtiiii f>-/(i Ici'i. Ann. .^ Mag-, ^at. Hist. s. 1. vol. xvii., p. 440 (1870). Fawn-coloured, legs and ventral surfaci^ testaceous ; dorsal segments gradually narrowing towards eacli extreniily, with two longitudinal depressed lines and a lateral marginal ridge; terminal segment subconical behind. Lengtb of body 1 ineb, including bind legs 1 iucli S lines ; widlb in Ibe centre 4-2 lines. (Coll. by Slater.) Only one specimen was obtained. Scolopendridae. Scolopeudra, Liniueus. Scolopendra mossambica, Peters. IVenty-one specimens. Scolopendra ferruginea, Koch. Thirty-three specimens. The two preceding species look much like difierent ages of the same animal ; but Dr. Peters seems to have tigured his species from an adult example, which has alone ])revcntcd my luiiting them. Rhoinbocephalus, Netoport. Ithoniboccphalus smaraydiii us. Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. 446 (1876). Anterior half of body emerald-green, posterior half olive green ; segmental incisions deeper-coloured ; legs jiale greenish, in twenty pairs ; posterior pair without denticles ; preanal plate oblong. Length of the body 10^ lines. (Coll. by Slater.) One specimen. Mecistocephalus, Neioport. Mecistoceph a I us gu lUveri. Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. 446 (1876). Amber-coloured, legs paler ; head reddish eastaneous ; forceps of mandibles black, the latter with about four teeth ; head below deeply excavated ; dorsal segments with a deep longitudinal sulcus ; forty-eight jiairs of legs. Length 1 inch 1 line. (Coll. by Gulliver.) This species, of which only one example was found, is evidently nearly allied to the Geophilus insularis of Lucas ; indeed it is just possible that the two may be varieties of one animal. ARACHXIDA. — ARTHUR GARDINER BUTLER. 501 ARACHNIDA. re N 1 & •a ss . Previously obtained elsewhere. Names of Species obtained in Rodriguez. 1 O •e 1 Other Localities. |.2 o § "% " S n s Clubiona insularis, Vi7iso)i X Scytodes thoracica, LutrciUc X Pholcus elongatus, Vi/isoit X Pholcus vexillifer, Butler X Theridium liorboiiifuiii, Jliisoii X Theridiiim diurmim, JIdsoh X LiUrodectus lUfiiavodi, Jlitsoa X Liuypliia arjiyrodes, Tf'alc/ienaer - X Epeira opuntiao, Dtifour - - - X X Spain and Algeria. Epeiia Isabella, var., Vinson X Ai-giope luauritia, U'alckenaer X X Soutli Africa and Madeira. Meta uudulata, Vinson X Aleta vafillans, Biitlcr . . - X Nepliila iuauiata, JValckenacr X Nephila ardentipes, Butler X Kephila iiistigans, Butler - X Tetragnatha protensa, TValchcnaer X Tetragnatha nero, Butler - X Gasteracantha mauritia, Vinson X X Uloborus vanillarum, Vinson X Miagrammopes Gulliveri, Butler - X Olios Icucosius, JVuhkenaer X X X " Magdelcine." Olios capliosus, JValehenaer X X Sphasus lucasii, Vinson - - . X Sphasus estensipes, Butler X Salticus baptizatus, Butler X Salticus scabellatus, Butler X Salticus rodericensis, Butler X Holothynis testudiaeus, Butler X Tityus niarmoreiis, Koch * - - X * Locality not pre\iously recorded. Araneidea. Clubiona, Latreille. Clubiona insularis, Vinson. One specimen of this species was obtained by Mr. Gulliver. Scytodes, Walckenaer. Scytodes thoracica, Latreille. Seven examples. It is "common under leaves, bark, &c." — G. GuUicer. 502 . ZOOLOGY OF KODKIGUEZ. Pliolciis, Jfulckcitoer. Tholcus clonf/ofiis, Vinson. Five examples obtained. PholCHs rexlllifer. (Plate LII., figs. 6, 6«.) Ami. & Mag. Nat. Hist. s. 4, vol. xvii., p. ill (1870). ? . Cephalothorax inverted licart-shapcd, waved at the ends, bright reddish castaneoxis; caput ascending, witli tlie front margin and a central marking yellowish; eyes upon elevations ol' the surface, those of the lateral groups J'orming triangles, anterior pair of eyes small ; falces blackish, yellowish at base ; maxilla) and labrum tawny ; pectoral shield pyriform, yellowish ; legs pale amber-yellow, clothed with whitish hairs and black spines ; palpi testaceous, with pieeous tips : abdomen sub- cylindi'ical with a wavy outline, above testaceous, with a central stellate streak, several oblique lines, and the borders brown, posterior extremity testaceous ; under- side the same colour, with the base and a central streak greyish brown. Length 10 millims. ; relative length of legs 1, 2, 4, 3. (Coll. by GuUiver.) Of the lateral groups of eyes, those nearest the mesial line are smallest, the claws of the legs are black ; the spinners are testaceous. Two examples were obtained. Theridiiim, Walckcnaer. Theridium horbonicimi, Vinson. One specimen found "on leaves." — Gulliver. One specimen. Two specimens. Theridium diurnum, Vinson. Latrodectus, Walckenaer. Lotrodeckis menavodi, Vinson. Linypliia, Latreille. Linyphia urgyrodes, Walckenaer. Foiu' specimens foimd " on leaves." — G. Gulliver. Epeira, Walckenuer. Epeira opuntice, Dufour. Pifty-eight specimens, exhibiting numerous variations, among which is the form named var. flava by M. Vinson. This species according to Mr. Gulliver is " common " and " lives in a web of irregular construction." ARACHNIDA. — ARTHUR GARDINER BUTLER. 503 Ei)eira Isabella var. Vinson. Two specimens. Argiope, Koc/i. Argiope manritia, Walckenaer. One example was obtained. Meta, Koc/i. Mela unclulata, Vinson. Foiir examples were obtained. Meta va^cillaiis. (Plate LII., figs. 4, 4.«-c.) Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. Ul (1876). (? ? . Ccplialotborax smoky testaceous, obovatc, truncated at tbe ends, depressed behind ; caput ascending, convex, with a marked depression on each side ; a central longitudinal ridge ; eyes black, in two slightly arched series : abdomen cylindrical, testaceous, with two central lines and a broad undulated longitudinal dusky band, bounded by thi-ee black spots on each side : the sides silvery ; underside black, with a silvery line on each side : legs long, slender, testaceous, paler below, blackisli at the joints, sparsely dotted with long spines ; palpi slender, testaceous ; male palpus with a clavus resembling the head of a vulture, covered with long bristles ; falces subcylindrical, testaceous, about nine teeth on their inner margins ; pectoral plate pentagonal scutiform, testaceous. Length of body 6 millims. ; relative length of legs 1, 2, 4, 3. (CoU. by Gulliver.) The cephalothorax is slightly convex in front, and concave behind, with the centre of the thoracic region depressed. Eyes black, arranged in two slightly convex series across the anterior part of the caput, those of the posterior series rather larger than those of the anterior. The falces have a long curved terminal movable fang, the maxillae and lab rum arc sordid testaceous. Fovu" examples were collected. Nephila, Leach. Nephila inatirata, Walckenaer. Seventeen specimens, including both sexes, were obtained. Nephila ardentijyes. (Plate LII., figs. 1, la-c.) Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. 443 (1876). ? . Cephalothorax black, haiiy, oblong, with convex margins, contracted behind the caput ; two central shining black tubercles ; a deep excavation at the back of the thoracic region ; caput ascending, laterally excavated ; oculiferous tubercles 504 ZOOLOGY OF KOlJKlLaEZ. shini'nc: ; piil])!* cyliiulrii-al, inalio2:any-red, hairy at tlio base, llie last lliroo joints l)lac'k anil hairy : ahilonion i;Tayish testaceous, silvery pnhescent ; Ihe dorsal ren'ion suhoehnu'oous, eiu'losini^ a quadrangle of lour hlaek spots in front and two parallel rows of hlaek spots behind, where it is also l)ound(>d by six hlaek spots, three on eaehside; sides covered with parallel, iri\'L,'iilar. iiili'rruplfd hlaek lines; underside brownish, dark towards the base and crossed by a yellow haiul, Ix'yond which is an ruva enclosed by hlaek dots and shaped like an heraldic shicdd : leys robust, hairy, mahogany- red ; the apices of the femora, knees, apices of tibitx), the tarsi, and metatarsi black, the black parts densely hairy ; maxilla*, black, w ith nMldish margins ; pectoral plate heptagonal, with eight tubercles, three of Avbieb ai'e prominent and form a triangle ; falecs black, with three marginal teeth. Length 36 inillims. ; relative length of legs 1, 2, 4, 3. (Coll. by Gulliver.) !Most nearly allied to X. nigra of Vinson ; but w ith the abdomen smaller and not clouded with black, and the legs altogether redder. I have examined thirty- seven examples, but have seen no variation. " Web strong and sticky ; used by the natives Avliere we use waxed ends." — G. Gulliver. Nephila instigans. (Plate LIT., lig. 10.) Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. 442 (1876). ? . Cephalothorax oblong, wider behind than before, laterally convex, contracted behind the caput, dull mahogany red, but the back of thoracic region shining black, with a deep central depression ; caput ascending, covered with black bristles, which also form a central longitudinal line ; legs bright ochre-yellow, sprinkled Avith black spines ; the tarsi and metatarsi castaneous, clothed with brown hair ; proximal extremities of the femora blackish; basal half of palpi yellow, nearly smooth, apical half castaneous, clothed with black bristles ; falces dark reddish castaneous, with three teeth on each inner margin ; pectoral shield heptagonal subpyriform, black, with central orange longitudinal streak ; abdomen sandy brown, with a quadrangle of four punctures ; region of epigyne blackish. Length 19 millims. ; relative length of legs 1, 2, 4, 3. (Coll. by GuDiver.) This species does not possess the usual tufted legs. Two examples were collected. Tetragnatha, Latreille. Tetragnatha protensa, Walckenaer. One specimen. ARACHNIDA ARTHUR GARDINER BUTLER. 505 Tetragnatha nero. (Plate LII., tigs. 3, 3rt, 6.) Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. M2 (1876). $ ? . Ceplialotliorax fusiform, truncate at each end, narrowed in front, witli a slender marginal ridge ; caput ascending, bounded behind by one or two depressed diverging lines, and with a central depressed spot ; centre of thoracic region depressed ; eyes black ; palpi long and slender, of the male with a globose clavus with white cap, and held by fiugcr-Hke processes ; falces with ten central teeth on each inner margin, and two terminal teeth, the longer one of which projects for- wards on each side of the movaljlc fang, the latter black ; but the remainder of the cephalothorax and its members tawny ; the palpi, cox93, and trochanters pale ; pectoral plate deep castaneous ; abdomen pale brown, krorated with teneous, sub- cylindi'ical, a darker dorsal line. Length of body, S 8 millims., ? 12; relative length of legs 1, 4, 2, 3. (CoD. by Gulliver.) Tlu'ee examples were found by Mr. Gulliver, who says that it " lies with legs " extended on a single thread placed over or near streams." Gastracantha, Latreille. Gastrucantha manritia, Vinson. Two examples. Uloborus, Latreille. TJldborus vanillarum, Vinson. Six specimens of this species were obtained. • It " makes a web in crevices of " rocks." — G. Gulliver. Miagrammopes, Cambridge. Miagrammopes gulliveri. (Plate LII., figs. 5, 5«, h.) Ann. «fc Mag. Nat. Hist. s. 4, vol. xvii., p. 443 (1876). AUied to M. timaitesii of Cambridge, but differing as follows :— Cephalothorax longer and broader, more angular when seen laterally, and with convex rather than concave margins; caput longer, rather higher than the thoracic region, more convex in front, testaceous, with silvery reflexions and a whitish hind border ; eyes black, forming a convex series across the back of the caput ; lateral eyes twice as large as the central pair, the latter far apart; legs amber-yellow, the femora notched at the distal extremity, the intermediate joints shorter than in M. timaitesii, the tarsi of the hind legs broader, claws black ; palpi short, slender, hairy, testa- 3s 500 ZOOLOGY OF KODKKJlKZ. ccous ; fall-OS small, moval)lo fang piccous; stcnuiin smncwlial dillVront in ronn ; aluloiucn nearly twk-o as long as the ccphalothorax, testacoous, AviUi a I'ontral blai'kisli lint' widor llian in M. f/ncai/rsii, and not Ihrowint;' olV downward l)i'anclics ; basal roirion less contraotod, lateral niargii\s subangiilaled in the middle, whence they eonverge towards tho hinder extremity; ventral siiriaco slightly concave, testaceous, wit li two longitudinal brown lines; spinners concealed by dense hairs, which converge towards the central line. Length 0 millinis. ; relatiA-^e length of legs 1, i, 2, 3. Two examples of this rare and interesting species were " swept from grass " by 'Ml'. Gulliver. Olios, Walckenaer. Olios lencosins, Walckenaer. This species, of Avhich 20 specimens were obtained, is said by Mr. Gulliver to be " common in and out of doors." Olios captiosiis ? Walckenaer. One specimen, obtained by Mr. Slater. Spliasus, Walckenaer. Sphasus liicasii, Vinson. Four specimens. Sphasus ? extensipes. (Plate LII., figs. 2, 2a.) Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. 439 (1876). ? . Cepbalothorax testaceous, irregularly pentagonal, almost flat ; a longitudinal ridge from behind the eyes to beyond the caput : eyes black, placed upon a whitish elevation on the front of the caput ; the anterior pair very minute, in the centre of the face, the others considerably larger and forming a transversely hexagonal figure : front margin of face projecting in tlie centre, behind the projection a V-shaped depression : abdomen testaceous, subcylindrical, flattened below, tapering slightly towards the posterior extremity : legs pale ochraceous ; the tibiae and metatarsi of the two anterior pairs fringed with brown bristles ; front pair very like those of Miagrammopes ; palpi short, pale ochraceous, with blackish tips ; maxillae and falces pale brown ; pectoral plate elongate hexagonal ; ventral svirface of abdomen wdth three central longitudinal grey lines. Length 7 millims. ; relative length of legs 1, 4, 2, 3. Mr. Giilliver obtained one specimen of this little species by sweeping grass. It has the general aspect of the genus Iliogramniopes, but the structiu-c of a Sphasus. ARACHNIDA.— ARTHUR GARDINER BUTLER. 507 Salticus, Latreille. Salticus scahellalns. (Plate LII., fig. 9.) Aim. & Mag. Nat. Hist., s. 4, vol. xvii., p. Wl (1870). ? . Oculiferous region of ccphalothorax black, bounded by a crescent-shaped redcUsli castaneons belt ; thoracic region dark piceous ; eyes opaline ; palpi testa- ceous ; falces, labrum, maxillae, and anterior tibi;p dark piceous ; legs above fawn- coloured, pectoral plate and legs below paler ; abdomen grey, subpyriform, its base black ; a central pale brown longitudinal stripe, bounded laterally by four black spots, which form a square ; venter pale In'own, speckled with dark brown, a central line and the margins dark grey. Length 8 raillims. ; relative length of legs 4, 1, 3, 2. (Coll. by Gulliver.) This species seems to be allied to S. exilis of Cambridge ; only two examples were brought home. Salticus haptizatus. (Plate LII., fig. 7.) Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. MO (1876). ? . Ccphalothorax inverted bell-shaped, dark shining piceous, with a whitish sub- marginal line, bordered within laterally by a rather wider tawny line ; a minute cruciform amber-coloiu'cd spot in the centre of the caput, two lateral crescent-shaped fawn-coloured spots (one on each side) clothed wdth white scales, and a leaf-like fawn-coloured patch in the centre of the thoracic region, the front of which is depressed ; the hinder part and apex of the thoracic patch clothed with whitish scales ; falces, maxillfe, labrum, femora, tibite, and metatarsi of anterior legs castaneous ; pectoral plate fusiform, testaceous ; last three pairs of legs and tarsi of anterior pair amber-yellow, claws black ; palpi whitish : eyes opaline, forming an oblong across the caput ; relative sizes as follows — anterior central pair, posterior, lateral anterior, lateral intermediate : abdomen elliptical, dark brown, with a lateral whitish line and a broad, white-edged, irregular, central testaceous streak ; ventral sui*face whitish, striated with brown at the sides ; a central ferruginous streak ; region of spinnerets testaceous. Length 9 millims. ; relative length of legs 1, 4, 2,3. (Coll. by GulHver.) This species has the general coloration of the much smaller species figured in Lucas's " Algeria " under the name of S. testaceoUneatus. Only one specimen was obtained. Saliicm rodericensis. (Plate LII., fig. 8.) Ann. & Mag. Nat. Hist. s. !•, vol. xvii., p. 1-10 (1876). ? . Ccphalothorax dark piceous, clothed with testaceous hairs ; a central longi- 8s 2 508 ZOOLOGY OF RODRTGUKZ. tudinal reddish castaneous streak from the back of the caput to the posterior margin ; eyes opaline ; falecs, labnim, maxillce, legs above, femora and tibiso below, and epigync castaneous, clothed with testaceous bail's ; tarsi testaceous, with blac-k claws ; abdomen ovate, dark broAAni, pilose, with a dorsal testaceous streak in- terrupted by angulatcd lines at its inferior extremity, and intersected by a black line ; two arched black lines spotted with testaceous from the sides of the testaceous streak to the spinnerets ; pectoral plate, coxse, trochanters, and venter fawn- coloured, base of venter white ; a central tapering grey steak. Length 10 millims. ; relative length of legs 1, 2, 1, 3. (Coll. by GulUver.) Apparently allied to S. intentus of Cambridge: Five specimens were obtained. ACAREDEA. Holothyrus, Gervais. Holothijrii^ ? testudineus (Plate LII., figs. 11, 11 a-c). Ann. & Mag. Nat. Hist. s. 4, vol. xvii., p. 441 (1876). Carapace ovate, convex, conical in front, fawn-coloured, clouded with brown, smooth and shining, with well-defined lateral carina ; legs rather short and robust ; ventral surface flattened and shining behind, with an apparently hinged oblong projecting anterior plate, wider in front than behind ; body in front concave, with only three pairs of legs, their relative size 3, 2, 1 ; antcnnse with hah'y terminal joint. Length f millim, (Coll. by Gulliver.) Although fourteen examples of this species have come, not difi'eriug in size or structure, I have been unwilling to erect a new genus for its reception. Erom many points which it appears to have in common with H. coccinella of Mauritius, I think it possible that it may tm'n out to be an immature condition of that species. SCORPIOlsIDEA. Tityus, Koch. Titytis marmoreiis, Koch. Twenty-four examples were collected by Messrs. Gulliver & Slater. ARACHNIDA.— ARTHUR GARDINER BUTLER. 509 Fig. 1. Nephikt ardentipes, 16- ,, „ ■ic. „ „ 2. Sphasus exteiisipes '2a. 3. Tretragnatha nero „ oOi. „ „ ,j 3o> » >j » 4. Jlfeto vadllans „ '*'^' n » » ■*^* » )j j> *C' » » J) 5. Miogrammopes gulliveri,, ■'f*' » » >) ■"'• )> » )) 6. Pholcvs vexillifer „ VCC. ,, ,, „ 7. Scdticus haptizatv.s „ 8. „ rodericensis „ 9. „ scabellatus „ 10. NepJdla instigans „ 11. Holothyrus teshulineus „ 11a. XXva }j J, i; Description of Plate LII. Butl. Upper surface. „ Profile \-iew enlarged. „ Ventral surface of abdomen, sho^ving scutiform area. „ Pectoral shield, showing arrangement of tubercles. „ Pi'ofilc view, enlarged three times. „ Upper surface of cephalothorax and base of abdomen greatly enlarged. „ Upper surface enlarged. „ Falx opened to show the dentition. „ Male palpus. „ Profile view enlarged three times. „ View of upper surface without the members. „ Falx opened to show the dentition. „ Male palpus, as seen from the side. Profile view much enlarged. Upper surface, enlarged about 8 times. Profile view of cephalothorax. Upper surface, enlarged. „ „ enlarged about 3 times to show form, pattern, and arrangement of eyes. Upper sui'face enlarged. „ ,, natm-al size. „ „ gi'eatly magnified. Under surface „ „ Apparently hinged plate projecting from flattened posterior area of the ventral surface. Lea:. 510 ZOOLOGY OF KODRIGUEZ. CoLEOPTEH.\. — By C. 0. TFafrrhouse, Assistqual, shorter than the first; claws with a membranous lobe beneath each. Body pubescent. The species upon which this genus is founded resembles Haplocnenils, but is of a shorter form, the head is short, and the apical joint of the maxillary palpi is very sti'ongly securiform. Felecophorus is described as having the basal joint of the tarsi very short, shorter than the second ; this I cannot apply to the insect here described, or I should have placed it in that genus. MaUhacodes pictus (PI. LIII., fig. 7). C. TTaterh., Ann. & Mag. Nat. Hist., 1876, xviii., p. 116. M. oblongus, leviter convexus, nigro-scneus, griseo-pubescens ; antennis nigris, articulo basali piceo ; thorace longitudine duplo latiore, sat crebre subtilitcr punc- tate, lateribus arcuatis, tcnuissime flavo marginatis, basi utrinque leviter sinuato ; clytris thorace vix latioribvis at 2J longioribus, crebre sat fortiter punctatis, fasciis duabus flexuosis rufo-testaceis ; femoribus piceis, tibiis tarsisque pallide testaceis. Long. If lin., lat. f lin. A single example only was found of this interesting new genus. Cleridas. Tarsostenus tmivittatus, Spin. The single specimen found does not differ from the European examples. Ptinidae. Xylodes, C. Waterh. Ann. & Mag. Nat. Hist., 1876, xviii., p. 116. General form nearly that of Hedohia. Antennse rather thick, not approximate at the base ; the basal joint oblong, the second shorter and smaller, a little narrowed at its base, the third nearly as long as the first, parallel-sided, the fourth to tenth nearly quadrate, scarcely longer than broad, eleventh nearly twice as long as the preceding, rounded at the apex. Thorax arched in front, as long as broad, slightly narrowed in front and behind. Scutellum pentagonal. Elytra nearly twice as broad as the thorax, and twice and a half as long, parallel, rounded at the apex, punctate-striate. Legs rather short and stout ; tarsi rather short and stout, the COLEOPTERA C. O. WATERHOTJSE. 527 basal joint scarcely elongate, tlie second to fourth a little shorter, the fifth elongate and slender. Body velvety. Xylodes albovaria (PI. LIII., fig. 8). C. Waterh., Ann. & Mag. Nat. Hist., 1870, xviii., p. 117. X. niger, velutinus; thorace supra albo, medio nigro annulate, marginihus nigris ; elytris basi fasciaque post medium dentata albis, humeris nigris. Long. l\-2 lin., lat. f-* lin. Six specimens of the pretty new species were found ; they vary much in size. Cioidse. Lyctus rugicollis. Walker. Bitoma rugicollis, Walker, Ann. & Mag. Nat. Hist., 1858, ii., p. 206. Lyctus rugicollis, C. Waterh., 1. c. 1876, xviii., p. 117. This species, which was originally described from Ceylon by Mr. Walker as a Ditoma, is closely allied to L. ohsitus. Well. A single specimen was found. Lyctiis brunneus, Steph. Tlu'ee examples. Cis insularis. C. Waterh., Ann. & Mag. Nat. Hist., 1876, xviii., p. 117. C. ohlongus, convexus, piceo-niger ; capiter leviter convexo, crebre sat fortiter punctato ; labrum palpisquc testaceis ; thorace longitudine paululo latiore, convexo, antice paulo augustato, crebre sat fortiter punctato, margine antico leviter arcuato, supra caput vix superante, angulis anticis omnino rotundatis, lateribus leviter arcuatis, reflexo-marginatis, angulis posticis obtuse rotimdatis basi marginata ; elytris thorace vix latioribvis, at 2\ longioribus, convexis, ad apicem arcuatim attenuatis, sat crebre fortiter punctatis ; antennis pedibusque piceis, tarcis testaceis. Long. 1^ lin., lat. \ lin. This species has the elytra rather vmusually narrowed at the apex ; the punctua- tion is very distinct, thick but not crowded ; in the middle of the forehead there is a small, ahnost imperceptible fovea ; the elytra are covered with a slightly metallic bloom. A single specimen only. Cis sexcarinatus. C. Waterh., Ann & Mag. Nat. Hist., 1876, xviii., p. 118. C. ohlongus, ater, vix nitidus, fortiter confertim rugoso-punctatus ; capite piceo ; antennis testaceis, clava picea ; thorace longitudine \ latiore, lateribus fere pai'allclis (vix arcuatis), angulis obtusis, basi bisinuata; scutello la^vi; elytris basi thorace 528 ZOOLOGY OF RODRIGUEZ. hand latioribiis at diiplo longioribus, latcribus fere rectis ad apicem obtuse rotun- datis, sutura pannu clovata, singulo elytro ad apicem carinis tribiis instructo ; corporo sxibtus baud erebiv fov1itov]iiin('1ato ; podibus ])iciMS. Long. 11 mill., lat. | mill. Thirteen specimens found in fungus. HETEROMERA. Opatrinae. Opatrum ccesum, Stev. Eight examples, which agree with the description of this species, Cistelidae. Clstela hrunnea (PI. LIII., fig. 9). C. Waterh., Ann. & Mag. Nat. Hist., 1876, xviii., p. 118. C. oblongo-ovalis, parum convexa, sat nitida, hrunnea, brcviter aureo-pubescens ; capite triang'ulari, crebre distinctc punctato ; autcnnis thoracc duplo longioribus; thorace obscuriore, longitudine fore duplo latiore, antice arcuatim parum angustato, confertim fortiter pimctato, angulis anticis omnino rotundatis, posticis rectis, basi recto-truncata, medio solum vix lobata ; elytris thorace vix latioribus at 3^ longiori- bus, ad apicem arcuatim attenuatis, irrcgularitcr crebre punctatis. Long, -i lin., lat. If lin. The elytra are punctatc-striatc at the apex, the stria next the suture extends for a little distance towards the scutellum. This species, of which four specimens were brought, may eventually have to be separated from Cistela proper. RHYNCHOPHORA. Anthribidae. Balanodes, C. Waterh. Ann. & Mag. Nat. Hist., 1876, xviii., p. 118. Head as long as broad ; rostrum a trifle narrower, very short. Antennae placed close to the eye, a little longer than the head and thorax together, very slender ; the first and second joints scarcely stouter than the following ; third to seventh a trifle longer than the second, subequal, the eighth rather shorter but a little elongate ; the ninth to eleventh subequal, forming an elongate, slender, loose club. Eyes a little prominent, slightly ovate. Thorax nearly semicircular, truncate in front for the width of the head; the basal ridge very slight, scarcely separated from the posterior margin, only visible near the shoulders. Scutellum minute. Elytra at the base scarcely as broad as the thorax, but a little more than twice as long, gradually narrowed towards the apex, which is rounded, punctate-striate. Tibi^ COLEOPTERA.— C. O. WATERHOUSE. 529 cylindrical, not widened at the ajjex; the anterior pair rather long- and curved beyond the middle, minutely denticulate beneath the base. Tarsi moderately broad, the basal joint a little elongate, the second transverse, the third short, bilobcd ; claws with a distinct mesial tooth. Body evenly convex, thickly clothed witii pubescence. Closely allied to Arceocori/iius, but, besides the difference of form it differs in having slightly narrower tarsi and the tibiae are not Avidened at the apex. Balanodes tomentosus (PI. LIII., fig. 10). Ann. & Mag. Nat. Hist., 1876, xviii., p. 119. B. oA'alis, convexus, dense breviter flavo-griseo-pubescens, obscure piceus, elytris pedibusque brunneis ; thorace creberrime baud fortiter punctato, angulis anticis obHteratis, posticis rectis, basi utrinque vix sinuata ; elytris fortiter striato-punctatis, interstitiis alternatis vix latioribus, subtilissime coriaceis ; antennis articulis tribus apicalibxis piceis. Long. 2| lin., lat. 1^ lin. The pubescence on the elytra is slightly interrupted by the rows of punctui'es, which gives them a striped appearance. Another of the interesting new genera, of which only a single example was found. Caranistes annulipes. Ann. & Mag. Nat. Hist., 1876, xviii., p. 119. C. oblongus, nigro-piceus, dense piceo-tomentosus, flavo variegatus; pedibus pallide brunneis nigro aunulatis. Long. 2i-3 lin., lat. 1^-1.} lin. Front of the head and the rostrum clothed with yellowish pubescence, that on the vertex brown ; eyes rather widely separated above, very slightly emargiuate in front. Antennge obscure testaceous, the club only dark, twice as long as the thorax ; the third to eighth joints very slender and gradually diniinisliing in length ; the ninth to eleventh elongate, forming a distinct but slight club. Thorax very finely granular, nearly one-third broader than long, much narrowed in front, the sides nearly straight, rather narrowed behind the basal ridge, which is very distinct, well separated from the base (especially at the sides) ; the pubescence is brownish. Scii- tellum yellowish. Elytra as broad as the thorax, one-fourth longer than broad, not veiy convex, rounded at the apex, rather strongly punctate-striate, the interstices very slightly convex ; the pubescence is yellowish, interrupted by small brown square spots ; a few of these spots are larger and darker than the others ; and the yellow pubescence has a tendency to form a sutural spot beliind the middle. The legs are 3x 530 zoor.OGY OF KoniJir.rEz. very pale browu; a liug uii the femur, two ou the tiltia, the apieal half of the basal tarsal joint, and nearly all the secoud joint l)laekish. Some large punctures arc scattered over the Hanks of the prosteriuun and the metastermun. Three examples, found under hark hy 'Slv. Gulliver. Curculiouidae. Ci'otopns iiioriialKS, Ann. et Mag. Nat. Hist., 1S70, xviii., p. 120. C. elongatus, niger (vel nigro-pieeus), nitidus, setis albidis In-evissimis paree adspersus ; autenuis piceis ; rostro basi sat crebre, apice pareius loevius punctulato, fronte discrete distincte punctata ; oculis oblongis, baud prominentil)us ; tliorace lougitudine \ latiore, antice angustato, vix crebre sat fortiter punctate ; scutello la?vi ; elytris thorace paulo latioribus at 3^ lougioribus, ad apicem regulariter acuminatis, fortiter striato-punctatis, interstitiis vix irregularibus, punctis parvis adspersis, apice tuberculis parvis obsitis, marginibus dimidio basili distincte cariuatis, ad apicem serrulatis ; pectore breviter griseo-pubescente ; femoribus anticis subtus dente parvo armatis. Long. 5-6^ lin., lat. 2-2i liu. Seven specimens. Cratopus viresceiis. Ann. & Mag. Nat. Hist., 1S7G, xviii., p. 120. C. elongatus, pioeus, dense virescenti-pubescens ; capite rostroque crebre sat fortiter punctato ; oculis vix prominulis ; thorace longitudine vix latiore, antice angustato, crebre irregulariter granulate, lateril)us arcuatis ; scutello viridi-albo tecto ; elytris thorace paulo latioribus at 3^ longioribus, apice angustatis, sat fortiter striato-punctatis (piuietis rotundatis, medio tuberculo minuto instructis), interstitiis baud convexis, parce suljtilissime asperato-pvmctatis, margini])us hand carinatis ; femoribus anticis subtus dente magao armatis. Long. 5-G lin., lat. 2-2f lin. Nine specimens. Cratopus magnifictis. Ann. & Mag. Nat. Hist., 1876, xviii., p. 120. C. elytris piceo-niger, dense squamulis rotundatis parvis tectis ; squamulis Isete viridi vel aurato vel cupreo nitentibus ; sutura kevi. Long, elvtr. 5i lin. Only t'O'o elytra of this most beautifid insect were found by Mr. GulliA-er in a spider's Tveb. The colours are most brilliant, reflecting bright copper, golden,. CULEOPTEHA.— C. O. WATERHOUSE. 531 emerald green, or pale bluish-greeu in stripes, according to tlic position iu which the insect is held. Mr. Gulliver has this note on the species : "I was told that this beetle occurred " very rarely, but is found occasionally concealed in the sheaths of the leaves of " the maize plant." SitophUus linearis, Hbst. Five examples. Sitophilm ori/zllcctocl by Messrs. G. Ciulliver and H. H. Slater, twelve of whieli belonu; to the Acnleata, the remainder to the Tere- bnnititi or lehueumouidoiis division ; five of tlie iirst of these divisions ap])enr to be undi^seriVied. Six species of Formiciihe "were obtained, four being- new to science: the other two have a cosmopolitan range; one is the species described by Heer as the Ilonse Ant of Madeira, Pheidole pnsilla ; the other is the Trenolepis graciUenceiit! of Nylander, also foiuid in ^Madeira, but common in conservatories generally ; it is found abundantly in the Botanical Gardens both at Kew and Paris. The most interesting species of Apidce is the Apis itnicolor of Lati'cille, first dis- covered in Madagascar, and subsequently in the island of Bourbon. Two species of leaf-cutting Bees were found : both have a wide geographical range, as has also the only species of "Wasp, Polis/es hehrivus ; this lias occurred in India, China, Mauritius, 31adagascar, and in several of the islands of the Eastern i\.rchipclago. Six species of Diptera were obtained, two belonging to the TiptiUdce ; these latter were in a fragmentary condition, and coiild neitlier be identified nor described ; of the remaining foiu" one is a described species, the others being apparently new. Of Neuroptera three species were foimd, all of which are described by Bambur from specimens obtained in ^Mauritius. HYMENOPTERA. Hitherto found in Rodriguez only. Previously obtained elsewhere. Kumc of Species found in Uodriguez. .Mauritius aud Bourbon. X X X X X X ilada*ras- car. Other Localities. Tapinoma jjallipes. Smith ., fragile, Smith Mononiorium elongatuni, Smith - ,. iinpressum. Smith - Preuolepis gracilleseens, Xi/land. Pheidole pusilla, Heer - - - Scolia (Dielis) rufa, St. Fart/. - Pelopanis convexu?, Smith Megacliilc nifiventris. Gucr. ,, lanata, Fahr. - - - Apis unicolor, Latr. Polistes hebrsiis, Fair. - Evania laavigata, La/r. - - - Paniscus perforator. Smith Ophion rufiis. BruUt . - - X X X X X X X X Cosmopolitan. Co?mopolitai). India. India. India, Java, China. Australia, Angola, Egypt, Eniope, Mexico. Bengal, Java, Africa. UYMENOPTERA. — F. SMITH. 535 FormicidsB. TcquiWiHu, roei'stei'. Topiuouia pdUipes. Ann. & ilag. Nat. Hist., 4th sev., vol. xvii., p. 447 (187()). TTorker. Leugth \\ Hue. Head aud thorax: reddish bro^ii, abdomen (hu-k brown ; eves black ; mandibles and antennre pale testaceous. Thorax rounded anteriorly, slightly narrowed posteriorly, with the metathorax obliquely truncate ; the legs, petiole, and squama, pale testaceous, the latter rounded above. Ab- domen smooth, shining, and irapunctate. Coll. by Gulliver. 20 specimens taken. Taphioma fragile. Ann. & Mag. Nat. Hist., 4th ser., vol. xvii., p. 447 (1876). Worker. Leugth 1 line. Head brown, smooth, and shining ; the eyes black ; mandibles white and pellucid, antenuoe also white. Thorax pale brown, the sides and metathorax darker brown ; the prothorax rounded in front ; the meta- thorax oblique and concave ; the legs, petiole, and squama, pale testaceous, nearly white. Abdomen brown, smooth, and shining, palest at the base. Coll. by Gulliver. Three specimens taken. Found under bark ; very like the preceding species, but smaller ; the legs and antennae being pellucid. Myrmicidae, Smith. Monomorium, Mayr. Ilonomorium irtqyressiim. Ann. & Mag. Nat. Hist., 4th ser., vol. xvii., 447 (1876). Female. Length Ij line. Head and abdomen dark fusco-ferruginous. The clvpcus, mandibles, and antennte flavo-testaceous ; ocelli distinct, with an im- pressed line in front of the anterior one, extending to the insertion of the antennte ; the apex of the scape and the club of the fiagclium slightly rufo-fuscous. Thorax narrower than the head and one-third longer ; the prothorax rounded in front, the sides nearly parallel, slightly narrowed towards the metathorax ; the legs pale testaceous, the femora rufo-testaceous ; the tarsi colovu-less and pellucid ; the thorax testaceous, as well as the nodes of the abdomen. Abdomen oblong-ovate, as long as the head and thorax. Male. Length 1^ line. Dark fusco-ferrugiuous ; mandibles and antennte rufo-testaceous ; ocelli prominent, with a deeply impressed line in front of the anterior one. Thorax oblong, narrowed posteriorly ; the wings colourless hyaline ; 53G ZOOLOGY OF KODIUGUEZ. legs wiiito, witli tho J'cniora slightly fuscous in the middle. Ahdomen o])long- ovate, smootli, and shining. Nino specimens found under havlc. Coll. by Gullivev. Jloiioiiioriuiii. cloiigofinu. Ann. & Mag. Nat. Hist., Ith s(>r., vol. vii., p. dlS (1876). Fen\ale. Length 2 lines, lleddish brown ; head oblong, slightly narrowed behind the eyes, trmicato posteriorly; ocelli glassy, the anterior one placed in a deep fossulet ; mandibles and antcnucc pale rufo-testaceous. Thorax oblong, narrowed behind ; the metatliorax truncate ; legs pale rufo-testaceous, the femora slightly fuscous. Abdomen oblong-ovate, smooth, and shining, with a thin scattered pubes- cence. "Worker. Length 1 line. Kcddish brown, shining, with the mandibles, antenna), and legs pale testaceous ; the head oblong, the sides nearly parallel, truncate behind, ■with the angles rounded. Thorax strangulated a little beyond the middle ; the metatliorax without spmes. Abdomen ovate, smooth, and shining. Fifty-three specimens collected by Gulliver. SCOLIID^. Scolia, Fabr. ScoUa rufa, St. Fargeau. Z^Iale. Length 7^ lines. Black, the pubescence fulvo-ferruginous ; the face densely pubescent ; the clypcus and mandibles yellow, the former with a large triangular black macula in the middle, the latter rufo-piccous at thcu' apex and narrowly so on their lower margin ; the head pubescent behind. Thorax pubescent, the metathorax densely so ; wings fusco-hyaline, the nervures black, slightly irides- cent ; the anterior legs, with the tips of the femora above, also the tibiae and tarsi above, yellow ; the tips of the joints of the tarsi black ; the intermediate and pos- terior tibiae yellow above, their tarsi entirely so, with the tips of the joints black ; all The calcaria pale testaceous. Abdomen with broad yellow fasciae, the fascia; emarginate in the middle and at the sides ; the apical segment black, smooth, and shining, with a few strong puncttu'es at the base, the rest of the abdomen yni\\ shallow punctures ; beneath, the segments have narrow yellow apical marginal fasciae, and are fringed with fulvous pubescence. Two specimens. The female only of this species was known, until Mr. Gulliver took both sexes. Sphegidae. Pelopceiis, Latreille. Telopaiis convexus. Ann. & Mag. Nat. Hist., 4th ser., vol. vii„ p. 449 (1876). Male. Length 5-6J lines. Head and thorax blue ; abdomen and legs blue HYMENOPTERA.— FIIEDEKICK SMITH. 537 black. The face witli silvery pubcsceuce ; on the cheeks a thiu gi-iseous pubes- cence. The )nesothorax evenly punctured and convex, with a central impressed longitudinal line, without any other trace of longitudinal channels; rather strongly but not very closely punctured, on the disk a few transverse irregular carinne; wings subhyaline, with then- apical margins clouded; the thorax has a thin sparing griscous puljcsccnec ; the apex of the metathorax with bright silvery pubescence ; the tibiie with the calcaria black. (Coll. by Slater.) Three specimens obtained, which differ so much from the males of all the blue species of the genus, in having the thorax very convex, that the male is described as that of a new species : the captiu-e of the other sex will probably prove the correct- ness of doing so. Apidae. Megachile rufiventris, Gu6r. A single specimen taken by Mr. Slater. Megachile lanata, Fabr. A single specimen taken by IMr. Gulliver. Apis nnicolor, Latr. Thirty-two specimens taken by Mr. Gulliver. VespidaB. Folistes hebrceics, Linn. Seven examples taken by Mr. Gulliver. Ophiouidae. Oj)Mon rufus, Brulle. Two specimens taken by Mr. Gidliver. Taniscus perforator. Smith, Ann. & Mag. Nat. Hist. 4th ser., vol. vii., p. 449. Female. Length 6 lines. Rufo-ferrugmous ; the face and scape of the antennje in front, yellow ; the space between the ocelli black. Thorax : the sides, the sutures on the mcsothorax, and the scuteUum paler than the disk ; wings hyaline and iridescent, the stigma and nervures rufous ; the legs and abdomen rufous, the latter slightly fuscous at the apex. A single example taken by Mr. Slater. This species is like P. melanoptts, Brulle, but it differs in having all the tarsi rufous, as well as the nervures of the wings : in P. melanopus they are black. 3 Y 538 ZOOLOGY OF UODIJIGUEZ. EvauidsB. Evania hcvigata, Latr. A sm2:le specimen taken by !Mi-. Gnlliver ; the genus Evania being parasitic on BlattiiUo, some of the species are conveyed Avith merchandise to all parts of the world. U. hcvigata is one of the species that has a Avido distribution, and is found occasionally in warehouses in the docks in Loudon. DIPTERA. Tabanidse. Tahanus seqiieiis, Walker. A single specimen taken by Mr. Gulliver. Muscidae. Pollenia hasalis, Macqnar. Smith, Ann. & Mag. Nat. Hist., Ith ser., p. 119 (1876). Length 1 lines. Dark blue, with shades of green on the disk of the thorax ; the face and cbeeks luteous, and clothed with short, dense yellow pubescence ; at the end of the facialia a black bristle on each side ; antennaj in deep facial grooves, slightly yellow at the base, third joint long, arista longly phunose ; palpi luteous, with black bristles ; frons black and more than a quarter the width of the head ; wings hyaline with then- base fuscous ; the transverse vein at the end of the wing nearly rectangular below and much cm'ved above, leaving the cell open ; the lower transverse vein very oblique. Abdomen with a little short luteous pubescence at the apex and also beneath ; alulce dirty white. A single example taken by Mr. Slater. Sarcophaga niutata. Smith, Ann. and Mag. Nat. Hist., Ith ser., p. 150 (187G). Female. Length 5 Hues. Black, thinly clothed Avith black haii-s and bristles ; the face and cheeks Avith a covering of bright silvery pile ; the margin of the facial groove fringed anteriorly with short black bristles ; eyes dark reddish broAvn ; the base of the antennae black, the thu-d joint tawny and Avith a long plumose black bristle at the base; frons AA'ith a broad black stripe rimning from the antennte to the vertex. Thorax taAAniy above and Avith three black longitudinal stripes ; the lateral margins set AA'ith black cm'ved bristles ; legs black and with a number of black bristles ; wings sub-hyaline, the veins black ; the tranverse vein at the end of the DTPTERA AND NEUROPTER A.— FREDERICK SMITH. 539 yrins: stvaifyht al)oro, ohliquo mid curved below ; the lower tvausrersc vein evenly wared ; the akilfo white. Abdomen tessellated with black and silvery pile, covered with short blaclc hairs, and having a few long black bristles at the apex ; there are also foiu' long black bristles on the hinder margin of the scutellum. Two examples taken by Mr. Gulliver. Sapromyza squalUda. Smith, Aim. and Mag. Nat. Hist., 4th ser., vol. vii., p. 950 (187G). Male. Length 2 lines. Pale brown ; frons yellow, with a dark brown central line ; two long bristles at the hind corner of the eye, and two between them and the antennre ; face pale yellow ; peristoma with a row of small bristles. Antennae yellowish, brown above, the second joint with long black bristles beneath, at the end ; arista black, yellow at the base and plumose ; the palpi yellow. Thorax with obscure pale (almost white) longitudinal stripes ; the scutellum pale brown, its outer margin and a central longitudinal pale yellow line. Legs yellow ; the tibife have all a preapical bristle ; the anterior femora with scattered bristles, the intermediate pau* with a row of bristles in front near the tip. Wings faintly yellowish, middle transverse veinlets darkened, and yellowish spots at the tip of the lower transverse veinlet and at the tips of the second and thii'd veins, and also on those veins, some distance from the tips, and on the second vein opposite where the first vein ends in the costa. Abdomen brownish yellow and thinly sprinkled with short black hairs. A single example taken by Mr. Gulliver. NEUROPTEEA. Hitherto found in Kodriguez only. Previously obtained elsewhere. Name of Species found in Rodriguez. Mauritius and Bourbon. Madagac car. Other Localities. Termes ? - - - Myrmelion obscui-us Libellula manritiana Anax mauritianus Agrion ferrugineum X X X X X Libellulidse. Libellula manritiana, K-ambur. Four specimens obtained by Mr. Gulliver. 3 Y 2 540 ZOOLOHY OF PvODKinrEZ. Anax mauritianns, Rambur. Four specimens taken by Mr. Gulliver. Agrionidae. Agrion fcrruginenmt Rambur. A single example only obtained by Mr. Gulliver. Termcs, sp. One buudred and seven specimens of a small siiecios of Termite were in the collection, they were snow wlute, but only Avorkers Avcre obtained ; in the absence of tbe females and males it was impossible to draw up distinctive specific characters. Obtained by Mr. Gulliver. Jjltjrmelion ohscurus, Eambur. A single specimen only obtained by Mr. Gulliver. 541 Lepidoptera. — By Arthur Gardiner Butler, F.L.S., F.Z.S., ^-c. Tw'cnty-ono species of Lejiidoptera were obtained by Mr. George Gulliver; of these, seven are Ehopalocera and fourteen Heterocera ; of the latter, four arc new and probably endemic forms. Twelve of the species, including the whole of the Butter- flics, have been previously known from Mauritius and Bourbon, and eight of them, including six of the Butterflies, from Madagascar. All the genera, with the exception of Lavei^na, and eight of the species, have been recorded from India or Cevlon. The species here enumerated evidently form but a very small part of the Lcpi- dopterous fauna of Rodriguez : in all probability many of the species which occur at Mauritius and Bourbon will be found to exist there also, of these Terias Jloricola and Atella columhlna among the Bhopaloccra, and Ch(crocanipa celerio, Daphnis nerii, Frotoparce convolvuli, Deiopeia jiulchclla, Fliisia chalcills, and Ueliolhis armigera, wdth several other wide-ranging species, amongst the Ilcterocera, may be confidently looked for. It is interesting to find Acherontia atropos in Rodriguez ; the Mascarene Islands probably form the extreme limit of this species, which (with Frotoparce convolcull) seems to be common throughout Europe and Africa, but does not occur in Asia, the Indian representatives of A. atrojws being all perfectly distinct. Name of Species fonnd in Rodriguez. Hitherto found in Rodriguez only. Previously obtained elsewhere. Mauri tins and Bourbon. Madagas- car. Other Localities. Danais chrysippus, Linneeus Melanitis ismcne var., Cramer Junonia rhadama, Boisduval Diadema raisippus, Linnisus Lampides telicanus, Jlerbst Hesperia forestan, Cramer Pampliila borbonica, Boisduval Aclieiontia atropos, Linntcus Aigina cribraria, Clcrc.k - Laphygma cycloidos, Gitenec Prodenia littoralis, * lioisditval Caradrina cxpolita, liutlcr Cosraophila xautliiiKlyraa. Boisduval Dioinca bryophiloidcs, Biif/er Homoptera tuibida, Butler Acliaja catilla, Gitcncc Ophiusa properans, H'al/icr Trigonodes acutata, Guenee Remifjia conveniens, Walker A.sopia vulgalis, Giienie - Laverna plumipes, Butler X X X X X X X X X X X X X X X X X X X X X X X X Africa, Asia, S. Europe. Africa, Asia. Africa, Asia. Africa, S. Europe. Natal, E. Africa, Sierra Leone, India. Europe iind Africa. India, C'liina, Malaysia. S. Africa. India, Ceylon. Asia, ? Australia. Senegal, Abv.ssinia. S. Africa, Natal, W. Africa. S. Africa, Natal, Senegal, Sierra Lcono. Natal, Congo, Sierra Leone. Africa, Ceylon, West Indies. Identical with P. testaceoidet, Gu^nec. 5^2 ZOOLOGY OF liODRTGUEZ. Rhopalocera. Dauais chrysippns, L. Five specimens, of both sexes, were colloctcd. Melanin^ ismeiie var., Cramer. Two specimens. Jnnonia rhadama, Boisd. Fifteen examples of botli sexes. Diadem a misippns, L. Only cue male. One female. One male. One male. Lnmpkles /elicaiins, Ilcrbst. Jlespo'la forestan, Cramer. .VampJnla horhonica, Boisd. Heterocera, Acherontia afropos, L. Six examples of both sexes. Argina crihraria, Clerck. Five specimens, one of which is the variety figured by Cramer (pi. 208, fig. C), and another is the P. astrea of Drury. Lapliygma cycloides, Guenee. Three males. Prodenia littoralis, Boisd. One male. Caradrina expoUta. Ann. & Mag. Nat. Hist., s. 4, vol. xvii., p. 407 (1876). Primaries glossy brown, reniform spot indistinct, a brown streak below it ; a waved dark brown transverse line towards the base ; a transverse postmedian Kne, arched, with its lower extremity straight, followed by a series of black points ; an m-egnlar dnsky discal band, zigzag externally ; a marginal series of black dots ; the margin indicated by a clay-coloured line ; fringe shining brown, darker at base : secondaries shining pale bro^ii, with deeper-coloured outer border; marginal characters and fringe as in primaries : body pale olive-brown, abdomen greyish ; LEPIDOPTEKA.— ARTHUR GARDJNER BUTLER. 513 palpi black, with wliity-browu tips : wiugs below shining brownish, a dusky spot ou the discoceUuIars ; au obUque black costal litm-a in primaries, followed by a continuous discal series of black dots upon the nervm-es ; an oehraceous margmal line and a marginal series of black dots ; secondaries paler than primaries : body shining whitish. Expanse of Avings 1 inch. (Coll. by GuUiver.) Allied to C. co»formis. Cosmophila xanthindyma, Boisd. Two specimens. Diomea hryophiloides. Ann. & Mag. Nat. Hist., s. 4, vol. xvii., p. 408 (1876). Primaries rounded at apex, smoky brown ; external area darker, crossed by a blackish tapermg baud bounded externally by whitish dots ; several costal spots of the same colour : secondaries pale hxavm ; margin, veins, and a spot on the dis- cocellulars dusky : body smoky brown, abdomen pale ; palpi whitish above, blackish below ; primai-ies below paler than above, markings indistinct, a marginal series of black dots ; secondaries sordid white ; costal area irrorated with brown ; basal area oehraceous ; a black discocellular spot ; two irregidar discal brown lines ; outer border grey ; a marginal series of black dots ; legs spotted with ochi-aceous ; coxae tufted : body below whitish varied with broA\Ti, sides of pectus oclu'aceous. Expanse of wings 11 lines. (Coll. by GuUiver.) One example. Homoptera turbida. Ann. & Mag. Nat. Hist., s. 4, vol. xvii., p. 408 (1876). Primaries cinereous, crossed by undulating brown lines ; a blackish line at the base, followed by a dark oblique brown baud, the outer edge of which is straight, and bounded towards costa by a white streak ; reniform spot ill-defined, confounded with a brown costal patch crossed by pale lines ; the latter bounded externally by an abbreviated narrow Avhite streak ; postmedian line slender, black, siuuated at its upper end, regularly undulated from the first median to the inner margin ; outer border broadly grey; a sub marginal black line, a marginal series of yellowish spots ; margin indicated by a reddish brown line ; fringe brown, with basal ochi'aceous line : secondaries pale buff, internal area (excepting the abdominal region) and the outer border cinereous, crossed by parallel undulated broMii bands ; two parallel subanal black lines, between which is a brown line ; marginal characters as in primaries : head, palpi, and collar pale reddish brown ; tcgulio and thorax greyish brown ; abdomen pale brown, three dorsal greyish tufts ; two subanal dusky bands, the inner one broad ; anal tid't tipped with broAVu : wings and body below pale 511 ZOOLOGY OF RODUIGUKZ. bull", markincfs only indicated en \hv apical half of ]n'imarics and cosla of secondaries, the latter covered Avith hair; an apical niar;j,iiial sinuatcd black line. Expanse of uinns. 1 inch 10 lines. (Coll. by Gulliver.) -Vllied to, but ij^uitc distinct from, II. rinsouii of Gucnoe. Aclicea catllla, Gu6n6e. Two specimens, rather more strongly marked than Abyssinian examples (whicsh in all prol)ability would agree with the type from Senegal), but not sufficiently distinct to warrant their separation. One example obtained. Five specimens. One specimen. One specimen. Oplmisa properans. Walker. Trigonodes acutata, Guen6e. JRemigia conveniens, Walker. Asopia vnlgalis, Gucnec. Iiccverna plumipes. Ann. & Mag. Nat. ffist., s. 4, a^oL xvii., p. 409 (1876). Primaries shining brown, irrorated with black ; a subapical black spot ; fringe grey : secondaries sericeous pale brown, the apical area dai'ker ; fringe pale brown : head and thorax olive-brown, abdomen pale brown : wings below shining i)ale brown : body and legs below shining whitish ; tibia3 of second pair of legs with a long terminal pencil of hair-scales betAveen the spines ; tibioe of hind pair densely clothed with long projecting hairs. Expanse of wings 10 lines. (Coll. by Gulliver.) 545 Orthoptera and IIemiptera.— i?// Arthur Ganlincr BiUler, F.L.S., F.Z.S., SfC. (Plate LIV.) Fifteen species of Orthoi^tera were obtained by Messrs. Gulliver and Slater, one of these was, however, in an ininiatiire condition, and therefore only its genus could be ascertained ; four of the species appear to be eudciuic and new to science, the species of most interest being P/m'«w spmlfera,^ form nearly allied ioT.pec- tinata from the Samoa Islands, Molluccas, and Ceylon. Of the OrthoiDtera previously described, seven are recorded from Mauritius and Bom-bon, and three from Madagascar ; two of the latter are, however, cosmopolitan species. Amongst the Jlemiptei'a of which there are 20 species, five are new to science ; two are immature and cannot be specifically identified. With the ex- ception of three species, the whole of the recognisable forms are referable to the Memiptera Seteroptera. Their geographical distribution seems to be as follows : five endemic, five occurring in Mauritius and Bourbon, and seven in Madagascar ; two of these, Nezara viridula and Conorhhuts ruhrofasciatus, appear to be cosmopolitan or nearly so, and one species, Ser'metha lateralis, seems to be generally distributed over Africa ; in addition to these, four species have hitherto only been recorded from South Africa, and two from Algeria. ORTHOPTERA. Hitherto found in Rodriguez only. Previously olitained elsewhere. Name of Species found in Rodriguez. Mauritius and Bourbon. Madagas- car. Other Localities. Forficula auricula, Linn. - - - Forficula varicornis, Smith Pauclilora coiticiini, Scrville Phyllodromia gcrmanica, Linnaus Pcriplaiiota amcricana, Linnaus Periplancta rliombifolia, StoU Polyzosteria latipe.s, JVaUicr Grvllus capensis, Fnhricius Mogoplistes sp. (immature) Neuiobius lutoolus, Butler Phisis spinifera, Butler - - - Xiphiiliuin iris, Scrvi/lc . - - Conocephalus diflerciis, Scrville - Pachytylus cinera^ceus, Fair. Epachroniia roilericcnsis, Butler Jljuitis variegata, Olivier Bacillus incommodus, Butler X X X X "x X XXX X XXX X X X Cosmopolitan. Congo, Whydah. Cosmopolitan. Cosmopolitan. China, Bivizil. Sierra Leone. Almost cosmopolitan. Old World generally. Natal and Sierra Leono. 3z 546 ZOOLOGY OF RODHIGUEZ. Forficula auricula, Linn. A single specimen taken by Mr. GuUiTcr. This insect is now almost cosmopolitan. Forftcula varlcornis, Snuth, Ann. & Mag. Nat. Hist., dth sor., vol. vii., p. 150 (1870). Length 1^ lines. Nigro-pieeons, sinning; the palpi, mouth, and legs pale testaeeons, the prothorax rul'o-pieeons ; antenna? 15-jointed, joints 12tli and 13th Avliite, five or six of the hasal joints rulo-pieeous, as arc also the palpi ; the prothorax oblong-quadrate, narrowly mai-gincd, and with a longitudinal im- pressed line not extending to the posterior margin; apterous ; the abdomen smooth and shining ; the basal half of the segments with fine shallow punctures ; the apical segment with a deeply imi)ressed longitudinal line, its posterior margin emarginate ; the forceps very stout, triangular' at the base, curved iuAvurdly beyond the middle, acute at the apex, their inner margin crenulatcd. Ten specimens obtained by "Mr. Gulliver. This species would probably come into the genus Brachylabis of Dohru. Tanchlora cortlcnm, Servillc. Six examples were obtained by Messrs. Gulliver and Slater. This species has been considered svnonvmous with P. Surinamensis, but the African and Mascarenc examples seem to be constantly shorter in the wings than those from Tropical America. JPkt/llodromia germanica, L. One example. Periplaneta americana, L. Mr. Slater collected five specimens of this species. 'Periplaneta rhoinhifoUa, Stoll. The fom* specimens obtained differ in no respect from Chinese or Brazilian examples. Polyzosteria latipes, Walker. Eight examples, slightly varying in punctuation. Gryllv.s capensis, Fabr. Ten examples, varying slightly in colour. Mogoplistes sp. (immature). Two specimens. ORTHOPTEUA.— ARTHUR GARDINER BUTLER- 5J:7 Nemohius luteohts. (Plate LIV., figs. 2, 2^) Ann. & Mag. Nat. Hist., s. 4, vol. xvii., p. 409 (187G). Stramineous; vertex of head greyish ; pronotum iiTorated Avith hroviiish, with a lateral slender ridge and a central depressed line ; posterior abdominal segments greyish ; cerci as long as oviduct, hairy ; oviduct castaueous helow and at the tip ; tegmina nearly extending to the end of abdomen, pale stramineous ; tibiaj of hind legs -with three pau*s of spines, also several terminal shorter spines. Length of body 5 millims., of oviduct 2. (Coll. by Gulliver.) The single example is somewhat shrunken, and may have been killed before attaining its full coloration. The head is obtusely produced between the antennae, and is slightly depressed in the centre ; the oviduct, as usual, is curved upwards ; the wings arc whitish and extend when closed to the end of the abdomen. Phuis sjnmfera. (Plate LIV., figs. 1, 1".) Ann. & Mag. Nat. Hist., s. 4, vol. xvii., p. 410 (1876). Female. Testaceous, pronotum oblong, greyish, with carinatcd margins ; abdomen greyish ; oviduct stramineous ; the femora of the two anterior pairs of legs with two converging longitudinal ferruginous lines above, of the first pair Avith six exterior and five interior lateral spines, distal end terminating on each side in a conical denticle ; tibiae with seven spines on each side ; femora of second pair with seven exterior short spines, inferior margin denticulated ; tibia) with six exterior and five interior spines, longer than in the femora ; femora of hind pair reddish, spinulose below ; tibiae denticulated. Length of body 19 millims., of oviduct 9. (Coil, by GulUver.) More robust than P. pectinata, rather differently coloured; anterior femora with strong terminal conical denticles. The oviduct is blackish at the tip, compressed, and curved upwards ; the tibiae are more strongly denticulated on their superior than on their inferior surface. Xiphiclium iris, Serville. Two specimens of this delicate species were collected by Mr. GulUver. Conocepludus differens, Serville. Eight examples. The species obtained under this name from M. Erunucr de Wattenwyl, and occurring in the Philippines, is entirely distinct ; it is a much 3z 2 5-18 ZOOLOGY OF RODRIGUEZ. more elougatod species, with the head nuieh more produced aud acuminate. M. Servillo says of liis type " trouYc>s i\ rilc-dc-Erauce," aud his description in all respects agrees Avith examples from llodriguez. Pachytylns cinerasccns, Fahr. Twenty examples in various stages. JEpachi'omia rodericensis, (Plato LIV., figs. 3, 3".) Ann. & Mag. Nat. Hist., s. 4, vol. xvii., p. 410 (187G). Male. Pale carneous ; pronotum and head abovo brown, adorned with an X-shaped marking, intersected by a longitudinal fascia, palo carneous edged with black ; pronotum with a lateral oblique black streak, meso- and metanota testaceous, with lateral black spots ; knees black at the sides ; tibiae with three black spots below, spines black ; tegmina with the basal half coriaceous, brown, inner border whitish, base Avhitisli-speckled, a white subquadrate costal spot ; apical hair pale brownish hyaline, crossed by ill-delined ii'regidar white bars ; veins dark brown ; wings hyaline white, veins black. Length of the body 12 to 18 miUims. (CoU. by Gulliver.) This species does not vary in coloration. Eleven specimens in various stages were taken. Mantis variegata, Olivier. Fifteen examples in various stages. Bacilhis incommodus. (Plate LIV., figs. 4, 4.«-'.) Ann. & Mag. Nat. Hist., s. 4, vol. xvii., p. 410 (1876). " Green, with red antenna, when alive." Ochreous, clouded with purplish brown ; eyes black ; head truncate-ovate, with central and lateral longitudinal depressed lines, longer than the pronotum ; antennai Avith twenty-fovir joints, basal joint broad, depressed; pronotum quadrate, with obtusely carinated margins ; tAVO longitudinal series of four obtuse well-developed tubercles ; a central depression ; mesonotum dorsally carinate, covered with tubercles ; a series of lateral tubercles between the paii"s of legs ; tegmina small ; AA"ings minute, coriaceous ; anterior abdominal segments roof -like, laterally carinated ; anal styles lanceolate ; anterior legs considerably the longest, the femora strongly excavated and rugulose at the base, with two internal longitudinal marginal series HEMIPTERA, — ARTHUR GARDINER BUTLER. 549 of denticles ; all the legs parallelopipcdous, obtusely carinated ; posterior femora denticulated below. Length 3 inches 10 lines. (Coll. by Gulliver and Slater.) The pronotum might perhaps bo more strictly described as irregularly lobulate, the lobes mammoidj capped with prominent tubercles, and arranged in two longitu- dinal series of foiu", between them a depressed mesial line. The general aspect of the species is that of the genus Lopaphus, but its structm-al characters agree in all respects with those of Bacillus. Eivc specimens Avere found in the island. It is " common on leaves of fan-palm, which it resembles in colour." — G. Gulliver. HEMIPTERA. Hitherto Previously obtained elsewhere. Name of Species found in Eodrigucz. foimd in Mauritius Madagas- Kodriguez and Other Localities. only. Bourbon. car. Phricodes hystrix, Gcrmcir X X Caifraria. jEthus sp.* (immature) - X ? Nezarii viridula, ^•a^• leii, Hope - X X S. Europe, S. Asia, Malaysia, Columbia. Serinetha lateralis, Siguoret X X Caffraria, Senegal, Sierra Leone, Calabar. LctliEeus tartareus, Stal - - - Cape. Rhyparochromus liictuosus ? Lucas Algeria. Dysdeicus fasciatus, Signoret X Sierra Leone, Zanzibar. Mezira caffia, Sm Cape. Conorhinus rubrofasciatus, De Gecr X X Sierra Leone, India, Malaysia, China, Brazil. Oucocepbalus cancellatu?, Stal • Caffraria. Reduvius laniger, Butler X Emesa barbara ? Lucas - - - Algeria. Velia iufernalis, Butler - X Sigara scutellaris, Stal - - - Country of the River Kuisip, Namaqua Land. Sigara sulcata, Signoret - X Sigara felix, Butler X Cicada sp. (larva) - - Tylana carinata, Fabricins X Coccus ccratiformis, Butler X Aptinothrips fasciatus, Butler - X Hemiptera Heteroptera. Phricodes hystrix, Germar. Four specimens were obtained l)y Mr. Gulliver at Eodriguez. ^thus sp. (immature). One specimcit. * Jlay be JE. iiiiiiliit/iisciiriiusis. 550 ZOOI.OCY OV KODKICUEZ. Nezara virklula, Linn., car. Ic'ti, Hope. If the various representatives oi" this species from all parts of the vt^orld (din'oriug in form, sculpturing, and colour) bo considered conspocilic, I'entaloma plicaticoUls of Lut-as may be added to the synonymy. Que example of N. Icll was taken by Mr. Gulliver " on ilowers." Serinetha lateralis, Signorct. Twenty-one examples, ia various stages of development, wer(> Jouud l)y Mr. Gulliver '* under stones." Lethccus (artarceiis, Stal. The seven examples obtained by Mr. Gulliver differ in no respect from the Cape form. Uhl/parochromns hictuosus ? Lucas. Two examples. Judging by the figure I cannot distinguisb this species from that described by M. Lucas. Di/sderciis fasciatns, Signorct. Twenty-five specimens, in various stages of development, found " on Malvaceous trees." — G. Gulliver. Mezira caffra, Stal. Fifteen examples. " Common under bark of trees." — G. Gulliver. Conorhimis rubrofascialus, De Goer. One specimen was obtained by Mr. Slater, Oncoceijhulns caiiceUutus, Stal. Ten examples were collected, Rediivitis laniger. (Plate LIV., figs. 6, 6^) Ann, & Mag. Xat. Hist., s. 4, vol. xvii., p. 411 (1876). Densely clothed with testaceous hair ; antennae slender, the three basal joints amber-yellow, remaining joints blackish; head castaneous ; eyes blackish ; rostrum pale castaueous, dark at the tip ; thorax testaceous, with two nearly parallel longi- tudinal piceous bands, anterior lobe with a central longitudinal fossa ; j)ostcrior lobe with a central depression ; scutellum piceous, pale at the tip ; corium of hemelytra testaceous, with two blackish spots and a cuneiform whitish spot between them ; membrane dull pui'plish black ; abdomen above amber-yellow, with marginal brown IIEMIPTERA.— ARTHUR GARDINER BUTLER. 551 spots, below piceous, with a yellowish l)aml on each side ; pectus shinincf l)lack ; legs amber-yellow ; the femora with ceutral aud tormiual browu bands. Length G^ millims. (Coll. by Gulliver.) One example brought home; it is said to be "common under cow-dung." — G. GuUiter. It. laniger seems to be of about the same size as B. nanus, of which Stal says : " ^Minima mihi cognita species generis." It cannot, however, be confounded with that insect, as, apart from its different coloration, its hairy character would alone be sufficient to distiuEruish it. '»' Emesa harhara ? Lucas. Two specimens, apparently immature. " Common in the hottest time of the year, February and March." — G. GitUiver. Telia wfernalls. (Plate LIV., fig. 5.) Ann. & :Mag. Nat. Hist., s. i, vol. xvii., p. 411 (1876). Velvety black above, two grey spots on the front of the thorax, slaty grey below ; sides of pectus Avhitish, anal segments of venter brown ; legs black above, brownish below ; the base of the antennse, the coxae, and the base of the femora orange- yellow. Length 1 millims. (Coll. by Gulliver.) Allied to V. nigricans of Burmeister, but distinguished by its dull black colour- ing, and the two grey thoracic spots above. Sixteen adult examples and 15 larvae were obtained. Two examples. One specimen only. Sigara scutellaris, Stal. Sigara sulcata, Signoret. Sigara felix. (Plate LIV., figs. 8, 8".) Ann. & Mag. Nat. Hist., s. 4, vol. xvii., p. 112 (1876). Ovate, piceous ; head rather wider than the thorax ; vertex of head and sides of tegmina whitish testaceous, below greyish brown ; legs fawn-colour ; face rugose, vertex with slender transverse irregular impressions ; thorax transversely striated ; tegmina punctm'ed, setose. Length 2-3 millims. (Coll. by Gulliver.) 552 ZOOLOGY OF RODRlGUliZ. Sirailav in oolorafion to ^S*. minnfissimo ; but smaller ami compavath'^cly liroadcr. " Common in streams." — O. Gulliver. IVcnty-lbiir specimens •were collected. Hemiptera Homoptera. Cicada sj}. (larva). Poiu' specimens. " Common." — G. Gulliver. Tjilana carlnata, Fahricius. One specimen. Coccus ceratiformis. (Plate LIV., fig. 9.) Ann. & Mag. Nat. Hist., s. 4, vol. xvii., p. 412 (187G). Female. Irregularly elliptical ; above witb posterior central depression ; thinly covered -with a fine powdery, silvery, waxy efEorescence, whicli conceals the coloration; wlien this is removed the insect is bright yellow, often with the central area largely piccons; about nine segmentations traceable; under surface somewhat concave ; head sinuate in front, Avith an angular frontal depression, in front of which is an obtuse I'idge ; rostrum situated in the fore part of a deep obovatc depression ; antennae 9-jointed, pale testaceous, sparsely clothed with setae ; legs pale testaceous, rather long (extending beyond the sides of the body in small examples), apparently with three tarsal joints and two terminal claws ; pediferous lobes very prominent. Greatest length 3 millims. (Coll. by Gulliver.) I camiot identify this species with any of those described by Messrs. leery and Signoret. Fourteen examples. Physopoda. A2)tinofhri2)s fasciatus. (Plate LIV., fig. 7.) Ann. & Mag. Kat. Hist., s. 4, vol. xvii., p. 412 (1876). Blackish piceous, glabrous ; -nings hyaline ; bases of antcnnal joints, eyes, ocelli, and five broad abdominal bands crystalline white ; frons fulvous ; antenncc 7-jointed, basal joint conical, the second to fourth obconical, fifth to seventh fusiform; the HEMIPTEKA. — ARTHUR GARDINER BUTLER. 553 terminal joint terminating in an acute point; head rounded, truncate in front and behind, with a central obtuse carina and an oblique stria behind each eye. Length 3 millims. (Coll. by Gulliver.) One example. Fig. 1. „ 1". „ 2. „ 2-. „ 3. „ 3". „ 4. „ 4". „ 4*. « ^ * „ 5. „ 6. „ C". „ 7. „ 8. „ 8". .. 9. ORTHOPTERA AND HEMIPTERA. Description' of Plate LIV. Phisis spinifera, Butl. t3 >J >J Nemobius luteolus, Butl. - Profile view enlarged. - Upper surface of head and thoracic shield. - Upper surface enlaiged. „ „ „ - Head and thoracic shield in profile, greatly magnified. Epachroiuia rodericensis, BiUl. - Upper surface enlarged. „ „ „ - Head and thoracic shield in profile, greatly magnified. - Upper surfiice. - Head in profile, greatly enlarged. - Terminal segments of abdomen, ventral surface, Bacillus incommodus Velia infernalis Reduvius laniger Aptinothrips fasciatus Sigara felix - J) » Coccus ceratiformis - enlarged. Terminal segments of abdomen, profile view. Upper surface, greatly magnified. j» » Profile of head, „ Upper surface, „ j» J? Front of head, „ Anterior portion of ventral surface. 4 A 551 ZOOLOGY OF RODRIGUEZ. Annelida. — By Frofessor Ed. Griibe. The Annelids obtained are few in nunibor and belong to tyro species : — Pericliaeta, Sc/dihL Ferlchaeta roclericeiisis. Longius vcrmiformis, anteriora versus citius magisque att(^nuata, teres, ex earneo flavesccns, la)te iricolor, paulo pellucida, scff mentis fere 110, anie 17"""" sitis majoribus, psene 2-plo latioribiis, posterioriljus 3-plo fere latioribus quam longis. Lobxs c(ipit(dis, quoad observare licuit, semiovalis, latior quani longus, linea paene recta a scgmento buccali scpositus. Scgmentiim huccale illo ut proximo longius. Segment inn 11"'""' et 15"'""' (clitclla)) sordide crocea, satis libera ut buccalo, nee vero tumida. Setae apicc quasi curvatro, sogmcntorum ante clitcllas sitorum paulo majores, 32-nae ad 3G-nas, annulis Icviter clatis collocatse, posteriorum minores, stepe tegre distinguendoe, ad CO-nas, postrcmorum 40-nae. Segmentum 17'"""' subtus pone cingulum setarum utrinque cminentia planiore orbicular! munitum, centro distinctiv;s circumscripto. Apertiu'as genitales anteriores baud observatae. Longitudo fere 100 mm., partis ante clitcllas sitaj 21 mm., clitellarum 2 mm., latitudo maxima (proximo ante clitcllas) 4, 5 mm., partis postremse 3 mm. Eatber common in damp earth and moss. — {Gulliver). Several examples were received, the longest of which measured 110 mm., and the shortest 45 mm. ; they were in all cases particularly soft, but, as in but a few cases the integument had pealed off to a very small extent, this did not appear to be due to the mode of their preservation, but to the thinness of the muscular layer. The wall of the body is so transparent that it was possible to see the viscera and lar^e vascular trunks quite distinctly ; thus, in the 9th and lOtb segments tbe stomach may be seen shining through the wall of the body ; in the 11th and 12th the two pairs of organs to wliich L6on Vaillant* has given the name of testes (1. c. Fig. 4 h h') and in the 16tb to the 19th inclusive, as also in the 20th segment, his " Glandes " accessou-cs du Canal deferent." Dissection revealed the presence in the 18th segment, of a minute C-sbaped Ccecum ; but it was in one example only that I noted tbe distinct presence of tbe above-mentioned circular protuberances with then.' sharply projecting centra (in which I was in no case able to find an orifice) . There are no longitudinal costulse on the most anterior segments ; the whole animal is romided, the hinder end not being either flattened out or * Leon Vaillant. — Note sur I'Anatomie de deux Especes du Genre Pericliaeta. — Montpellier, 1869. ANNELIDA. — ED. GRUBE. 555 dilated, but also -well rounded, aud gradually tapering off to the extremity of the body. It was impossible to detect tlic presence of any genital orifices. The cephalic lobe was very indistinct, as the pharynx was partially [exserted, but I believe I discerned an almost perfectly straight line of boimdary between it and the buccal segment. The animal under description would seem therefore to belong to the genus Mlo- cris, Kbg.,* inasmuch as the setae on the post-clitellar segments are distinct, and, as a rule, much more numerous than those on the anterior segments ; it is true that Kinberg in his diagnosis, says " cingulum nidlum " ; perhaps, because he regards the presence of a swelling as an essential character of the cingulum (clitellaj) ; such a swelling is absent also in our PericJiaeia, but I am, ncTcrtheiess, of opinion that the tAvo segments which are so markedly distinguished by their colour, and by the absence of setae must be regarded as forming a cingulum. As however tliere is some doubt as to the presence of this organ, I think it best to describe this worm from Rodriguez imder the more general name of Pet'ichaeta (as defined by Schmarda). Only one species of Nitocris is known, N. ffracilis, Kbg., from Rio de Janeu'O ; this has as many as 91 segments, and is from 42-66 mm. long ; there is a cephalic lobe which is said to be nearly as long as the buccal segment. In our species it is, I think, shorter ; Kinberg does not state the number of seta3 on the anterior or posterior segments, or any other distinctive characters ; but, in any case, it would be against all experience, if the two forms were identical. Ampliinome, Brug. A. {Eiirythoe) pacijica, Kb. When I was determining the examples of this species I was imdecided whether they should be referred to E. pacijica, or E. capensis, Kbg. ; as in both of these species, there are four buccal segments, and the caruncle does not extend to the fifth segment. In E. capensis, however, it is said to extend beyond the line which separates the thii*d from the fourth segment, and not to be lobate ; in our specimen, the caruncle does not extend so far, and its basal portion is lobed. I did not observe any distinctly finer setse in the ventral fascicle, where the number of bristles was about 25 ; the setse of the superior fascicles are either linear, having theu" points long and somewhat distinct, or they are much stronger and provided with about 11 denticles at their points, otherwise they are smooth ; all white, but those in the ventral bvmdles are of a more yellowish colour. The dorsal cirrus, which Kinberg states to be shorter than the superior seta?, I find to be of the same, or even of a somewhat greater length. The branchise are most intensely coloiu-ed, nearly ferrugineous, which I have not observed in any other examples of A. pacifica ; the filaments are grouped rather in the form of the fan, than in several rows, and, as a * Kinberg, Annulata nova. Ofvers. Kongl. Vetensk. Akad. Forhandl., 1866, p. 102. 4a 2 656 ZOOLOGY OF RODRIGUEZ, rule, arc divided into four branelies ; the fii'st gill had only four, but the second as many as 15 filaments ; the fourth and iifth reach those opposite to them, but this is not the case ^Yith the succeeding ones. The anus is dorsal in position and occupies the three last segments. In one example, in wliirli there are 73 segments, the length is 71 mm., the greatest breadth of the median segments, Avliieli are about 4-^ or 5 times as broad as long, is on the ventral siu'face 9 mm., or, adding the phai'etrae, 10, orthesetoe 13 mm. Kinbcrg's figiu'e* (1. c. Tal". XII., Fig. 11) represents the animal as much more slender, but in his description no ■weight is laid on the proportions ol' the segments, and it is possible that the figure A)as taken from a living specimen. The caruncle is 1" L mm. long; its basal portion has six descending grooves; the upper portion is swollen and undivided ; the azygos tentacle is 0'5 mm. long, and somewhat shorter and thinner than the others. * Eiijienies Ecsa. Aiinulata. 557 TuRBELLARiA. — By Gcorge Gulliver, B.A. (Plate LY.) Two forms of considerable interest belonging to tbis group were found to be tolerably abundant. Tliey belong to tbe genera Tetrastemma and Gcoplana, and I can scarcely entertain a doubt tbat tbcse animals will finally prove to liavc a mucb Avider geograpbical range, or, in other words, tbat it would be premature to regard them as peculiar to the island of Rodriguez. I may also express it as my belief, that the Laud Plauarians are much more numerous than has been supposed hitherto ; but, unfortunately, with but few exceptions, Zoologists who have visited this region, have paid but little attention to them. The species of Tetrastemma is, I believe, the third Land Nemertean which has been discovered, one having been found by Semper in the Pelew Islands, and another quite recently by the late Dr. Von Willemoes Suhm in the Bermudas, which he has described in " Annals and Magazine of Natural History," 1873, xiii., p. 309. He expresses it there as his opinion that land Nemerteans are much more common on tropical islands than has commonly been supposed, and the discovery of one at Rodriguez goes a long way towards justifying his supposition. Dr. Suhm's, like mine, is a species of Tetrastemma [T. agricola), but the present species is much more truly a terrestrial species than his, inasmuch as it occiu-s in woods — in damp rotten wood — and similar situations. This was the case also with that discovered by Semper. The animal was killed almost immediately by immersion in salt-water, for which it showed its distaste by rapidly exserting its proboscis, as it does when placed in alcohol. Fresh water, on the contrary, was by no means so distasteful to the animal, and T have had them remain alive for a whole night in it. It never, however, was observed to attempt to crawl on the upper surface of the water, though it would ascend the sides of the glass, and, if possible, settle itself outside the water. All the specimens seemed to be exceedingly torpid, and showed no disposition to move. If disturbed in their lodgment in rotten wood they appeared to shrink aAvay into some recess, but never attempted to move for any length of time. It would seem to be very probable that this season Avas one of torpidity AA'ith tliem. Tetrastemma rodericamim. — (Plate LV., fig. 2.) Specific character. — Body narrow, elongated : dorsal surface dark green with a narrow white streak along the median line, and one at each side near the ventral 558 ZOOLOGY OF RomnorEZ. surface. Ventral surface white. Head o\' a lighter green colour than the dorsal sxu-face of the body, and having foiir white spots on the snout, visible when the animal is looked at irom before, ^farginal stylet sacs four. Cephalic sacs absent. Jlohilat. — On rotten wood, under decayed leaves, &c. in the Island of llodrigucz. The animal is from one to three inches in length. The dorsal surface is convex, the ventral Hat. The colour of the dorsal surface is generally a very dark green, but occasionally (ho shade is a good deal lighter ilian usual. The lateral white lines are broader than the median one. Tlie ht^ad is bilid anteriorly. Of the four white spots one is situated laterally on each lobe of the snout, and of the remaining two, one corresponds to the apertm'c of the proboscis, and one to the mouth. Anaiomi/. — Tetrastemma Rodericauum corresponds very closely in the arrange- ment of its internal organs with the other members of the Enoplan division. It presents however certain peculiarities which arc worthy of mention. Cephalic Jissitres and sacs. — No trace of the cephalic iissures or of the sacs to which thev lead, could, after careful examination, be detected. This was also the case in the Bermudian Tetrastemma described by Dr. Suhm. Proboscis. — As in Au/pJii2)orm hastatus, there are four marginal stylet sacs situated in the same plane. The structm-e of the proboscis corresponds very closely with that of other members of the group. There is lioAvever one respect in which it appears to be peculiar, namely in the possession of an additional system of circular fibres which traverses the great longitudinal columns which constitute the so-called " beaded laver " (Plate LV., fig. 8). As in the other Enopla, each of the great lon- gitudinal bands is svirrounded by fibres which are perhaps derived in part from the external, and in part from the internal, circular coats, but the system of fibres just mentioned is additional to these, though mingled with them. It cuts through the lono-itudinal fasciculi, dividing each fasiculus into an inner and outer half, constitut- ing a distinct cu-cular coat (Plate LV., fig. 9, F). These fibres are wavy, as are also the others just spoken of as surrounding the longitudinal fasciculi. None of the fibres are distinctly traceable among the bundles of longitudinal muscular fibres. ■Whilst the internal circular and external circular coats, and the external lon- (^itudinal and internal longitudinal muscular coats, are readily stained by carmine, the lon"-itudinal elastic bundles, with the fibres which surround, and those which traverse them, do not take the staining fluid at all. Nervous system. — The general arrangement of the nervous system presents no peculiarities. The distinction between the fibrous and cellular elements in the lateral nerve trunks is, however, remarkable from its clearness and constancy of the arrano-ement of the two elements. Well marked groups of nerve-cells ' are present also in the ganglia, and are continuous with the cellular elements of the TURBELLARIA.— GEORGE GULLIVER. 559 lateral cords. 0\^^n£? to the difficulty of obtaining symmetrical sections of the ganglia, it is very hard to gain anything like a notion of the arrangement of the cells in those bodies. In size and form the ganglionic cells apijcar to be identical with those of cords, the general ai-rangement of which can readily be studied in transverse sections. Tn the present sjiccies the cells have a definite arrangement which does not appear to have been the case in any of the species examined by previous writers. The arrangement is in all the more easy to study, as the cells arc deeply stained by carmine or logwood, whilst the fibres are comparatively unstained. If a transverse section be made through a lateral nerve trunk at any part of the body, appearances similar to those figiu-ed in Plate LV. figs. 4, 5, 6, will be seen. The trunk itself is somewhat kidney shaped in section, the convexity being towards the integumental, and the concavity towards the visceral, aspect of the body. It is bound down to the body-wall by well marked fibres. There are seen to be two cell ]-egions, and two fibrillar regions. The ujipermost or dorsal cell region (Fig. 6 A.) crosses the cord, obliquely slanting inwards and downwards. Above it, is the upper fibrillar region C, and below it the lower fibrillar region, D. The upper cellular region is, allowing for slight variations, constantly expanded at its external, and to a much, less degree at its internal, extremity. The lower cellular region B. occupies the venti'al part of the cord, and is in contact below with the sheath. It has a more or less crcscentic form, ten dins to have its outer horn prolonged to meet the external and lower extremity of the upper cellular region. In some sections the two regions may be seen to meet, but always by a very narrow belt of cells, whilst in most cases the two regions are, as in Fig. 6, separated by the lower fibrillar region. The arrangement here mentioned and figured is so constant that there is never any difficulty in distinguishing the dorsal from the venti-al, or the inner from the outer aspect of the cord, in the abscence of any other guide but the arrangement of its cells and fibres. The cells themselves are of small size, granular in structm-e, and never present an apj)earancc of a dis- tinct nucleus and nucleolus as do those fi-om the ganglia of Leptoplanu figiu'ed by Moseley in the " Phil. Trans.," 1874 Plate XV., nor can any sign of j)rocessess arising from them be detected, though in some cases the matrix in which they lie suggests the presence of them from having a fibrillar appearance. Fine bands of fibres, Fig. G E., may be constantly seen running into the masses of neiTe fibres. They appear to be in a large measure fibrous septa running into these masses, but may in part be fibres derived from the cells, though no distinct evidence of this has been obtained. "With regard to the structiu'e of the fibrillar masses themselves, as migiit be expected, no tiling definite can be made out. In transverse section they appear granular, having the fibres just mentioned running into them at parts. 500 . ZOOLOGY OF l^OniJKil'EZ. ]£yes. — The eye consists of an aiitcrior clrar, and a posterior ijio^montcd, portion. In form it rosonibles that figured by Loydig as occurring in TIanaria lactca (" Vom Ban dcs Thierisclien Korpers," j)!. 1.). Tn spooimons ])roscrvod in alcohol, the oyc piijnu'nt lias a tawny bn)\vn colour, ll was suggested to uic by Dr. ilollcstou that this might be due to tlic solvent action of the alcohol, and, accordingly, having examined some specimens which had been preserved in cliromic acid, and found the eye pigment of these to be of a deep black colour, it was at once obvious that such was the case. ^Foseley, /. c. \^. 121, appears to doubt the solubility of the pigment of rianarians in alcohol, and mentions that it is especially not the case with the eye pigment of LeptopJana. Nor is it the case in the planarian, Geoplana whartoni, about to be descril)ed, nor with regard to the body pigment of the present species of Tetrastemma. It is liow(»ver most imdoubtedly the case that the eye pigment of the latter is soluble in alcohol. Cephalic Glandular Mass. — Commencing at a very short distance from the snont of the animal, persisting throughout the whole oesophageal region, and, if one may so express it, occupying the space, Avliich after the termination of the ocsojihagus is filled up by the voluminous digestive tract is a mass, which is here described as "•laudular, without, it must be confessed, much knowledge as to its true nature. It is figured as it appears under a low power in PI. LV. fig. 3, "P. Under a high power this mass consists of spaces irregular in shape and size, and enclosed by a tissue which is in part homogeneous and in part fibrous. In the spaces thus enclosed there is amorphous granular material, but no distinct appearance of cells. Connective Tissue.— Dv. Hubreclit has, loc. cit., insisted Avith great justice on the importance of the " fibro-muscular connective tissue," which serves to support the various internal organs by bauds of fibres which pass from the body-wall to these organs. In the present species this system is well developed, and, as in other species, is most marked where spaces intervene between the internal organs and the bodv-wall. Thus fibres constantly bind down the lateral nerve-trunks to the body wall (Fig. 6, P.), and in like manner pass from the body-wall to the dorsal surfaces of the proboscis sheath and digestive tract. But the connective tissue in this species presents a featiu-e which has not hitherto been described, namely, the presence of cells in addition to fibres. These cells "are represented in figvu-es 4 and 5 P., and, highly magnified, m Fig. 6 G. and in Fig. 7. These cells are large and of fairly uniform size, though, scattered amongst them, are, as may be seen in Fig. 7, a few of much less magnitude. At parts these cells present the appearance of being con- nected with fibres, as is the case with the cell B, in Fig. 7. They present a well- marked nucleus and nucleolus, and might at first in female specimens be mistaken for voung ova, as they are especially numerous in the vicinity of the lateral nerve trunks. They may with care be traced all round the body internal to the inner muscular coat, being recognizable in a single layer, and havmg a flattened appearance, TURRELLARTA.— GKORGE GULLIVER. 561 in i^hicos whore the voluminous (lig-(v<;tive tract is in close apposition with the hody wall, whereby it makes the recognition of sejiarate fibres out of the question. Vasmlar System.— T\\g opacity of this species makes the study of this system very (lifTicult, as no general view can be obtained. In Pig. 3. D. the two lateral vascular trunks are seen in section. Generative Si/ateni. — The arrangement of the generative organs presents nothing peculiar. The animal is dioecious. Fig. 4 represents the general arrangement of the testicular masses, and fig. 5 shows an ovum lying in its sac. In all cases the male organs seem to bear a much greater proportion to the size of the body than do the female. In some sections I have observed the appearance of ducts passing from the testis through the bodj'-wall, but not with sufficient constancy or distinctness to justify any lengthened description. Digestive Tracts. — Tlie digestive tract has tlie usual arrangement. Tlie mouth and aesophagus present well marked cilia. The contents of the post-;csophageal portion of the digestive tract has a uniform granular appearance, many of the granules being like oil globules. In none of the sijecimens examined has anything definite been observed that would lead to a knowledge of Avhatthe animal feeds upon, and, in fact, at the time when the specimens were collected, it is very probable that the animals were torpid and not feeding at all. Beside the Nemertine, I was fortunate enough to obtain specimens of a small species of Land Plnnarian, which occurs in situations similar to those in which the Nemertine lives, and, indeed, is often found together with it. Wishing to ascertain where this animal should be placed, and referring to the arrangement of Mr. Stimpson (Proc. Acad. Phil., 1857), I found that he states, under the characters of his sub-family Geoplatiidcp, that the eyes are two or many situated in the anterior part of the body (" Oculi duo vel plurimi in capitc dispositi.") This part of the diagnosis Mr. Moseley (loc. cit.) has very properly altered in consequence of his discovery of eyes in various parts of the body in Bij)alium. The genus Geoplana, the first \inder !Mr. Stimpson's sub-family, has amongst its other characters this " ocelli nuraerosi marginales, v. in acervos submarginales, in capite dispositi." This genus is composed of all the species described by Mr. Danvin (Ann. Nat. Hist., Series i. vol. xiv.), except those which have no eyes, and of one species found and described by Mr. Stimpson himself. But there appears nothing in Mr. Darwin's description to justify Mr. Stimpson using the words " in capite dispositi." For instance, in describing Flnnaria raginuloides {Geoplana vaginnloides, Stimpson), Mr. Darwin says, " Ocelli numerous, placed at regular intervals round the anterior extremity of the body and irregularly round the foot." Again, in the description of Tl. tamianica, " Ocelli scattered round the entire margin of the foot, but most frequent at the anterior extremity," Mr. Stimpson's words " in capite dispositi '' must mean, " in capitc solum dispositi ;" at least Mr. Moseley has taken them to 1. 1? 562 zooLonY OF Tiom^TonEZ. mean this, and it would lio difficult to put any other construction on them. This explanation is necessary, as tlu> planarian Avhich exists in Rodriguez has, like P/. fasniaiiicd, eyes situated roiuid the entire margin of the foot, hut more fr(Mpu>nt at the anterior extremity. In other respects this agrees in its characters with the genus Geoplana ; and it seems to he necessary then to alter the characters of this genus, as Mr. Moseley has done those of the genus Bipalium, in order that the former may include Mr. Darwin's species and the present one. The latter I propose to name G. whartoni, after my friend Commander "Wharton, E.N., lately in command of n.:Nr.S. " shearwater." Geoplaxa (altered from Stimpson). Corpus depressiim, vel depressiusculum, clongatum v. lincare, capite continuo. Ocelli numerosi, marginales, v. submargiuales ; v. in parte anteriori corporis solum, vel passim circa corpus, singulatim plerumque, nonnunquam in acervos dispositi. G. tvharioni, — PI. LV., fig. 1. Body elongate, pointed at both ends. Eyes numerous, placed singly all round the margin of the foot, hut most numerous at the anterior extremity. Body cream coloured, and marked with three dark longitudinal stripes, one median and two lateral, which latter converge towards the extremities and meet the median one. IMouth placed behind the centre of the body. Generative orifice nearer to the posterior end of the body than to the mouth. Body one inch to an inch and a half long when extended, much less when at rest. Habitat. — In the island of Uodriguez on rotten wood. Mr. Darwin mentions {loc. cit.) that he found a Land Planarian at Mauritius. In the island of Mahe in the Seychelles group, I observed an animal of similar size and colour, which unfortunately became dry before I had time to examine it. Description of Plate LV. Fig. 1. — Geoplana whartoni, sp. nov., about twice the natm-al size, from a specimen preserved in alcohol. Fig. 2. — Testrastemma rodericanv/m, sp. nov., about three times the natural size, from a sjiecimen pi'eserved in alcohol. Fig. 3. — Transverse section through the anterior part of the body of a specimen of T. Roderi- canum. Hartnack's Obj. 2. DraAvn with the camera lucida, as were also the succeeding figures. A. Cuticular and muscular layers of the body-waU. B. (Esophagus. C. Proboscidean sheath. D. Vessels. E. Lateral nerve trunk. F.F. Cephalic glandular mass. Fig. 4. — Transverse section through the middle part of the body of a male individual of the same species. Hartnack Obj. i. TURBELLARIA. — GEORGE GULLIVER. 563 A. Cuticle. B. Circular muscular coat. C. Longitudinal mu.scular coat. E. Testicular masses ha\'ing well defined walls, enclosing cells and spermatozoa. F Connective tissue, consist- ing of cells and fibres. G. Portion of intestinal tract with granular contents. Fig. 5. — Similar section through a portion of the body of a female. A, B, C, D, F, G, as in the preceding figure. E. an ovum contained in its sac, and possessing a well mai-ked nucleus and nucleolus. Fig. G. — Transverse section of latei-al nerve trunk, highly magnified. Hartnack's Obj. 8. A. Dorsal cellular region. B. Ventral cellular region. C. Doi-sal fibrillar region. D. Ventral fibrillar region. E.E. Fibres running into the fibrillar masses, probablj- consisting mainly of connective tissue, but pei-haps in part of processes from the cell regions. F. Fibrous bands binding do^vn the cord. G. Connective tissue cells. H. Internal muscular coat. Fig. 7. — Portion of a transverse section highly magnified, to show the connective tissue. A. Int. muscidar coat shown in outline. The fibres represented passed between it and a por- tion of the digestive tract. D. Fibres of connective tissue. B. A large connective tissue ceU, with fibre attached, other cells mostly large, but some small are also seen. Fig. 8. — Section through a portion of the extended pi'oboscis to show the an-angement of the " beaded layer." Hartnack. Obj. 8. A. Internal circular muscular coat. B. External cii'cular muscular coat. C. Internal lonsl. muscular coat. D. External longitudinal muscular coat. E. One of the great longitudinal bands surrounded by fine wavj' fibres, and also divided by similar fibres which pass through it ; all these fibres go to constitute the cii'cular elastic coat. F. The cii'cular elastic coat. •1 B 2 564 ZOOLOGY OF KODKIGL'EZ. ECHINODEKMATA. — By Edffar A. Siiiilli, F.Z.S. (J'kitc Li., ligs. 1-3.) The collection of Sea-urchins and Starfishes from the ishuul ol' llodiiguez, although cuniprisiug ])ut few species, contains some of considerable interest, among- which may be mentioned a very remarkable new Coniatitla, two apparently nnde- scribed species of Oph'iocunia, and tlic liiu- OpIiiouKt.sti.v cciioaa, described by Dr. Peters from Mozambii|iu\ The rest of the forms are well-known inhabitants of tlie Indian Ocean, certain of them boins,' also eomnion in the Paeiiie. Crinoida. Comalula iiidica. (Plate LI., figs. 3-3b.) Ann. and Mag. Nat. Ilist., 187G, xvii., p. 106. Ravs 30 ; dorsal disk, small, convex, subpcntagonal, sculptured with little contiguous and well-delined pits which are the sockets of the cirri; these are about 15 in number ; radial joints two, the axillary without a syzygium ; between tliis and the next bifurcation are two articulations; after two more joints the two outermost arms bifurcate ; none of the brachial axiUaries Avith a syzygium ; each third segment n1)0ve an axillary has a syzygium, and on the arms the next joints with syzygia are at very unequal distances, the most proximate being separated by as many as 20 joints, but usually by about 14 ; then nearer the extremity of the rays, they are rather closer together, the interlying joints varying in number from G to 10. The second pinnulai are very long and composed of much elongated joints. The colour of the specimen is pale purplish brown, with the sutures of the ray-articula- tions blackish brown. Diameter of disk 6 mill., leugtli of rays about 6 inches (153 millim.). The only example of this interesting sj^ecies Avas collected by Mr. Slater. Unfortunately it is imperfect, all the dorsal cirri being wanting.* The remark- able elongate second pinnula^ (17 mill, in length) are situated on the first and second segments above those joints which always have a syzygium, and which are the thml joints above the last brachial axillaries, or in other words, they rest uj)on the fourth and fifth joints above these axilhiries. They consist of about 17 joints, * The figured dorsal cirrus was found among debris at the bottom of the bottle containing the Comatula, but it is not absolutely certain that it belongs to the present species, although the evidence points to that concUision. ECHINODERMATA. — E. A. SMITH. 5G5 W'hicli are elongute, cylindrical, those at the middle bciu- the longest, and the two or three basal ones the shortest. la the short diagnosis given in the " Annals and Magazine of Natural History," May 1870, p. 400, ])y an oversight it is stated that "the two hiHerjnu.^t arms hifurcate; '' it should be outermost or exterior. Ophiuroida. Ophiomastix venosa. Peters, Wiegm. Arch. Natm-gesch., 1852, p. 83 ; Von Martens, in Von dcr Deeken's Reisen in Ost-Afrika, p. 130 ; Lyman, Opliiui-idse, p. 92. Three fine examples of this remarkable species were collected by Mr. Gulliver. It has also been previously found at Mozambique and Zanzibar. Ophiocoma erinaceus. MiiUer and Troschel, System der Asteriden, p. 98 ; Dujardin and Hupe, Echi- nodermes, p. 200 ; Liltken, Additamenta ad Hist. Ophiurid, p. 101 ; Von Martens, in Von der Deeken's Eeisen in Ost-Aii-ika, III., p. 129; Lyman, Ophiuridai, p. 83. Seven specimens were obtained by Mr. Gulliver. This species has a wide geographical range, having been met with in the Red Sea, at Mozambique, Zanzibar, Xatal, Seychelles, Mauritius, Sandwich Islands, Kingsmills and Society Islands. Ophiocoma variegata. (Plate LI., figs. 2-2d.) Ann. & Mag. Nat. Hist., 1870, vol. xviii., p. 39. Disk circular, finely granulated above and beneath ; oral shields rather longer than broad, slightly octangular, with the sides rectilinear ; adoral shields narrow, extending along the lateral margins of the orals ; three mouth-])apilhe on each side of each oral angle, the outermost being the largest and somewhat squarish; above the teeth at the apex of the angle is an irregular cluster of about 12 smaller papilltu ; teeth foui", strong, arcuate at the tips. Arms rather stout, about four times as long as the diameter of the disk ; ujiper plates transversely narrowly oval, a\ ith the outer margin faintly angulated in the centre, about twice as broad as long ; lower plates siquarish, the aboral angles rounded, the side margins excavated ; arm-spines four, subequal, the uppermost a trifie the shortest and stout and the lowest but one a little the longest, about as long as the width of the dorsal arm-plates (the tenth) ; ambu- lacral scales two in number to the extremity of the arms, short and compressed. Colour of a specimen in alcohol : — disk above and beneath uniformly pm-plish brown ; upper arm-plates dirty brown, at intervals varied Avith tliree or fom' con- tiguous pale ones, which arc marked with the dark outline of a subquadrate figure, particularly observable towards the outer ends of the arms ; lower arm-plates pale, 566 ZOdl.OCiY OF RODKKUIEZ, "blotched with dirty brown, especially towards the outer extremity of the rays ; arm- spines of a unit'orm colom-, hut lighter than the upper avni-i)lates ; oral shields pale, mottled with dirty brown. Diameter of disk 38 mill., length of rays 105; width oL" hroadcst vipper arm- plates i. The nearest ally of this species is 0. scolopendrina, Lamarck, from which it differs in coloration, form of the arm-shields, oral shields and the hrachial spines. The dark suhquadrate outline on the pale upper arm-plates is very characteristic, as also is the dirty brownish mottling on the oral and lower hrachial shi(dds. Ophiocoma hrevispinosa. (Plate LI., figs. 1-lf.) Ann. Mag. Nat. Hist., 1876, vol. xviii., p. 40. Disk suh-circular, flat ahove, minutely granulated above and beneath; oral shields heart-shaped, broader than long ; adoral plates small, crcscentiform, bordering the sides of the orals ; three or four mouth-papillte on each side of an oral angle, and a group of about twelve at the apex ; teeth foiu', the two intermediate ones larger than the two exterior. Arms a trifle more than three times as long as the diameter of the disk; upper plates transversely oval, about twice as broad as long; lower plates (twelfth from the base) a little longer than broad ; aboral margins arched and a little pointed in the middle, lateral edges rather deeply concave ; oral margins somewhat convergently slopnig and interrupted by the outer margin of the adjacent precedino- plate ; tentacle-scales two, short, and comj)resscd ; brachial spines short, four (sometimes five on a few plates Just beyond the disk), the two upper ones shorter than the others, broad and flattened ; the two inferior ones (of which the second or upjier is rather the larger) are slightly conical, and not so long as the width of the broadest dorsal arm-shield. Colom* of a specimen in alcohol : — disk du-ty white, irregularly mottled with green above and beneath ; arms of the same coloiu- as the disk, with a narrow green line above, more or less distinct, down the centre; lower shields, ambulacral scales (and two lower series of arm-spines for the most part) uniformly dirty white, and the two upper series of spines ornamented with one or two narrow greenish rings and dots ; oral shields spotted with green. Diameter of disk 17 millim. ; length of rays 54. This species is well distinguished by its peculiarity of colouring and the character of the short brachial spines. Asteroida. Scytaster variolatus, Retzius. A sin""le fine large specimen collected by Mr. Slater. Other localities where this species has been met with are the Ked Sea, Reunion, Mauritius, Zanzibar, Bourbon. ECHINODERMATA.— ]■ . A. SMITH, 507 Ziiickia mulfifora, Lamarck. All four examples collected by Mr. Gulliver have five rays iu diilereut stages of development. This common starfish also inhabits the Red Sea, the shores of Mozambique, Eeimion, Eoiu-bon, Seychelles, Mauritius, Xcw Caledonia, and the Sand^vich Islands. Echinoida. Phyllacantlms imperialis, Lamarck. Of this species a fine large example with the ends of the primary spines well fluted, was collected by Mr. Gulliver. This species is also recorded from the Eed Sea, Zanzibar, Aru and Molucca Islands, Sulu Sea, New Holland and Tonga. Hipponoe variegata, Leske. The single specimen of this polymorphous and widely-distributed species is of a depressed form, roundly pentagonal, of a brownish colour, and clothed with wMte spines. It was collected by Mr. Gulliver. This species appears to be found in all parts of the Indian Ocean, at Zanzibar, Mam'itius, Seychelles, Nicobar Islands, in the Red Sea, the Pacific, and at the Philippines, Moluccas, Piji Islands, &c. Seto'ocentrohis mammillatus, Klein. There are two varieties of this most variable sea-urchin brought home from Rodriguez, one with pale brown spines with dark tips and one or more pale rings, and the other with the primaries of a rich chocolate their entii-e length, and also having pale annulations near the extremities. Mr. Gulliver reports that this species is " very abimdant." The geographical range of this species is like that of the preceding. Ecliinometra lucunter, Leske. Of this extremely variable species two varieties were met with at Rodriguez ; one with almost uniformly violet spines, but palish towards the base, and the other with light greenish-olive spines. The geographical range of this species extends over the Red Sea, Indian Ocean, Chinese and Japanese Seas, and several parts of the Pacific. 568 ZOOLOGY OF T^ODRTOT'KZ. Ech'nioiioi^ ci/closfoiinif), Loske. A siiii^lo example" of tlii-^ spocios, about an indi in Icn^'lli, was dolUH'tcd by Mr. Slater. Other localities for the species are Mozainbi([ue, Zanzibar, Ri^union, Bourbon, Australia, Amboina, Flores. Manila, Hood's Island, and lb(> Ivingsmills Islands. Explanation of Pi.atk LI. (Figs. 1 to 8). Fi^'. 1. Ophicoma hrevisj^inosa, enlarged view n{' upper side. la. „ „ „ „ of lower side. M). „ „ „ „ of an oral sliield. Ic. jj „ „ „ of the madreporic shield. IJ. „ „ „ „ of two upper arm-plates. le. „ j, „ „ of a lower arm-jilate. !/■' J, „ „ „ of an arm-segment (seen from the exterior side and showing the relative length of the spines). •2. Ophicoma variegata view of upper side, natural size. •2a. ,, „ „ lower side •2b. „ „ enlarged view of an oral angle. 2c. „ „ „ of a lower arm-plate and ambulacral papilUu. 2(1. . „ „ of an arm-segment (seen from the exterior side and showing the relative size of the spines). 3. Comatula indica, enlarged view of the disk and a portion of the rays showing the manner of furcation. 3f^. .^ „ of a dorsal cirrus (probably belonging to this species). 3/j. jj „ lateral view of an outer arm, showing the position and form of the pinnnhe. 569 Corals. — By Dr. F. Bruggemonn. The specimens of Corals procured at Rodriguez amount to 102 and belong to 49 species. They were taken by Messrs. Slater and Gulliver evidently at moderate depths ; at least no deep-sea forms, especially no TitrbinoUidcB or Oculhiidce, are represented amongst them. As might have been anticipated, the Rodriguez coral fauna bears a close affinity to that of Mauritius, Madagascar, and the Seychelles Islands, and there can scarcely be any doubt that the species described as new in the present paper -will prove to occur likewise in the localities mentioned as -well as in others of the Indo-Paciflc Region. Many of the species are common to Rodriguez and the Red Sea ; a less considerable number have been recorded fi-om the north-eastern part of the Indian Ocean ; some of them extend their range into the Pacific. However, the coral fauna of these regions is hitherto comparatively but little known, and the present collection most likely contains only a small mino- rity of the Rodriguez species ; so that it would be premature to offer any further remarks on their geographical distribution. ALCYONARIA. Alcyoniidse. Aleyonium sphaerophorum. Lobularia sphaerophora, Ehrenberg, Abhandl. Akad. Wiss. Berl. 1832, I., p. 281 (1831). Aleyonium sphaeropliora, M. Edw. Hist. Nat. Corall, I., p. 119. Sarcophytum latum. Aleyonium latum, Dana, U.S. Expl. Exp. Zooph. p. 623, PI. 58, fig. 6 (1846) ; M. Edw. Hist. Nat. CoraU. I., p. 121. Milleporidae. Millepora verrucosa. Millepora verrucosa, M. Edw. Hist. Nat. Corall. III., p. 227, PI. P. 2, fig. 1 a, b, c (1860). Millepora for ska li. Millepora forskaU, M. Edw. Hist. Nat. Corall. III., p. 228. Several specimens, belonging to different varieties, in one of them the terminal branches being much prolonged, stout and nearly cylindrical. This species I believe to be hardly separable from the West Indian M. alcicornis\ * This communication oa the Corals of Roiliigucz was one of the last works of the author, who died before the manuscript passed through the press. Shortly before his death he expressed a wish to make some alterations, but without specifying them. Therefore the editors have considered it best to print the manuscript as it was left by the author. 4 C <<^ 570 ZOOLOGY" OF KODRIGUEZ. MiUepora goHagra. MiUcpora gouagm, M. Edw. Hist. Nat. Corall. III., p. 230, PI. F. 3, fig. 1 a, b (ISCO). ZOANTHARIA. Actiuiidae, Pah/thoa argtis, Pali/fhoa argils, Elircnb. Abli. Alcad. AYiss. Berl. filr 1833 I., p. 373 (1831); M. Edw. Hist. Nat. Corall. I., p. 305. Palijthoa violacea. Corallum thiu, cxplanatc, spreading, incrusting cocnenchyma rather rigid, coriaceous, its siu-face covered with numerous grains which are distant by once to twice their diameter. Polyps scattered, small, wholly immersed when contracted, theu" disk surrounded by six to nine short prominences, a little larger than the small verrucce on the cocnenchyma. Iloight of corallum l of a mm. Diameter of polyps, about 2 mm. Colour (in spu-its), violet ; the prominences of the cocnenchyma somewhat lighter, whitish. There is only one small specimen of this apparently new sj)ecics in the collection, spreading on the surface of a fragment of Madrepora monticulosa. It is not well preserved, and, therefore, I am not able to give any details respecting the structure of the polyps. Stylophoridae. Stylophora palmata. Skleropora palmata, BlainviUe, Diet, des Scienc. Nat. LX., p. 360 (1830). Stylophora palmata, M. Edw. & II. Hist. Nat. Corall. II., p. 137. AstraeidsB. Galaxea laevis. Surface of corallum slightly convex. Corallitcs crowded, parallel, subcylindrical, compressed, only slightly angular, united by a rather dense peritheca, the cells of which measure from \ to 1 mm. Walls comparatively very thick, opake, smooth, with excessively slight indications of costse for every cycle. Calicular fossse oblong, sub-ano-ular, deep, Three cycles complete, a fourth one represented only by a few septa in some of the calicles. Septa smooth, unequal, the primary ones very thick, becoming suddenly thin towards the centre, much projecting, regularly truncate at the top ; secondary septa a good deal smaller, less exsert ; those of the third cycle very thin, narrow and hardly extending beyond the border of the calicle. Greatest diameter of calicular fossge 8 mm. ; distance between the coralKtes 2 to 5 mm. ; height of corallitcs about 15 mm. This species comes near to G. botigahivillci ; it is readily distinguished from aU its congeners by its thick smooth walls and septa. CORALS. — F. BRUGGEMANN. 571 Galaxea fascicular Is. Madrepora fascicular i^, Linn. Syst. Nat. Ed. XII., p. 1278 (1767). Galaxea fascicularis, M. Edw. & II. Hist. Nat. Corall. II., p. 227. Galaxea laperouseana. Sarcinnla laperousiana, M. Edw. & H. Ann. Scienc. Nat. 3" s6v. Zool. X., p. 311, PI. 6, fig. 5 (1818). Galaxea laperouseana, M. Edw. & 11. Hist. Nat. Corall., II., p. 231. Mussa iimbellata. Corallites vcntricoso-tui-binate, arising from a common stem, united tQl near tlic summit, then suddenly isolated and much diverging, forming a sort of umbel ; few of them remain associated to form short series consisting of two or three corallites. A rudimentary cpitheca ascends near to the borders of the calicles. Costae distinct from the base of corallum upwards, projecting but little, smooth for the greater part of their length, then tuberculate, and near the top spinose. Eossae rather shallow, triangular or quadrangixlar, rounded. Columella well developed, very spongiose. Eom- cycles complete, a fifth one indicated only in some of the systems. Primary and secondary septa rather thick, inflated, not much projecting ; their margins pro- vided with fom- to six coarse teeth of moderate size, most of which are truncate at the top. Inferior septa much thinner, tapering towards the centre, with delicately dentate margins. This species is distinguished by its mode of growth as well as by the dentition of the septa. The latter are less projecting than in most of the other members of the genus. Diameter of calicular fossae 20 to 30 mm., theu* depth about 7 mm. Leptoria tenuis. Maeandrina tenuis, Dana, tJ.S. Exj)!. Exp. Zooph. p. 262, PI. 11, fig. 7 (1846). Leptoria tenuis, M. Edw. & H. Hist. Nat. Corall. II., p. 407. Flatygyra esperi. Astroria esperi, M. Edw. & H. Ann. Scienc. Nat, 3" ser. Zool. XI., p. 298 (1849). Coeloria esperi, M. Edw. & H. Hist. Nat. Corall. II., p. 47. The two generic names, Coeloria and Astroria being given by Milne Edwards & Haime to the present genus in 1848 (Compt. Rend. Acad. Sc. XXVII., p. 493) ought to give way to the earlier appellation JPlatijcjijra of Ehrcnberg (Abhandl. Bcrl. ATvad. Wiss. fllr 1832,1., p. 323,1834; type: Madrepora lahyrinthica oi Ellis and Solander). Tliis is a very variable species ; there are specimens with short and long, mcean- 4c 2 572 ZOOLOGY OF KODKIGUEZ. di'ous and straight vallovs; iu oue nearly all the calieuhirfosscc arc as long as broad, plenty of calielcs behig tluis cu'cnniscrilKHl. Walls either thin or thick, even indated. Septa in most instances very narrow, hnt in one specimen (the one alluded toa])ove) much projecting towards the centre; their edg(>s faintly or coarsely dentate. Columella rudimentary or well developed, consisting of confluent frabecles or of vertical lamella^ the latter being placed in the cUrcctiou of the valleys, and sometimes entirely confluent, so as to render such specimens hardly distinguishabh; from the genus Leptoria. It is true that most of these dillereiu'cs are of what is called " specitie " or even " generic" value; but I found it impossible to draw any precise lines of distinction. In the present case the beautiful series procured by the Expedition comprises every stage of transition, and what is most significant, there are frequently dilVercnt forms represented in dlirerent places of the very same corallun\. I have made similar observations on the true Coelorice ; in one specimen of this genus taken in the Red Sea by Professor Hacckcl, there arc really two " species " on the same specimen, one occupying the centre of the surface, the other two opposite sides, and both jiassing into each other. Sydnophora microcona. Monticularia mio'oconos, Lamarck, Hist. Anim. sans Vert. II., p. 251 (1816). Hydnophora microcona, M. Edw. & II. Ilist. Nat. Corall. II., p. 423. Astrcea denticulata. Madrepora detiticulata, Ell. & Sol. llist. Zooph. p. 100, PI. XLIX., fig. 2 (1786). Favia denticulata, M. Edw. & II. Hist. Nat. Corall. II., p. 428. Astrcea affinis. Parastrea affinis, M. Edw. & H. Ann. Scienc. Nat. 3^ Scr. Zool. XII., p. 167 (1850). Favia affinis, M. Edw. & H. Hist. Nat. Corall. 11. , p. 429. Astrcea lobata. Parastrea lobata, M. Edw. & H. Ann. Scienc. Nat. 3^ Ser. Zool. XII., p. 171 (1850). Favia lobata, M. Edw. & H. Hist. Nat. Corall. II., p. 434, pi. D. 8, fig. 3. P lesiasti'cea qiiatrefagesa n a . Plesiastrcea quatrefagesana, M. Edw. & H. Ann. Scienc. Nat. 3" Ser. Zool. XII., p. 119 (1850) ; Hist. Nat. Corall. II., p. 491. The single Eodriguez specimen may prove to belong to a new species ; the CORALS.— F. 15RUGGEMANN. 573 caliclcs are remarkably smaller than in the true J'/, qualrcjagescfiid, and their borders less well defined. Aca)ith(tst)'(Ca aiigidosa. Corallum low, nearly cxplanate, with an even slightly undulate sui-facc. Calicles unequal, irregularly tetragonal or pentagonal, with simple rather thin borders and shallow fossae. Columella feebly developed, more or less papillose, parietal. Three cycles complete, a foiu-th one irregular ; well developed in several systems of the larger calicles, rudimentary or entirely wanting in others. Septa unequal, thin, slightly granulate, narrow, hardly projecting ; their edges provided with long thin si^iniform teeth at the summits, and Avith shorter and feebler ones towards the centre. Greatest diameter of calicular fossse, 10 to 14 mm. ; their depth 4 to 6 mm. Prionastrcea rodericana. Corallum convex, somewhat lobate. Calicles polygonal, mostly quadrangular, subequal ; their fossa? of moderate depth. "Walls simple, rather thin. Gemmation takes place at the very borders of the calicles. Columella pretty well developed, trabecular. Three cycles complete, a fourth, one very incomplete and obsolete. Septa crowded, unequal, slightly projecting, narrow, thin, smooth, with their edges deli- cately and equally dentate ; those of the first two cycles suddenly enlarging below and there armed with slender pointed, but not paliform, teeth. Greatest diameter of calicles averages from 7 to 8 mm. ; depth of fossa) 4 to 5 mm. Differs from the closely allied P>\ seycheUensis in its thin walls, the regularly denticulate, smooth septa, the smaller calicles and mode of gemmation. From Tr. melicemm, a species I have not yet seen, it may be distinguished by the presence of a fourth cycle, and the form of the columella. Frionastrcea scabra. Corallum explanate, somewhat convex. Calicles polygonal, slightly ii-regular ; fossse of medium depth. Walls simple, ratber thick, roughly spinose. Columella well developed, distinctly papillose. Three cycles complete, with some additional septa of a foiu-th one. Septa (about 30 in number) subequal, rather thick, delicately granulate, narrow, hardly projecting ; theii- edges with scarce, spiniform, sharp- pointed teeth. Greatest diameter of calicles, 9 to 12 mm, depth of fossu), G mm. Very similar to Fr. tesserifera, but the columella is crowdedly papillose instead of being spongious, the calicular fossae are deeper, the septa thicker and their teeth stouter and more pointed, giving the whole corallum a spinose aspect. Echinopora spinulosa. Corallum foliaceous, explanate, slightly undulate, unifaeial. Costaj indistinct, replaced by irregular scries of small, acute, delicately granulate spines. Calicles 57X ZOOLOOY OF RODRIOUEZ. oblong, rarely circular, inodoratcly crowded, very lit tie prominent, with shallow fossa?. Columella Avell developed, delicately spongious. Eour cycles complete. Septa crowdod, altogether thin, spinulous on their surfaces, the primary, secondary, and tertiary ones nearly equal, projecting, unevenly trimcate and toothed at the summit, irregularly denticulate and deeply emarginate on the inner edge ; those of fourth cycle much thinner, smaller and not prominent. Under surface of corallum faintly striate. Greatest diameter of ealicles, 5 to G mm. This species should he placed next to E. gemmacea with which it agrees in having the fourth cycle complete and in being partly lissiparous. It differs from this as well as from the other species by the condition of the costaj and by the very slight prominence of the corallites. It bears a general resemblance to E. ehrenhergi in which, however, the fourth cycle is entirely wanting, the costse being well marked and the ealicles more projecting. With the coarsely spinose E. hemprichi it cannot be confounded ; from E. helli it is distinguished by the absence of paliform lobes, and by the number and description of the septa. Fuugidse. Ftmgia haimei- Fiaigia " disons," M. Edw. Hist. Nat. Corall. III., p. 9. Fimgia haimei, Verill, Bull. Mus. Comp. Zool. I., p. 51 (1861). Lobactis dentigera. ■ F/nigia dentigera, Leuckart, Zooph. corall. spec. s,cn. Fung. p. 48, PI. III. fig. 1, 2 (1811) ; M. Edw. Hist. Nat. Corall. III., p. 17. Favonia cristata. Madrepora cristata, EU. & Sol. Hist. Zool. p. 158, PI. XXXI., fig. 3,1 (1786.) Lophoseris cristata, M. Edw. Hist. Nat. Corall. III., p. 66. Eupsammiidae. Dendropki/llia ehrenbergiana. Coenopsammia ehrenbergiana, M. Edw. & H. Ann. Scienc. Nat. 3* s^r. Zool. X., p. 109, PI. 1, fig. 12 (1818) ; Hist. Nat. Corall. III., p. 127. Madreporidse. Madrepora pharaonis. Madrepora pharuonis, M. Edw, Hist. Nat. Corall. III., p. 143 (1860). Madrepora pustulosa. Madrepora pustiaosa, M. Edw. Hist. Nat. Corall. III., p. IM (1860). CORALS.— F. BRCGGEMANN. 575 30. Madrepora plantaginea. Madrepom plantaginia, Lamarck. Hist. Aiiim. s. Vertebr. II., p. 279 (1816) ; M. Edw. Hist. Nat. Corall. III., j). 119. Madrepora acervaia, Daua, U. S. Expl. Exp Zoopli. p. i60, pi. 31, fig. 4 (181<6), M. Edw. Hist. Nat. Coi-all. III., p. 15-1. In two examples the covallum consists of a thin spreading base, covered with numerous low, mammilliform branches. This may be either a younger state or a variety. 3Iadrepora haimei. Madrepora haimei, M. Edw. Hist. Nat. Corall. III., p. 151 (18G0). Madrepora gonagra. Madrepora gonagra, M. Edw. Hist. Nat. Corall. III., p. 151 (1860) ; Briiggem. Abh. Ver. Brem. V., p. 398. In the Mam-itius specimen mentioned in my paper, the coralhim shows a smaU compact base from which half a dozen of simple, straight, sub-cylindrical branches arise. One of the Eodriguez specimens nearly agrees with it, only the branches are more irregular and partly ramified. The second is apparently very old, the base of the corallum being entirely worn, the branches altogether confluent, and their summits coalescing, forming a nearly continuous plane and bearing the general aspect of the top of a cauliflower. This condition is evidently due to some immediate outer in- fluence ; most probably the coraUum had approached too near the sui-face of the water. Madrepora seriata. Eeteropora seriata, Ekrenb. Abh. Berl. Akad. 1832, I., p. 337 (1834). Madrejiora sei'iata, M. Edw. Hist. Nat. Coral. III., p. 152. A low, slender branched variety, in which the seriate disposition of the calicles is not weU pronounced. Madrepora corymbosa. Madreptora corymbosa, Lamarck, Hist. Anini. sans vert. II., p. 279 (1816) ; M. Edw. Hist. Nat. CoraU. III., p. 154 In the Bodriguez specimens, the stems and branches show an extreme tendency to coalesce and form bulky masses. Madrepora jlabelliformis. Madrepora Jlabelliformis, M. Edw. Hist. Nat. Corall. III., p. 156 (1860). 576 ZOOLOGY OF T^OnT?Tr.ITF.Z. Miidrepora microchidns. Hctci'opora microchuhs, Elurnb. Al)h. lierl. Akad. 1832, I., p. 383 (1831). 3[adrcpora snrculosa, Dana, U.S. Expl. Exp. Zoopli. p. 115, PI. 32, lig. 1 (181G) ; M. Eihv. Hist. Corall. III., p. l.")S. Dana consiilors EhnMil)crL,''s species to lie closely allied to, or identical with, M. spirifera ; iMilne Edwards -who seems to have examined the type specimen places it near to the Jf. iiiiUeporo. In my opinion, it should he imitcd to the 31. surcnlosa, the descriptions agreeing -well -with each other. 3Iadrepora monticulosa. Corallum solid, explanate, spreading. Surface nearly even, covered for the greater part of its extent with large, stout, rounded, suheonical prominences. Coen- cnchyma f\untly developed, spongious. Calicles much crowded, short, suhlabcllate or nearly tubiform, equal on the cones, intermixed with smaller or immersed ones on the plain parts of the corallum. Walls rather thin, reticulate, slightly echinu- late. Septa nearly entirely wanting. Apical calicles always indicated, but not very conspicuous, with thicker walls and more distinct septa, the latter being generally six to eight in number. Height of corallum, from under surface of base to the summits of largest cones, GO mm. ; greatest height of cones, 30 mm. ; greatest diameter at base, 20 mm. Diameter of ordinary calicles 2 mm. ; their length the same. Diameter of apical calicles, 3 mm. ; height, Ijr to 2 mm. ; widtli of their apertures, 1 mm. This well-marked species is, in its general aspect, similar to 31. conigera, but it seems to be always spreading, and therefore, tmifacial. Besides, it is distinguished by its lower and everywhere rounded cones, by its much shorter subequal calicles, by the absence of the septa, etc. 3Iadrepora cdces. 3Iandre2)ora alces, Dana, U.S. Expl. Exp. Zooph. p. 437, PI. 31, fig. 12 (1816) ; M. Edw. Hist. Nat. Corall. III., p. 160. One of the specimens is provided with long, erect, cylindrical branches rising at nearly a right angle from the palmato-digitate main stem. There are similar examples in the collection of the British Museum, one of which is stated to be from the Louisiade Archipelago. Turbinaria mesenterina. Explanaria mesenterina, Lamarck, Hist. Anim. s. Vert. II., p. 255 (1816). Turbinaria mesenterina, M. Edw. Hist. Nat. Corall. III., p. 166, pi. E. 1, fig. 1 a, b. CORALS.— F. BRUGGEMANN. 577 Poritidae. Porites Intea. Forifes lidea, M. Edw. & H. Ann. Scienc. Nat. 3'' sev. Zool. XVI., p. 28 (1851) ; M. E(hv. Hist. Nat. Corall. III., p. 180. Milne Edwards quotes as a synonym, the Forites conylomeratu var. hdca, from the " Voyage de I'Asti-olahe." Quoy and Gainiard, indeed, mention a variety of P. conglomerata, hnt without naming it. Porites arenosa. Madrepora arenosa, Linn. Syst. Nat. Ed. XII., p. 127G, sp. 17 (1767). Porites arenosa, M. Edw. Hist. Nat. Corall. III., 180. Montiporidae. Ilontipora divaricata . Corallum ramose, subarborescent. Branches diverging, more or less angular, enlarged towards the summit ; branchlets compressed, obtuse. Calicles immersed, distant by about twice the diameter, subequal, hardly larger on the in-ominent parts of the corallum. Coenenchyma moderately spongious, with even, delicately echinu- late surface. Two cycles ; septa subequal, very narrow, rather stout. Diameter of calicular apertm-es, nearly three quarters of a millimeter. Allied to J/, rits and M. monasteriata ; differing by its mode of growth, and by the absence of the prominent verrucae. Montipora foliosa. Madrepora foUosa, Pallas, Elench. Zooph. p. 333 (1706). Montipora foliosa, M. Edw. Hist. Nat. Corall. III., p. 212. Montipora incrnstans. Montipora incrnstans, Briiggemann, Abhandl. Ver. Brem. \., p. 398 (1877). Tliis species was established for a specimen from Mauritius. The condition of the coenenchyma and the excessively small calicles render it easily recognizable. Montipora explanata. Montipora n.sp., Briiggemann, Abb. A^er. Brem. V.,p. 399. Corallum explanate, spreading, incrusting. Surface uneven, in some places slightly tuberculate. Coenenchyma rather dense, consisting of confluent trabecles covered with small spinulous grains. Calicles immersed, equal, distant by about twice the diameter. Twelve septa, but little developed, thin and narrow, one of the primaries always stouter and larger ; the secondaiy ones sometimes rudimentary. 4 D 578 ZOOLOGY OF RODRIGUEZ, Height of corallum averages 7 mm. ; cliamotcr of calicular apertures, three-quarters of a mm. llodri'niez {Slater) ; Mam-itius (JSrit. Miis., Jena Mns.). Distini^uishetl from the preceding by its much larger calicular apertm*es, by the coai-ser coenenchymal tissue and less delicately spinulous svu'face; closely allied to M. scabriciila, but the latter is a more massive species, with uniformly tuberculate surface. JH. lichen, with which I was formerly inclined to unite it, has ])rominent calicles, a character totally absent in J/, explanata. Montipora lichen. Manopom lichen, Dana, U.S. Expl. Exp. Zooph. p. 492 (1816). Montipom lichen, M. Edw. Hist. Nat. Corall. III., p. 218. There are several fragments in the collection which seem to belong to the above species. The corallum is explanate, foliaceous, partly bifacial, partly incrusting. Sm"f ace more or less even ; margin in one fragment reflexcd, in another straight and someM'hat tumid. Coenenchyma spongious, with most delicately echinulate surface, the minute spines being generally imited to small packets. CaKcles very small, unequally distant on the upper surface, crowded below, nearly everywhere encircled by a coenenchymal ring, which is either continuous or separated into six roimded papilltc. Primary septa well-developed, thin ; secondary ones rudimentary. PociUiporidse. FocilUpora brevicornis. Focillipora brevicornis, Lamarck, Hist. Anim. s. Vert. II., p. 275 (1816) ; M. Edw. Hist. Nat. Corall. III., p. 304. The present collection contains the true P. brevicornis, as well as a singular varietv, which may be termed P. brevicornis var. cerebrum. The corallum is hemi- spherical, brain- like. Tlie principal branches are much enlarged and compressed, confluent, mteandrous, subsimple, with nearly smooth summits and obtu.sely verrucose lateral sm-faces. This form would, according to the classification of Milne-Edwards, be placed in quite another section of the genus, viz. close to P. inceandrina, etc., and I shou.ld never have ventui'ed to regard it as a variety of the P. brevicornis, but for the complete series of transitional examples in the collection. Pocillipora favosa . Focillipora favosa, Ehrenb. Abb. Berl. Akad. Wiss. 1832, I., p. 351 (1834) ; M. Edw. Hist. Nat. CoraU. III., p. 305. CORALS.— F. BRtJ'GC4EMANN. 579 Pocillipora grandis. Pocillopora grandis, Dana, U.S. Expl. Exp. Zooph. p. 533, pi. 51, fig. 2 (1846) ; M. Eclwd. Hist. Nat. CoraU. III., p. 307. The principal branches are, as a rule, more elongate, and their summits more mceandrous than in the specimen figured by Dana. The lateral verructe are very variable : either crowded or more remote, or (on the outer sm-face of lateral branches in some of the coralla) even totally wanting ; either appressed, or more horizontal ; generally pointed and angular, but in one example obtuse and rounded. One specimen approaches the P. verrucosa in having the summits of minor branches covered with verrucse. Perhaps the Rodriguez coral is a distinct species ; but since I have no specimens of the Pacific P. grandis for comparison, I abstain from characterising it under a new specific name. In my opinion, moreover, the species of Focilli])ora have akeady been far too much multiplied. LONDON: Printed by George Edwaud Etuk and William Spottiswoodk, Printers to the Queen's most Excellent Majesty. Trans oi^ Venus Exped. Kerguelen Island, Botany, PL I. J.NFitcb litk RANUNCULUS TRULLIFOL.IUS, Hook.f. Trans, ol Ven-us Exped. Kerouelen Island. Botany, n.ii i^ ;v-\ line Trans, of Venus Exped. Kergukj.en Isla>id Botany. PI v. ij ,in : mfi J.N Filch lull Fi Ich imp ■■oo o o Q izi S5 p u CO ^ ■»!?:^. -^«»^' CO C '"1 Trans, of Venus Exp e d . Kerouelen Island. Zoology PI. VII >/ 2. Cjrril Davenport (Ul.etlitk t Mmtern Bro'imp Trans, of Vemis ExpecL Kerguelen Island Zoolo^^y Pi. MUi. i^avanpnrt. ^.j!^. iii>, Miii**>j-n Bxoit iTi >^ .> ■^^ - -^ ^ %!^. ,«' ,<■■*. ^^^^z ^K^-^- 1^ 'If J> ^3 "^ ■6r •-, •a;'- '..i' \J' I — i 1^ . Sli^TlT'lY,:^ :liii* /// "* ->*•. ^:; t»3 -:% m i*! ♦^ -?»«- (fj § :sn: V ^^ X' /-/-'-'^'^i^'- x/-,- ^. % -> 4 "t^- ^■|ti' m Trans, of Venus Exp e d. Kerguelen Island Zoolo^ PL XI. G B Sowerby del etlitb. Mintam Bxo* inyo. ,,,^.ot VemisExped Kergxielen Ist^^^ Zoology P1_X11. b I V t MmtemBroa C. B Sowirkyl"* 1_10. SappWna ^^^^^^^a^^u. jj^_1.9. Centropa.ge3 brevi IvHTLS. of Venus Exped. Kkrotie-l'f.'n Islam p Zoology PI. xm MintemBro* iTi O.B S<»w.«-by lni^ - 1 / K \;,.r-r ■<'■'' ■< H w ^v ,'^' ^--l' --^>:'\ >>*:-•;- ^-'W'^-. / 1- f ;V ^' .-\ J Trans, of Venus Exped Kerguelen Island Zoology pi.:xvi. CB^rjeoo lit>i Mmt«ri\ Bro« -.n-.f Trans^ofVenuiE^^A IVER r.lTELEN IsLAHP ZooIo^VlZL™. Mint*m Bro» imf ;, Bcri^a^ ^■■'■^ •"; Q<- y \t ^ \ a ^^. . \ /i^' ^*^ti — W ^ ,• i/ ^, /■^ \/ "A^' y > ) \\ ^ v- \ i \( ^v^ Trans of Venus Exped. Rod RIGUEZ BotanjPJIIX. JWFitch bUi QUivi5iALAClNlATA,^^ M. fl'cJi imp Trans of Yenus Exped Rod HIGUEZ BotanjTIXIX. JJf.Fitch liUi QUIVISIA LACmiAlk.Sal/'M. Fitch unp. Trans, of Vemis Exped. E OX>RIGUE2;. Botany, PI XX JtlFa'.i. Iiih SCLEROCARYA CASTAISTEA, Baker. FibcK I rap Trans, of Venus Eiped HODRIG-UEZ Boiany.PUXl. FilcK >iaf jHri'.cik si>> MATHUKINA PE^DU^IFLOBA, ^^^^^ Trans, of Venus Exped R ODRIGUEZ BoUnyPlXXII. JNFitehKtli RANDIA HETEROPHYLLA, Balf/H. F.tch imp tans, of Venus Exped. HoDRIG-UEZ. Botany.PlXXlll 1 FikcK in J n Fitch litl> FEKHE n lull. A ABROTANELLA RHYNCHOCARPA, ^a//-/j7.. B CARISSA XYLOPICRON. /><•/ 7"/. Trans of Veims Exped. Rodriguez Botany PI. XXV Jil JNBtchLiU TANULEPIS SPHENOPHYLLA BaU\f\l. Titchlnip. Trans of Venus Kxped iioDRl G-UE2 Botanj.PIXXD: JtrPiu), Kik 1-3.SARC0STEMMA VIMINALE,^^,- Trans, of Venus Exped Rodriguez Botany.Pl.XXX JKFiUhlith HYPOESTES ROBRIQUESIAUA. J^a^M Kllrli imp T, Frans of Venus Exped. Rodriguez Botany.Pl XXXI JNFitcIiKtl, NESOGENES DECUMBENS, Bdlf R.I. I'ltcK Trans of Venus Exped RoDHIGU EZ. Botany. PI XXXJI .'NPilciktl, CLERODENDRON ^.KZmiklMn , £aif M Frtck , tans, of Venus Exped. E ODRIG-UEZ. Botanj.PlXXXUl. JirFitch liih T.»-'r.- — r PISOl'ITA VTP-COSA. B^df/il. Trans of Venus Exped. "RoDH IGUEZ Botany Pl.XXXIV J.HFitcli litk. PILEA iiALi'0\JRl.^aA:er. ^,iA 1 fraTis. ol Venus Exped. R OD-RIGVEZ. Botany PI. XMV J.N.Rl<:Klili. PEPEBOMIA RETICULATA,^fl//;aJ Til.), nsp Trans, of Venus Exped. BoDRlOUEZ Bolany,Pl.X]©/I. JB.Fitcli lith LISTROSTACHYS APHRODITE, BalfM.&SMoori' Ftbck iicf 'frans of Vetius Exped. RODRFGUEZ A. BARTRAMIA VVK&EKS, MM^ Botany Pi. KOfVE. f» r [1 CALICO STELLA L^VTOSCULA, AfUt. PTEHOOONiL'M CUR\aPOUUM M/y.. n SEMATOPHYIIUM FULVIFOLIUM Mi/f J >; JUcii. LitiL ikii^ Trans o!' Vevras Exped. "RODT^TGUEZ PLJOQCVm A ..^Jijki^ SEMATOPlrA,I.UM INCURVUM, i/l^. KCTROPOnCECiUM DOLIAR£,Afi«. O LOPHOCOLEA SUBROTUlfDA, 3/?irt'. D LOt'llOCOLEA dPFOSlTA,^^^ JNT^iidilnlh Jitzhlmp rails of Venus Exped. Rodriguez Botany Tl,3gKIX. FHRAGMICOMA EMERRENS. Mm c LEJEUNIA BAI.FOURJ VlW, FHRAGMICOMA CARINATA, Mutl. D LEJEUMW ABORTIVA,M>K Trans of Venus Exped. LEJEUNIA FURVA. MiU Rodriguez BotanvPl.XL A mfi^/f lEJUKIA SUBCILIAIA Mlil, c FRULLANIA OBSCURJFOLIA. Afttt. [A ,/ / L' PRUiLANIA APICALLSA/d/ JJIMALitii FlUl Imp Zoology^H_5i- of VenusExpf^ RODRIGUE MinlemBro. vmp Ri.iiut«T da.eu.ih A— F Canne murivora G. Nycticorax megacepi ,Kalus. H.NycUcorox griseus ■ans. of Venus Exped Rodriguez. XLII. R MiTji«m del ot.Klh. MinieiTi Bron inip. A-G. Necropsar rodpricanus. HrN.NecropsitUcua roderica^nus. Trans, of Venus txped. Rodriguez. IXLlll. Aphanapteryx Jcguati '''""^TM Bi-o» imp RODRIGUEZ. Zoolo^ PI XLIV ^ |°|;i^,S.^V~rj^O^HAPS _SOI.ITAR tebra lA 5.-6 Tw^i^n/;, K..r/,i,r 7 i.^^< JTo^/. MinUrn Bro« HUi Trajis of Venus Exped. RoERIGXJEZ Zoology PiXLV 10 \ N 8. 14- . l< Mintonidfll S OL.il Aire E,- ., pte,^j,o.U« 1 <;,^/;i and sevcMf^i st«>-"«i- rU>s u, .r,w,at 2-6 ^W.fc^ r^a of n^M. srde, , '2 ^-Xt^uy <-/• .-c>u««, u,.J ,.ui- 7-9 SterruxL niis , ? /4-. Brawt • Minion M..-* !•! Trans, of Venus Exped. R ODHIGUEZ. Zoology PI. XLVi Muii«m Bro" hth. SOLl'I'AIRE. PKXOPHAPS S01.1TARTA 1 Pelvis oF mxUe sUeUton, . C Side view oP coccyx aiis. of Vejnus Exped. RODRIGTJE/^ Zoology Pl.XLVlI. \ .:^ \ vi .1 o~ ^^^. rintern del. SOLITAIRE 7 Pelvit! ^ 2 i'elvvB t PEZOPHAPS SOLITARIA. 3 'FufcvJia:. 4-. Ston^ swaiiowirl hv Ihi- SoUtairt ^ MiiilHru Bn<'- IiiIl Tr>£ms. of Venus Exped. Rod RiauEz . ■* ■Bi'Sfc-, Zoology PI. JOiVUI ■* Minium, d*- SOLITAIRE. . PEZOPHAPS SOLITARIA. •*- ■ 3t£j~rtu^ Muvtecn firo* Uth ns. of Venus Exped Rodriguez. Zoology PI. Xi.lX ■^^F "X