ee er + SO he el tone mee oe Poe atte wh AOeh Sar on prerre SN RAN ON Ok et A ~~ ena SONA A TS Tob Aa RAN AN a Ao ee ee oe en De Te . arate ane inde hte tat ae en ie, We Te RN Hee Na Matha th ok inh RTE Ne be So OR ee) o-: ne ge BAO OM BES es eerce & ana ea ©. Aw me Bom Bp ign, oh eon Ay Sow een 68 NA Cannel = aN + arts eat waweP@Penmrs oe Hees Mays sate me Met Re Nar ae A Nad tee wt he prea res . . < : 2 we eee A : * . 26S BS Ti Sera s as . « «: . . ‘ * r . “ * soe . ca - * HANDBOUND AT THE ke S UNIVERSITY OF TORONTO PRESS ae Me ie , ae none Mi Ny A o hil a Che ee we ad THE ANNALS G2 AND MAGAZINE OF NATURAL HISTORY, INCLUDING ZOOLOGY, BOTANY, ann GEOLOGY. (BEING A CONTINUATION OF THE ‘ANNALS’ COMBINED WITII LOUDON AND CHARLESWORTH’S ‘ MAGAZINE OF NATURAL HISTORY.) CONDUCTED BY ALBERT C. L. G. GUNTHER, M.A., M.D., Ph.D., F.B.S., WILLIAM CARRUTHERS, F.R.S., F.L.S., F.GS., AND WILLIAM FRANCIS, Ph.D., F.L.S. 7? ( Fé Deets VOL. XX.—SIXTH SERIES. PIII Oe LONDON: PRINTED AND PUBLISHED BY TAYLOR AND FRANCIS. SOLD BY SIMPKIN, MARSHALL, HAMILTON, KENT, AND CO., LD.; WHITTAKER AND CO.: BAILLIERE, PARIS: MACLACHLAN AND STEWART, EDINBURGH : HODGES, FIGGIS, AND CO., DUBLIN: AND ASHER, BERLIN. 1897. “Omnes res create sunt divine sapienti et potenti testes, divitie felicitatis human :—ex harum usu donitas Creatoris; ex pulchritudine sapientia Domini ; ex ceconomid in conseryatione, proportione, renovatione, potentia majestatis elucet. Harum itaque indagatio ab hominibus sibi relictis semper eestimata ; a veré eruditis et sapientibus semper exculta; malé doctis et barbaris semper inimica fuit.”—Linnavs. “Quel que soit le principe de la vie animale, il ne faut qu’ouvrir les yeux pour voir qu’elle est le chef-d’euvre de la Toute-puissance, et le but auquel se rappor- tent toutes ses opérations.”—Brucnner, Théorie du Systeme Animal, Leycvu, 1767. oe ee ow ee oe se + Lhe sylvan powers Obey our summons; from their deepest dells The Dryads come, and throw their garlands wild And odorous branches at our feet; the Nymphs That press with nimble step the mountain-thyme And purple heath-flower come not empty-handed, But scatter round ten thousand forms minute Of velvet moss or lichen, torn from rock Or rifted oak or cavern deep: the Naiads too Quit their loved native stream, from whose smooth face They crop the lily, and each sedge and rush That drinks the rippling tide: the frozen poles, Where peril waits the bold adyenturer’s tread, The burning sands of Borneo and Cayenne, AU, all to us unlock their secret stores And pay their cheerful tribute. J. Taytor, Norwich, 1818, ANNOUNCEMENT. Tue present Number completes the Sixth Series of the ‘Annals,’ and I avail myself of this occasion to pass the responsible editorship over to my Son, who for some years past has largely assisted me in the management, and who is fortunate in retaining the co-operation of the two co-Kditors whose names appear on the Titlepage, and whose advice has always been at my service during the twenty years I have been associated with them. It is now more than 60 years since, fresh from my studies at the University of Berlin, I was consulted by Mr. Richard Taylor as to the probability of such a Journal meeting with sufficient support. The two Natural History Journals then existing in this country were Loudon’s, subsequently Charlesworth’s ‘ Magazine,’ and the ‘ Magazine of Zoology and Botany’ conducted by Sir William Jardine, Dr. Johnston, and Mr. Selby. This latter Journal was about to be discon- tinued, and the same fate seemed jikely to result in the ease of Sir William Jackson Hooker’s ‘ Botanical Companion.’ In these circumstances it was suggested to Mr. Taylor that by amalgamating the two there would be a greater chance of success. The two most important Journals on the Continent at that time devoted to Natural History were the ‘ Annales des Sciences Naturelles’ and the ‘ Archiv fiir Naturgeschichte,’ recently founded by my friend and teacher Professor Wiegmann. At a meeting arranged by Mr. Taylor, at which Sir W. - Jardine and Sir W. J. Hooker were present, and which I was iv invited to attend, it was considered that a Journal conducted on lines similar to these Continental Journals would be welcomed by British Naturalists, and the 120 Volumes are evidence that the expectation then entertained was well founded. Sir W. J. Hooker’s connexion with the ‘Annals’ terminated with Volume IV., as it was impossible to find room for the long communications from Botanical ‘Travellers, which, although very interesting, did not appear to be suitable for publication in its pages. Charlesworth’s ‘ Magazine’ was absorbed into the Annals’ in 1840. In the Preface to the First Volume of the Second Series the Editors acknowledge “ the aid of Dr. W. Francis, as from the commencement of the work they have had the advantage of his constant and valuable assistance in its regular produc- tion;” and in January 1859 my name was added to the list of Editors. Although the publication of this Journal has been by no means a source of any considerable profit, I have been amply rewarded, as it has procured for me the friendship of most of the eminent men who have contributed so largely to the vast progress of Natural History during Her Majesty’s reien. May I hope that the same good fortune will attend my successor, and that he will receive the same friendly assistance from those who are now devoting themselves to the advancement of those departments of Science for the promotion of which this Journal was founded. WILLIAM FRANCIS. CONTENTS OF VOL, XX. [SIXTH SERIES. ] NUMBER CXV. I. The Actiniarian Family Alicitide. By J. E, Durrven, A.R.C.Se.(Lond.), Curator of the Museum of the Institute of Jamaica. CRIBS T.) ccc cece c cece nteeccrssscsccetecccssccccsssccess II. Ona Collection of Heterocera made in the Transvaal. By DP RIPIRTANT Soca codes acncewerssotneestesececscesenvesoas Ill. Cicadide from the North Chin Hills, Burma. By W. L. cose cen cere enc sacetsswesanceresessnccenenns : IV. Contributions from the New Mexico Biological Station.— No. IL. (continued). On a Collection of Diptera from the Lowlands of the Rio Nautla, in the State of Vera Cruz. LU. By C. H. Tyter MOWMARND, FES. 5... c cece ccecsecccasseerecesvsscvsvesesves V. The Species and Subspecies of Zebras. By R. I. Pocock, of the British Museum of Natural History ...........sececeeevvecs VI. Aquatic Rhynchota: Descriptions and Notes.—No. I. By GeORGE W. KIRKALDY ......cscsecccccceresnssenns dis sustaie VIL. New Cyclostomatous Bryozoa found at Madeira. By James Prd SOHNEON, OOIr. Mi B.: eer escent eee seesssecnances VIII. On Lepidoptera Heterocera from China, Japan, and Corea. By Joun Henry Leecu, B.A., F.LS., F.Z.8., &c.—Part I. Family Geometride ; Subfamilies Gnochromine, Orthostivine, Larentiine, Acidaliina, and Geometrin® .iccsseseccsersccerteveneeesesnnes IX. On new Species of Rhopalocera from Toungoo, Burma, and the Battak Mountains in Sumatra. By Major J. M. Fawcerr .... X. Description of a new Rat from China. By OtprreLp THomas New Book :—Ueber die Palpen der Rhopaloceren. Ein Beitrag zur Erkenntnis der verwandschaftlichen Beziehungen unter den Tagfaltern, Von ENz2I0 REUTER ..........sseeeeeeeeeeee Proceedings of the Geological Society ..sseeeesseneeereeenaees Page 19 33 52 vi CONTENTS. NUMBER CXVI. Page XI. Notes, Morphological and Systematic, on the Madreporarian Subfamily Montiporine (Montipora and Anacropora), with an Account of the Phylogeny of the Madreporide. By H. M. Bernarp, MoASCantabs = (Plate 2) ac sreterere ence cvs treteinatte ke ais 01s ateRageee 117 XII. Contributions from the New Mexico Biological Station.— VI. The New Mexico Bees of the Genus Heriades, and a new Halictus. By T¥. DA, COCKEREMI: g..myete etc tee are 155 XIII. Revision of the Pierine Butterflies of the Genus Delias. By A. G. Borrier, Ph.D., &c., Senior Assistant-Keeper, Zoological Department, British Musewmy 2 strc. wie eictalete laleloiciste wis,» eh eee eae 143 XIV. Notes from the Gatty Marine Laboratory, St. Andrews.— No. XVIII. By Prof. M‘Inrosu, M.D., LL.D., F.R.S. (Plate II.) 167 XV. On new Species of Histeride, and Notices of others. By G. aewis, FLAS. ..: <00 5 ct. «denier ereer eee ei oie ee 179 XVI. On a Collection of Heterocera made in the Transyaal. By Wi LJDISTANT™ & accindd ede eee n Gee Ce eC eee 197 XVII. Descriptions of Two new Species of Amphidromus. By Hueu Fuiron.. .(Plate Wi) ye gai. oem see een 211 XVIII. On supposed new Species of Oleacina, Trechomorpha, and Bulimulus, By HueH Furron. (Plate VI.) ..........0.004, ». 212 XIX. On some small Mammals from Salta, N. Argentina. By OLDELELD: THOMAS, secemteaies a isia-isle wise Who #0 whe oe «(tana eee 214 XX. Descriptions of Four new South-American Mammals. By OLDFIELD: THOMAS « <5. cites cece ce rome emilee le nnd ee nn 218 XXI. On the Anatomy of Apera Burnupi, E. A. Smith. By Water E. Cotiiner, F.Z.8., Assistant Lecturer and Demon- strator in Zoology and Comparative Anatomy, Mason University College, Birmingham. (Plate V.) ..000..2.4- 6. «003+ ace ih epee XXII. Description of a new Papilio from Bali of the nox group. By LionEL DE NICEVILLE, F.E.S., C.MLZS., &e. 2. ss deel . 225 XXIII. Description of a new Ceratopterine Eagle-Ray from Jamaica. By G. A. BoULENGER, PVRS. ©... 2... .sleeeeeee 227 XXIV. On Lepidoptera Heterocera from China, Japan, and Corea. By Joun Henry Leecu, B.A., F.LS., F.Z.8., &.—Part I], Family Geometride ; Subfamilies Gnochromine, Orthostivine, Larentiine, Acidaliine, and Geometrine. (Plates VII. & VIII.) ............ 228 Proceedings of the Geological Society .............. =. SReeerarreray aan aee 248 CONTENTS. Vil NUMBER CXVIL. XXV. On the Genera and Species of Tropical African Arachnida of the Order Solifuge, with Notes upon the Taxonomy and Habits of the Group. By R. I. Pocock, of the British Museum of Natural GOL suis ses's iewesesviivesssscccctWemenanceace 240 XXVI. Contributions from the New Mexico Biological Station.— No. IL. (continved). On a Collection of Diptera from the Lowlands of the Rio Nautla, in the State of Vera Cruz. Il. By C.H.Tytrr EN EGO fiiala wis le pibie,k) oR cv denon ao Sait oy mow sae aes 272 XXVII. Two new aay of the Genus Xanthospilopteryx, Wallengren. By W. J. Hotztanp, LL.D., F.ES., &. ......eeee 291 XXVIII. On a Collection of Fishes from the Island of Marajo, eee eer A. BOULENGER, FLRAG. occics cccccaaccccvisen tues 294 XXIX. Descriptions of some new Species of Heterocera from Tropical America, By Herpret Deuce, F.LS. &c. .........- 299 XXX. Description of a new Gymnotine Fish of the Genus Sternopygus. By G. A. BouLENGER, F.R.S, 2... . cc cece ee ee eee 305 XXXII. On the Reptiles of Rotuma Island, Polynesia. By G. A. PRENSA aor, oc ay «ale aed oe eine ed Od ce bev in dle wre 306 XXXII. On two new Rodents from Van, Kurdistan. By Oxp- a tie) + tay nee bundads aephclenm kT bas a eghss 308 XXXII, On new Species of Forfieularia. By Maucotm Burr, NN cediaih coh nigid Hig’ ea Kidwe pied T Sabi! AE tin OTE 08d 057m © 310 XXXIV. On a Collection of small Mammals from Uganda. By AN I sine Wallon, #.d o's) nce maxillary and fronto-nasal sutures runs straight across froy, side to side Ann. & Mag. N. Hist. Ser. 6. Vol. xx. s 8 114 Bibliographical Notice. instead of being bowed backwards ; the supraorbital rims are more developed (though still small for so large an animal), and are continued along the parietals to the outer corners of the intermaxillary. Anterior edge of anteorbital plate more slanting. Posterior nares wider and more open than in the allied form. Incisors broad, pale yellowish, finely but irregu- larly striated in front. Dimensions of the type (an adult female in skin) :— Head and body (probably stretched) 310 millim. ; tail 290; hind foot (moistened) 60 *. Skull: basal length 51°33; greatest breadth 27:4; nasals 23 x 5°6 ; interorbital breadth 8°2; breadth of brain-case 21°3 ; interparietal 7x16:2; palate length from henselion 26; diastema 17; anterior palatine foramina 10°6 x 45 length of upper molar series 9-1. Hab. Kuatun, N.W. Fokien. Lipe> va. Nieno..9t40:622: This fine rat is evidently very closely allied to Mus Govwerst, which is a native of Burma and Tenasserim, and is therefore widely distinct from it geographically. It differs, however, by its rather larger size, especially its longer hind feet, and by the various cranial differences above enumerated. It is named in honour of Mr. J. de La Touche, of Foochow, to whom, in conjunction with Mr. Rickett, the British Museum is indebted for a considerable number of valuable Chinese mammals. Among these may be specially mentioned examples of the rare Typhlomys cinereus, M.-EKdw., specimens which have enabled me for the first time to show the proper position of this interesting genus f. BIBLIOGRAPHICAL NOTICE. Ueber die Palpen der Rhopaloceren. Kin Beitrag zur Erkenntnis der Verwandschaftlichen Beziehungen unter den Tagfaltern. Von Enzio Reuter. (Acta Societatis Scientiarum Fennice, tom. xxii. No. 1.) (Helsingfors, 1896.) Tue early writers on insects used to complain that the Order Lepidoptera was one of the most difficult of all to classify, on account of the deficiency of characters. But with the increase of our know- * The two other specimens have hind feet 57°5 and 58 millim. long. Three spirit-specimens of M. Lowers: in the Fea collection have feet respectively 51, 51, and 52 millim, in length, tT P. Z. 5. 1896, p. 1016. Bibliographical Notice. 115 ledge we find that characters abound, and it is now rather a question of the real value to be attached to the structure of different organs than a matter of complaint that characters cannot be found. Nor shall we arrive at a really satisfactory system of classification of Lepidoptera until the structure of the principal organs has been worked out in all the various stages of the insects; and this is a lifelong study for a great number of observers. The author of the present treatise has directed his attention chiefly to the structure of the palpi in butterflies, paying special attention (1) to the outward structure and form, (2) to the hairy or scaly clothing of the palpi, and (3) to the basal spot, which is a bare space on the inner side of the basal joint, which is striated, pitted, and set with numerous conical hair-scales. For the purposes of the present work 3557 palpi have been examined, belonging to 670 species and 302 genera, the result of this long and patient study being embodied in the elaborate treatise before us. It is illustrated by 6 plates, the first five representing structural details and the sixth containing a genealogical tree of the evolution of the Lepidoptera. The Hesperiide are regarded as a distinct suborder from the Rhopalocera, under the name of Grypocera, which is certainly an improvement on the more usual course of treating them as an aberrant family of the latter. The first portion of the work consists of a description of the general form and clothing of the palpi under the various genera, and especially of the basal spot; the description of the latter some- times exceeds in length that of all other structures noticed. Having concluded this, the author generalizes his results, and discusses the comparative relations of the various families and smaller subdivisions which he admits. Here his observations are not confined to the palpi, but extend to the neuration and other morphological characters of the insects; and he shows himself to be thoroughly acquainted with the extensive and not always easily accessible literature of the Order Lepidoptera, and compares his own conclusions with those of other writers to great advantage. This is followed by general observations on the origin and classification of the Lepidoptera, not without reference to palzonto- logical considerations ; and by an extensive Bibliography, filling 11 closely printed pages. We congratulate the author on the completion of a valuable and meritorious work, which marks an epoch in the study of the particular structures to which it is devoted. Much good work has previously been accomplished in Finland in other orders of insects; and the Finns are fully entitled to claim as high a rank as entomologists as they have long occupied as philologists. 116 Geological Society. PROCEEDINGS OF LEARNED SOCIETIES. GEOLOGICAL SOCIETY. March 24, 1897.—Dr. Henry Hicks, F.R.S., President, in the Chair. The following communications were read :— to} 1. ‘On the Association of Sigillaria and Glossopteris in South Africa.’ By A. C. Seward, Esq., M.A., F.G.S., University Lecturer in Botany, Cambridge. In this paper the Author describes in detail several specimens of fossil plants submitted to him by Mr. David Draper of Johannesburg. His conclusions as to tne geological age of the plant-bearing beds differ from those arrived at by Mr. Draper from stratigraphical evidence ; the plants point to an horizon which may be referred to what is now termed the Permo-Carboniferous age. The diffieulty of distinguishing between various forms of Glossopéeris-leaves is discussed at some length: and the opinion expressed that it is practically impossible to separate the Indian, Australian, and African forms of G@. Browniana, G. indica, and others. ‘The chief interest as regards the plants centres round the specimens of Sigillaria; these are fairly well preserved impressions, and are referred to the well-known species, S. Brardi. In addition to various forms of the genus Glossopteris and the specimens of Sigillaria, the following plants are recorded :—WNoeygerathiopsis Hislopi, Gangamopteris eyclopteroides, Phyllotheca, Conites sp., Cardiocarpus sp., and Sphenoptercs sp. The paper concludes with some general remarks on botanical provinces in the Northern and Southern Hemispheres, and the relation of the Glossopteris-flora to the Coal-Measure vegetation of Europe. 2. ‘Notes on the Occurrence of Sigillaria, Glossopteris, and other Plant-remains in the Triassic Rocks of South Africa.’ By David Draper, Esq., F.G.S. The Author gives a brief description of the geology of four localities, within a comparatively short distance from Johannesburg, from which several fossil plants have recently been obtained. He considers the plant-bearing beds to belong to the Lower Stormberg Series of Dunn, and to the horizon known as the Molteno Beds. The most important locality described in these notes is that of Vereeniging, 30 miles south of Johannesburg, where the Author found several specimens of Stgillaria associated with Glossopteris and other plants in iron-stained sandstones. The significance of this discovery of Sigillaria is briefly discussed. The several species of plants have been described by Mr. A. C. Seward in a paper recently sent to the Society. JED. del. Ann & Mag. Nat. Hist.S.6 Vol XX.PUL. iS age » (B23 Uta | ese ; Vintern Bros Jith. THE ANNALS AND MAGAZINE OF NATURAL HISTORY, (SIXTH SERIES.] No. 116. AUGUST 1897. XI.—Notes, Morphological and Systematic, on the Madre- porarian Subfamily Montiporine: (Montipora and Anacro- pora), with an Account of the Phylogeny of the Madreporide. By H. M. Bernarp, M.A. Cantab. [Plate IL] IN continuation of the work of cataloguing the National Collection of Corals, I have devoted nearly two years to the study of the genus Montipora. I propose in this paper to give a short summary of the morphological results arrived at. Before doing so, however, I should like to take this oppor- tunity of expressing my gratitude to the Director, Sir William Flower, F.R.S., for the friendly interest he has taken in the progress of the work, and also to my friend Prof. F. Jeffrey Bell, who has charge of the coral collection, not only for valuable advice and criticism, but also for much active and willing help. The first volume of the official catalogue dealt with the genus Madrepora, and was written by the late George Brook ; the second volume, containing two smaller genera— Turbinaria and Astrewopora—is the work of the present writer. The morphological results arrived at during the preparation of that volume will be found in its introductory chapters and in two papers in this Magazine *. The present study of the Montiporine, which, with Madrepora, Turbinaria, and * Vol. xv. 1895, p. 499, and vol. xvi. p. 273. Ann. & Mag. N. Hist, Ser. 6. Vol. xx. 9 118 Mr. H. M. Bernard on the Astreopora, form the family Madreporide, enables me to summarize the conclusions arrived at as to the inter-relation- ships of these genera. That summary will constitute the concluding part of this paper. The Genus Montipora. This genus differs in many remarkable points from either Madrepora, Turbinaria, or Astrewopora, and doubts as to its affinities are, as we shall see, very prominent in the works of former students. Though founded by Quoy and Gaimard, the name appeared first in print in de Blainville’s ‘ Diction- naire des Sciences naturelles,’ t. 1x. (1830), the author having seen it in MS. He placed the new genus between Gemmi- pora (= Turbinaria) and Madrepora and among a number of other genera, including Porites. In 1834 Ehrenberg* suppressed the genus, distributing the species among the Porites. Dana, in 1848, re-established and greatly enlarged the genus, but changed its name to Manopora. He objected that the word Montipora referred to the coenenchymatous elevations of the surface, which were not universally present. He thought that the genus Mano- pora was closely allied to Madrepora, and, in fact, could be deduced from it by the degeneration of the protuberant calicles. In 1849 Milne-Edwards and Haime removed it from the Madreporide and placed it among the Poritide, and to this arrangement they kept in their monograph of the Poritide in 18511}; they there noted, however, that Montipora showed certain structural resemblances (‘‘ quelques rapports de forme”’) with the Madrepores. Prof. Verrill at first adopted Milne-Edwards’s arrangement, but eventually followed Dana in placing the genus among the Madreporide. Briggemann apparently came to no conclusion. In two papers which appeared after his death the genus is variously laced. In one the Montiporide followed the Madreporide and Poritide ; in the other Montipora occurs with Porites, Turbinaria, &c. under the Madreporide. Prof. Studer, in 1878, followed Milne-Edwards, but in 1880 adopted Dana’s classification so far as to class Monti- pora with Madrepora. Klunzinger also foliows Dana in this respect. * © Corallenthiere des rothen Meeres.’ t Ann. d. Sci. Nat. ser. 3, xvi. p. 21, Madreporarian Subfamily Montiporine. 119 In 1884 Ridley * denied the close relationship between Montipora and Madrepora claimed by Dana. The forms of the latter which, owing to the obscuration of the apical polyp, Dana thought might constitute a connecting-link between the two, lent, according to Ridley, no support to such a con- clusion, inasmuch as the apical polyps in these types are not really absent, but only inconspicuous owing to their multi- plication. Further, it was claimed that a far-reaching difference in the method of budding separated the two. In Madrepora the budding is said to be centrifugal, the fresh buds forming below the central apical polyp, while in J/onti- pora undifferentiated ccenenchyma takes the lead and the fresh polyps appear above one another. I shall endeavour to estimate later on to what extent this is a true diagnosis of the morphological difference between Madrepora and Montipora. Duncan, in 1884, in his revision of the Milne-Edwards and Haime system, followed these authors in placing Monti- pora with Porites. Quelch, in 1886, in his description of the ‘Challenger’ Reef Corals, placed Montipora among Madreporide, as does Miss Ogilvie in her recent ‘ Microscopic and Systematic Study of Madreporarian Types of Corals” t. Lastly, in 1889 Dr. Ortmann §, after following Dana in 1888, classed the Montiporide with the Madreporide, Pori- tide, Turbinariide, &c. as independent families of the Madre- poracea. The conclusion here arrived at on this point, viz. that the Montipora belong unmistakably to the Madreporide, is based upon a study and comparison of nearly 400 specimens, divisible into some 120 types, of which more than half are new. The youngest colony that I found is contained in a small oval epithecal saucer, 3:5 millim. long diameter (Pl. II. figs. 1, 2). This saucer is filled with a spongy ccenenchyma. One polyp, about °25 millim. in diameter, opens in the highest part of the coenenchyma and near the centre, while a few smaller ones open between it and the epitheca. It seems to me that there is no escape from the conclusion that this largest and most central polyp is the parent polyp of the colony, and that the ccenenchyma stretching from it to the epitheca in which the other polyps open is, or more correctly was, before the other polyps appeared, its thick porous wall. * Ann. & Mag. Nat. Hist. (5) xiii. p. 284. + Journ. Linn. Soc. vol. xviii. } Phil. Trans. vol. clxxxvii. (1896). § Zool. Jahrb, vols, iii. and iv. (syst.). 120 Mr. H. M. Bernard on the It will, perhaps, be remembered that I found just such a young colony of Astrwopora, and, further, I concluded that the morphological differences between Turbinaria and Madre- pora could be best explained by postulating such a young colony in each case; in Madrepora the central polyp grew up with tiers of daughters springing from its side, while in Turbinaria the ring of daughters shot ahead and formed a cup. We are then, I think, justified in deducing Montipora from a parent polyp opening in a mass of spongy ccenenchyma, @. e. with a very thick porous wall, contained in a saucer-like epitheca. In JMontipora, unlike the Madreporide just men- tioned, the parent polyp formed no projecting cone, but opened level with the upper flattened surface of its thick wall. Similarly the young polyps, opening laterally, do not form protuberant cones; they appear as mere openings in the ceenenchyma, often in contact with the epitheca, in which case the latter may form part of their outer walls. The ccoenenchyma in the young specimen actually examined consists, as seen from the surface, of jagged flakes, which may be twisted in all directions, but which tend to lie horizontally ; the apertures of the polyp-cavities are bounded by the edges of two or more such flakes: in this particular instance the ccenenchyma has been secondarily specialized. From the edges of the flakes points project into the polyp-cavity and form vertical series of spines. ‘These series of spines are, in many types, the only remains of the septal apparatus. Examination of all the types, however, shows that the septa were originally lamellate, as in the other Madreporide. The large directives are often continuously laminate, and here and there some of the other primaries also; while, again, what appear to be traces of laminate coste can be found in the coenenchyma of very many types. ‘lo these important points, as also to the cause of the widespread degeneration of the septal apparatus, we shall return. Starting, then, from such a young colony, with its parent polyp surrounded by a ring of daughters all immersed in a mass of spongy reticulum contained in an epithecal saucer, the stock may develop along different lines. But while in the other Madreporide the calicles lead and the canenchyma fills up the interstices, in Montipora the caenen- chyma takes the lead in the formation of the corallum. I am aware that these expressions may, at first sight, appear very loose; they are, however, sufficiently useful to render their employment justifiable, provided we are quite clear as to what they really mean. We must clearly recognize that the Madreporarian Subfamily Montiporine. 121 term ccenenchyma applies, in the Madreporide, to the fused orous walls of the individual polyps forming the colony. i using the expressions “ the calicles lead”’ and “ the ccenen- chyma fills up the interstices,” what is really meant is that the walls of the individual polyps are distinguishable as such above the level of fusion; where the walls fuse together to form the ccenenchyma they cease to be distinguishable. While, then, in the other Madreporide the walls of individual polyps are typically recognizable in so far as they keep above the level of their fusion, in the genus Montipora as soon as the parent polyp has budded to forma colony, no matter how small, we can no longer speak of any porous walls except theoretically, for, fusing right up to the level of their aper- tures, they together form an expanding mass of coenenchyma. It was this last-mentioned fact that struck Ridley as presenting such a contrast to the method of growth in Madrepora; but the true explanation of the difference is not to be found in his “centrifugal” and “ centripetal’ methods of budding, for it is obvious that any generalization affecting a genus which leaves all the more primitive explanate growths out of the reckoning must be unreliable. The true significance of these comparisons will be still further discussed in the section on the affinities of the genus, as will also the fact that the coenenchymatous edge of the rim of the cup in Turbinaria is closely comparable with the coenenchymatous edge of a foliate Montipore. In view, then, of this great development of the coenenchyma and of its prime importance in building up the Montiporan coralla, we have, it seems, no choice but to utilize the variations presented by the ccenenchyma as the basis of our classification. In so doing we emphasize the fact that the Montipore are ccenenchymatouscorals par excellence, that, whereas the coenen- chyma of the Madreporide is primarily merely the tissue arising by the more or less limited fusion of the porous walls of adjoining polyps to form a mutual support, in Montipora it ismore than this. Resulting from the complete fusion of the walls, it has in many cases taken on other functions as well as that of a supporting and cementing tissue, for, rising above the level of the polyp-cavities, it is specialized in various ways for their protection. A study of the variations which the ccenenchyma presents supports this assumption of its taxonomic importance. The specimens admit of being divided in the most natural way according to the specialization of the coenenchyma. We also have the additional satisfaction of finding that transition forms reveal the lines along which the leading specializations 122 Mr. H. M. Bernard on the have travelled. Hence, although one would never have selected a tissue like the coenenchyma, which all experience shows to be dangerously variable, as a basis for classification, there is in the present case really no choice, ¢. e. if our mor- phological diagnosis is correct. Hitherto the variations of the surface ccenenchyma—very superficially handled—have been accorded only a secondary place. Dana, deducing Monti- pora (Manopora) from Madrepora by the degeneration of the calicles, classified its species accordingly into those in which the protuberant calicles persisted and those in which they had quite disappeared, the latter group being further subdivided according to the form of the corallum and the character of the surface. Milne-Edwards and Haime divided the Monti- pore primarily according to the form of the corallum. We may at once dismiss this latter classification as purely artificial. Returning, however, to Dana, it must be noted that there is no evidence whatever to make us believe that Monti- pora is deducible from Madrepora by gradual degeneration of protuberant calicles. The only Montiporan forms which Dana adduced as transitional hardly support his contention : one—M. gemmulata—has been removed by Verrill to the Turbinarians, while the protuberant calicles in the other— M. caliculata—are not true calicles in Dana’s sense, but a peculiar specialization of the interstitial coenenchyma which will be referred to again. We have, then, no choice but to accept the variations in that tissue, the specializations of which are essentially the peculiarity of the genus, as the basis of classification. Be- neath all its baffling superficial variations the laws of its growth can be made out and the main lines along which it has diverged can be traced. ‘This serves to divide the genus into groups which have some claim to be natural. Uncertainty, however, comes in when, in further subdividing these groups, we come within range of the superficial varia- bility due to accidents of position and nutrition. It must therefore be at once confessed that many of the assumed specific variations are not to be relied upon. ‘The “ species ” established are in many cases only descriptions of individual specimens the surface characters of which give no clue as to their affinities with other specimens. Of course in many cases there are other characters sufficiently striking to justify us in confidently claiming new and distinct types. The following analysis of the development of the coenen- chyma was only very gradually arrived at after studying series of sections revealed by fractured specimens. It will be best understood if we reverse the process of its discovery, Madreporarian Subfamily Montiporine. 123 and, assuming our conclusions to be correct, start from the hypothetical parent-polyp of the genus. This polyp, as we have seen, differed from that of Madrepora, Turbinaria, and Astreopora in that the porous theca did not rise up into a cone, but was low and thick, filling up, but probably not much overtopping, the rim of the epithecal saucer. Primarily the porous walls consisted of radial lamine joined together by synapticule *, so that the ecenenchyma surrounding this parent polyp may be said to radiate outwards towards the edge of the epitheca. When the first ring of buds appeared just within the edge of the epitheca their walls would also radiate outwards, either con- tinuing to be supported by an extension of the primitive epitheca or shooting out freely beyond its edge. From this initial stage in the development of Montipora we should expect to find the coenenchyma consisting of laminate plates standing at right angles to the epitheca and radiating out- wards on all sides. We should expect to find this because the coenenchyma is nothing but the resultant of fusion of the porous walls of the component polyps, and their laminate coste would necessarily be arranged in the manner described. Now this initial stage in the growth of the Montiporan corallum is traceable in almost every type. A surface of fracture through any explanate Montiporan will almost inva- riably reveal a thin basal layer streaming outwards towards the growing edge. While this basal ‘ streaming layer” is of fairly uniform thickness and the direction of its fibres is always outwards—7. e. in the line of growth—its texture may vary. (1) It may be composed of ribbon-like bands running outwards more or less at right angles to the epitheca, but so united as to form a system of flat canals apparently freely communicating with one another. ‘This laminate reti- culum, occurring as it invariably does in the ‘streaming layer,” may, | think, safely be regarded as a vestige of the primitive laminate coste which were once the most important element in the calicle walls. (2) The primitive band-reticulum may, owing to the extensive perforation of the amine, have lost this character and have become a filamentous reticulum. In this case also the direction of the threads is typically very pronounced, streaming outwards towards the growing edge. There are, lastly, a few cases in which the reticulum shows no special streaming; these would appear to have been secondarily modified. Here it should be remarked that only * T use this term generally for all outgrowths from the faces of septa for their mutual support, irrespective of their form and position. 124 Mr. H. M. Bernard on the fractures in the direction of growth at any time show the appearance here called streaming; all sections across the streaming show what might be called, in contradistinction, a “* stationary ” reticulum. In the figures 3 and 4d (PI. II.) the artist—Mr. Percy Highly—has well shown by dots the usual appearance of the streaming layer. If these dots are thought of as pores in radial (septo-costal) plates lying in the plane of the paper, the reader will get a fair idea of what I believe to have been the origin of the streaming layer. The relative thickness of the streaming and of the thickening layer in figs. 3 a, 36, 3¢ should be reversed. ‘The thinness of the streaming layer is probably indicative of the very early budding of the polyps. We have, then, in all Montiporans (with a few secondary modifications) a basal layer of reticulum streaming outwards (no matter how large or small the colony may be) and forming the growing edge, with or without the supporting epitheca. This streaming layer, which expands the corallum, for some reason or other cannot, as such, thicken it, and the new formation of ccenenchyma for this purpose is in most cases sharply marked off from the streaming layer. An explanation of this thickening layer, shown in figures 3a, 36, 3c, as compared with 4 d, will be suggested later on. The thickening layer, wherever the epitheca accompanies the growing edge, is confined to the upper surface ; but if the growing edge is free, thickening layers may be added to both upper and under surfaces. ‘These layers are derived from the threads or jagged edges of the surfaces of the streaming layer. On the upper surface they grow upwards at right angles to the direction of the streaming, and, uniting among themselves, form a filamentous reticulum. Similarly a layer of reticulum developed from points of the streaming layer bent down at right angles to that layer may cover the under surface. Whereas the upper layer may develop to almost any thick- ness and give rise to a very great variety of beautiful surfaces, the lower layer seldom thickens much; the individual threads soon tend to thicken, and thus to form a very dense reticu- lum, and sooner or later the epitheca grows out, covering over the calicles and leading to the more or less complete solidification of the lower surface. This description of the thickening and solidification of the lower surface does not apply to erect leaves or to branches which may be regarded as thick rounded leaves. In these cases the thickening layer may develop evenly on both sides of the leaf or all round the branches, and show all the surface specializations which in the horizontally growing specimens Madreporarian Subfamily Montiporine. 125 are confined to the upper surface. In the case of the branched gai the tips of the branches which correspond with the growing edge of the explanate forms consist entirely of the streaming layer, and this may be seen forming the axis ofallthestems. ‘The cortical layer, which gradually thickens the branches, can in most cases be seen to be formed of threads bending outwards at right angles to the direction of the axial streaming layer. Leaving, then, the primitive basal streaming layer, we have to consider the variations presented by the upper thick- ening layer just described :— (1) It may merely thicken the streaming layer gradually as a filamentous reticulum, the surface in which the calicles open remaining all the time smooth. In reference to the level surface, 1 have called this group ‘ glabrous,” and under it I have arranged some thirty different types (fig. 3a). (2) The thickening reticulum may grow faster than the ealicles, causing the interstices to swell up into ramparts surrounding pits, in the bases of which the calicles open. I have called this the “ foveolate” group, after the most extreme type, M/. foveolata of Dana. Between these two come specimens which are foveolate while in rapid growth, but eventually become smooth ; these I have called “ glabro-foveolate.” I have found some twelve foveolate types and five glabro-foveolate. One specialization of these ramparts has a curious resem- blance to true protuberant calicles. This appears to have misled Dana in his ascription of true calicles to IW. caliculata. (3) The thickening reticulum shoots up into papille which rise up above the general surface. There are several more or less distinct variations of the “papillate”’ specialization, which at the moment of writing is shown by at least thirty- three types (fig. 30). The leading differences are as follows:—(a) reticular uprisings froth up the interstices over irregular patches of different sizes; (4) the papille are always in some relation to ealicles, forming hoods or mounds, on the outer faces of which calicles open; (c) the papille run together to form either nearly parallel series in the direction of growth, or else more or less gyrating ridges; (d) lastly, as the extreme type, the papillz rise as nipple- or nearly symmetrically dome-shaped processes scattered more or less thickly over the surface, but not arranged in radial series (fig. 30). (4) ‘The thickening reticulum undergoes a change in its texture ; the threads which bend up vertically become differen- tiated from the rest of the elements of the reticulum and 126 Mr. H. M. Bernard on the become stout solid trabeculae. The rest of the reticulum merely forms the cross pieces which support these trabecule. Every stage in the gradual differentiation of these trabecule. can be traced. In many cases the more vertical elements of the thickening reticulum run in nearly straight lines, but without thickening. Comparison of specimens shows that the thickening was due to the rising up of the tips of these vertical threads above the surface, perhaps at first as echinule. These became stouter and stronger, probably for protective purposes, and thus, as they sank beneath the rising surface, became thick trabeculee (fig. 3c). This group, showing the rising of stout trabecule above the surface to form protective ‘“ tubercles,” is very large and contains more than forty types. The distribution and shapes of the tubercles are very varied: they may be densely crowded as minute rounded granules or tall and lancet- shaped; they may be grouped in rings round calicles, or, again, they may run together to form thin keels or ridges. This group is called the “‘ tubereulate ” group. We thus have four main divisions of the genus—glabrous, foveolate, papillate, and tuberculate—each term having refer- ence solely to a peculiar specialization of the ccenenchyma, While the first three of these terms need no comment, the last requires justification. In all the earlier descriptions of Montiporan types the terms papillae and tubercle seem to have been used indiscriminately. It is often impossible to tell whether a writer was describing a specimen belonging to group 3 or to group 4. The most important use of the word tubercle occurs in Lamarck’s description of the specimen Porttes tuberculosa, Lk. (= Monti- pora tuberculosa). In fixing the use of the word tubercles to mean the small solid tips of individual trabecule: when they project above the surface, I have been led to do so by the conviction that these were Lamarck’s “ tubercles”’ as seen on his type “ tuberculosa.” Certain expressions in Lamarck’s text point clearly to this. In describing P. tuberculosa* he speaks of “les tubercules dont la surface est parsemée” as being “ graniformes ou columniformes ;”’ and, again, on the next page he speaks of interstices being “ hérissés de tuber- cules.” Both these expressions are quite inapplicable to the much larger swollen reticular knobs here called papilla. One other remark on these tubercles with their trabecula- like sunken portions. It was the presence of these trabecule which appears to have misled Milne-Edwards. He compared * ¢Animaux sans Vertébres,’ ii. 1816, p. 272. Madreporarian Subfamily Montiporine. 127 them with the entirely different trabecule of Porites, and accordingly placed Montipora among the Poritide. So far as my own observation goes, the two are morphologically distinct ; the trabecule in Porites are primitive structures, rising straight up from the epitheca, while in MJontipora they are, as we have seen, quite secondary. Turning from the ccenenchyma to the polyps, polyp- cavities, method of budding, and to the forms of the corallum, points which are, as a rule, of prime importance, we find that the special development of the ccenenchyma has, as it were, overshadowed them. With regard to the last-named, we find all the typical methods of growth in each of the four chief divisions based upon the specializations of the coenenchyma. The polyps themselves are minute and their tentacles are little more than papille or crenulations of the edge of the oral disk. The polyp-cavities are also very small and the septal apparatus as a rule degenerated into mere vertical rows of projecting spines. The largest or directive septum with a tew of the larger primaries may be more or less interruptedly laminate. I look upon these as survivals of a primitive lami- nate condition of the septa and coste in the thick porous walls. The cause of this degeneration of polyps and septal apparatus may perhaps be correlated with the great development of the ccenenchyma, the production of which must be a strain on the resources of the living organism, leading to the fixation of the polyp at a very undeveloped stage. In contradistinction to this extreme we may cite in support of our suggestion the cases of the Alcyonarta and Actinia, in which the polyps reach a very high level of development, while the skeletal matter deposited is either scanty or altogether absent. The character of the budding, as also a few further points on the degeneration of the septal apparatus, will be referred to in the concluding section on the interrelationships of the Madreporide. In that connexion such matters can be more advantageously discussed comparatively. We shall there also summarize the description of the genus above given, and in that way emphasize the arguments in favour of classing Montipora with the Madreporide. The Genus Anacropora. This genus, founded by Ridley in 1884 (/. c.) to contain a branched coral from Keeling Island, was said to be distin- guished from Madrepora by the method of budding and from Montipora by its protuberant calicles, The method of budding in Madrepora, in which smaller 128 Mr. H. M. Bernard on the daughters appear from the sides of larger parent (“ apical’) polyps, was described as centrifugal, while that in Anacro- pora, in which an apex ‘of undifferentiated coenenchyma takes the lead and the young polyps appear in it as it grows, was ealled centripetal. The distinction was thought to be funda- mental. On the other hand, the new genus came very near Montipora, differing from it chiefly in the fact that the calicles in Montipora are typically immersed, while in Anacropora they bulge up the surfaces of the branches into mounds or eminences. The new genus was accepted at once by Duncan in his revision of Milne-Edwards and Haime’s system, and he allied it with Montipora. The ‘ Challenger’ expedition brought home two new types, which Quelch classed under Ridley’s genus, and in 1892 Rehberg * added another specimen and type, bringing the number up to four. The following notes are based upon the study of the specimens and fragments (twenty-two in all) in the National Collection. These include all the existing types except that of Rehberg (A. spinosa), which is in the Ham- burg Museum. The examination has resulted in the establishment of two new types, one being represented only by fragments, the bulk of the specimens being in the Vienna Museum. Full details will appear in the official catalogue, which is in the press. I was for some time quite uncertain as to the validity of the distinction made by Ridley between Anacropora and Montipora. Slight mounds or elevations on which the calicles opened might and do, indeed, occur in Montipora, wherever the corallum is very thin, while, on the other hand, we have in Anacropora the streaming axial layer leading the growth, and forming, as in Montipora, the tips of branches, and a further cortical layer formed just as in Montipora. It seemed to me, therefore, that while the fundamental identity in the structure of the colonial skeleton showed that Anacropore were really Montipores, the presence of protuberant calicles, which might be a slight return to primitive conditions, hardly justified the establishing of a new genus. Comparison with other types and with the undescribed material in the collec- tion has, however, revealed other characters which are important enough to warrant our retaining the genus, but uniting it with Montipora under a subfamily Montiporine. While, then, the fundamental identity in the structure of the ccoenenchyma shows that Anacropora has branched off * Abh. Nat. Ver. Hamb. xii. p. 46. Madreporarian Subfamily Montiporine. 129 from Montipora, we may assume that the protuberant calicles, which may, in some cases, be even tall and conical, suggest that this branching off took place at a very early stage in the development of that genus. ‘That these protuberances are primitive, and not secondary returns to primitive conditions, may, perhaps, be gathered from the very important fact that the primary septa in the more protuberant calicles are laminate, and, further, that these laminate radial structures may even project down the outer wall of the protuberance as costal ridges (PI. II. fig. 5). It is specially worthy of note that the less protuberant calicles, or those which open flush with the surface, have the degenerated septal apparatus characteristic of Montipora, while those which grow taller and _ slightly larger develop radial skeletal laminz, septa and costae. While it is of course quite possible that this is a secondary return to primitive conditions, there is no reason why we should not assume it to be the persistence of such conditions. The burden of proof, 1 think, rests with those who prefer the former suggestion. I have been much struck by noting that many of the pro- tuberant calicles with costal ridges running down their sides show the tendency to a spiral twisting of the whole calicle which I have already referred to in Zurbinaria and Madre- pora. ‘This fact, again, seems to me to suggest that the protuberance of the calicles is primitive and not atavistic. Hence, then, we conclude that Anacropora branched off from Montipora before the degeneration of the calicles and of their laminate radial skeleton had gone as far as it now has in the latter genus. In this connexion it is worth noting that the axial streaming layer is typically laminate or band-like, and that, in those cases In which it appears most filamentous, examination shows that this is a secondary condition due to the formation of large perforations in the primitive longitudinal bands. This band-reticulum, as we have seen above, can be best traced to the outward streaming of the primitive laminate radial structures composing the chief portion of the thick walls of the parent and daughter polyps in the earlier stages of colony formation. In addition to this important laminate structure of the walls of the more protuberant calicles, the method of branching is quite peculiar, All the known types are composed of rather thin cylindrical stems more or less knotted (by the protuberant calicles) like a thorn-stick. While the stems are generally slightly curved, the branches come off suddenly at rather wide angles, the stem at the same time bending 130 Mr. H. M. Bernard on the away from the branch. It is, in reality, a kind of forking, only the stem remains the more important and less diverging prong. ‘The result of repeated branchings with free fusions between parts that touch is to form a rather closely matted tangle low down near the ground, the meshes in the tangle being more or less angular. This angular character of the meshes is, however, frequently obscured by curvings of the branches. Broken fragments falling down into the tangle freely fuse on again, and help to make the net thicker. In claiming this very peculiar method of growth as characteristic of the genus I am aware that it is not immediately evident in all the types. It is very marked in Ridley’s original type (A. Forbest), in Quelch’s types (A. gracilis and A. solida), and in one of the new types (A. echinulata *), whereas it is not so marked though traceable in A. erecta *, and apparently least visible in Rehberg’s type (A. spinosa). In these last two forms the branching does not come off at such a wide angle, and hence the whole corallum is more symmetrically arborescent. But in A. erecta, so far as I remember the photographs shown me by Dr. Marenzeller, the larger clumps were very close tangles of thin knotted stems, and Rehberg’s figure of A. spinosa (1. c.) appears to show distinct traces of a tendency to sudden angular bendings of the stems and branches. These points, then, the protuberant calicles, showing distinct lamination of their radial structures, and the peculiar character of the branching, serve, I think, to separate Anacro- pora from Montipora, with which genus it is, however, funda- mentally associated in the structure of the ccenenchyma and in the presence of calicles with degenerate septal apparatus exactly like those of Montipora. Interrelationships of the Madreporide. As we have above seen, the only argument for allying Montipora with Porites, as was done by Milne-Edwards and Haime, and later by Duncan, falls to the ground as soon as the secondary character of the trabecule is established +. Hence we have no hesitation in claiming the genus with its ally Anacropora as together forming a subfamily of the Madreporide. I shall now endeavour to show that the remaining three accepted genera—Madrepora, Turbinaria, * Full descriptions of these are given in the Museum Catalogue, + In 1889 Dr. Ortmann suggested, without going into details, that Montipora might be deduced from Porites through Alveopora (Zool, Jahrb. (syst.) iv. p, 584). Interrelationships of the Madreporide. 131 and Astrwopora—can be usefully united in a second sub- family ; so that, for the future, the Madreporide will consist, so far as we at present know, of two subfamilies—the Madre- poring, comprising three genera, and the Montiporine, comprising two genera. The strongest argument in favour of this classification lies in the fact that the five genera can be deduced from a common ancestral form. In describing this form we are, for obvious reasons, confined to a consideration of its parent polyp, and not of its colony. Every colony starts from a parent polyp, and, indeed, receives its chief characteristic from the structure, growth, and method of budding of this individual, directly developed from the attached larva. Hence it is enough if we can trace any group of colony formations back to a common ancestral parent polyp. Reference to the analyses already given in this and in the earlier papers on Turbinaria and Astreopora shows that this common parent polyp possessed the following leading charac- teristics :—(1) a porous wall, with laminate radial structures ; (2) a well-developed saucer-shaped epitheca; (3) the habit of very early budding while the parent polyp was still very small; (4) the production of true buds, starting from the smallest beginnings out of the sides of the polyp, and forming their skeletons, at least in the first stages, upon and with some slight modification of the radial symmetry of the porous wall of the parent polyp*. From such a form we may deduce the genera under dis- cussion along the following lines of specialization :— Madrepora.—The skeleton of the parent polyp grew in height, and consequently somewhat in size, shooting upwards in a tall cone with thickening base (fig.4a). The buds grewout in tiers from its sides, remaining comparatively small. The radial structures persist as lamine, and those septa of the buds would be largest which could start at once upon, and in the same plane with, one of the radial laminate structures (costee) of the parent; hence the “ directive” septa of the buds are typically radially symmetrical with those of the parent. The epitheca is left behind. Lurbinaria.—A ring of buds shoots up round and from the sides of the parent polyp, together forming acup, the wall of each bud rising up as a distinct cone above the level of the fusion of their walls to form the common ccenenchyma (fig. 46). * For Miss Ogilvie’s alternative derivation of the Madreporide see Phil. ‘Trans. vol. clxxxvii., 1696. This has been criticized by me in the Geological Mag. vol. iv. 1897, p, 170. 132 Mr. H. M. Bernard on the The parent polyp dies away, and its primitive protuberant cone is immersed under the ccenenchyma formed from the fusion of the walls of a ring of daughters. These daughters carry on the colony, the budding of the daughters being limited to their free or outer sides, 7. e. to the sides turned away from the axis of the cup. Hence the fact referred to above, that in Turb¢énaria as well as in Montipora the young buds appear in the undifferentiated coenenchyma which forms the growing edge of the cup. This edge represents morpho- logically the outer sides of the combined porous walls of the last-formed ring of polyps, and differs from the porous wall of the parent polyp mainly in the facts, (1) that the laminate radial structures are more or less obscured, and (2) that the epitheca has been left behind. The polyps forming the Turbinarian colony develop equally, and there is no such disparity in size as is seen between the axial polyp of Madre- pora and its daughters. Principal or directive septa occur and can be accounted for in the same way as in Madrepora. Astreopora.—The budding is promiscuous; a new bud develops wherever there is room for it, each one typically carrying up its wall into a protuberant cone (fig. 4¢). Asa result of this crowding the known forms are, without exception, thick encrusting, or massive. The costal radial structures of the original parent ceased to be laminate, but broke up into radial series of spines, the tips of which formed protective echinule. One apparently natural consequence of this was a considerable degeneration of the septal apparatus in the daughters of the colony. Montiporine.—The original parent polyp was distinguished by great thickness of its porous walls, which apparently early arrested the development of the polyp, and by a tendency of the whole skeleton to be low, and even perhaps disk-like, and not to rise up into a cone as in the last three genera (fig. 4d). In the modern Montipores this has reached its extreme limit, but in Anacropora the habit of forming conical walls is not yet lost. The synapticular con- nexions between the radial structures reached far in towards the centre, so that the visible septal apparatus tended to be limited to rows of septal spines; when the calicles protrude (Anacropora), and hence grow a little in size, laminate septa appear. The tendency to enormous thickness of porous wall was inherited by the daughter polyps. Hence the two chief characteristics of the genus—(1) minuteness of the polyp-cavities, (2) great richness of coenenchyma, which is nothing but the result of fusion of the greatly thickened porous walls of the individuals of the colony. The budding Interrelationships of the Madreporide. 133 of the daughters seems, as a rule, to be limited to their free or outer sides ; the fresh buds turn upwards if the growing — is accompanied by an epitheca, but may turn up or down indifferently if the growing edge is free. In the diagrams the former case is, for the sake of simplicity, alone illustrated. In Montipora we have almost all possible growth-formations resulting from this aggregation of small thick-walled polyps, aided by the secondary additions of tissue, above described as the “thickening layers.” These begin to form at varying distances from the growing edges or apices, 7. e. after the budding of the polyps has ceased. May not this fresh ee be correlated with the very early budding of the ontiporan polyps and their subsequent continued but limited growth—limited, that is, by the abundant secretion of skeletal matter—which is the characteristic feature of the genus ? In Anacropora the growth-form is highly specialized. We may thus look upon Anacropora as a survival of a special growth-form of some more primitive Montipore, 7. e. of some Montipore in which the degeneration of the protuberant conical wall had not gone so far as it has in the modern representatives of the genus. While in Montipora the lami- nate radial elements of the calyx have almost entirely disap- peared, being only occasionally found ina few large primaries, directives and others, and, again, in the streaming layer of the coenenchyma, in Anacropora laminate septa and coste appear in the more protuberant calicles in addition to the lamination of the streaming axial layer. It is further worth noting that not only does the occasional presence of laminate directives support the deduction of Montipora from an ances- tral polyp with laminate radial skeleton, but the mere presence of directives points also that way, that is, if the explanation of the rise of directives above given is correct. The primi- tive epitheca, which is lost in Anacropora, persists and plays a great part in the formation of many Montiporan coralla. In these different ways all the genera which are at present included in the Madreporidz can be deduced from a common parent. The two last mentioned are associated by the pecu- liar structure of the ccenenchyma, which, as we have seen, is traceable to the great thickness of the porous walls of the individual polyps. These, then, form the subfamily Monti- porine. The remaining three genera are also united by one character in common, viz. the typical upgrowth of the polyp- walls into freely protuberant calicles, their basal portions alone being fused together to form a ccenenchyma. I can see no reason why this character should not unite Madrepora, Turbinaria, and Astreopora into a second subfamily—the Ann. & Mag. N. Hist. Ser. 6. Vol. xx. 10 134 = On the Interrelationships of the Madreporide. Madreporine. The chief objection to this rests in the specialized character of the laminate radial structures of Astreopora, which ought, perhaps, to separate that genus from Madrepora and Turbinaria. In the meantime, how- ever, they can be usefully united in the manner suggested. I therefore suggest the following arrangement of the family :— Family Madreporide. ae vf SS