TRANSACTIONS OF SECTION D. 751 



each bottle containing a specimen the depth from which the net in which the 

 specimen was found had been drawn up. In many instances, from the nature of 

 the specimen, it is impossible that it can have come from anywhere but the 

 bottom, but in many others it is quite possible that a particular specimen may have 

 entered the net at any intermediate depth, or close to the surface, and this is a 

 matter on which the author of the monograph in which the specimen is described 

 can form the best conclusion, if one can be formed at all from his knowledge of 

 the animal itself. In all doubtful cases the mere record of the depth must ba 

 received with caution. 



Just as before the commencement of the present period of deep-sea research 

 there was a strong tendency amongst naturalists, owing to the influence of the 

 views of Edward Forbes, to refuse to accept the clearest evidences of the existence 

 of starfish and other animal life on the sea-bottom at great depths, so there seems 

 now to have sprung up in certain quarters an opposite tendency, leading to the 

 assignment of animals possibly of surface origin to great depths on inconclusive 

 evidence. 



With regard to the constitution of the deep-sea fauna, one of its most 

 remarkable features is the general absence from it of Palaeozoic forms, excepting 

 so far as representatives of the Mollusca and Brachiopoda are concerned, and it is 

 remarkable that amongst the deep-sea mollusca no representatives of the Nautildce 

 and Ammonitidce, so excessively abundant in ancient periods, occur, and that Lingula, 

 the most ancient Brachiopod, should occur in shallow water only. 



There are no representatives of the most characteristic of the Palaeozoic corals, 

 such as Zaphrentis, Cystiphyllum, Stauria or GoniophyUum. Possible representa- 

 tives of the CyathoncmJd<B have indeed been obtained in Guynia, described by 

 Professor Martin Duncan, and Haplophyllia and Duncania, described by the 

 late Count Pourtales, but the Cyathona.vidce are the least aberrant and characteristic 

 members of so-called Rugom. Pourtales justly felt doubtful whether the arrange- 

 ment of the septa in four systems instead of six could in itself be considered as a 

 criterion of the Rugosa, 1 and in the cases of Haplophyllia and Duncania the 

 septa may be described rather as devoid of any definite numerical arrangement 

 than exhibiting any tetrameral grouping. Further, I have lately examined by 

 means of sections the structure of the soft parts of Duncania in a specimen 

 kindly given to me by Mr. Alexander Agassiz for the purpose, and find that with 

 regard to the peculiar arrangement of the longitudinal septal muscles and the 

 demarcation of the directive septa the coral agrees essentially with the Hexactinian 

 Caryophyllia and all other modern Madreporaria the anatomy of which has been 

 adequately investigated. 



There are further no representatives of the ancient Alcyonarians, forming 

 massive coralla, the ffeliopoi-idte and their allies, in deep water, no Palcsocrinoids, 

 Oystidea, or Blastoidea, no Palechinoidea, no Trilobites, no allies of Limvlus, no 

 Ganoids. Further, other ancestral forms, certainly of great antiquity, although 

 unrecorded geologically, such as Amphiaxus, do not occur in deep water. It might 

 well have been expected that, had the deep sea been fully colonised in the Palaeozoic 

 period, a considerable series of representative forms of that age might have 

 survived there in the absence of most of the active physical agents of modification 

 which characterise the coast regions. 



From the results of present deep-sea research it appears that almost all modern 

 littoral forms are capable of adapting themselves to the conditions of deep-sea 

 life, and there is no reason why Palaeozoic forms should not have done so if the 

 abyssal conditions were similar to those now existing, just as a considerable number 

 of forms of the chalk period have survived there. In fact, however, most of 

 survivals of very ancient forms — Heliopora, Limuhis, Amphioxus, Dipnoi, Ganoids- 

 occur in shallow seas or fresh water. 



With regard to the origin of the deep-sea fauna, there can be little doubt that 

 it has been derived almost entirely from the littoral fauna, which also must have 



1 ' Zoological Kesults of the Hassler Expedition.' See Cat Jfits. Conip. Zool. Har- 

 vard, No. viii. 1874, p. H. 



