BRYOZOA 



fauna as distinct as is the Mediterranean. This exaggerated use of the term raises a difficultv, 

 when we wish to compare the Antarctic fauna, as the narae has already been extensively 

 emplo}'ed for something quite différent, which we must now speak of as a Subantarctic 

 fauna. 



Classification 



Since I commenced the examination of this collection, made by the Belgica, three 

 important works hâve appeared, of which the first was Calvet's « Bryozoaires Ectoproctes 

 marins », igoo, giving résumés of our knowledge of the structure of Biyozoa, and adding 

 some results of the author's own work. Then S. Harmer (') has shown the importance 

 of the compensation sac, and in G. M. R. Levinsen's « Studies on Bryozoa ( 2 ) » the charac- 

 ters used in classification receive attention, especiall)- the front wall, the ovicells, the rosette 

 plates and pores; and with the more careful examination of a number of minute characters 

 we may be sure, that the next few years will see some important modifications in oui- 

 classification ( 3 ). We look to see how Levinsen will carry out his ideas, for he thinks, 

 that the structural features of the rosette plates and the ovicells are most to be relied upon, 

 besides which he considers, what he calls the « oral bow » of great classificatory value. He 

 would make one of the leading characters the pores of the rosette plates, namely the main 

 divisions would be based on there being one or mari)' pores in the rosette plates ; but although 

 I was the first to call attention to the value of the rosette plates in classification, I am 

 not prepared to go as far as Levinsen, until he shows us how he carnes out his ideas. 

 li the number of pores is such a leading character we should expect, that the rosette plates 

 oi the distal and latéral walls would be of the same character, whereas we may find one- 

 pored rosette plates on the one wall, and rosette plates with numerous pores on the other. 

 For example in Rctepora lata B. we hâve a distal rosette plate with several pores, whereas 

 the latéral plates hâve only one pore, and there are similar cases in the Flustridae. 



As the resuit of many years examination of the opercula ( 4 ), and the préparation of 

 the chitinous appendages of not far short of iooo species I am more and more convinced, 

 that no character will give as much help in the détermination of the Bryozoa, and in making 

 the classification more natural. This it will do, by enabling us to see in smaller groups, 

 where the relationship is close, and thus to form larger groups. 



I hâve more than once said that Schizoporella, Lepralia and Ccllcpora require revision, 

 and hâve no doubt that the studv of the opercula will give much assistance in doing so, 

 but it is not only the shape that must be examined, but also the muscular attachments, 



(i) On the Morphology of the Cheilostomata. {Quart. Journ. Micr. Se, vol. 46, n.s., p. 263-35o, pi. XV-XVIII.) 



(2) Vidensk. Medd. fra den Naturh. Forai, i Kjobenhavn, 1902. 



(3) A valuable paper « Notes on the Natural History of East Finmark .1 in the Aimais and Magazine of 

 Natural History, s. 7., vol. XI and XII, by Canon A. M. Norman has appeared too late for me to make use 

 of it for the présent memoir. It contains many important suggestions as to classification 



Julliex and Calvet, Bryozoaires provenant des campagnes de I'Hiroxdelle. (Camp, scientifique du Prince 

 de Monaco, fasc. XXIII) was received just before the manuscript was sent off. 



(4) The Use of the Opercula in the Détermination of the Cheilostomatous Bryozoa. (Proc. Manchester 

 Lit. and Phil. Soc, vol. iS, pi. I, 1878.) 



