INTRODUCTION. 23 



groups among the Cyclostomata is wasted. But, so far as I am. 

 able to judge, this has not happened so frequently as might be 

 expected : an occasional specimen of a Berenicea from the Miocene 

 may be so much like a Jurassic species, that it is impossible to 

 separate them by a verbal diagnosis ; but I have never met with 

 any case in which a series of Miocene specimens could not be easily 

 recognized as different from a Jurassic series. 



2. The Comparison of Equivalent Zooecia. In estimating the vari- 

 ations we must remember that the Bryozoa are colonial, and that 

 colonial animals have always been recognized as especially difficult 

 of specific diagnosis. Thus, most recent workers on the corals 

 admit that different parts of the same mass vary so greatly, as 

 to throw grave doubts on the use of defining species or genera. 

 The same difficulty is experienced with animals which, like the 

 Pteropods, live in vast shoals, subject throughout to precisely the 

 same conditions. Similarly with the Bryozoa; the colonies are 

 constituted of swarms of zooecia, most of which live under almost 

 identical conditions. Thus, one specimen of Theonoa bowerbanki, 

 Haime, in the British Museum Collection, consists of over 17,000 

 zooecia. But a Bryozoan colony not only includes a shoal of 

 individuals, but zooecia in many different stages of development. 

 It is impossible, therefore, to compile a diagnosis which shall be 

 equally true for every member of so extensive a series of zooacia, 

 including individuals of all ages, and some which have been 

 dwarfed or aborted by overcrowding in the zoarium. 



Barrois, 1 when advocating the taxonomic value of the larval 

 stages in the Bryozoa, remarked that " il me semble en effet tout- 

 a-fait indispensable, pour des animaux SL caracteres si difficiles 

 a saisir que les Bryozoaires, de tenir compte en meme temps de 

 toutes les formes a la fois. Tine classification basee sur la seule 

 forme des zooeciums est encore, quelle que parfaite qu'elle soit, tres- 

 insuffisante." The principle of Barrois's argument is sound, but 

 a system which would necessitate the inclusion of larval characters 

 in diagnoses is too complex for general use. It is sufficient, 

 however, to compare equivalent zocecia in a colony; just as in 

 determining worms, echinoids, or amphibians, the comparisons are 

 made between equivalent setae, plates, or stages. 



1 J. Barrois. Recherches sur 1'Embryologie des Bryozoaires, p. 251. Lille, 

 1877. 



