1868.] DR. J. S. BOWERBANK ON SPONGES. 1 19 



all adopted the system of division and subdivision of these numerous 

 and protean animals by means of their chemical constituents and 

 their external forms ; and the results of their attempts at classification 

 by such means have universally been unsuccessful in leading stu- 

 dents to a ready recognition, or, indeed, to scarcely any recognition at 

 all, of the species described by the authors who have had recourse to 

 such systems. The natural result has been that every naturalist who 

 has attempted to recognize the species of bis predecessors has found 

 himself to so great an extent unsuccessful, through the vagueness and 

 uncertainty of the system that was to have been his guide, that he has 

 therefore naturally commenced his career of the study and record of 

 species unknown to him by a new method of arrangement, which, 

 although perhaps sufficient for his own limited circle of subjects, 

 becomes, when applied to a fresh series of them, quite as inapplicable 

 to a general and extended view of these singularly protean forms as 

 those of his predecessors. 



This was precisely the condition in which I found myself at the 

 commencement of my own career of investigation ; and I naturally 

 asked myself the question. Is there no means of escape from these 

 various and inefficient modes of registering the examination of these 

 animals, through a natural division of them into classes, orders, 

 and genera, by means of their internal and external organization, 

 after the manner pursued in other departments of zoology, and espe- 

 cially in botany ? Strongly impressed with this idea, I com- 

 menced an investigation of a large collection of British andforeign spe- 

 cies in my own possession ; and I soon found that sponges, like other 

 organized being?, were always provided with a skeleton, and that, as 

 in other branches of zoology, the materials and mode of its structure 

 varied very considerably in different species, and that those peculi- 

 arities of its structure were remarkably uniform and persistent 

 through a considerable number of species in which they occurred. 

 Here, then, was a foundation for the primary division of these crea- 

 tures, in perfect accordance with the rules of zoological science as 

 established by Linnaeus, Cnvier, and other laborious and talented 

 authors of modern times ; and I had the satisfaction of finding that 

 the more widely I extended my observations the more uniform and 

 certainly available these primary parts of the organization became ; 

 and in addition to their characters of uniformity and constancy, there 

 was this strong recommendation of them as bases for the foundation 

 of classes, orders, and genera, that, however imperfect in form or 

 dilapidated by external injuries or partial decomposition, the most 

 persistent and last surviving part of the animal was always the 

 skeleton. 



In botan)"^ and some branches of zoology the principal difficulty 

 that meets the student is the correct determination of the genera; but 

 this is not the case to so great an extent in the study of the Spon- 

 giadse when their characters are founded on their structure and 

 organization. As far as the genera have been established on these 

 principles, they are so well marked, both by the material and the 

 peculiar modes of the arrangement of their component parts, that they 



