CLASSIFICATION OF THE ANIMAL KINGDOM. 201 



groups. We are reduced to speculation— to the formation of 



more or less probable hypotheses; and, though I believe that phylo- 



genetic speculations are of great interest and importance, and are 



to be reckoned among the most valuable suggestors of, and guides 



to, investigation, I think it is well to recollect, not only that they 



are at present, for the most part, incapable of being submitted to 



any objective test, but that they are likely long to remain in that 



condition. For the ultimate test of the truth of a phylogenetic 



hypothesis is the historic record of the succession of living forms 



contained in the fossiliferous rocks ; and the present state of 



geology gives no encouragement to the supposition that even the 



whole series of fossiliferous rocks represents a period coextensive 



with the existence of life on the earth. In speculating on these 



subjects, it is constantly needful to remind oneself, even now, tliat 



there is every reason to believe that all the leading modifications of 



animal form were existent at least as early as the close of the 



Palaeozoic epoch ; and though it is true that the fossiliferous 



Palaeozoic rocks are thicker than all the rest put together, yet 



the amount of progress in evolution from a mouer to the fully 



differentiated. Vertebrata of the Trias bears an enormously larger 



ratio to the amount of progress from the Triassic vertebrates to 



those of the present day. All such comparative measurements 



as these are but rough aids to the imagination ; but the Inverte- 



brata yield even stronger evidence in the same direction. The 



larger divisions of the Arthropoda were completely differentiated 



in the Carboniferous epoch ; so were those of the Mollusks and 



those of the Echinoderms. The great desideratum is the discovery 



of estuarine and. freshwater formations of Silurian, Cambrian, and 



Laurentian date. At the present moment, I do not think that 



any one is in j^ position to form even a probable guess as to 



what will be found in such deposits. 



Taxonomy should be a precise and logical arrangement of veri- 

 fiable facts ; and there is no little danger of throwing science into 

 confusion if the taxonomist allows himself to be influenced by 

 merely speculative considerations. The present essay is an attempt 

 to set a good example, and, without reference to phylogeny, to 

 draw up a classification of the animal kingdom, which, as a fair 

 statement of what, at present, appear to be well-established facts, 

 may have some chance of permanence, in principle, if not in 

 detail, while the successive phylogenetic schemes come and go. 

 No doubt the increase of our knowledge of embryology will largely 



