of the phylum was a long one before this character was acquired. 

 If the turbellaria-like ancestry be the theory advocated, then the 

 evidence from comparative anatomy is positive that the shell is a 

 comparatively recent acquisition. On the other hand, if the trocho- 

 phore theory be defended, the interpretation of the shell, so well 

 stated by Korsheldt and Heider (1900, vol. iv, p. 323), would un- 

 questionably have to be adopted. "The very early rise of this 

 organ (the shell gland), which may in a few cases be found even 

 before the trochophore form fully develops, must be regarded as a 

 shifting back to an early period of the embryonic development of 

 this feature, which was only a recent acquisition." 



So we might go through nearly the whole list of the fundamental 

 types of animal organization and show the extreme improbability, 

 if not the impossibility, that their earliest ancestry will ever be 

 accessible to the paleontologist. 1 This view is of course only the 

 rigorous application of the " Law of the Unspecialized," apparently 

 first definitely formulated by Professor E. D. Cope. That this prin- 

 ciple has not been sufficiently recognized at all times by morpho- 

 logists there is no doubt. The effort to make out the transformation 

 of a Limulus-like arthropod into a marsipobranch-like vertebrate 

 can but be regarded as one of the most extreme cases of disregard 

 of the principle. 



It may be held as practically certain that paleontology will never 

 be able, by direct discovery, to bridge the gaps between any of the phyla 

 of animals. All it will be able to do in this direction will have to be 

 by inference, and when resort is had to this method, paleontology 

 is worse off than comparative anatomy, since the range of its avail- 

 able data for any particular problem is less. The paleontologist 

 must rest his case on the testimony of a single organ system, the 

 skeletal, while the morphologist has at his command all the systems 

 of organs. True it is that frequently the testimony of the different 

 systems is so conflicting that the difficulty of balancing it up and 

 deciding just what the total signifies is exceedingly great; and in 

 such cases the morphologist is somewhat prone to feel that he has 

 too much, or, rather, too many kinds of evidence; that he could 

 do better could he be rid of some of his facts. But, of course, the 

 paleontologist's seeming advantage here is of a perilous sort. It is 

 similar to that of the systematist's, whose species of plants or 

 animals are beyond question so long as he does not have too many 

 to fit into his descriptions. Since, then, it is by the morphological 



1 I do not forget in this connection the extremely interesting observations made 

 in recent years on fossil remains of soft tissues, as for example, of medusa? (Walcott) 

 and of striated muscles (Dean, 1902). It is, however, hardly conceivable that 

 remains of this sort will ever be found in sufficient abundance and condition of 

 preservation to make them of real importance in the solution of phylogenic 

 problems. 



