328 COMPARATIVE ANATOMY 



problems are here. It is probable, too, that fruitless theory has 

 reached its climax in this field. That paleontology is the court of 

 final appeal here for every case that can be carried before it there 

 is no question. Since, however, a class of cases is ever on hand 

 that can never be taken to this court, I propose to make these 

 the centre of consideration, with the understanding, though, that 

 most of the general conclusions set forth have application more 

 or less strictly to other cases as well. I have in mind the problems 

 of the relationships between the primary divisions of the animal 

 kingdom. 



It is well to remind those disposed to value lightly efforts to 

 trace phylogenies through morphology and embryology that if 

 we are ever to know the interrelationships of the primary sub- 

 divisions of animals, the knowledge will have to be gained from 

 these sources. A well-known paleontologist has lately remarked 

 that "Perhaps the most disappointing element in paleontological 

 results thus far is the lack of all information concerning the origin of 

 the great subkingdoms, or phyla, of animals." (Woodward, 1898.) 

 These results, or rather lack of results, ought not to be counted 

 against paleontology, for, as we now see, paleontology should never 

 have been held responsible for this task: That science is, of course, 

 able to say almost nothing about extinct animals that did not possess 

 hard, imperishable parts. Now, looked at comprehensively, the 

 evidence of paleontology, morphology, and of embryology concur in 

 support of the belief that the phyla of the animal kingdom all had 

 their origin in ancestors much simpler and smaller than any repre- 

 sentatives of these at present known to us, in ancestors so small and 

 simple, in fact, that preservable hard parts had not yet arisen. A 

 well-established skeleton, even of a simple type, must be regarded 

 as always marking a comparatively advanced state of evolution. 

 Thus, even such simple representatives of the ccelenterata as Dycto- 

 rema or Diplograptus of Hallfrom the Silurian must be supposed to 

 have but gradually acquired a sufficiently chitinous hydrotheca 

 to enable them to leave even such remains of them as we have; and 

 consequently, that they must have had a long line of ancestors 

 about which paleontology can never expect to get much direct 

 information. Similarly with the actinozoa, all the evidence we have, 

 both from comparative anatomy and from embryology, leads to 

 the unequivocal conclusion that the coral-producing members of 

 the class acquired the skeleton now so characteristic of them only 

 after a long evolutionary course. Or consider the mollusca. Which- 

 ever of the two prevailing theories of the origin of the phylum be 

 favored, constant and characteristic as is the shell, and early as is 

 its appearance in ontogeny, there is no possibility of its being a 

 really primitive structure. In other words, the evolutionary career 



