No. 3.] GROWTH AND METAMORPHOSIS OF TORN ARIA. 447 



With the adult Echinoclerm it is different, as there are many- 

 points, besides the similarities of the larvae, common to the two 

 groups. If we assume with Semon that the Pentactula comes 

 nearest to the ancestral type of the adult Echinoderm, we have 

 found a form with which it is not difficult to compare the adult 

 Balanoglossus. Briefly these resemblances may be pointed out. 

 In each there is a diffuse nervous system formed by the ectoderm 

 with a nerve-fibre layer beneath. Certain parts of the ectoderm 

 have been specialized to a slight degree, in each forming more 

 central nerve paths. The body cavities present close similari- 

 ties both in their origin and fate ; the musculature being formed 

 from its walls and the peritoneal epithelium giving rise to the 

 generative products, the gonads, opening by very short simple 

 tubes to the exterior. More important is the close connection 

 between the locomotor water system of each, which comes from 

 the anterior enterocoel, and while I do not think that we can 

 directly compare the proboscis of Balanoglossus with an ambu- 

 lacral foot of Echinoderm, as Metschnikoff has done, still 

 the fundamental arrangement of the two systems of organs is 

 the same. Moreover, the madreporic plate and the dorsal 

 water pore of Balanoglossus are practically identical ; also the 

 blood system and its close connection with the anterior entero- 

 coel in the two forms is very similar. 



In Balanoglossus we have discovered an animal outside the 

 group Chordata, showing very primitive structures and relation- 

 ships and having already a mouth opening on the ventral or 

 abneural (natural) side of the body. Now if the evidence 

 brought forward to show its resemblance to the Chordata be 

 valid, there is no necessity to believe, nor any reason for assum- 

 ing, that the present chordate mouth is a new structure and 

 that an old mouth of some kind or other has been lost during 

 the past. And if Balanoglossus be related through its larva 

 with the Echinoderms, as I have attempted to show in the 

 preceding page, we see how old a phylum that of the Ver- 

 tebrates must be, and hence the futility of attempting to derive 

 them from any such highly specialized animals as the Annelids 

 of to-day. 



Johns Hopkins University, April 10, 1891. 



