VOL. III.] Balanoglossus. 199 



raised a doubt that all belong- to one genus — which is the 

 more primitive way ol development, directly without the larval 

 stage, or through the larva? Did the first Balanoglossus reach its 

 developmental goal by the long, indirect tornaria road, and did a more 

 modern one, imbued with the rapid transit idea, cut across lots leav- 

 ing the ancient roundabout way? Or did the older forms go across 

 while the younger ones have taken to the longer road? No one 

 has discussed this question at any length, and I am n.ot going to un- 

 dertake it at present. In fact without a knowledge of the first stages 

 of development of the Tornaria, it would probably be impossible to 

 arrive at any very satisfactory conclusion on the subject. It is sug- 

 gested by Korschelt & Heider^' that the direct development is 

 the more primitive, their reason for this conclusion being found 

 in the fact that the mouth and anus do not form in this larva till a 

 comparatively late stage — a condition which would seem to be in- 

 compatible with a free swimming of larva. 



There are, however, some quite serious difficulties in the way of 

 this suggestion, one of which is that the circumanal ciliated band 

 appears very early in the directly developing species, while it forms" 

 quite late in the Tornaria. 



Eor the solution of this question, as well as of several others, it 

 is of the utmost importance that we fill up the gap that now exists 

 in our knowledge of the earliest embryonic stages of Tornaria; and 

 to this end the more species we have access to, the better become 

 our chances of being able to do this. It is quite probable that 

 somewhere on our great extent of sand and mud beach a represen- 

 tative of the genus will be found. 



Explanation of the Figures of Plate xxii. 



Fig. 1. Balanoglossus kowalevskii . (After A. Agassiz, from Korschelt 

 and Heider.) 



Fig. 1. Sagittal longitudinal section through the proboscis and collar of 

 Balanoglossus sarniensis. (After Kohler, from Korschelt and Heider.) 



Fig. 3. The young Balanoglossus, shortly after its transformation; under 

 the compressor. 



Fig. 4. The anterior portions of a Tornaria shortly before its transforma- 

 tion to Balanoglossus. The larva was flattened down somewhat by the com- 

 pressor. The outlines drawn with a camera lucida. 



Fig. 5. A Tornaria at a somewhat older stage than Fig. 6, to show internal 

 structures. 



"I.e. 



