540 PROCEEDINGS OF THE ACADEMY OF [July, 



in the test cells on the ventral side. The median one, of a transitory 

 character, is said to represent the stomodseum, but as a stomodseum 

 exists in Halomenia in the normal position I am strongly of the opinion 

 that Pruvot is in error regarding this point. The lateral invagina- 

 tions unite, forming a transverse band, and posteriorly are prolonged 

 to meet the mesoblastic bands of the trunk. Some of the more 

 dorsal elements constitute the cerebral ganglia. In addition, the 

 ectoderm of the head appears to arise wholly from these same lateral 

 invaginations. Such an origin of mesoblast elements is certainly 

 unique, and I have only to state that I believe this entire group is 

 ganglionic. Its posterior union with the ventral cords certainly 

 indicates its nervous character. The anterior enlargements, the 

 future cerebral ganglia, are of unusual size in Halomenia, but there 

 is nothing whatever to indicate that they comprise any mesoblastic 

 elements. Furthermore, there is nothing in Halomenia to suggest 

 the development of head ectoderm from any of these cells; and the 

 counter theory, that at least in part it may arise from one or two 

 large cells located beneath the cells responsible for the development 

 of the cerebral ganglia, has been noted in a preceding paragraph.^ 



If on the basis of comparative anatomy it is impossible at the 

 present time to definitely place the solenogastres in their proper 

 systematic position, it is obvious that this is more emphatically true 

 where scanty embryological data are the sole criterion. However, 

 it is evident that their development is more in accord with what we 

 find among the moUusks than with any other phylum. The resem- 

 blance of the embryo shortly before its metamorphosis is strikingly 

 similar, in several important details, to Yoldia or Dentalium or to 

 the chitons if we neglect size differences with respect to the test. 

 Plate ( '92) has shown that there are good reasons for the belief, long 

 ago expressed by Blainville ('25), that the scaphopods are most 

 closely related to the prosobranchs rather than to the lamellibranchs. 

 The excessively developed tests encountered in certain species of 

 the first- and last-named classes are therefore not of fundamental 

 importance, and furthermore its small size in the chitons is accord- 

 ingly not a serious obstacle to the theory expressed by several authors 

 that the solenogastres and the chitons are derivatives of a common 

 ancestor. Such a conclusion has Ijeen based almost entirely on 



1 A full account of tho development of Halomenia will appear in connection 

 with a report on the solenogastres from the eastern coast of the United States, 

 but as considerable time must elapse before its completion it has seemed desirable 

 to publish this preliminary account. 



