366 



Although I*have given considerable attention to the point, I have 

 as yet not been able to fully convince myself that there is any com- 

 munication whatever between the ascidiozooids and the vessels; certain 

 it is that there are many ascidiozooids without such communication. 

 Since, however, there are quantities of blood -cells in the vessels, 

 these undoubtedly do communicate with the blood spaces of at least 

 some of the ascidiozooids. 



5) The development of the organs of the blastozooids is in general 

 quite similar to that which takes place in Botryllus. I therefore shall 

 not here even sketch the process as a whole but shall restrict myself 

 to a few statements concerning the development of the central nervous 

 system, the hypophysis, and the heart. 



6) The common neurohypophyseal anläge arises at an 

 early stage as a wide evagination from the dorsal portion 

 of the "endoderm". At a slightly later stage it is a simple, wide, 

 short tube, communicating with the branchial cavity in front and end- 

 ing blindly behind. I cannot determine that it communicates with 

 either of the peribranchial cavities at any time, as is said to be the 

 case in Botryllus. The ganglion is differentiated in the ventral wall 

 of this tube, as it is stated by Hjort to do in Botryllus, 



7) The pericardial vesicle is present at an early stage in the 

 development of the ascidiozooid, but I have not yet been able to deter- 

 mine its origin ; since, however, there is an epicardium as in Botryllus, 

 it probably is formed from this. The heart developes from the peri- 

 cardial vesicle in the usual way. 



8) The great similarity between the blastogenesis in Goodsiria 

 and Botryllus, considered in connection with the close resemblance 

 in many essential points of adult structure of the two genera, strongly 

 suggests that a critical comparison, embryologic, blastogenic, and ana- 

 tomic, of the two families to which these genera belong, will prove 

 them to have closer relationships than have been supposed heretofore. 



2. Perophora. 

 At the outset my chief purpose in undertaking a study of the 

 bud development in this genus, was to satisfy myself, if possible, on 

 the much controverted question of the origin here of tbe nerve gan- 

 glion. From the numerous times that the blastogenesis of Perophora 

 has been a subject of investigation, I had supposed that in all points 

 excepting the development of the ganglion and possibly of the heart, 

 our knowledge here needed neither augmentation nor modification. 

 However, in the course of my attempts to reach definite conclusions 



