BUDDING IN TUNIC AT A. 53 



would gradually predominate to the exclusion of the other, 

 so that the entire water-channel on each side would 

 eventually develop from one instead of both of the layers. 

 There is no apparent reason why the predominating 

 element should in all cases have been the ectoderm ; and I 

 regard Botryllus, and possibly Salpa too, as one of the 

 forms in which the endodermal portion of the primitive 

 gill-slit has enlarged to form the whole of the peribranchial 

 sac. 



It will be admitted that if the theory which I have here 

 set forth upon the phylogeny of the cloaca and peribranchial 

 sacs is true, the variable origin of these structures in the 

 buds of Tunicata ceases to offer any serious difficulty. In 

 the buds of all fixed Ascidians so far investigated these 

 organs arise as outgrowths from the endoderm ; while in 

 Pyrosoma, Anckinia, and Dolckinia they undoubtedly arise 

 from the ectoderm. If the theory in question be admitted, 

 not only does this diversity of origin cease to be incon- 

 sistent with the doctrine of germinal layers, but it actually 

 furnishes an additional and striking argument in its support. 



With regard to the development of the stigmata in the 

 bud, I have elsewhere shown that it represents a simple 

 abbreviation of the phylogenetic mode of origin, which is 

 recapitulated in the embryos of certain types (8). 



The endodermal origin of the pericardium and heart in 

 the buds of certain Tunicata [e.g., Clavelina, Circinalium, 

 Botryllus (?), Anckinia, Dolckinia) can scarcely be regarded 

 as a very serious difficulty. Not only is a similar origin 

 ascribed to these organs in the embryos of Clavelina 

 (Seeliger, Van Beneden and Julin); but, as the mesoderm 

 itself is derived from the endoderm, the independent origin 

 of the heart from the same layer can be readily interpreted 

 as an instance of precocious segregation. It is desirable, 

 however, that future investigators should examine this point 

 with particular caution. In the buds of Botryllus the 

 development of the heart has been described independently 

 by Hjort, Oka, and Pizon, and there is a considerable 

 difference between their accounts. Pizon (20) states that 

 the pericardial vesicle is derived as a diverticulum from the 



