BUDDING IN TUNIC AT A. 51 



layer theory to the process of blastogenesis in Tunicata. The 

 ectodermal origin of the peribranchial involutions mPyrosouia 

 and many Ascidians, and the analogy of Amphioxus, have in- 

 duced many authors to regard these organs as universally 

 ectodermal. On the other hand no fact has been more 

 frequently demonstrated than the endodermal origin of the 

 peribranchial chambers in the Ascidian bud. The incon- 

 sistency of this fact with the above-mentioned view of the 

 ectodermal nature of the organs has accordingly led many 

 writers to give up any attempt to reconcile the phenomena 

 of budding with the germ-layer theory which has sprung- 

 out of modern embryological research. There is, however, 

 a less desperate way out of the difficulty before us. It is 

 possible that the peribranchial sacs are not entirely ecto- 

 dermal by nature : there still remains the view, propounded 

 in the first instance, I believe, by Fol (7), that the peribran- 

 chial sacs of Ascidians correspond to the branchial canals 

 of Appendicularians, and, therefore, to a pair of gill-slits 

 formed by the union of a pair of ectodermal involutions with 

 a pair of endodermal outgrowths. The nature of the peri- 

 branchial sacs, according to this view, is neither purely 

 ectodermal nor purely endodermal, but a mixture of both. 

 Van Beneden and Julin (3) endeavoured to establish this view 

 as a result of their classical researches upon the development 

 of Clave Una ; in embryos of this form the peribranchial sacs 

 seemed to them to be developed from both germ-layers. 

 Their statements upon this point, however, have recently 

 been controverted both by Willey (32) and Seeliger (30), who 

 maintain that the peribranchial sacs are formed from ecto- 

 dermal involutions alone. Yet the mistake must not be made 

 on this account of imagining that Fol's suggestion as to the 

 primitively double origin of the peribranchial sacs derives 

 no further support from embryology. In Salfta, according to 

 the recent work of Korotneff (14) and the older researches of 

 Salensky (23), the cloacal cavity is directly derived from that 

 of the pharynx ; and since its walls are stated to be formed 

 from the same layer as those of the pharynx (the endodermal 

 histogenes of Korotneff), there seems to be some evidence 

 that in this form at any rate the cloacal and peribranchial 



