66 



appears a patch of cells, the source of which was not clearly 

 made out. This is the rudiment of the heart and pericardium, 

 but the exact steps of its subsequent development were not 

 satisfactorily observed. 



Water enters the branchial sac from without through the 

 orifice which was the primary mouth, and flows through the 

 branchial slits into the space which is now found between 

 the epidermis and the gelatinous envelope, and which ex- 

 tends far back to an orifice, which appears close to the anus, 

 and through which excrementitious matter is ejected. 



As development jnoceeds peculiar organs, forming ap- 

 parently a network of blind tubes, arise round the intestines ; 

 they possess a complicated structure, to the study of which 

 the author hopes to return. 



Professor Kupfier, of Kiel, has addressed the following 

 letter to Professor Max Schultze, of Bonn, on this matter. 

 The letter is pubHshed in Schultze's ' Archiv,' Part iv, 1869. 



" You, of course, know Kowalevsky's work on the develop- 

 ment of tlie simple Ascidians, which brought to light facts 

 such as nothing else has done before, to span the gulf be- 

 tween vertebrates and invertebrates, and has given positive 

 foundation to the doctrine of a phylo^enetic connection be- 

 tween apparently entirely different circles of life. As far as 

 I know, no one but Haeckel has paid any particular atten- 

 tion to the work, and it does not seem to have been gene- 

 rally taken as trustworthy. I own I did not myself belong 

 to those who believed in it. It behoves me the more to 

 acknowledge this, as through continuous observation during 

 this summer of the Phallusia canina which is found in the 

 Bay of Kiel, my opinion has been entirely clianged. The 

 first phase of development, the formation of the free 

 swimming larva from the Gg^,, shows in such elementary 

 clearness the chief features of the development of a verte- 

 brate animal, that the observation is quite convincing. The 

 animal which I examined does not seem to have been any 

 one of those on Avhich KoAvalevsky worked. Apart from his 

 first stages of development having been described from Phall. 

 mammillata, his Ph. mtestmalis, Linn., is not to be identified 

 with the canina of O. P. Miiller. I leave out of considera- 

 tion, among other things, that he speaks of the peculiar 

 appendages of the egg-membrane in the forms used by him, 

 as soon falling off", whilst the analogous structures in our 

 species present characteristic cellular tufts Avhicli remain 

 fastened in regular order to the egg-shell until the larva is 

 hatched, so that the emptied egg membrane is still recog- 

 nisable by these appendages. The chief or fundamental 



