RETROGRESSIVE DEVELOPMENT. 461 



just as in the Amphioxus. The excretory opening of the 

 former corresponds fully to the abdominal pore of the latter. 

 In the adult Ascidian the gill-intestine and the heart rest- 

 ing on the ventral side of the latter, are almost the only 

 organs that recall the original relationship to Vertebrates. 



In conclusion we will glance at the development of the 

 curious external gelatinous mantle, or cellulose sac, in which 

 the Ascidian is afterwards entirely enclosed, and which 

 characterizes the whole class of Mantle Animals (Tunicata). 

 Very various and remarkable views have been entertained 

 as to the formation of this mantle. For instance, it was the 

 opinion of Kowalevsky, that the animal does not itself 

 form the mantle, but that the latter is produced by special 

 cells from the maternal body, which surround the egg. 

 According to this the mantle would be a permanent 

 egg-envelope. This is contrary to all analogy, and d 

 priori highly improbable. Another naturalist, Kupifer, 

 who has confirmed and extended the researches of the 

 former, assumed that the mantle develops from cells which, 

 even before the impregnation of the egg-cell, form from the 

 outer portion of the yelk, and separate entirely from the 

 inner portion. This seems very doubtful and unlikely. 

 Hertwig's researches, which are confirmed by my own 

 observations, first showed that the mantle develops as a 

 so-called "cuticula." It is an exudation from epidermic 

 cells, which soon hardens, separates from the real body of 

 the Ascidian, and condenses so as to form a strong envelope 

 round the latter. The matter of these cells is chemically 

 indistinguishable from the cellulose of plants. While the 

 epidermic cells of the external horn-plate are secreting this 

 mass of cellulose, some of them drop into it, continue to 



