THE POSITION OF SPONGES, ETC. 451 



but criticise what he puts before us. I must therefore ven- 

 ture to point out that he has figured in the plainest manner 

 the immigration of the cihated cells, and the formation of the 

 chambers from them. On plate xvi., figs. 26, 27 and 28, 

 show in surface view the growth of the inner mass over the 

 ciliated cells ; in fig. 30 we see the ciliated cells heaped up 

 at one pole and becoming internal : figs. 39, 40 and 41 

 of plate xvii. show groups of these cells forming chambers, 

 though their cell outlines are not represented ; and figs. 

 44, 45, 47 (plate xviii.) and 50, show later stages with 

 chambers formed of cells with small nuclei, similar to the 

 larval ciliated cells. On plate xvii., again, fig. 36 shows 

 some of the ciliated cells not yet absorbed, others scattered 

 throuoh the interior, and even arranoed to form chambers, 

 and figs. T,S and 43 represent later stages in which the 

 chambers are well shown to be formed from the small 

 cells. Finally attention must be drawn to the most im- 

 portant stage figured on plate xxi., fig. 85, showing in a 

 larva just fixed the commencing inversion of the layers. 

 Knowing how easy it is to represent in drawings what one 

 believes the preparations to show, it is remarkable to find 

 in Wilson's figures so much evidence for a view opposed to 

 that which he holds, and the fact is certainly a great testi- 

 mony to the accuracy of the author's observations. Wilson 

 does not always represent the disproportion in size between 

 the nuclei of the two layers as distinctly as has been done 

 by Maas and Delage, no doubt because the point never 

 occurred to him as one of importance. While, therefore, 

 Wilson's figures must be cited as a most valuable contribu- 

 tion to the theory — or rather the facts — of the inversion ot the 

 layers, it may further be pointed out that he admits in the text 

 that the ciliated cells at least partially travel inwards. On 

 p. 299 after pointing out that "all the columnar cells cannot 

 be transformed into flat cells," he says that the appearances 

 suggest "that the ectoderm [i.e., ciliated] cells are migrat- 

 ing at this point into the interior of the larva," and again 

 on p. 319 he admits that " some of the ectoderm cells of the 

 larva migrate into the interior during the metamorphosis ". 

 On p. 375 he makes the confession : " My observations on 



