THE POSITION OF SPONGES, ETC. 443 



cells became in part transformed into flat epithelium, 

 and in Sycons, which start developmentally from an 

 Ascon-like condition, this transformation must be con- 

 tinually occurring during growth. If, however, we exa- 

 mine the grounds other than theoretical upon which was 

 founded this theory of the transformation, alleged to be of 

 such frequent occurrence, of collared into flat epithelium, 

 we find direct evidence of any kind, whether histological 

 or embryological, to be entirely lacking. As is shown in 

 the figures published by all trustworthy observers, and 

 especially in those of Schulze, and as can be seen in any 

 good preparations of sponge histology, the collar cells linino- 

 flageliated chambers or radial tubes always pass abruptly 

 and without the slightest gradation into the flattened epi- 

 thelium ot the incurrent or excurrent canals. This was 

 especially well shown by Schulze (1875) ^^ his investiga- 

 tions upon just the very sponge in which direct evidence 

 of the modification of collar cells into flattened cells might 

 be expected to be forthcoming, namely, Sycon raphanus. 

 Cells intermediate in form between the two kinds of epithe- 

 lium do not occur, and the many figures of such transitions 

 given by Lendenfeld must be characterised as erroneous. 

 The fact, sufficiently obvious in itself, that a collar cell if 

 cut obliquely may present in sections the appearance of a 

 rounded or columnar cell without flagellum or collar, per- 

 haps will account for some of these incorrect descriptions, 

 though the considerable differences both in the structure of 

 the nucleus and the nature of the cytoplasm between collar 

 cells and flat epithelium cells ought to have been sufficient 

 to prevent any such confusion. 



Schulze was led by his observations upon Sycon raphanus 

 (1875, p. 258) to regard the flat epithelium as a layer of 

 uniform nature throughout, on account of its great simi- 

 larity, amounting to identity, both at the outer surface of 

 the body and in the interior of the canal system. His 

 embryological investigations upon Plakina (1880) first led 

 him to express the contrary view referred to above. It 

 cannot, however, be said that his figures of the latter 

 form afford any direct evidence of an " endodermal " flat 



30 



