206 WILSON. [Vol. IV. 



the proliferation of which the paired mesoblastic bands are pro- 

 duced. The teloblasts are often differentiated at a very early 

 period, — .sometimes even prior to the gastrulation, — arising 

 near the region corresponding to the posterior lip of the blasto- 

 pore. In no case do they arise from the ectoblast ; in some 

 cases they seem to arise from, or at least to be closely associated 

 with, the cells of the archenteron. In still another class of cases 

 the mesoblast appears to arise neither by delamination from 

 the ectoblast, nor from teloblasts, but from a central mass of 

 "mes-entoblast," the lateral portions of which give rise to the 

 mesoblast-bands, and the central portion to the entoblast. This 

 would seem to be the case, for instance, in the development of 

 EjichytrcEoides as described by Roule ; ^ and it is apparently the 

 case also in Sprorbis, according to the investigations of my 

 former pupil. Miss H. Randolph, whose preparations I have 

 carefully examined.^ 



How are these various modes of mesoblast-formation to be 

 reconciled.^ It is impossible to doubt the homology of the mes- 

 oblastic bands in Polygordius, Etipomatiis, Lumbriciis, Clepsine, 

 Lopadorhynchns, and E7ichytr(Boides — a series that includes rep- 

 resentatives of the three modes of mesoblast-formation I have 

 mentioned. It must, therefore, be possible to reduce these 

 modes of development to a common type, and it is a remarkable 

 illustration of the elementary state of our knowledge of annelid 

 development that no one, as far as I am aware, has made even 

 a suggestion as to how this is to be done. The only exception 

 known to me is an hypothesis put forward in my recent paper 



1 Ann. d. Sci. Nat., VII., 18S9. 



2 Gotte has described the mesoblast of Spirorbis nautiloides as arising from a pair 

 of large teloblasts, derived from the archenteron. This, however, is certainly a 

 blunder. As Salensky has suggested, and ^Sliss Randolph has conclusively proved, 

 the rounded body, mistaken by Giitte for a pair of primary mesoblasts, is simply the 

 rounded, hollow posterior part of the mesenteron. The mesoblast first appears in 

 the form of two lateral masses which join each other posteriorly, and in some cases 

 seem to fuse with the walls of the mesenteron. Salensky describes these masses 

 as delaminating from the ectoblast in Pileolaria, a form nearly related to Spirorbis. 

 In the latter form, however, the mesoblast-cells are quite distinct from the ectoblast 

 from the earliest period at which they are distinguishable, while they are so intimately 

 related with the entoblast, as to form with it, apparently, a common mass of mes- 

 entoblast, as in Enchytrcsoides. Miss Randolph's researches are, however, still 

 incomplete, and a more thorough study of the earlier stages may lead to a different 

 interpretation. 



