436 Wilson: [vol. vi. 



merely on sections and on general views of embryos prepared 

 by unsuitable methods ; and the errors into which he has fallen 

 are a striking illustration of the necessity of employing a 

 method which will give clear and definite surface-views. 



2. Comparison with Other Annelida. 



Many embryologists have made observations on the early 

 stages of cleavage among the Polychaeta, but almost without 

 exception these stages were not brought into sufficiently defi- 

 nite connection with the larvae and adult structure to admit of 

 any detailed comparison with Nereis. The most complete of 

 these studies we owe to Hatschek, whose account of Eupomatiis 

 uncinatns gives a clear view of the general relations of the 

 cleavage stages to the gastrula and the larva. Even in this 

 case, however, the history of the individual blastomeres is im- 

 perfectly known, and neither the relation of the mesoblast to 

 the four-celled stage nor the origin of the ventral plate was 

 determined. 



I turn, therefore, to the OHgochaeta and Hirudinea, the early 

 stages of which have been far more thoroughly studied than 

 those of the Polychaeta. As a basis, I shall take Whitman's 

 work on Clepsine (Nos. 27, 28) and Vejdovsky's on RJiynchelmis 

 (No. 25), and my task is much simplified by the fact that Whit- 

 man has himself given a full comparison between these two 

 forms, and between Clepsine and the other Hirudinea. 



The eight-celled stage has the same structure, and in all prob- 

 ability arises in the same manner, in the three forms, the only 

 difference being the much greater relative size of the micro- 

 meres in Nereis. (I agree with Whitman that Vejdovsky is 

 probably in error in his account of the second cleavage. There 

 is no doubt of the exact agreement of Clepsine, Nereis, and 

 many other annelids on this point.) There is, however, a 

 remarkable difference in the axial relations, since in both Clep- 

 sine and Rhynchelmis the large macromere marks the posterior 

 region, as in the polyclades, and the first two cleavage-planes are 

 inclined approximately at an angle of 45° to the median plane 

 (see p. 454). 



In Clepsine the large macromere first gives rise to a large 

 cell (" neuro-nephroblast ") and then divides into two "primary 



