EMBRYOLOGY OF CLEPSINE. 297 



can the origin of the mesoderm in this place be referred to one 

 of the primary germ-layers, to the exclusion of the other ? In 

 such cases the resemblance of the cells, to those of the ectoderm 

 or entoderm is usually regarded as evidence of relationship ; but 

 such a test is by no means decisive. The difficulty in the case of 

 most fishes, birds, and mammals, is even greater. In a very few 

 instances the mesoderm has been traced to distinct cleavage-cells, 

 and in no case has this been done more completely than in | 

 Clepsine. Yet the question, which lamella gives it origin ? admits 

 of a difference of opinion. One might be inclined to think that the 

 two mesoblasts arise from the entoderm, judging from their size, 

 position, and composition. Like the entodermic blastomeres {a, i 

 h, c,) they contain a large amount of yolk, which later serves as 1 

 food for the embryo. But this is no criterion, as the neuro- i 

 blasts are also loaded with the same food-material. 



On the other hand, the origin of the mesoblasts from that one 

 of the four original blastomeres (.r) which is preeminently ecto- 

 dermic, would seem to favour the opinion that it is derived from 

 the ectoderm. The question is therefore debatable. I incline , 

 to the former opinion for this reason, that the mesoblasts repre- \ 

 sent the lower pole (entodermic pole) of the blastomere {x) while ' 

 the neuroblasts represent the upper pole (ectodermic pole). 

 This view, it seems to me, is most in harmony with known facts. 

 The first two meridional divisions do not completely separate the 

 elements of the future lamellae in Clepsine ; for the upper pole 

 (oral pole) of each is ectodermic, and the lower, if we except the 

 fourth [x) is entodermic. The first sharp separation of ectoderm 

 from entoderm begins with the parallel division, which produces 

 the four original ectoblasts at the oral pole. 



I shall enter into the historical part of this subject only in so 

 far as it, in conjunction with my own observations, bears directly 

 upon a few important points, the most significant of which is the I 

 bilateral origin of the mesoderm. ' 



The first, so far as I am aware, to trace the mesoderm to 

 to a single pair of mesoblasts was Kowalevsky (1871). He 

 stated (^4) that during the invagination (Lumbricus) a single 

 cell {m), on each side the median line, steps out of the entoderm 

 into the blastocoel. His figures (10-16, PI. VI.) show that these 

 two cells a])pear behind the blastopore, and that they produce 

 fonoard two longitudinal masses of mesoderm-cells. The next 

 case of this kind was established by Rabl (1876). Rabl (3%^) 

 traced the origin of two mesoblasts, situated at one end of the 

 blastopore, to a large entoderm-cell, and was the first to empha- 

 sise the general importance of the double origin. Hatschek 

 (1877, -ro-V) found also two mesoblasts, derived from entoderm, 

 at the hind end of the blastopore in Pedicellina. In addition to 

 these four cases (Clepsine included) where it is certain that the 



