936 The American Naturalist. [November, 
mic diverticula. In other words, even if Lankester’s entero- 
coele hypothesis holds true of the Vertebrates, they would still 
show only a very remote resemblance with Amphioxus so far 
as the early stages of the mesoderm are concerned. But there 
is yet another important difference between Amphioxus and 
the Vertebrates, namely, in the former the segments are total, 
in the latter partial. In Amphioxus the whole of the meso- 
thelium is segmented, in Vertebrates only the dorsal regions of 
the mesothelium are segmented, the ventral regions or walls 
of the splanchnocoele are not segmented. That there is an 
impressive similarity between the formation of the mesoder- 
mic bands in Amphioxus and Annelids was shown by the 
discoveries of Hatschek, and in both types the segments are 
total. Itis perfectly clear, therefore, that the early history of the 
mesoderm indicates a much closer affinity of Amphioxus with 
the Annelids than with the Vertebrates. The existing differ- 
ences may be explained by the assumption that the cavities 
of the segments are formed in Annelids after, in Amphioxus 
during, the morphological separation of the mesoderm from 
the entoderm. ? 
2. Sexual organs.—The sexual organs of Amphioxus are con- 
fined to the branchial region, they are segmentally arranged 
and they are lateral of the excretory organs. Those of Verte- 
brates do not appear in the branchial region, they are not seg- 
mented and are mediad of the excretory organs. The view 
that the sexual organs are developed from segmental anlages 
has been advanced, and such anlages have been named gono- 
tomes. I have shown elsewhere, that this view is based upon 
an incomplete consideration of the facts, and recently C. 
Rabl has fully confirmed my arguments. In Annelids the 
sexual organs are segmented, and are so disposed that we can 
understand how, from an Annelid ancestor, they may have 
given rise to the disposition found in Amphioxus, but, unfor- 
tunately, no careful study of such possible comparisons 1$ 
known to me. Evidently a direct comparison of the sexual 
organs of Amphioxus with those of Vertebrates involve the 
most serious difficulties, while the comparison with Annelids 
promises well. 
