234 s. HATTA; FORMATION OF THE 



free edge may help its ventral growth, although I have never observed 

 any thing which ^vonld justify such a conclusion. But in the anterior 



portion where the lateral plates of the invaginated gastral mesoblast 

 early reach to, and meet, in the ventral median line, there is no need 

 of free cells to be added to the ventral edges to lielp their growth. 

 As far as my observations go, these cells are all used in building 

 up the vascular s^'stem. It seems most proljable that the lateral and 

 A^entral parts of the mesoblast are formed solely by the downward 

 growth of the dorsal part ; for many mitotic figures are, even in much 

 later stages, seen in the mesoblnst, so that it needs no contribution 

 from the free cells in question. 



Fi-om its first appearance, the heart is a single endothelial tube ; 

 there is no reason to think it has a paired origin. 



My deepest thanks .are due to my esteemed teacher. Prof. K. Mitsu- 

 kuri, for his kindness in looking through the manuscript of this 

 paper. 



Biological Lahovatorij, tlw Galaishuin (Kohles School), April 1897. 



