86 BULLETIN OF THE 



and a half days, shows the first evidences of union between the edges of 

 the lateral plates, and Figure 11 shows their fusion almost completed. 

 The union of the phites first takes ])lace in the region of the otocyst, and 

 thence proceeds both posteriorly and anteriorly, but it does not proceed 

 as rapidly forward as it does backward. The fusion is effected first 

 at the dorsal part of the fundament of the heart. The completion of the 

 closure I have traced in many cases, but as it has little bearing on Ihe 

 present question, I give no figures of the completed heart. It will be 

 seen in Figures 10 and 11 (Plate III.) that the cells, i. c. m., which have 

 been shown to be present between the approaching lateral plates in all 

 stages, are enclosed between the facing walls of the plates when the 

 union of the latter is completed. It is these enclosed cells, now in a 

 solid mass, which after a time arrange themselves as a lining to the thick 

 walls of the heart by the gradual appearance of a lumen in the middle 

 of the mass. They form the endocardium, and it is the origin of these 

 cells, therefore, which is still in controversy. This is the problem to 

 which we now have to direct our attention. 



As Riickert ('88) has pointed out, many investigators have erred in 

 not giving attention to these cells at a suflBciently early stage in their 

 history. 



In the later stages there is nothing but the form of the individual 

 cells and their position on whicli to base a theory of their origin, — the 

 first being uncertain, the second delusive, as investigation of early stages 

 shows. 



Observers have seen this mass of cells lying between the two lateral 

 plates at a stage when the plates were too well defined to allow one 

 to attribute to them the production of the mass of cells in question ; 

 since they are, beside?, distinctly separated from the rest of the meso- 

 derm, they have been supposed to originate from the less shai'ply defined 

 cells of the endoderm, or from the parablast. 



Rabl ('8G) stated it as his belief that the cells, judging from their po- 

 sition, are endodermic. He however said that, before making a final 

 statement, he should like to investigate the matter further. Hofl^man 

 ('82), on the other hand, is less guarded in his conclusions ; he announces 

 in his work on Bony Fishes that the cells both in jiositton and form are 

 so much like the parablast, nnd in places are in such close relation with 

 it, that in his mind there is no doubt about their arising as a prolifera- 

 tion of parablastic cells. 



Returning now to the case of the cod, there is no difficulty in tracing 



