664 JOHN BEARD, 



phenomena in both, and in the peculiarities of their nuclei point in 

 the same direction. 



Lastly, in the later embryos of upwards of 20 mm the practical 

 replacement of the "megaspheres" by undoubted germ-cells in abnormal 

 situations is a fact of like significance ^). 



The writer quite anticipates his interpretations being deemed by 

 some to be absurd: but he is equally prepared to witness the com- 

 plete failure of any attempt — based on actual work! — to prove 

 the "megaspheres" to be anything other than germ-cells. 



The interpretation has been tested in every way; and, finally, 

 adopted as completely explaining the facts. No other interpretation 

 does this. In fact, there is no other construction in existence. 



If the "megaspheres" always occurred in like numbers, sizes, and 

 places in every embryo, they might represent, as Hoffmann thought, 

 stones of the embryonic edifice. They are never alike either in 

 position, size, or number in any two embryos. The writer has seen 

 them again and again in the past twelve years; more recently 

 he has studied them to the period, when as "megaspheres" they 

 cease to exist. Never once has the slightest indication 

 been remarked of any attempt on the part of one of 

 them to form any portion of an embryonic organ. On the 

 other hand, the figures drawn and the many instances studied all go 

 to show the above interpretation to the correct, as it also is the only 

 possible one. 



If "megaspheres" be not germ-cells, then many apparent germ- 

 cells are not such; as examples, those of Figs. 8, 9, 16, 24, 25, 28, and 

 34 to 36. And, indeed, it is to be feared, that we shall never be in 

 a position to determine with any degree of certainty at all, what is a 

 germ-cell in a skate-embryo! 



The final possibility of a genetic connection of "megaspheres" 

 and "merocytes" is disposed of by the twin-nuclear condition of the 

 former, especially as revealed in those which degenerate, thus in 

 Figs. 46 and 47. 



The "megaspheres" within the embryonic area being now identi- 



1) A comparison of the primary germ-cells and megaspheres of 

 Torpedo on the one hand, and those of Raja on the other leads to 

 interesting results. Reserving details, it may be mentioned, that in 

 size and other characters the first or smallest megaspheres of Torpedo 

 correspond exactly with the primary germ-cells of Raja, while the 

 latter cells of Torpedo are smaller. 



