Beard, Heredity and the epicycle of the germ-cells. 359 



zoon is fully differentiated, and, of course, before there is any trace 

 of the embryo. 



For reasons to be fully given in my memoir on the germ-cells 

 the division of U. K. Z. the primitive germ-cell, is considered to go 

 back to about the tenth cleavage products, and in the skate there are 

 either eight or nine divisions. 



The publication of the present writing has been delayed for several 

 months, in' order that time might be gained for the tabulation and 

 counting of the primary germ- cells in a series of embryos. This has 

 now (March, 1901) been done in 18 embryos of Raja batis and in 8 

 of Scyllium canicula. 



The number of primary germ- cells in the embryology of R. batis 

 may be taken at 256 in the male and 512 in the female. 



It may be added, that the number appears to be much smaller 

 in Rana esculenta and in Petromyzou planeri. In the former 8 and 

 in the latter 32 primary germ-cells would seem to arise. 



These latter numbers have not yet been confirmed on a material 

 large enough to afford any certainty of their correctness. 



The division of the primitive germ-cell into primary germ-cells is a well- 

 marked epoch in the life-cycle, and one of the greatest possible moment. 

 Hitherto its import has been overlooked by every ernbryologist, and the 

 record of it is now made for the first time as the result of my work. 



From every point of view it is as important as the phenomena of 

 maturation; and, probably, its essential necessity in development will 

 not need to wait long for ample recognition. 



The number of the products of the primitive germ-cell is very 

 large in the skate, as many as 512. But it must be pointed out, that 

 this number furnishes no criterion for other animals. There may be 

 cases, in which it is larger; though, I imagine, the occurrence of many 

 such is unlikely. Undoubtedly there are instances, in which it is much 

 smaller: and, probably, these are well represented among the Inverte- 

 brata. In short, it may be as low as two; but, as the sexual gener- 

 ation or embryo must arise from one product, and as this must con- 

 tain some sexual elements, it can never be lower than two. In other 

 words, the primitive germ-cell must divide at least once, yielding two 

 primary germ-cells, of which one will give rise to the embryo and the 

 other will supply the ,,sexual products". Apparently it divides once 

 in Cyclops and Ascaris megalocephala, twice in Cecidomyia, and thrice 

 in Chironomus 1 ). 



In other chapters of my work the essential similarity, the equi- 

 valence of all the primary germ-cells, whether their number be 2, 16, 

 128, 512, or anything else, has been insisted upon. The point is one 

 of the utmost importance, and ; therefore, it may be well to once more 

 briefly indicate the grounds for the conclusion. 



