55__ 



mesentery, i. e., the splanchnopleure, it varies considerably, and may 

 be as low as 2 or 3, and as high as 16. In the younger embryos, 

 liowever, it is difficult or impossible in many instances to determine, 

 whether many of the germ-cells should be entered in the lists as nor- 

 mally placed, or as upon the mesentery; and, lastly, as others, in- 

 cluding Rabl, have noted, they may be met with in the nephridial 

 tubules, or within the myotomes. The majority of these have been 

 regarded in my lists as normally placed germ -cells; for it was not 

 part of the programme to increase the number of vagrant germ -cells 

 by limiting too narrowly the positions of their normal distribution 

 and situation. 



Surveying broadly the observations yet made in embryos of Pristi- 

 urus of 5 — 16 mm, they reveal a percentage of vagrant germ -cells 

 as low as 2"/o and as high as 11%, whilst it is still higher in the 

 youngest embryos of the series. No single embryo examined 

 was devoid of such germ -cells in impossible places! 

 The percentage is not as high as in Raja, but against this must be 

 set the smallness of the embryos of Pristiurus and the less number 

 of germ-cells. Even in the 13 embryos studied germ-cells have been 

 found in almost all the situations already recorded for Raja batis. 

 None have been seen far forward in the head, or in the nervous 

 system, but almost everywhere else in some one or other of the em- 

 bryos they have not been wanting. 



The history of the germ-cells in embryos younger than those above 

 treated of is of great interest. Here, as in Raja, the answer to the 

 question "how do the peculiarities in the distribution of the germ- 

 cells in older embryos come about?" is given by the study of very 

 early phases in the development of the embryo. Rabl ('96, p. 748) 

 has already recorded the presence of germ-cells in Pristiurus- embryos 

 of 18 somites; but, while recognising the great importance of their 

 very early appearance, his finds are slight. 



The number of Pristiurus-embryos of early periods as yet studied 

 by the writer is very limited, and it is not such as to bridge over com- 

 pletely the gap between the conditions in later embryos and our know- 

 ledge of the egg-cleavage, which we owe to Rückert's researches ('99). 

 Of such embryos three may be here described. 



The largest of these (embryo no. 6, 4 mm, circa 40 somites) throws 

 light upon one of the smallest (no. 12) of the 13, previously referred 

 to. Embryo no. 12 is 5,5 mm in length, and possesses about 55 so- 

 mites posterior to the last gill -pouch. In it 131 germ -cells were 

 counted. Of these 29 were in abnormal situations. But in the enu- 



