132 BULLETIN OF THE 



eggs of April 2d were only about 0.4 mm. in diameter, and therefore 

 scarcely exceeded in size those of October and November, the largest 

 ovarian eggs of May 2d measured over 0.8 mm. Between May 1 and 

 June 1, — by which time the eggs have reached their full size, — the 

 growth is still more rapid. The egg membrane of early May eggs meas- 

 ures about 6.5 fx in thickness, and has distinct pore-canals. 



There exists an exceedingly thin outer membrane overlying the zona 

 radiata. It was discovered in the examination of fresh ripe eggs, in 

 which the striation of the zona itself could be seen much better than 

 in sections of hardened eggs. In one instance, in which the zona of a 

 fresh ripe egg was ruptured, this overlying membrane was left intact. 

 It is with this membrane that the bases of the filaments are continuous. 



In view of this condition in Fundulus, and of the fact that other pro- 

 cess-bearing eggs (Cyprinidse and Gasterosteidse) possess a thin outer 

 membrane, it would be interesting to re-examine the eggs of the Scom- 

 beresocidse, of Menidia, and of Mugil to find whether they do not also 

 possess this structure. 



The outer surface of the fresh ripe egg of Fundulus heteroclitus shows 

 a network of lines (Fig. 7). This appearance is doubtless due to the 

 presence of superficial ridges, which in radial sections have the appear- 

 ance of minute projections fitting in between the bases of the granulosa 

 cells (Fig. 9). Where two or more lines meet, there is a thickening. 

 The whole arrangement bears a superficial resemblance to the appear- 

 ance presented near the surface of the zona in the perch (Fig. 31). In 

 the case of the latter, however, the thickenings correspond in position 

 to filaments, each of which corresponds to the middle of a granulosa 

 cell, whereas in Fundulus heteroclitus the thickenings correspond in po- 

 sition to the boundaries between granulosa cells. From the position of 

 the filaments in Fundulus it is probable that, like the ridges, they are 

 outgrowths of the outer structureless egg membrane. It is evident from 

 what has been said that there is a fundamental difference between the 

 filaments found in Perca and those in Fundulus. In Perca they owe 

 their origin to the granulosa, and are formed after the zona has nearly 

 reached its full growth ; in Fundulus, on the contrary, they owe their 

 origin to the activity of the egg itself, and they begin to be formed 

 before the zona. 



Pygosteus pungitius. 



After Haeckel had described the long filaments peculiar to the eggs 

 of the Gobiesocida3, Kolliker ('58) described external appendages in the 



