FERTILIZA TION 1 73 



a small percentage of the eggs of the starfish are able to develop without 

 any external cause or agency being applied. The number of these eggs 

 varies in the eggs of different individuals, but is, as a rule, very small, 

 e.g. a fraction of i per cent. The rate of segmentation in these "natu- 

 rally" parthenogenetic eggs is slower than the rate of development of 

 the eggs fertilized by sperm, and the blastula begins to swim consid- 

 erably later than the blastulae coming from fertilized eggs or from eggs 

 in which a membrane had been produced artificially. Moreover, the 

 blastulae of the spontaneously developing eggs differ somewhat in ap- 

 pearance from the blastulae coming from the two latter. 



Neilson and I found that the number of eggs which develop without 

 a membrane formation can be increased by treating the eggs transitorily 

 with acidulated sea water. Delage* simultaneously obtained the same 

 result by treating the eggs of Asterias with CO^ ; I am inchned to be- 

 lieve that the CO^ acts as an acid, although Delage is not willing to admit 

 this. 



The fact that the unfertilized eggs of the starfish may develop with- 

 out any external cause has often been overlooked, and this has led some 

 authors again to state that any "stimulus" may cause the development 

 of this egg. Acids, indeed, increase the number of eggs which will 

 develop; the same is possibly true for mechanical agitation, as A. P. 

 Mathews has observed.f He is inchned to beheve that in this case the 

 mechanical agitation is the direct cause of development (by producing 

 coagulation). It is, however, necessary to state that the taking up and 

 dropping of eggs with a pipette suffices. In eggs of Amphitrite, an 

 Annehd, I have convinced myself that the number of eggs which develop 

 does not bear any relation to the extent of the mechanical agitation. I 

 consider it possible that some secondary factor connected with the agi- 

 tation, such as the diffusion of gases into or from the egg, e.g. CO^, m,ay 

 be the real factor involved in this case. 



The experiments thus far mentioned indicate that the process of 

 membrane formation, or some process underlying this, is of importance 

 for the complete physicochemical imitation of the developmental in- 

 fluence of the spermatozoon. The question arises, What is the nature 

 of this process? It seems to me from my observations on Echino- 

 derms that the essential feature of this process is the squeezing out 

 under pressure or the secretion of a fluid from the protoplasm of the 

 egg. As a mechanical effect, the surface film of the egg is lifted and 

 separated from the protoplasm by a hquid secreted by the egg. When 

 the secretion of this hquid occurs very slowly, the liftmg up of the sur- 



* Delage, Archiv. de Zool. experimmtale. Vol. 10, p. 213. 

 t A. P. Mathews, Am. Jour. Physiology, Vol. 6, p. 142, 1901. 



