174 DYNAMICS OF LIVING MATTER 



face film will not occur; only when the secretion of the liquid is rapid 

 enough will the secretion result in the membrane formation. According 

 to this view, the secretion of a Uquid from the egg is the essential feature, 

 while the membrane formation itself is possibly only a secondary, mechan- 

 ical effect of the sudden secretion. If this be true, the essential feature 

 in fertilization in Echinoderms is not the membrane formation itself, but 

 the secretion of a liquid from the interior of the egg. This conception 

 is corroborated by an observation I made several years ago. I found 

 that the unfertilized eggs of a sea urchin could be kept ahve in sterihzed 

 sea water for a week, or possibly more. When sperm was added to 

 such eggs, they developed, but without the formation of membranes. It 

 is quite possible that a process of secretion may be produced in every 

 egg through the entrance of a spermatozoon, while the actual separation 

 of the surface film of the egg from the protoplasm is only a secondary 

 mechanical consequence of this secretion, which may or may not occur. 

 Since I have only recently recognized the importance of the process 

 of membrane formation for the complete physicochemical imitation of the 

 developmental effect of the spermatozoon, I have not yet found time to 

 see whether it holds good only for Echinoderms. An observation re- 

 cently made by Professor Lefevre on artificial parthenogenesis in a 

 marine worm, seems to indicate that the artificial membrane formation, 

 or rather the process underlying it, is of more general importance. 

 Lefevre found that about 50 per cent of the eggs of Thalassema develop 

 into normal larvae, after having been exposed to sea water (to which a 

 little acid had been added) for a few minutes. After they were taken 

 out of the acidulated sea water, they formed a membrane and developed. 

 The case seems to be similar to that of Asterina. The development of 

 the eggs seemed to be normal, and the vitality of the larvae seemed to 

 be the same as that of the larvae originating from fertilized eggs. I, as 

 well as others, had, before Lefevre's observations, produced artificial 

 parthenogenesis in the eggs of worms, but without artificial mem- 

 brane formation. In all these cases the larvae had always a diminished 

 vitality, and the development was often different from that of the egg 

 fertihzed by sperm. I had found that the unfertilized eggs of Chatopte- 

 rus, a marine Annelid, can be caused to develop into swimming larvae* 

 with certainty by adding a small but definite amount of a soluble potas- 

 sium salt ; but the vitality of these eggs was considerably less than that 

 of the larvae originating from fertihzed eggs. I may also add — although 

 this does not belong to our problem — that I noticed that the eggs of 

 Chatopterus, which had been caused to develop parthenogenetically by 

 KCl, reached the trochophore stage and began to swim about seem- 

 * Loeb, Am. Jour. Physiology, Vol. 4, p. 423, 1901. 



