The first method used in the production of larvæ from the unfertilized eggs did not lend itself to an analysis of the activating effect of the spermatozoön upon the egg, since nothing was known about the action of a hypertonic solution, except that it withdraws water from the egg; and there was no indication that the entrance of the spermatozoön causes the egg to lose water. No further progress was possible until another method of artificial parthenogenesis was found. When a spermatozoön enters the egg of a sea urchin or starfish or certain annelids, the surface of the egg undergoes a change which is called membrane formation; and which consists in the appearance of a fine membrane around the egg, separated from the latter by a liquid (Figs. 4 and 5). O. and R. Hertwig and Herbst had observed that such a membrane could be produced in an unfertilized egg if the latter was put into chloroform or xylol, but such eggs perished at once. It was generally assumed, moreover, that the process of membrane formation was of no significance in the phenomenon of fertilization, except perhaps that the fertilization membrane guarded the fertilized egg against a further invasion by sperm. However, since the fertilized egg is protected against this possibility by other means the membrane is hardly needed for such a purpose. [Illustration: FIG. 4. Unfertilized egg surrounded by spermatozoa (whose flagellum is omitted in the drawing). FIG. 5. The same egg after a spermatozoön has entered. The fertilization membrane is separated from the egg by a clear space.] In 1905 the writer found that membrane formation, or rather the change of the surface of the egg underlying the membrane formation, is the essential feature in the activation of the egg by a spermatozoön. He observed that when unfertilized eggs of the Californian sea urchin _Strongylocentrotus purpuratus_ are put for from one and a half to three minutes into a mixture of 50 c.c. of sea water+2.6 c.c. N/10 acetic or propionic or butyric or valerianic acid and are then put into normal sea water all or the majority of the eggs form membranes; and that such eggs when the temperature is very low will segment once or repeatedly and may even--if the temperature is as low as 4°C. or less--develop into swimming blastulæ[87]; but they will then disintegrate. On the other hand, if they are kept at room temperature they will develop only as far as the aster formation and nuclear division and then begin to disintegrate. It should be mentioned that the time which elapses between artificial membrane formation and nuclear division is greater than that between the entrance of a spermatozoön and nuclear division. [87] The reader will find a description of the development of this egg in the next chapter. It was obvious, therefore, that artificial membrane formation induced by butyric acid initiates the processes underlying development of the egg but that for some reason the egg is sickly and perishes rapidly.