3 2 4 AUGUST WE ISM ANN [apkil 



course of the score of years which have almost elapsed since I first 

 brought them forward, simply because I did uot then know all that I 

 know now. I may be permitted to give an example. 



In 1885 I directed the attention of biologists to the importance of 

 the " polar bodies " in relation to the theory of heredity. The persist- 

 ence with which these bodies, in themselves apparently insignificant, 

 appear throughout the whole animal kingdom seemed to me to indicate 

 that they played an important physiological role which must be associated 

 with their liberation from the egg, and in connection with the theory 

 of the continuity of the germ -plasm — then only developed in rough 

 outline — I suggested that they might have to do with the extrusion of 

 the " ovogenetic " idioplasm, that is, with the removal of the nuclear 

 substance determining the histological structure of the egg-cell. 



This supposition of mine — my first hypothesis concerning the im- 

 portance of the polar bodies in relation to the theory of heredity — was 

 opposed to the opinions expressed earlier by Minot and E. van Beneden, 

 who regarded the immature ovum as a hermaphrodite cell which became 

 female only after the extrusion of its male elements, the polar bodies. 

 Observation showed me that even parthenogenetic eggs form polar 

 bodies, and consequently Minot's hypothesis was overthrown, while my 

 own was strengthened. But further investigations by Blochmann and 

 myself led to the discovery that parthenogenetic eggs only form one 

 polar body, while those which require fertilisation always form two ; 

 and then, supported by E. van Beneden's fundamental discovery of the 

 numerical equality of the paternal and maternal chromosomes in the 

 fertilised egg, I modified my first hypothesis so far as to suggest that 

 the " ovogenetic " idioplasm is expelled in one of the polar bodies, while 

 the other implies a reduction of the ancestral germ-plasms, or ids as I now 

 call them. The necessity of this process I attempted to prove, and 

 postulated for its accomplishment a special mode of nuclear division, 

 a reducing division, in which the usual longitudinal splitting of the 

 chromosomes is suppressed, so that only half the number of these migrate 

 to each daughter-cell. What was at the time only a theoretical con- 

 clusion merely indicated by the observations then on record and not 

 deducible from or capable of proof by them, has been fully justified by 

 later investigations. There is a reducing division. 



In this second hypothesis, however, an error was also involved, for 

 Boveri's l observations on the egg of Ascaris have shown that the 

 chromosomes expelled from the ovum in the first polar body are not 

 essentially different from those remaining in it, because in certain 

 circumstances they are interchangeable with them. I therefore altered 

 my second hypothesis to a third, according to which both divisions of 

 the ripening ovum together imply a reduction in the number of ids, 



1 "Zellenstudieu. fiber das Verhalten der chromatischen Kernsubstanz bei der Bildung 

 der Riclitungskorper und bei der Befruchtung," Jcnaische Zeitsehr. f. Naturwiss. 1890, 

 Bd. xxiv. p. 314. 



