April 21, 1905.] 



SCIENCE. 



633 



that show male characters in certain parts of 

 the body and female characters in other parts 

 — the so-called gynandromorph condition — 

 has long been known, and several suggestions 

 have been made to account for the result. 

 The view recently proposed by Boveri has met 



with the most favor. Before stating this view 

 it may be well to recall that in bees it is gen- 

 erally admitted that unfertilized eggs give rise 

 to male individuals (drones) as a rule, while 

 fertilized eggs give rise to females (queens or 

 workers). Boveri has suggested that the 

 gynandrous condition may be due to the 

 spermatozoon failing to unite with the egg 

 nucleus, but subsequently pairing with one of 

 the products of its first division, as shovra in 

 diagram A. If a union of this sort should 

 occur, all of the cells that are derived from 

 the paired nucleus might be expected to give 

 rise to female characters (as in the case of the 

 fertilized egg) ; while all the cells that come 

 from the unpaired half of the nucleus might 

 be expected to produce male parts. Hence 

 the individual that develops from such an egg 

 might be expected to show the characters of 

 the two sexes combined in different ways ac- 

 cording to the positions of the descendents of 

 the two kinds of cells. On purely theoretical 

 grounds I offer an alternate hypothesis which 

 will explain the facts equally well. More- 

 over, if this view, rather than Boveri's, should 

 prove to be the correct one it will furnish im- 

 portant data in regard to the value of the 



spermatozoon in determining the sex of the 

 bee. I venture to suggest this alternation, es- 

 pecially, as the two views can be put to the 

 test of actual observation by any one in posi- 

 tion to obtain the necessary material ; and also 

 because the possibility of this interpretation 

 appears to have been entirely overlooked as an 

 explanation of gynandromorphism. I suggest 

 in brief that the results may be due to two 

 (or more) spermatozoa entering the same egg, 

 one only fusing with the egg nucleus, and the 

 other not uniting, but developing without 

 combining with any parts of the egg nucleus, 

 as shown in diagram B. The products of 

 division of the paired nucleus will account 

 for the female parts of the embryo, while the 

 products of the division of the single sperm 

 nucleus will account for the male characters 

 of the other parts. The assumption of 

 polyspermy on my view is not arbitrary, for it 

 has been often described for the bee and 

 other insects, and seems to be of frequent oc- 

 currence. As a rule it appears that the 

 spermatozoa that do not unite with the egg 

 nucleus fail to develop, but under exceptional 

 cases they may do so. In fact, several cases 

 of ' male parthenogenesis ' in other forms have 

 been described in recent years. 



It will be observed that on Boveri's hypoth- 

 esis the male characters will be derived from 

 the egg nucleus, while on my view they will 

 come from the unpaired sperm nucleus. If, 

 therefore, a queen of one race and a drone of 

 another should produce one of these gynandro- 

 morphs we ought to be able to decide which 

 of these views is correct; for, on Boveri's 

 hypothesis, the male characters of the gynan- 

 dromorph would be those of the race to which 

 the mother belongs, while on my view they 

 would be those characteristic of the race of 

 the father. Thus if an Italian queen-bee 

 were to be fertilized by a German drone and 

 a gynandromorph produced, the male parts 

 should be Italian on Boveri's view, and Ger- 

 man on mine.* 



* I am aware, of course, of the prolonged dis- 

 cussion that has taken place in regard to the char- 

 acter of hybrid bees, but despite these difficul- 

 ties the test might still be made, especially as the 

 drones are generally described as purely paternal. 



