690 ANNUAL REPORT OF THE Off. Doc. 



From these facts it is posisible that the sex may be determined 

 in the ovaries before fertilization. Male eggs do not require fer- 

 tilization, and therefore can develop when laid by a drone-layer, 

 but the female eggs of a drone-layer require fertilization, and since 

 they do not get it they die. I am as yet unable to give an exact ratio 

 between the number of eggs which develop and those that do not, 

 owing to difficulties in observation, but of the fact that some do not 

 develop I am sure. 



Of course, it will be recognized that this is but a theory with a 

 somewhat small basis of fact, but the facts observed seem to me to 

 be enough to throw doubt on the second part of the Dzierzon theory 

 — that sex depends on fertilization. For fear of being misunder- 

 stood, let me repeat that my observations confirm the view that 

 drone eggs are unfertilized, so that the first part of the Dzierzon 

 theory remains unchallenged, as far as I am concerned. The entire 

 subject of the parthenogenetic development of the drones is still 

 but little understood. A few facts are well known, but around 

 these facts there has been woven a mass of good and bad guesses 

 which must be cleared up. If the theory could be stripped of 

 these surmises, the whole subject would be much clearer; and one 

 who undertakes to work on this line must drop all but well-verified 

 facts. 



There is one other line of work on bees in which I have been 

 interested for some time and on which there is yet considerable 

 work to be done. According to the views of the majority of zoolo- 

 gists, the variation of animals is the result of crossing of two lines 

 of heredity. In other words, worker bees would tend to vary all 

 the way between their two parents while drones would tend to be 

 like their single parent. This is certainly logical, but by this time 

 we know that it is not possible to figure out in advance what animals 

 are going to do. To test this I have measured something over a 

 thousand each of drones and workers. In this work I chose cer- 

 tain characters on the wings, for reasons which need not be dis- 

 cussed here. Briefly my results are as follows: Drones vary con- 

 siderably more than workers, rather than less, as we would logically 

 conclude; and furthermore, this variation depends more on the en- 

 vironment under which they are raised than on any inherited ten- 

 dency. Some as yet unpublished measurements confirm this view 

 most strongly. 



I have mentioned but relatively few of the habits of the bee, and 

 if I seem to have taken the view that our present knowledge is 

 meager I hope you will overlook it if you think me pessimistic. The 

 study of the habits of the bee are of the utmost importance to api- 

 culture and since so much remains undetermined, let us hope that 

 many will be enough interested to take up the work. My acquaint- 

 ance among bee-keepers is not as wide as I would wish, but let me 

 say that the best and most successful that I know are the ones who 

 most carefully study their bees. On this account I urge the neces- 

 sity for still more work on the habits. 



I have carefully avoided a discussion of modern appliances in 

 bee-keeping, and especially cut very short any mention of queen- 

 rearing, since this subject will be ably discussed to-morrow by a 

 man who knows that subject better than I do. It is not because I 

 undervalue the practical side of bee-keeping that I have confined 



