-64- 



circumstances give but imperfect offspring, and that there would be a 

 very wide range in the degrees of the imperfection of the f)lants produced 

 from these imperfect kernels. 



As a matter of fact, the farmer in planting, selects with care the 

 most perfect ears, and the most perfect parts only of the ears so selected, 

 and yet we have the males, the females, and the neuters or undeveloped, 

 for the result, as I have described them. 



Indian corn is so extremely variable in this matter of sex, that care- 

 ful experimenting in this direction would be likely to give most interest- 

 ing results in a single lifetime. 



Having now illustrated some principles of variability, and given 

 some idea of the extent to which it may go, under our own observation, 

 we must deal with the question before us by way of hypothesis. 



Let us suppose a primitive or typical Bee among the honey seeking 

 insects of early days. She is necessarily a creature having such attributes 

 as Txre common to all species of bees which are her offspring, but in 

 many respects she is very unlike our Hive Bee of to-day We see her 

 at a time when this typical species has already learned the wonderful 

 lesson of thrift. She stores honey in times of plenty to provide for times 

 of want. She is feeding her offspring from her stores. As the keen com- 

 petition of life goes on, she must provide for the wants of her offspring 

 for an ever increasing period, and, as her powers in this respect are taxed 

 to the utmost, her powers of reproduction are of necessity diminished; 

 .she produces some imperfect eggs, and she produces fewer eggs. Still 

 the vast majority of her offspring perish, either for lack of sufficient food 

 or as prey to natural enemies befc)re their power of self defence are suffi- 

 ciently developed for successful flight or resistance. 



It is quite reasonable to suppose that the bee has been subjected to 

 such vicissitudes as these. The extraordinary differences in the sizes of 

 the various living species of bees would indicate the truth of the theory 

 of insufficient food as far as we have yet followed it. If we have a species 

 of bee only one-eight of an inch in length w-hile some others are an inch 

 and a quarter in length and stout in proportion, it will take one thousand 

 (looo) of these Lilliputian bees, to weigh as much as a single specimen 

 of one of these largest species. Is it not most reasonable to suppose that 

 this tremendous variation in size, is chiefly due to the matter of food 

 supply, as is the well known fact in the very laige variation we can thus 

 make in the size of an individual fish? 



Now when the food supply is so very scant that the size of the off- 

 spring is necessarily much dwarfed, evidently the weakest will die in the 

 process of rearing; evidently also, the mother-bee whose reproductive 



