INSTINCTS OF NEUTER INSECTS. 237 



and treatment ; yet, though one would be sorry to set 

 limits to the convertibility of food and genius, it seems 

 hard to believe that there can be any untutored food 

 which should teach a bee to make a hexagonal cell as 

 soon as it was born, or which, before it was born, should 

 teach it to prepare such structures as it would require 

 in after life. If, then, food be considered as a direct 

 agent in causing the structures and instinct, and not an 

 indirect agent, merely indicating to the larva itself 

 that it is to make itself after the fashion of neuter 

 bees, then we should bear in mind that, at any rate, it 

 has been leavened and prepared in the stomachs of 

 those neuter bees into which the larva is now expected 

 to develop itself, and may thus have in it more true 

 germinative matter — gemmules, in fact — than is com- 

 monly supposed. Food, when sufficiently assimilated 

 (the whole question turning upon what is " sufficiently"), 

 becomes stored with all the experience and memories 

 of the assimilating creature; corn becomes hen, and 

 knows nothing but hen, when hen has eaten it. We 

 know also that the neuter working-bees inject matter 

 into the cell after the larva has been produced; nor 

 would it seem harsh to suppose that though devoid of 

 a reproductive system like that of their parents, they 

 may yet be practically not so neuter as is commonly 

 believed. One cannot say what gemmules of thigh and 

 proboscis may not have got into the neutral bees' 

 stomachs, if they assimilate their food sufficiently, and 

 thus into the larva. 



Mr. Darwin will be the first to admit that though a 

 creature have no reproductive system, in any ordinary 



