THE ANTS OR PISMIRES 



workers produced can be similarly altered. With one nurse 

 for each grub, provided enough food is given to the colony, 

 all may become queens, whereas with one nurse for ten 

 grubs, all will become workers. A similar result can be 

 obtained by partial starvation of the half-grown grubs and 

 in this way, in certain conditions, intermediates between 

 queens and workers were obtained. 



These experiments have not yet been described in full, 

 and it is not known why the grubs vary so much in size at 



Fig. io 



Male-worker mosaic of the ant, 



Formica sanguinea, found at 



Bewdley, Wore, 20.vii.1909 



(Bondroit). Right half male. 



the beginning of the winter. There is always the fundamental 

 difficulty that worker ants may recognise the caste of newly 

 hatched grubs although we cannot do so. This might lead 

 them to treat each type of grub in the appropriate way. On 

 the other hand, in the leaf-cutter ants, as Wheeler pointed 

 out, fungus appears to be almost the only food of all grubs 

 which have access to an almost unlimited supply, and yet 

 the workers are highly polymorphic. 



What has been regarded as the best evidence for the 



M3 



