1907.] Wheeler, The Polymorphism of Ants. 25 



the anterior region, and especially in the head, while the abdomen is unduly 

 expanded by the worm. Since the thorax of the normal worker, as a por- 

 tion of the body already reduced to a minimum, is incapable of further 

 diminution, it remains unchanged. In the head, however, the results of 

 imperfect nutrition are more strongly felt; there is no longer sufficient sub- 

 stance to form the powerful muscles that go to move the mandibles, so they 

 remain small, and with them the muscle-containing occipital portions of 

 the head are unable to acquire their normal proportions. Hence the char- 

 acteristic narrowing in the posterior region of the head. 



"The explanation of the appearance of the ocelli and the greater devel- 

 opment of the eyes in the parasitized microcephalse seems to me to present 

 more serious difficulties. It might be supposed that, owing to the small 

 development of the occipital region of the cranium, a greater quantity of 

 ectodermal substance had remained over and thereby established more 

 favorable conditions of space and nutriment for the formation of the eyes- 

 and ocelli. Unfortunately, the preservation of the dry specimens did not 

 permit an accurate anatomical study of the organs contained in the cranium. 

 I feel certain, nevertheless, that we are here dealing with something like 

 a general law. 



"The effect of the parasite on the formation of the body, and especially 

 of the head, is not always equally intense: the extent of this effect depends 

 on the relations of the size of the worm to that of the ant. The larger the 

 worm, and hence the greater the volume of the abdomen of the ant, the 

 greater the aberration in the head. For this reason Pheidole shows the 

 greatest alteration, then come the two species of Odontomachus and Pachy- 

 condyla fuscoatra, then Neoponera inversa, and finally Paraponera; in this- 

 gigantic ant the increase in the abdomen had no appreciable influence on 

 the shape of the head." 



This adaptation of Roux's 'Kampf der Teile in Organismus' to the 

 mermithergates seems very plausible at first sight, but on closer examina- 

 tion it proves to be less satisfactory as an explanation of the phenomena. 

 It would seem to be necessary to reach some definite conclusion, first, as to 

 the morphological relations of the mermithergates to the normal castes of 

 the species, and, second, as to which parts of the mermithergates are enlarged 

 at the expense of other parts. In the case of Pheidole there are three possible 

 answers to the first question: the mermithergates may represent small- 

 headed, i. e., defective .soldiers; second, they may be very defective females, 

 and third they may represent excess development of the worker caste. 

 Emery rejects the third possibility, and while considering the two otiiers, 

 is inclined to regard the mermithergates as having arisen from soldier larvae 

 in which the gaster has appropriated the nutriment of the head region so 



