22 Bulletin American Museum of Natural History. [Vol. XXIII, 



of laying its eggs. Or the eggs may, perhaps, be discharged into the ali- 

 mentary tract of the ant and escape with its excrement. It is not difficult to 

 conceive how they may find their way into the young larva. They may 

 adhere to the body of the workers till gathered up by their strigils or tongues 

 and transferred to the buccal pocket. Thence they could be fed inadvert- 

 ently to the larvae, either with the regurgitated liquid food or with pieces of 

 insects that had been malaxated and distributed by the workers. 



The living mermithergates were easily studied in my artificial nest and 

 exhibited several interesting peculiarities in their behavior. On exposure 

 to the sunlight they hurried, like the females, to the dark chamber, thereby 

 evincing a much higher degree of negative phototropism than the workers 

 and soldiers. They never carried the brood, although even the soldiers 

 were seen to do this occasionally. They never fed the larvae, workers or 

 females, and, had earth been present in the nest, it is very probable that 

 they would never have shown any inclination to excavate. They never 

 visited the manger of the nest but were fed exclusively by regurgitation. 

 As befitted animals containing such enormous parasites they were in a 

 chronic state of hunger. It was impossible at any time to uncover the nest 

 without finding one or more of them either being fed by a worker or eagerly 

 begging for food. And as soon as one was offered food, three or four of 

 them would rush up and put out their tongues for a share of it. Once I saw 

 a single worker trying to feed five of her huge parasitized sisters simultane- 

 ously. While imbibing their food the mermithergates stridulated, either 

 continuously or at regular intervals. Sometimes they were so impatient 

 to be fed that they would hold down a worker's head with their large fore 

 feet and compel the little creatiu'e to regurgitate. Under these circum- 

 stances the larvtie must have been poorly fed, for the insatiable mermither- 

 gates continually intercepted the workers on their way from the manger 

 to the brood chamber. The mermithergates in my artificial nest may 

 have been massacred or starved by the workers and did not die merely as 

 a result of parasitism for when the food supply becomes insufficient the 

 tiny Ph. commutata workers, like those of Ph. instahUis, probably rid the 

 colony of voracious and non])roductive mendicants, even when these belong 

 to their own species. Apart from their chronic hunger, the mermither- 

 gates exhibited no abnormal traits of behavior directly attributable to 

 parasitism. They ran about with as much alacrity as the workers and 

 soldiers. I^'heir other peculiarities, such as their strong negative photo- 

 tropism, their dependence on regurgitated food, and the absence of the 

 foraging instinct, like their large size, their ocelli aufl rich red color, may be 

 regarded as female or soldier traits. 



Mermithergates occur also, I believe, in two other Texan ants belonging 



