456 ANNUAL KEPORT SMITHSONIAN INSTITUTION, 1925 



Mr. Kershaw noted that the plants upon Avhich this species lays 

 its eggs are overrun bj^ a host of ants of two species {P olyrrhachis 

 dives and Dolichoderus bituberculatus) , both aphids and ants appar- 

 ently feeding on juice exuding from the plants. An}^ vegetation 

 subject to these secretions seems to be attractive to them, the ants 

 not appearing to use the aphids as ant cows, though of this he was 

 not quite certain. 



The ants do not appear to meddle either with the butterflies or with 

 their eggs, though ants are A^ery destructive to the eggs of most 

 butterflies; nor do they seem to interfere with the caterpillars. 



In contrast to the immunity of this butterfly from the attacks of 

 ants, Mr. Kershaw saw two small butterflies of other kinds (a skip- 

 per and a species of Neptis) seized by the tongue as they probed a 

 flower and dragged off b}' one of the same species of ants among 

 which the Gerydus live. 



In 1914 Mr. Edward Step published a full page illustration of 

 "A Butterfly Enemy of Green-fly " which was based upon the plate 

 and tlie data in Mr. Kershaw's article, though the larvae are shown 

 upon bamboo instead of on Bidens pilosa. 



Megalopalpus ztmna 



Mr. W. A. Lamborn Avorked out the life history of this species in 

 the Adcinity of Oni Camp, Southern Nigeria. Oni Camp is 70 miles 

 east of Lagos, at a low elevation, never more than 50 feet above 

 sea level. 



The caterpillars of this butterfly feed not upon aphids, but upon a 

 variety of Homoptera belonging to the families Jassidse and Mem- 

 bracidse which are invariably attended by ants. 



The mother butterfly, in depositing her egg, which is a very charac- 

 teristic one, exercises the same care in insuring an immediate food 

 supply for the newly hatched larva as do other butterflies for their 

 plant-eating offspring. She places it very commonly in the imme- 

 diate neighborhood of an ant shelter containing Homoptera, and 

 an eg;g shell is sometimes found attached to a stem actuall}'^ within 

 a shelter, having obviously been deposited before the Homoptera 

 attracted the attention of ants, and, indeed, probably on the egg 

 mass itself before hatching, since the membracid and jassid colo- 

 nies seem to remain and feed close to the spot where the parent 

 laid her eggs. 



The egg masses of the jassids {Nehela ornata) are attached to 

 plants in clusters much like those of the membracids {Leptocentrus 

 altifrons) ; that is, in parallel rows often superposed so as to form 

 oval masses. The lycpenid larva? do not interfere with these. 



