June, 1908. J 



Wheeler: Studies ox Myrmecophiles. 



7:: 



about. Then the ants at once pounced upon them with open mandi- 

 bles and curved their gasters toward between their hind legs, in the 

 attitude assumed by all the forms of F. rufa when irritated and about 

 to discharge their formic acid. There could be no doubt that the ants 

 were hostile and irritated but nevertheless the acid batteries were 

 never called into action, probably because there is a disinclination to 

 use these organs within the narrow con- 

 fines of the nest. They contented them- 

 selves with biting the beetles and pulling 

 their legs. It soon became apparent, 

 however, that the biting was largely re- 

 stricted to the posterior corners of the 

 beetle's thorax. The ant clung to the 

 side of the beetle in the position shown 

 in Fig. 2, seized the posterior angle be- 

 tween its mandibles in such a manner that 

 the toothed border of one mandible fitted 

 into the arcuate groove which separates 

 the angle from the more convex portion 

 of the thorax, and then moved its head 

 from side to side in a vigorous attempt to 

 tear away the triangular piece of chitin 

 bearing the cushion of trichomes on its 

 ventral surface. Several of the workers busied themselves thus for peri - 

 ods varying from one to twenty minutes, and one pertinacious indi- 

 vidual gnawed at the posterior thoracic angle of the same beetle for an 

 hour and a quarter. The anterior angles and the legs were also gnawed 

 and pulled, but much less frequently and persistently. With a pocket 

 lens it was possible to study the mouth-parts of the gnawing ant. 

 There was no attempt to lick any portion of the beetle's body. In 

 fact, the ant, while gripping the thoracic angle, kept its mouth tightly 

 closed and held its maxillary palpi back against the gula while the 

 labial palpi, with a rhythmic movement approached, but did not ac- 

 tually touch, the trichomes. The gnawing seemed to irritate or, at 

 any rate, to stimulate the beetle, for it stalked rapidly about the nest 

 carrying the ant on its back. Whenever the anterior part of its body 

 was touched by an ant, it quickly retracted its antenna, so that it 

 could never be seized by these appendages. Eventually it managed 

 to force its head and thorax into a corner or under the wet sponge. 



Fig. 2. Cremastockilus casta- 

 neee with Formica Integra worker 

 gnawing at one of its posterior 

 thoracic angles. X 4- 



