464 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1925 



Concerning the loose scales on this unique butterfly Mr. Dodd 

 says we have no evidence that the larvie are welcome inhabitants of 

 the ants' nests. However, it is highly probable that the ants have 

 no friendly feeling for the perfect insect, and would most likely 

 attack and kill it during its long rest after emergence if it were not 

 specially and wonderfully protected. So it will be seen that the 

 loose scales act as a perfect protection; for directly the ants en- 

 counter these they are in trouble; they fasten on to their feet and 

 impede their movements or, if their antenna3 or mandibles come into 

 contact with any part of the butterfly the scales adhere thereto, so 

 that the ant is soon in a bad way and has quite enough to do in 

 attempting to free itself of its encumbrances without taking any 

 further interest in the butterfly, from which it retreats as well as 

 possible. It is exceedingly ludicrous to observe the ants endeavor- 

 ing to free themselves; their legs move awkwardly, and their mandi- 

 bles are opened and closed in evident annoyance and perplexity, and 

 they are much concerned at the state of their antenna?, for the 

 obnoxious scales will not be shaken off. It is amusing to observe 

 the dejected change in an ant after its first spar with L. hrassolis^ 

 for it is such a pert, pugnacious fellow and perfectly willing to 

 tackle anything that moves if in proximity to its pets or nest. 



The small wings of the insect enable it to get through the nest 

 entrance. The scales on the forewing are necessary to its safety 

 while it is crawling out, for the ants might in some cases evade its 

 legs and get onto the thorax. But if the scales there did not van- 

 quish them, those on the forewings would. As the butterfly's 

 abdomen becomes strong enough it raises the tip to touch the sup- 

 port where it is hanging. In that position it is invulnerable; hun- 

 dreds of ants could not hurt it. As the forewings lengthen and 

 touch there is no further need of the scales thereon. The wind 

 doubtless dislodges the majority before the insect flies off, but the 

 other scales would not wholly disappear for some time. 



Mr. Dodd placed four larvae on a nest where he knew there were 

 none, and afterwards visited it and obtained from it two pupse. 

 On another small tree with several ant nests seven larvae were 

 placed, and going there some days later he found one chrysalis on 

 the outside of a nest, one within, several attenuated larvae wandering 

 about the tree — the ants, having become familiar with them, taking 

 no notice of them — and a dead larva on the ground. It would ap- 

 pear that the entrance to the nests was too small for the larger 

 caterpillars to gain admittance. Once he saw a larva on the outside 

 of a new nest, and he took a pupa on another. These were in locali- 

 ties where he had not interfered with the ants. These instances 

 serve to show, according to Mr. Dodd, that the larvae pass from one 

 domicile to another, presumably when their food is exhausted in 



