70 Journal New York Entomological Society. [Vol. xvi. 



12. C. planipes Horn. 



13. C. opaculus Horn. 



14. C. ineptus Horn. — Mr. H. W. Wenzel informs me that this 

 species was taken in the same locality as the three preceding species 

 and with the same ant by Mr. H. Kaeber. 



These records show that in the great majority of cases the hosts of 

 Cremastochilus belong to the genus Formica. Although C. spinifer 

 has been taken with P he i dole, C. variolosus with Aphoznogaster and 

 C. squamulosus with a species of Camponotus, it is practically certain 

 that these are accidental or irregular associations. Hence we should 

 expect the beetles to be found only within the geographical range of 

 the genus Formica, and this appears to be the case. Even the four 

 species cited in the " Biologia Centrali-Americana " (C. saucius, 

 planatus, mexicanus and crmitus) occur within the United States, and 

 have been taken in Mexico only on the high northern plateau where 

 Formica also occurs. It is not improbable that some or all species 

 of Cremastochilus prefer to live with particular species of Formica, but 

 this cannot be proved without a great many more records than I have 

 been able to obtain heretofore. 



Haldeman (18483 seems to have been the first to publish an 

 account of the occurrence of Cremastochilus in ant-nests, but the cor- 

 respondence of Hentz, Say and Harris published by Scudder in 1869, 

 shows that as early as 1825 Hentz had seen ants dragging a specimen 

 of the beetle over the ground. Horn (1871) says that he found C. 

 schau mi and angularis "very frequently in ants' -nests and in one in- 

 stance apparently eating the pupae. Several times I have seen large 

 black ants dragging specimens of schaumi along the surface of the 

 ground towards their nests and on examination have frequently suc- 

 ceeded in obtaining from nests specimens that had previously been 

 dragged there. Why these insects are found with ants is a question 

 to which I am not prepared to give a definite answer, unless, as I 

 suspect, the fossae at the anterior angles and the finely punctured and 

 apparently perforated patches under the hind angles are glandular and 

 yield some secretion grateful to the ants." The patches to which 

 Horn alludes are tufts of golden yellow or orange-red hairs, the trich- 

 omes, which characterize so many myrmecophilous insects, and are 

 most beautifully shown on various parts of the body in beetles of the 

 genera Adranes, Claviger, Paussus, Lomechusa, Xenodusa, etc. 



In 1886 Horn published some additional notes on the habits of 



