210 Journal New York Entomological Society. [ Vo1 - XV1 - 



fringes. The imaginal insects are treated as synechthrans, or perse- 

 cuted intruders, and are quite defenceless. The larval and probably 

 also the pupal Cremastochilus are synechthrans, or at any rate the ants 

 would destroy them if they did not protect themselves by hiding away 

 in the soil. The adult beetle may become indifferently tolerated, 

 though it seems generally to be treated with hostility. Still it is well 

 protected by its hard armor, so that it walks about the nest with im- 

 punity like the larval Microdon. Thus each of these myrmecophilous 

 genera has its Achilles heel. At first sight this seems to show a lack 

 of adaptation in a portion of the life cycle (larval and pupal stages 

 in Cremastochilus, imaginal stage in Microdon), but a broader view 

 suggests that what appears as a defect is really an advantage, both to 

 the parasite and its host, for it prevents the former from exceeding a 

 proper numerical relation to the latter. In other words, it is better 

 for Cremastochilus and Microdon to sacrifice individuals and thus 

 become rather rare forms, than to over-run the colonies and seriously 

 disturb the domestic economy of their hosts. 



Addenda. 



To the account of Cremastochilus and Hetozrius given in the former 

 studies of this series, I would add the following notes : 



During the past summer I succeeded in finding the larvae of Cre- 

 mastochilus casta nea; in a nest of Formica fusca var. glacialis Wheeler, 

 at South Harpswell, Me., and in rearing the beetles. I have described 

 the larval and pupal stages in a recent paper (The Ants of Casco Bay, 

 Maine, With Observations on Two Races of Formica sanguinea, Bull. 

 Amer. Mus. Nat. Hist., XXIV, 1908, pp. 619-645). 



In my summary of the records of Cremastochilus, I unfortunately 

 omitted an important paper by Wickham (On Coleoptera Found with 

 Ants. Fifth Paper. Psyche IX, 1900, pp. 3-5). This author cites 

 C. harrisi as occurring with Formica pallidefulva nitidiventris Emery 

 (Lowa, May 4) and C. saucius and knochi with Pogonomyrmex occi- 

 dentalis Cresson (opacicefs Mayr). C. saucius was taken by E. J. Oslar 

 at Denver, Berkeley and Salida, Colo. (April 10-July 11), C. knochi 

 at Berkeley (April 8) and Chimney Gulch, Colo. (June 18). Oslar 

 is quoted as stating that the former species occurs singly in the occiden- 

 talis hills at all times of the year, and Wickham adds that " it is diffi- 

 cult to get perfect specimens of this beetle, the legs often being much 

 mutilated supposedly by the ants, which are large and fierce creatures. " 



