208 Journal New York Entomological Society, [vol. xvi. 



very nests in which they have passed their larval and pupal stages. 

 Verhoeff (1892), however, has seen the adult flies hovering about and 

 apparently attempting to oviposit among the ants on the outside of the 

 nest. 



M. tristis, like other members of the genus, is highly variable in 

 size and coloration. To some extent this variability may be geograph- 

 ical, since this species is known to range from the Mackenzie River to 

 Virginia and from Maine to Washington and Oregon. Williston 

 (18S6) describes a var. ruficrus from Connecticut and regards M. 

 cotkumatus Bigot from Oregon as being merely a variety of tristis. I 

 have found still another variety which cannot be due to geographical 

 causes, since it was bred from one of the twenty-four puparia taken in 

 the schaufussi nest described above. The puparium from which it 

 emerged was indistinguishable from the others, but the adult insect 

 was covered with rich orange-yellow pile, whereas the seventeen other 

 specimens had the pale yellow or silvery pile of the typical tristis. I 

 have seen two other specimens of this orange variety in the American 

 Museum of Natural History. Both of these were collected by Mr. W. 

 Beutenmuller, one at Grant's on the Indian River, Fla., the other in 

 sweepings with several specimens of the typical tristis and globosus in 

 low ground at Katonah, N. Y. (June 4-14). This singular variety 

 suggests a number of questions : Does it indicate that M. tristis is 

 dichromatic like certain species of birds? Or is it a mutant? And 

 if a mutant, is it an adaptation to more intimate symbiotic relations 

 with the ants? In other words, does the orange- yellow pile have the 

 same significance as the trichomes of the true myrmecophiles (sym- 

 philes), and would its possessor be more amicably treated by the ants 

 than the typical tristis ? I could, perhaps, have answered this last 

 question, had not my specimen died soon after leaving its puparium. 



M. globosus Fabr. and M. fuscipennis Macq., two other North 

 American species not uncommon in collections, have not yet been 

 bred from their puparia. I have seen a larva and puparium which 

 are quite unlike the corresponding phases of tristis and probably 

 belong to one or the other of these species. The larva measures 8-1 1 

 mm. It is opaque, pale brown above and lighter beneath. The 

 dorsal surface is smooth and not covered with reticulations, but in 

 dried and alcoholic specimens may be vermiculately wrinkled from 

 contraction. The border has two fringes of golden yellow hairs, 

 separated by a rather broad space crossed by sparse transverse ridges. 



