SYSTEMATIC ACCOUNT. 81 



show that the eggs and spermathecsB are different. The two forms also 

 differ in food habits, since D. transversa breeds chiefly on fungi, this species 

 chiefly on fruit. I have bred it from banana, pineapple, tomato, and 

 potato. The species is to be found very commonly about tomato plants, 

 and is not uncommon about windfall apples. It can be bred in the labor- 

 atory on banana, but is not easily kept for many generations, and does not 

 breed well in pairs. 



The eggs have three filaments, as shown in plate 1. The chromosomes 

 have been described by Metz (1916, Amer. Nat., 50; see p. 39). 



Drosophila transversa Fallen. 1823. Dipt. suec. Geomyz., 2, 6. 



c^, 9 . Arista with four to six branches above and three below. Antennae yellow, 

 third joint darker. Front nearly one-half width of head, wider above; dull yellow. Second 

 orbital scarcely one-fourth first. Second oral bristle one-half to three-fourths length of 

 first. Carina broad, fiat; face dull yellow. Proboscis and palpi yellow. Cheeks yellow; 

 their greatest width one-sixth to one-fourth greatest diameter of eyes. Eyes with pale 

 pubescence. 



Acrostichal hairs in six rows; no prescutellars. Mesonotum and scutellum yellow, 

 sUghtly shining. Pleurae and legs yellow. Apical and preapical bristles on first and second 

 tibiae, preapicals on third. 



Abdomen yellow, with four posterior black spots on each of the first four segments, and 

 two larger spots on the fifth. 



Both cross-veins clouded, wings otherwise clear. Costal index about 3.5; fourth-vein 

 index about 1.8; 5x index about 1.1; 4c index about 0.9. 



Length body 2 mm.; wing 2.2 mm. 



Specimens examined: Hilversum, Holland (J. C. H. de Meijere); Machias, 

 Maine (C. W. Johnson); Hanover, New Hampshire; Boston (C. W. John- 

 son), Chester (C. W. Johnson), New Bedford (C. W. Johnson), Monument 

 Beach, Woods Hole, Massachusetts; Niagara Falls (C. W. Johnson), Ithaca 

 (S. W. Frost), Staten Island (F. Schrader), Cold Spring Harbor, New York; 

 Fort Lee, Riverton (C. W. Johnson), New Jersey; Bloomington, Indiana 

 (F. Payne) ; Flat Rock, Illinois (F. N. Duncan) ; Plummer's Island, Mary- 

 land (R. C. Shannon); Dead Run (R. C. Shannon), Falls Church (N. 

 Banks), Virginia; Athens, Tennessee; Kushla, Alabama. 



This species breeds on various kinds of fleshy fungi, where it is usually 

 to be found in great numbers, along with D. putrida Sturtevant. I have 

 also bred it from potato, but only with difficulty. 



It is very similar to D. quinaria when pinned, but can usually be dis- 

 tinguished by the characters given in the key. It is possible that some of 

 the records given above really are based on D. quinaria, but the range is 

 certainly a very wide one. I am certain of some of the records from Massa- 

 chusetts, from New Jersey, from Virginia, and from Alabama. The others 

 are not in great doubt; in cases where I could not be reasonably sure of the 

 identification, as between these species, I have omitted the record. 



This species is so variable as to suggest that we are really dealing with a 

 complex group. The specimen from Holland, that on this view is most 

 likely Fallen's D. transversa, has five branches to the arista above, second 

 oral three-fourths first, and cheeks one-fourth diameter of eyes. Specimens 

 from Monument Beach, Woods Hole, and Ithaca agree well with it. 



The eggs have three filaments (see plate 1). 



Drosophila putrida Sturtevant. 1916. Ann. Ent. Soc. America, 9, 339. 



cf, 9. Arista with about six branches above and two below. Antennae yellow brown, 

 third joint dark reddish-brown. Front nearly one-third width of head, wider above; 

 dull yellowish-brown, ocellar dot darker. Second orbital minute. Second oral bristle 

 nearly as long as first. Carina low, rather broad, flat; face yellowish brown, somewhat 



