No. 2368. AMERICAN SUBTERRANEAN GALLS ON OAK— WELD. 229 



ciilate area broader than high and narrowed gradually toward top, 

 petiole rugose. Mesopleura pubescent except on the more finely 

 rugose center. Hind leg with tarsus shorter than tibia, third shorter 

 than fifth, claws simple. Wings clear, apparently bare but in balsam 

 very short pubescent, ciliate only on margin of hind wing, veins 

 yellowish, areolet small or incomplete. Abdomen smooth and shin- 

 ing, slightly compressed, longer than high, second segment with two 

 densely pubescent patches on sides. Ventral spine tapering, in 

 balsam twice as long as broad, ovipositor when dissected out about 

 one and three-tenths length of antenna, ovarian eggs well developed. 

 Using width of head as base, the length of mesonotum ratio is 1.4; 

 antenna, 1.9; ovipositor, 1.9; wing, 3.5. 



Range in length of 10 pinned specimens, 3.7-4.3 mm. Average, 

 3.9 mm. 



This species can be separated from CaUirhytis ellipsoida Weld only 

 by color markings which, however, seem to be constant. The galls 

 also are similar but on a different oak. 



Type.—C&t. No. 22565, U. S. N. M. Type fly and gall. Four 

 paratypes. 



Host. — Quercus alba Linnaeus. 



Gall. — An abrupt ellipsoidal swelling on small rootlets found an 

 inch or two under the humus on forest floor underneath large trees. 

 Brown, smooth, thin-walled when mature, monothalamous, and 

 similar to galls of CaUirhytis ellipsoida Weld on Quercus bicolor but 

 the fly is different. 



Type locality. — The type fly was cut out alive from a gall found 

 at Highland Park, Illinois, October 22, 1916, on root of an unde- 

 termined oak. On May 11, 1919, five similar flies were collected at 

 Glencoe, Illinois, ovipositing on buds of Quercus alba. On May 23, 

 1919, found similar galls on roots of white oak at Ravinia, Illinois. 

 Some showed exit holes from which flies had recently emerged; 

 others were full grown but contained a thick translucent nutritive 

 layer and a barely visible larval cavity; others had a large cavity 

 and a third of the nutritive layer left and a nearly full-grown larva 

 which would probably transform in the fall and emerge next spring 

 in early May. In the United States National Museum are three 

 similar flies collected by J. G. Barlow at Cadet, Missouri, April 27 

 and May 5, 1883, ovipositing in buds of white oak. Also two from 

 Nyack, New York, collected by J. L. Zabriskie, April 21, 1885, on 

 buds of Quercus alba. 



These galls were also collected at Marianna, Florida, October 10, 

 1919, on the roots of Quercus alba growing in deep woods. A dead 

 adult was cut out on December 6, and mounted in balsam. This 

 agrees with the type material and proves the type gall to have been 

 on white oak. Galls collected in Washington, District of Columbia, 

 on alba contained living flies on October 31, 1920. 



