yellowish-tinged, hairy, the veins clear brown, those bounding the radial area darker; radial area closed, covered with a large, brown patch with a large, clear spot in the center; areolet moderately large, cubitus not quite reaching the basal vein below its midpoint; first abscissa of the radius distinctly angulate but without a distinct projection into the radial area. Length: 2.0-3.5 mm.
Male.—Similar to female but differing as follows: basal joints of antenne piceous brown; abdomen piceous, shorter, more compressed, and more slender basally; wings with only traces of the clouds on the radial areas, with a slight projection from the first abscissa of the radius into the radial cell; length, 1.5-3.0 mm.
[Redescription made from New England and New York material compared with cotypes.]
Galls.—Irregularly globose leaf-galls (Figs. 5 to 7), covered with a white, mealy powder. Each gall is about globose, monothalamous, but often several galls coalesce to form large, elongate, more or less entire masses sometimes 20 mm. long. On the terminal twigs, petioles, and stems of the leaflets of Rosa blanda, R. carolina, R. humilis, R. nitida, R. virginiana, and most likely other roses.
Cotypes.—Cotype females and galls in the Museum of Comparative Zoology, and "cotype" galls in The American Museum of Natural History.
Range.—Ontario: Toronto (Cosens). Massachusetts: Magnolia, Boston (Clarke); Westboro (Frost); Springfield (Stebbins). Rhode Island: Providence (Thompson). Connecticut: New Haven, Woodbridge (Britton); Waterbury (Bassett). New York: Albany (Felt); Nyack (Zabriskie); New York City (Beutenmüller); Staten Island (W. T. Davis). New Jersey: Fort Lee (Beutenmüller); Patterson, Bradley Beach (in Coll. Amer. Mus.). Pennsylvania (Beutenmuller). Maryland: Fareman (Osten Sackern). District of Columbia: Washington (Beutenmüller), North Carolina: Black Mts. (Beutenmüller); Asheville (Ashmead); Florida (Ashmead). Iowa (Beutenmuller). Colorado: Fort Collins (Gillette).
The galls of Rhodites ignotus are first noticeable about the middle of August, somewhat deforming the leaves of the roses on which they are formed. The galls overwinter, sometimes on the bush, but often on the ground, to which they readily fall when the leaves bearing them wither in the autumn. The mature wasps are known to emerge the following spring from May to August, most of the adults appearing about the last of May or the first of June. As with R. rosæ', this extended period of emergence is an unusual thing to find among the gall-wasps. Of thirty-seven of the wasps which I have bred, thirteen. i. e., 35% of them were males. which is a higher percentage than that known from other species of the genus. However, incomplete observations on other breedings I have made would indicate a much lower percentage of males to be more nearly normial. The number of parasites obtained from these galls is extremely high; I have found them to constitute about 90% of all the insects bred—another instance of the ineffacacy of highly developed "protective" devices. Most of these parasites are figitids, Syneiginæ, etc., the so-called "inquilines,' but many other parasitic Hymenoptera also attack the galls.
