S semen, n. sp.
On Salix nigra
A minute, monothalamous, hollow, irregularly hemispherical, greenish yellow gall, .02 — .04 inch in diameter, mostly on the upper side of the leaf, and often, but not always, with a pointed nipple on the middle of the hemisphere, always with a corresponding circular depression on the other side of the leaf, in the middle of which is a very minute, flattish hemisphere. Very frequently on one and the same leaf the position of the gall is reversed from the ujjper to the lower side, as in Q. pihiloe Walsh. On a single leaf scores of them may often be counted, generally with several masses among them, composed of two or more confluent galls. Commences its growth early in the summer, and by the last of August many are found to be burst open at top, yet at the same time very many of them, when opened, are found to be solid without any central cell. By November most of the galls from which I attempted to breed the insect had burst open into a ragged, wart- like shape on the hemisjiherical side, but no larvte had escaped from them and none were discoverable in them. As the leaves were kept too moist, so that they moulded badly, the larvae had probably perished in the galls. Described from 20 affected leaves. From its close homology with the much larger oak-galls St/mmrtrk-a 0. S. aqd Q. pilulse Walsh, iu the former of which Cecidomyidous larvae were detected by Osteu Sacken, and described (Dipt. X. A. p. 201) as having a Y-shaped breast- bone, and in the latter of which I found myself, September 14th, several orange-colored larvje, which, from the presence of a clove-shaped breast-bone, were undoubtedly Cecidomyidous, and from the fact of a similar leaf-gall on a Willow, S'. verruca n. sp., being inhabited by a Cecidomyidous larva, there can be no doubt, I think, that the gall S- semen is the work of a Grall-gnat. Prodigiously abundant and very common everywhere in Rock Island County, Illinois, on the Black Willow, the foliage of whole trees being thickly frosted over by it, so that the leaves look like nutmeg-graters. I have in a cursory manner noticed in July several specimens of what seemed a very similar gall ou S. discolor, but found no larva) in them ; and in a single instcince I found, August 20, two leaves of S. longifolia on a twig which grew out of a bunch of the galls S. brassicoides, covered so densely with some- what similar galls as to be intermediate in appearance between S. se- men and S'. penigma. On August 29 I discovered in one of the cells of this gall a minute, pale-colored, apod larva with a large, scaly head, and the disk of its dorsum, but not of its venter, fuscous. This so ex- actly resembled a much larger larva of which I have found many spe- cimens in the Cecidomyidous gall, Q. pilidx Walsh, and which I am sure, from comparing it with the larva of Anthonomus scutellatus Schonh., must be Curculionidous, that I believe it to be also Curculionidous, and inquilinous, like the othei* larva, iu the gall where it occurred. I have noticed towards the last of August galls about the size of the head of a large pin. similar to S. semen, and often similarly confluent, o'rowino- in con.siderable abundance on the leaves of the River Kirch (Betula nigra), chiefly or almost entirely on their upper surface. And on the leaves of the Button-bush (Cephalanthus) I have noticed at the same period of the year galls of the same character, in the same luxuriant profusion as S. semen occurs on the Black Willow, whole bushes being covered with them ; but in neither of the two kinds could I discover any larvae. I believe them both to be the work of Cecidomyia. It does not follow, because all these galls are so small, that therefore their Gall-gnats must be abnormally small. The gall S. rhodoides n. sp. is about 4 times as long and wide as the Grail S. gnaphalioides n. sp., yet the Gall-gnat produced from the latter is only \ shorter than the (rail-gnat produced from the former. Larva, pupa and imago unknown.
