leaving among their shrivelled bodies the shining, reddish-brown winter egg, either partially or entirely covered by the parent's skin.

On the 16th of June, 1877, I met with an isolated tree at Malvern, Iowa, belonging to Mr. H. K. Follett, which had been very badly infested with this species. The winged individuals crowded the trunk, and had perished in such quantities around the base of the tree as to lie in a matted mass three or four inches thick, being greedily devoured by their numerous enemies. One could not break off the smallest piece of the bark without finding the exposed interstices crowded with the salmoncolored sexual individuals.

Among the more prominent of the natural enemies of this species, I have noticed, of Coleoptera, Coccinella 9-notata, Coccinella sanguinea (munda) Say, Hippodamia convergens, and several species of Scymnus. I also found feeding upon them the perfect beetle of Podabrus modestus, and the Hemipterous Cyllocoris scutellatus, Uhler, and Capsus linearis, Beauv. A Lepidopterous inquiline, namely, the larva of Semasia prunivora, Walsh, is also quite common within the curled leaves, feeding both on the lice and on the substance of the leaf. A large green Syrphus larva and several Chrysopa larvæ also prey upon them.

DESCRIPTIVE.

SCHIZONEURA AMERICANA (Fig. 1).—Impregnated egg 0.5mm long, gamboge-yellow, inclining to brown in color, with no especial external sculpture.

First generation.—-Stem-mother: Pale yellowish-red, with black members when first hatched; the red deepening and becoming purplish or livid with age. When mature, averaging 3.5mm in length, globose or pyriform, with subobsolete honey-tubes and six dorsal rows of darker piliferous and tuberculous spots. Antennæ 5-jointed, joint 3 more than equalling 4 and 5 together in length.

Second generation.-Differing in no essential respect from the preceding, except that the individuals do not attain so great a size. Bright brownish-red when born, they soon become livid brown.

Third generation.-Mature, winged female: Alar expanse 5 to 5.6mm. Body dusky, the abdomen slightly reddish; legs either dusky or yellowish-red. Antennæ as long as head and thorax together, dusky, rarely yellowish, not pilose, but with a few short setous points; 6-jointed, the 1st and 2d joints slightly bulbous; 3d either surpassing or equating in length the 4th, 5th, and 6th together, which are subequal; the terminal joint usually the shortest, the apical sub-joint being normal, and in some cases sufficiently constricted to resemble an additional joint; joints 3, 4, and 5 rather distinctly annulated, the constrictions being generally quite deep, and producing a moniliform aspect, there being on an average 22 such on joint 3. Tarsi with the basal joint distinctly separated into a lobe, the claws strong, and in length twice the diameter of the tarsus. Wings hyaline: front pair with the veins becoming obsolete at tips; stigma subhyaline, either of a yellowish tinge or somewhat dusky; stigmal vein starting from the middle of the stigma and normally curved; cubital vein obsolete for nearly one-third its length, the furcal forming with it almost a point; the terminal distance between first and second discoidals equal to about five times that between their bases (often rather more); terminal distance between furcal and cubital and second discoidal veins subequal, that between stigmal and furcal slightly shorter, that between second and first discoidal one-third greater, and about equal that between stigmal and tip of stigma. Hind wings with the subcostal vein almost straight, there being no curve where it gives off the discoidal veins, which are obsolete

at their extreme base, and not confluent with it. [The wing-venation is very constant. Out of nearly 100 specimens examined, I have found only an unusual shortening of the cubital in two individuals.] The larva and pupa in this third generation differ from the winged insect in being more reddish and in having the antennæ ringed with less distinct constrictions, in the legs being paler, in the claws being stronger, and in the basal joint of tarsus being more connate with the terminal joint. They have a distinct annulated elevation at each side posteriorly-a sort of pseudo-honeytube. When first born, they are of a pale dull yellow, and the antennal joints are more nearly subequal in length.

Fourth generation.-That from the first winged females: Differs from the preceding in the promuscis being much longer, in the antennæ having but five joints, the third being somewhat longest and the first the shortest, but all often being of much more nearly equal length, with no annulated constrictions. The color is more decidedly orange. When newly hatched, the thickened end of the promuscis often extends onehalf the length of the body beyond anus. The legs are also long and stout, and the basal joint of the tarsus is distinct, but not separated. The capitate hairs are obsolete. It is born with an enveloping pellicle or pseudovum, and though of a bright red with pale legs at first, soon becomes brownish, with dark members.

Fifth generation.-The counterpart of the second.

Sixth generation.-Second winged: Resembles the third, but usually rather lightercolored, with the wing-veins, the spinous armature of surfaces, and the constrictions of antennæ less strong, and with the third joint of antennæ rather less in length than the terminal three together.

Seventh generation.—True sexual individuals: Born within an egg-like pellicle. With stout promuscis reaching to between middle and hind coxa; the antennæ 5-jointed, with the joints subequal. Bark-feeding. Orange in color. Undergoing one moult, and then being at once distinguished from the other forms by the brighter orangeyellow color, the rudimentary mouth, the more simple eyes (composed of three facets), by the shorter, 5-jointed antennæ, the joints subequal in length; by the shorter legs, with smaller claws to the tarsi, and more distinct terminal capitate hairs, or pulvilli. The skin is transparent, the body filled more or less with fatty globules. The female is nearly pyriform, and averages 0.4mm in length. A single egg is visible through the translucent skin, and, according to age, occupies more or less of the whole of the body. The male is narrower and smaller, the penis being bulbous, with a couple of spine-like genital clasps.

This species is very closely allied to the European S. ulmi (Linn.), and until I was able to compare it with actual specimens, I was in doubt whether to look upon it as a mere variety or a distinct species. Judging from Kessler's figure and description of the European leaf-curl, and by a figure sent me by Mr. Buckton, it differs from ours, 1st, in bending upward, i. e., the stem-mother settles on the upper instead of the under side of the leaf; 2d, in having a number of small, rounded or verrucose swellings. These differences in their dwellings are strongly presumptive of structural differences in the insects themselves; and the fact that S. americana does not attack the European Elms, either in Shaw's Botanical Gardens at Saint Louis, or in the grounds of the Department of Agriculture, points in the same direction. Differences are indeed easily enough made out if we take the more or less imperfect descriptions and figures of ulmi,* but are less apparent when the actual specimens are

* Koch's figure (evidently copied by Kessler) is faulty in several respects, and fails to indicate the hook-angle of hind wings, or the corresponding thickening of front wings, a fault that is, however, common to most of Koch's figures.

compared. The following are the more important differences, least subject to variation, between the winged females of ulmi as compared with those of americana: ulmi is a longer-winged species, averaging 7.3mm in expanse; the abdomen, wing-veins, and stigma are darker; the termi nal distance between 1st and 2d discoidals slightly greater; the 3d joint of antennæ is relatively longer; the annulations are less deep and more numerous (those on 3d joint averaging 30); joints 5 and 6 are smoother, i. e.. without annulations, but they are more setous; joint 5 is shorter than 4; the apical, narrowed part of 6th joint is relatively longer and more pointed; the subcostal vein of hind wings is less straight; the cubital vein is often continuous to very near the subcostal, while I have not found any tendency of the kind in americana, the tendency being in the opposite direction, or to become shorter; the 2d discoidal of hind wings shows a tendency to fork; the hooklets on costa of hind wings are 3 in number, while in americana there are normally 4;* the legs are more setous.

COLOPHA+ ULMICOLA (Fitch).

[Forming cock's-comb-like galls (Fig. 2, a) on the upper surface of the leaves of Ulmus americana, the galls appearing with the opening of the leaves, and turning brown and black in late summer.]

Another very common gall, which may be called the Cock's-comb Elm Gall (ulmi-ulmicola), is also found on the White Elm, and particularly, as in the case of the previous gall, on young trees. It was well described by Fitch‡ as an "excrescence or follicle like a cock's comb, arising abruptly on the upper side of the leaves, usually one inch long

*These hooklets get so easily broken off that they are not to be relied on; yet the normal number on most of the Pemphiginæ I have examined is 3, while in Hormaphis there are but 2. The fact that in Sc. americana there are 4 is therefore interesting, and of some value in this connection.

The bibliography of this species very well illustrates the confusion that too often surrounds the proper determination, not only of insects of this family, but of all orders. It is due to three causes, not easily removed: 1st, the miserably insufficient nature of the earlier descriptions and definitions; 2d, the isolation of the earlier English entomologists from those of the continent, and the dual nomenclature that has arisen from independent work; 3d, the want of a common ground for generic characterization. Walsh referred the species to Thelaxes, which has, however, 5-jointed antennæ. Vacuna, Heyden, is synonymous with Thelaxes, though Walker would restrict the former to alni, Schrank, and the latter to dryophila, Schrank ("The Zoologist," London, February, 1870, p. 2001), without pointing out generic differences, as the want of a fork in the cubital vein in Koch's figure is clearly an error of the artist. Mr. Monell founded the genus Colopha for ulmicola on the fact that the antennæ of the winged female are 6-jointed. Such a difference can hardly have generic value when we find ulmicola occasionally with but five antennal joints, and (if Huxley is correct in his determination) dryophila sometimes with six (Trans. Linn. Soc. xxii, pp. 203, 234). But, taken in connection with the fact that ulmicola is a flocculent species, the true female producing but one large egg, while dryophila is without flocculence, the female (according to Huxley) laying many eggs, Colopha, considering ulmicola as the type, may be accepted as a good genus.

Fifth Report on the Noxious Insects of N. Y. § 347.

and of an inch high, compressed and its sides wrinkled perpendicularly and its summit irregularly gashed and toothed; of a paler green color than the leaf and more or less red on the side exposed to the sun; opening on the under side of the leaf by a long slit-like orifice; inside wrinkled perpendicularly into deep plates." The gall is always found between two of the branching parallel veins, and those between which it grows are generally drawn closer together than the rest. The corru. gations and roughness, so characteristic of this gall, evidently result from the lesser susceptibility of the minute transverse veins to swell, compared with the more succulent tissue of the leaf. There is always a certain hoariness around the mouth of the gall below, while the base of the upper part is always contracted and compressed.

BIOLOGICAL.

The impregnated egg of this species is also to be found during the winter in exactly the same sheltered situations, in and under the bark of the White Elm, as that of Schizoneura americana. It is almost always sheltered by the dry and somewhat wrinkled skin of the true female, being seldom extruded, but occupying the whole of the body (Fig. 2, b). Occasionally the mother skin is more or less freed. The young stemmothers hatch from the winter egg about the same time, and are minute dark olive-brown specks, just visible to the unaided eye, and quite active during pleasant weather, crawling nimbly about over the tree, till they reach a tender leaf that is just unfolding, when they also settle upon the under surface, and begin to feed on and fret the same. They doubtless insert their beaks in various portions of the buds or expanding leaves ere settling, since, before the gall begins to form, the little architect has generally obtained twice the size it had when first hatched. By the middle of April, in the latitude of Saint Louis, the galls generally begin to show, at first as slight elongate ridges on the upper surface, with corresponding closed depressions on the lower surface. Upon drawing apart the lips of the wrinkle beneath, at this stage of the growth of the gall, the stem-mother, who still retains her glossy olive-brown appearance, is seen constantly running back and forth in the cavity, and inflicting rapid punctures with her beak, the inner surface of her dwelling being smooth and glossy, with a slightly blistered appearance, in contrast with the normal, more rough and pubescent texture of the under surface of the leaf. The development of the gall is very rapid, and, with favorable weather, the top part begins to bulge so as to give the contracted appearance of the base, and the tooth-like prominences begin to appear by the third day. The inmate likewise grows apace. After the first molt, she soon becomes more pyrifrom and paler, with transverse rows of powdery secretion. She is less active, but still marches about, incessantly fretting the surface with her short, stout beak. A second molt takes place, and by the time the gall has fully developed, or about two weeks from the time it commenced forming, the

process of reproduction commences, and continues for two or three weeks, until the stem-mother is exhausted, and the gall is absolutely crowded with this second generation in all stages of growth. The lice are more or less covered and interspersed with the mealy or cottony excretion, and with the various-sized globules of gummy liquid, which is sometimes so abundant that it will fall upon the ground like a shower of milky fluid, whenever badly infested trees are shaken. The insects comprising this second generation, or the immediate issue from the stemmother, thus born within the habitation which she had built up, are similar to their parent, but somewhat larger at the moment of birth than she was, and of a paler olive-green color. They are quite active within the gall, exploring its concavities, and obtaining their nourishment through its walls. After the second molt, they attain the pupa state, (Fig. 2, d), and in due time become winged. There is but one generation produced within the gall-a generation, however, that becomes very numerous under favorable conditions. They all become winged, and in this respect the species differs essentially from Schizoneura americana, as we have already seen. The winged lice carry their wings flat on the back while in the gall, but deflexed afterward. They issue from the slit on the lower surface of the leaf, which opens for their exit about the time they become fledged. They are all females, and give birth, in the course of a day or so, to upward of a dozen young, which, when first born, are enclosed in the usual delicate egg-like covering already alluded to, and which look like their immediate parent at a corresponding state of existence, except that their antennæ have five subequal joints, and the promuscis reaches to the hind coxæ (Fig. 2, c).

So far I have been able to trace the history of the species with absolute certainty, watching it for several years, and proving, by extracting the stem-mother soon after she had commenced reproducing, that the second generation, i. e., her immediate progeny, all become winged, the species agreeing in this respect with the gall-making species of Phylloxera that affect the Hickory. There is, however, a link yet wanting in our knowledge of the history of this species, between this third generation and the mouthless sexual individuals, the females of which so often perish while yet covering their solitary winter eggs. I have not been able to prove absolutely that there are two broods of the gall-making female, and my observations all tend to the conclusion that no galls are formed except by the stem-mother that hatches from the impregnated egg. I have never succeeded in obtaining galls either by enclosing the winged females in muslin bags tied on the living trees, or by similarly enclosing her immediate progeny, though I have succeeded in obtaining, without any difficulty, an abundance of galls by so enclosing the stem-mother. Moreover, all such succulent galls as this one are produced on the tender young leaves only, and I have failed to find them on any but those which develop early in the season. It is true that we may frequently find the galls quite fresh, and containing larvæ, pupæ, and winged insects as late