Parasites. Torymus abdominalis, T. incertus, T. regius, and T. auratus in May and June. Eurytoma rosae and Pteromalus immaculatus in July. Tetrastichus atrocaeruleus, Eupelmus annulatus, and according to Barrett the Tortrix, Zeiraphera communana.]

2. Neuroterus laeviusculus. Schenck1.

Gall. Cup-shaped, the edges thinned and incurved; in the centre there is a small but distinct boss surrounded by a circle of brownish hairs; diameter 2-3 mm. The form of the gall is often irregular, the rim bent; colour pale or reddish. The gall appears in July and matures in September. (Fig. 22.)

Rearing the Fly. When the mature gall falls from the leaf, its under surface will be found to be distinctly swollen. In order to observe the development of the flies indoors the galls must be kept on damp sand. Their progress may be forced, and in a room they will emerge in November, but in the natural course they do not appear until March of the following year. The earliest date on which I have found them in the open air was March 9.

Fly. Size, 2-4 mm.; black; thorax smooth and shining; abdomen much compressed, elongate; legs distinctly paler, white or yellowish, coxae and base of femora dark.

Experimental breeding. I have made experiments in breeding with Neuroterus laeviusculus in the same

[1 Neuroterus pezizaeformis, Schtdl.]

2 This gall is frequently confused with that of Neuroterus fumipennis. I myself made this mistake, and in my earlier publications these two names must therefore be transposed, but the facts remain unaffected.

way as with the former species, and have frequently succeeded in getting the flies to prick the small oaks. The first accurate experiments were made in March, 1875. Between March 14 and 26, thirty-six buds in all were pricked by a large number of flies. On the unfolding leaves there appeared in May the gall of a totally different fly, Spathegaster albipes. From the strict control exercised there could be no doubt that these galls proceeded from Neuroterus laeviusculus. From my first experiment I obtained thirty-six galls, but from others made in 1877 I only got two galls. As a rule experiments made with these flies are pretty certain to succeed.

[The smooth spangle gall is found on Q. pedunculata.

Inquiline. Synergus Tscheki in April.

Parasites. Torymus sodalis in March and April. T. hibernans. The spangle galls live about nine months.]

2a. Spathegaster albipes. Schenck1.

Gall. Size, 1-2 mm. in length; oval with a short apical point, of a greenish yellow colour, smooth or thinly set with solitary hairs. The galls are sessile on the leaves, which they deform more or less, causing indentations or sinuosities and often stunting them in their growth. This is due to the mode of origin of the gall which is formed on the rudimentary leaf while yet in the bud. The area occupied by the gall in the bud is of course small, but the effect upon the leaf when it expands is much more marked. (Fig. 2a.)

Fly. Size, 1-2 mm. long; black; thorax smooth [1 Neuroterus laeviusculus, sexual form, Cameron. Neuroterus albipes Mayr.]

and shining; abdomen distinctly pedunculate. Legs pale, only the coxae and bases of the femora dark. The flies emerge at the end of May or beginning of June.

Experimental breeding. I observed these flies for the first time on June 3, 1875, while they were busy in the open air pricking the under sides of the tender oak leaves. They are very delicate little flies, and can only be kept alive for a few days, but it is not difficult to observe them ovipositing if they are provided with very tender leaves. They are first seen to move about actively and examine the under surface of the leaves carefully with their antennae. They then direct the point of the abdomen perpendicularly to the surface of the leaf, the terebra is pushed into it, and an egg glides down into the channel thus pierced. This fly can deposit a large number of eggs in the leaf in a short space of time. The first traces of gall formation are found at the end of three weeks as little hairy spots which soon develop into the galls of Neuroterus laeviusculus. Of these there may be as many as 200 on a single leaf.

[Schenck's gall is found in May on Quercus sessiliflora and Q. pedunculata.

Inquiline. Synergus apicalis.]

Gall.

3.

Neuroterus numismatis1. Ol.

Very pretty circular galls, like buttons covered with brown silk, with a shallow depression in the middle. Diameter, 2 mm. They mature in the autumn with the preceding (Neuroterus) gall. (Fig. 3.)

[1 Cynips numismatis, Oliv. Neuroterus Réaumuri, Hartig.]

The flies are reared in exactly the same way as Neuroterus lenticularis.

Fly. 2-5 mm. in length; black; thorax dull, finely punctate; scutellum somewhat closely haired. The colouring of the legs variable, yellowish brown, bases of the femora mostly dark. Abdomen, looked at from the side, almost round; basal joints of the antennae dark, which is the only character by which this fly is distinguished from Neuroterus lenticularis.

Experimental breeding. Experiments, in the manner described above, were also made with this fly, the first being in March, 1875. From this first attempt, in which thirty-two buds were pricked, I obtained in all five galls which were formed under the leaf surface and proved to be those of Spathegaster vesicatrix. In 1876, I repeated the experiment with the same result. Later also an English entomologist Fletcher1 obtained from similar experiments the same species of Spathegaster.

[The silk button spangle galls appear in July on Quercus sessiliflora, Q. pedunculata, and Q. pubescens.

Inquiline. Synergus Tscheki, March to June.

Parasites. Torymus mutabilis, June-August, T. inconstans, T. fuscicrux, T. geranii in July. Platymesopus tibialis in June. Eurytoma curta, E. aethiops. Pteromalus domesticus in July. Eupelmus urozonus. Pleurotropus sosarmus.]

3. Spathegaster vesicatrix. Schltdl.2

Gall. These galls are inconspicuous and are embedded in the substance of the leaf, which they resemble. They project only slightly above the level of the surface.

1 J. E. Fletcher, Entom. Month. Mag., vol. xiv. p. 265 (May, 1878). [2 Neuroterus numismatis, sexual form, Cameron. Neuroterus vesicatrix, Mayr.]

Each bears in its centre a little conical projection from which rays run out to the margin of the gall. (Fig. 3a.)

The flies emerge in June, and are very easily reared if the galls are collected shortly before they mature.

Fly. Size, 2 mm.; black; thorax shining; legs yellowish, coxae and bases of femora dark. Male and female similar.

Experimental breeding. As it is very difficult to collect the flies of this species in large quantity, I have only once been able to make an experiment in breeding, and that was in the open air. On June 20, 1875, I observed several females creeping about on the under surface of the oak leaves and laying their eggs. I marked eight leaves which had been pricked, by tying threads upon them. After three or four weeks small round galls appeared which proved to be those of Neuroterus numismatis.

[The blister-gall occurs in May on Quercus sessiliflora and Q. pedunculata. A different but similar gall appears on Q. pubescens and Q. cerris.

Inquiline. Sp. ? of Synergus.

Parasites. Sp. ? of Torymus. This gall continues to live after its gall-maker has emerged.]

4. Neuroterus fumipennis. Htg.'

Gall. Generally circular, with the edges often incurved and emarginate. The gall is of a pale or reddish colour with delicate brown stellar hairs. (Fig. 4.)

This gall has a certain resemblance to that of Neuroterus lenticularis, but has more frequently been

[1 Spathegaster varius, Schenck.]