wasp that had made use of the burrows of a bee (Anthophora abrupta Say). The larva of Argyramoeba has very much the same appearance as that of Systœchus and Triodites, and the pupa is distinguished from the pupa of this last (Pl. vi, Figs. 5, 5 a), principally by its longer and more numerous hairs, longer anal spines, and more conspicuous spiracles.

Systropus also, in the larva state, preys on the larva of Limacodes, as has been observed by Walsh (Proc. Bost. Soc. N. H. ix, 300), and by Westwood (Trans. London Ent. Soc., 1876, p. 571), killing its victim only after the latter has formed its tough cocoon.

With these general remarks, we will now give a more full and descriptive account of the two bee-flies which, by rearing from the larva, we know to have this locust-egg-feeding habit. Our plate illustrates the insects as well as can be done in color printing, but the enlarged heads of the larva should be somewhat darker and less yellow.

SYSTOCHUS OREAS.-The character of the eggs and the manner in which they are laid have not yet been observed. The larva (Pl. vi, Fig. 1) is found in the locust egg-pods, or near them, of different sizes, during most of the year. These larvæ begin to transform to the pupa state early in the summer, and the pupa (Pl. vi, Fig. 2) pushes itself half way out of the ground in order to disclose the fly. These flies continue to issue during the summer months. As a rule, but one year is required for full development, but there is, in this respect, great irregularity, and the same tendency to retardation which we have called attention to in the case of the blister-beetles.1 We have had quite a number of the larvæ remain over unchanged till the second year, and all that we have said as to the philosophy of this retardation in the one case applies in the other. We are inclined to think that future obser1 Am. Entomologist, 111, 196.

2 This irregularity in the development of individuals is noticeable in many insects that are parasitic, and whose mode of life is precarious. In the case of our blisterbeetles, depending as they do on locust eggs, and especially in the case of those which feed particularly on the eggs of migratory species, it is not difficult to perceive how this trait may prove serviceable to the species possessing it. Migratory locusts occur in immense numbers, in some particular part of the country, at irregular intervals, and.there are periods or years of absolute immunity from their presence in the same regions. The young blister-beetles that hatch the year following the advent of the locusts in immense numbers may frequently find few or no locust eggs upon which to prey, and the great bulk of them would, as a consequence, perish; while the young from such exceptional individuals as should not develop till two, three, or

vation will show that there is a still further parallel, in that the newly-hatched larvæ of the bee-flies are much more active than the later stages, and somewhat different in structure.

The three later stages of the insect may be characterized as follows:

Larva (Pl. vi, Fig. 1).—We scription:

quote herewith our former de

"Average length, 0.50 inch. Body curved, glabrous, tapering posteriorly, swollen anteriorly. Color opaque whitish, with translucent yellowish mottlings, and some venous marks at sutures, especially along medio-dorsum. Sutures deep. A lateral row of swellings. Head small, flattened, dark brown, in five pieces, consisting above of a frontal ovoid piece and two lateral pieces of somewhat similar form, and each bearing near tip a minute, twojointed palpus; beneath of two broad, subtriangular jaws, having forward and lateral motion, and each, also, bearing near the center, in a depression, a two-jointed feeler. A spiracle each side in a fold between joints 2 and 3, and another on each side of the penultimate joint, 12. None otherwise perceptible."

With additional material we have been enabled to examine more fully the structure of the head. Underneath the median elevated piece which may represent the labrum, we find two stout spines (Pl. vi, Fig. I e), faintly notched on the outer edge, which are doubtless the mandibles, and correspond to the two dark lance-like mandibles of other Dipterous larvæ, for they are retractile and run back into the thoracic joints, and remain after the other trophi are detached. The pair of feelers upon the upper lateral pieces, which seem to have no motion, might then represent the antennæ, and the two lower jaws the maxiile with their palpi, while the labium is shown in a chitinous point visible only when the larva extends and raises the other parts. A peculiarity in the movement of the maxillæ or the lower pair of horny pieces is worthy of note. They move in alternation with one another in the forward and backward, i. e., up and down, motion. The palpus of these lower pieces when viewed from above is, as represented in the figure (Pl. vi, Fig. I d), circular, with two dark marks indicating minute appendages.

When the larva is fresh and plump it shows the greater swelling of the thoracic joints and the translucent mottlings mentioned more years after a locust invasion might stand a much better chance of finding appropriate food, and of thus perpetuating the species. In this case and in most other cases of retarded development with which we are familiar, the exceptional retardation may and does become a benefit to the species, enabling it to bridge over periods of adversity. And we can see how, by the preservation of such favored individuals, the habit of irregular development may have become fixed in the species as a consequence of surrounding conditions and circumstances which render it advantageous.

above. Toward the period of pupation, it becomes more opaque and more contracted.1

Color honey

Pupa (Pl. vi, Fig. 2).—Average length 8.5 mm. yellow, but varying with age, the head and thorax assuming a dark color with maturity. Head narrow, with two sets of three stout, dark spines on the top, all on a common prominence, the two lower ones of each set connected at base; a pair of smaller frontal spines near the base of proboscis, which is protuberant and laid along the breast, extending to near the tips of the wings; the face with two parallel depressions running from between the triple tubercles ending in two fossæ above the frontal spines; two basal, medio-dorsal tubercles. Thorax unarmed, the prothoracic spiracle very large and raised on a curved tubercle; mesothoracic spiracle on a swelling at base of wings; front tibiæ stout and curved; front tarsi reaching to tips of wings; middle tarsi to abdominal joint beyond, and hind tarsi to third abdominal joint beyond. Abdomen curved, with the ninth joint very small; across the middle, dorsally, each joint has a series of parallel, longitudinal, narrow, chitinous plates having at each extremity a spine, the posterior one stoutest; both plates and spines diminishing laterally, gradually aborted on the extreme basal and posterior joints, and replaced on the small ninth joint by a group of four converging and truncate tubercles; two stouter anal spines on the subjoint and a ventral lobe with two short, obscurely articulate processes; each abdominal joint with a circle of hairs, those on lateral ridge stoutest and one-third the width of abdomen in length; eight pair of abdominal spiracles (making ten with those on thorax), the first and last pairs rather difficult of detection.

TRIODITES MUS.-The habits of this insect in the larva state are precisely like those of the preceding :

Larva (Pl. vi, Fig. 4).—So greatly resembling that of the Systochus that it is well nigh impossible to separate the two with certainty. The head parts are somewhat broader, shorter and less flattened, the maxillæ more blunt, the labrum paler, and the mandibles sharper and with a smoother outer edge. The thoracic joints bulge less beneath and the thoracic spiracle is more sunken and less conspicuous.

Pupa (Pl. vi, Fig. 5).—Easily distinguished from that of Systochus in the broader and more bulbous head; in the two sets of three stout spines at top being well separated; in the frontal pair be

1So far as we can ascertain, there has hitherto been published no recognizable fig. ure of the Bombyliid larva. Dufour, in his articles above alluded to, describes that of Bombylius major very indifferently, and gives a dorsal view which shows little or no relation to the larva here described, while his description and figure of the mouth parts fail to indicate the different pieces we have observed in our larvæ. Yet in general form and structure the true Bombylius larva agrees very closely with those here described, as we know from Dr. Chapman's description.

1

ing stouter, each with a conspicuous bristle externally; in having a single spine or tooth above these, and another much stouter, erect, recurved spine, bidentate at tip, below them or at base of tongue, which is here represented by a cordate lobe. There is a spine on the front anterior border of each wing; the legs are all shorter; the prothoracic spiracles less conspicuous; the hairs on abdominal joints shorter; the transverse dorsal teeth smaller and in single row; the basal abdominal joint without spines, but with long stout hairs and the dorsal tubercles of abdominal joints nine replaced by a single spine.

EXPLANATION OF PLATE VI.

(Natural sizes indicated in hair-line.)

FIG. I-Larva of Systæchus oreas, from the side; 1 b, head from side, still further enlarged; I c, same from front; I d, left maxilla; 1 e, left mandible; If, mesothoracic spiracle; 1 g, pre-anal spiracle.

FIG. 2.-Pupa of Systachus oreas, ventral view; 2 a, same, side view; 2 b, dorsal part of anal end; 2 c, prothoracic spiracle; 2 d, form of dorsal horny plates and spines on the abdomen.

FIG. 3.-Systœchus oreas, Q; 3 a, head of same from side; 3 b, antenna of same from above; 3 c, antenna of same from side; 3 d, mouth parts separated. FIG. 4.-Larva of Triodites mus as it appears when contracted prior to pupation ; 4 a, head from side; 4 b, left maxilla; 4 c, left mandible.

FIG. 5.-Pupa of Triodites mus, ventral view; 5 a, same, side view; 5 b, dorsal view of anal parts; 5 c, form of dorsal plates and spines on abdomen.

FIG. 6.-Triodites mus, Q; 6 a, her head, front view; 6 b, her right antenna from above; 6 c, right antenna from side.

FIG. 7.—Triodites mus, ♂; 7 a, his head, front view.

-:0:

LATE EXPLORATIONS IN THE GABOON.1

BY HUGO VON KOPPENFELS.

"OWARDS the end of last year I again had the good luck to

TOY

kill an almost full-grown gorilla, whose length was 1.75 meters, and the width of the shoulders 0.89 meters. This gorilla, and a young female, should, by this time, be in possession of Dr. V. von Kraus, Stuttgart. I, by chance, shot the male gorilla in the vicinity of my stopping place on the Eliva-Comi (an inland lake), so that I was able to take a very good cast of it. I hope to be able to take a cast of the next animal I shall kill, but I cannot predict at what time and place I shall get a full-grown male gorilla, and whether the circumstances will allow me to take a cast of him.

1 From a letter to Mr. H. A. Ward, Rochester, N. Y.

I have already asserted, and I believe it is proved, that there are crosses between the male Troglodytes gorilla and the female Troglodytes niger, but for reasons easily understood, there are none in the opposite direction. I have in my possession positive proof of this. This settles all the questions about the gorilla, chimpanzee, Kooloo Kamba, N'schigo, M'bouvé, the Sokos, Baboos, etc.

The French savants seem to have a special predilection for creating new species from variations in the form of the skull, such as often occur in this group of animals.

There is but one district which forms the range of the gorilla, and this is situated in the western part of equatorial Africa, and here it exhibits no varieties, while the chimpanzee is found all over tropical Africa, and naturally exhibits considerable variation. The chimpanzee of Northern Guinea differs essentially from that of the southern portion of the same country, and, according to Livingston, the "Soko" differs from both, but is still a chimpanzee. Du Chaillu's Kooloo Kamba, N'schigo and M'bouvé are not distinct species, and this traveler, who is certainly a man of merit but is too credulous, has been imposed upon by the mendacity of the natives, which beggars description. The names N'schigo, M'bouvé, Koola, Baboo, Soko, Quia and Kooloo Kamba are only different designations of the chimpanzee by different tribes. The mongrel progeny of the male gorilla and female chimpanzee discovered by me, is found, but in individual cases, and as such deserves no special name.

I intend in a few days to start on an excursion to the Crystal mountains over the N'tampuny falls. My purpose is primarily to shoot elephants. Du Chaillu's journey to the Oschebas by the Munin-Tampnay did not extend far, as I accomplished, last year, the same distance in six days' marches. I found the population harmless though somewhat suspicious. The whole district is almost unknown, as I encountered, the second day, members of entirely new tribes on the Yoko, Manga and later the Akuke. The population of this region is continually migrating from the north-east to the south-west, and no one will ever succeed in sifting out the relationship of these commingling tribes.

As a point of departure for penetrating the interior from the west, this seems to be one of the most promising. In all directions are districts which are as yet entirely unknown. On the