256 GNATS OR MOSQUITOES — CHAPTER X a link between the gnats and midges, differ so markedly that there can be no doubt that they should be placed in a family by themselves, as though they closely resemble other Mosquitoes in the venation of their wings and in other general characters, their mouth-parts are formed like those of midges, so that they have no proboscis in the sense of the word as applied to other Mosquitoes, and have in place a short, bilobed organ, ending in a pair of blunt knobs not unlike that of the Indian " sand-fly," figured on page 82. The remaining four sub-families are distinguished by the old characters of their type genera, viz., by the relative length of their palpi, so that the five sub-families may be tabulated as follows : — Section A. — Proboscis formed for piercing. I.— Brilliantly coloured insects, with a very long, curved proboscis, (a) Palpi about as long as the proboscis in both sexes {Megarhina); or long in g, shorter in ? (Toxorhynchites). Megarhinina.
II. — Dull-tinted insects, with straight proboscis. {$) Palpi about as long as the proboscis in both sexes, those of the (J clubbed at the end, those of the ? linear. Anophelina.
(7) Palpi about the length of the proboscis in the <?, but much shorter in the J , being in her usually very short. CULICINA.
(8) Palpi very short in both sexes.
^DOMINA.
Section B.— Mouth-parts not formed for piercing, there being no true proboscis. Palpi small.
CORETHRINA.
The detailed characters of the various genera appertaining to each separate sub-family will be tabulated in their systematic order, but it will be well here to explain the peculiarities of the form and arrangement of the scales and bristles on which Mr. Theobald's new classification is mainly based.
On the opposite page, characters of this sort which have been found by him to be available for the purposes of generic distinction, are collected together in graphic form for ready
Form arul arrangement of ike scales of the 'nape in Megaiitina. CtUex. Stegomyia ank Ikies. ^"^ement of tfi^ scales of Graphic key to generic distinctions based on scalechaFacters Fig. 38.
17
258 GNATS OR MOSQUITOES — CHAPTER X reference, though many of the figures are, as a matter of fact, reahstic. The figures placed in the middle line of the figure are all those of the type genus of the family Culex, and on either side are placed figures of the structures of such genera as diverge from the type genus ; it being understood that in each instance, the corresponding regions of genera not figured conform more or less to the Ctilex type. For example, the clypeus is nude in all Mosquitoes except those belonging to the genera Stegomyia, Mdomyia, and Ti-ichoprosopon, in the two former of which it is scaly, those of the first genus being provided with a pair of tufts of scales of a peculiar sigmoid form ; while in the latter, it is clothed with a rather dense brush of hairs.
Commencing at the top of the illustration, the figures arranged in the first two lines will be found to refer to the arrangement of the scales of the nape. In this region, the clothing scales will be found to belong generally to one of three types, namely, (a) flat, spade-shaped scales, arranged overlapping each other in regular order, like the tiles of a roof, so as to form a smooth imbricated armour, which usually displays to the best advantage the pecuhar " metallic " interference tints exhibited by scales of this sort when illuminated at certain angles ; or (6) narrow curved scales, such as those shown in fig. 16/, which always give a characteristically woolly appearance to parts clothed by them, and owe such tints as they may possess to their natural coloration, independent of the point of view^ ; and (c) upright forked scales, narrow, stiff, bifid structures, usually black, which bristle out usually at an opposite angle to that taken by the associated scales of the other two types. In all Mosquitoes the sides of the head are covered with a pavement of scales of the first-mentioned sort, but in all except those of the Megarhinina sub-family and in Toxorhijnchites, in which they uniformly cover the whole of the back of the head, they are associated with one or both of the other types of scale. In the tj^pe-genus Ciilex the flat scales are confined to narrow patches, quite away at the sides, while the whole of the middle portion is covered with narrow curved scales, which constitute the entire tomentum on
ON THE CLASSIFICATION OF THE FAMILY 259 the vertex and back of the nape, but alternate with upright forked scales on the band between. In Anopheles we have again the small lateral patches of imbricated fiat scales, but the entire vertex and nape is covered with alternate ranks of narrow, curved, and upright forked scales, such as are found on the limited band across the Culex head. In Mdes, the lateral patches of flat scales are so large as to leave only a narrow band between them, which is occupied by a stripe of narrow curved scales, no upright forked scales being present. Lastly, in Stegomyia, just as in Megarhina, the whole vertex and nape are covered with imbricated flat scales, but instead of forming the entire tomentum, as in the latter genus, the back of the Stegomyia head is additionally provided with a band of scales of the upright forked type. Below the figure of each head will be found one of the corresponding type of scutellum, an examination of which will show that in Culex and all other Mosquitoes it is covered with long narrow scales, except in the sub-family Megarhiniiia, and the genus Stegomyia, in both of which this portion of the thorax is wholly covered with flat scales. Below these figures will be found the types of clypei already sufficiently described, and next beneath them three figures illustrating the nude metanotum of Culex, and of all other Mosquitoes, except the two small genera Trichoprosopoji and Wyeomyia. In the latter of these, placed to the left, it will be seen that the usually bare surface is armed with a few stout bristles, while in the former, in addition to similar bristles, there are also patches of scales. In searching for the bristles peculiar to these genera, it is well to remember that they are easily knocked off, but that in such case the minute black specks, showing the spot from which they have been detached, can usually be made out. Mucidus has certain peculiar head scales. It should be added that in the above remarks, what has been said of Stegomyia applies equally to the closely-allied genus Armigeres, as to the distinctness of which Mr- Theobald himself expresses some doubt.
At the bottom are given drawings showing the form and arrangement of the scales of the veins of typical Culex, and
260 GNATS OR MOSQUITOES — CHAPTER X of those genera which have been separated on account of differences of this kind. In examining the wing for these characters, it is best to choose the middle portion, as they are often less well marked in other parts, and it may be added that occasionally, as in Panoplites uniformis, the characters are not equally well marked in both sexes, so that it is well, if possible, to examine both sexes. If we examine the wing-veins of a typical Culex, it will be found that they carry two sorts of scales, which alternate with each other. On each side of the vein, placed close against the membrane, is a row of short, racquet-shaped scales ; while between each of these, but springing more from the free convexity of the vein, are rows of long narrow linear scales, which give to the wing its characteristic shaggy appearance. In all but four genera the veins, though differing sometimes in the density of their armature, and in various other details, conform generally to this type, but in Panoplites, the short lateral rows of scales of the veins of Cidex, in place of being small and of symmetrical outline, are so large that, when one is accustomed to their appearance, the genus can be recognised by means of a simple lens. These peculiar scales are bracket-shaped, forming a nearly right-angled, round-cornered triangle, whose base is formed by the free end of the scale, a much magnified figure of which is given on the right of the figure of the Culex vein, while to its left is represented the arrangement along a vein of the two sorts of scales, the linear scales being in shape and length much as in Culex, but owing to the comparatively larger size of the lateral rows of scales, do not show so much beyond them ; so that the wing, viewed with a hand lens, looks characteristically scaly instead of woolly. The only other genus that has scales at all like these is Mdomyia, which has been separated from the other Mdomina on this account.
Another type of leptotaxis is exhibited by the wing-veins of the genus Tcp-niorhynchus (as modified), illustrated by the figure on the extreme right. Here the distinction between lateral and linear scales is hardly to be made out, as the ranks of scales are disposed in half spirals, along and across
ON THE CLASSIFICATION OF THE FAMILY 261 the veins, and there are at least three sets as to length, all of which, however, are of essentially the same outline, viz., symmetrical and obovate. The large proportion of long scales and their rounded ends, gives a peculiar velvety appearance to the wings of this genus. Although separated from Psorophora on other grounds, the veins in the genus Mucidus are equally peculiar, being provided with large, thick, obovate scales, with relatively long stems, all of which are of the same form and size. They also present the additional peculiarity of some of these being particoloured, the basal half of the scale being yellowish, and the distal black ; the all-yellow and the parti-coloured scales being disposed in alternate groups along the veins, so as to give a characteristically brindled and rough appearance to the wing.
It is interesting to be able to note that the soundness of Mr. Theobald's judgment in selecting characters of this description as the basis of his new generic scheme is receiving early confirmation from the notes that are coming to hand as to the habits of the species of the new genera. Thus, the eggs of some at least of the Stegomyia are deposited separately, like those of Anopheles, and much resemble them in structure and form. Those of Panoplites, too, and Mucidus are also deposited separately, a circumstance which alone should suffice to establish their generic distinctness from the Culices, with which they have been hitherto associated.
One of the new genera, Deinokerides, has been constituted on the peculiar character of the antennae, which have their second joints as long as three or four of those that succeed it, instead of being of about the same length, as in all other gnats, except the queer little Mosquito already mentioned, that inhabits the domiciles of West Indian shore-crabs.
Certain other genera are also distinguished from neighbouring forms by peculiarities of the venation of the wings, but the consideration of these points will be better left to be taken in connection with the grouping of genera in each sub-family.
262 GNATS OR MOSQUITOES — CHAPTER X 111 endeavouring, then, to make out the sub-family to which any specimen under examination should be referred, the points to be first noted are : — (1) The structure of the proboscis, whether formed for piercing or otherwise, and whether strongly curved or quite or nearly straight.
(2) The relative length of the palpi in both sexes and its form, where specimens of one sex only are available. This being determined, species may be referred to their genera in many cases by noting the following points : — (a) The character of the clypeus, whether nude or armed with bristles or scales.
(b) The form and arrangement of the scales on the vertex and nape, and also on the scutellum.
(c) The character of the metanotum, whether nude or provided with bristles, or with bristles and scales. (d) The relative length of the second antennal joint. It is a labour-saving plan to in each case note down these points in pencil before proceeding to the more detailed examination of other characters.
In each genus the series of descriptions will be found to be preceded by a table, intended to serve as a key to the ready identification of the species belonging to it ; and the species will be found to be arranged in the same order as they fall in the table.
The plan of these tables is based on that adopted by Ficalbi for the European species, but has been somewhat modified owing to the much larger number of species that have to be accounted for.
It must be, however, distinctly understood that tables of this sort can only be employed as a sort of convenient index to facilitate the ready identification of species, and neither are nor pretend to be, in any sense, a natural classification. At the same time, as a matter of fact, although occasionally species will be found to be inconveniently separated from their nearest allies, the plan adopted generally places closely allied species in fairly consecutive order. The general principle of the tables is to place first the most elaborately adorned species, and to relegate those that
ON THE CLASSIFICATION OF THE FAMILY '268 are uniformly tinted to the end. In the first edition the incompleteness of the original descriptions, which were all that one had to go upon, led to great difficulties in compiling the tables, and large numbers of species had to be relegated to special doubtful groups. In the present, however, very few such instances remain, as with a few exceptions, we have seen and handled all the species now redescribed.
The earlier entomologists framed their descriptions so as to distinguish between the dozen or so of species alone known to them, and probably in no way foresaw the immense number that would be added by subsequent observers.
Hence, in all but a few species that are so common that a sort of tradition as to their identity has been handed down, there must always remain a doubt as to the identity of the specimens subsequently referred to the names they gave, and even in the case of that commonest of all species, Gulex pipiens, it does not appear to be by any means certain that the species commonly so spoken of in England and Southern Europe is really identical with the Scandinavian species described by Linnaeus.
Before proceeding to attempt to trace out a species of the genus CiUex, for example, the following points should first be noted on a scrap of paper : — (1) Whether the wings are spotted or unspotted. If spotted, the number and position of the spots, and whether due to difference of colour or merely to local accumulations of scales.
(2) Whether or not the tarsal joints are adorned with lighter bands, and if so adorned, the exact position of the markings.
(3) Whether the thorax is adorned with markings, or not conspicuously adorned.
(4) Whether or not the abdomen is adorned with marks or cross-bands, and if so, their character and position, especially the relative position of lighter cross-marks on the segments.
With these points carefully noted, I hope that even the