of the nervure of another species of Gryllus, viz., G. domesticus. With respect to the formation of these teeth, Dr. Gruber has shown" that they have been developed by the aid of selection, from the minute scales and hairs with which the wings and body are covered, and I came to the same conclusion with respect to those of the Coleoptera. But Dr. Gruber further shows that their development is in part

FIG. 13.-Chlorocoelus Tanana (from Bates). a, b. Lobes of opposite wing-covers.

directly due to the stimulus from the friction of one wing over the other.

In the Locustide the opposite wing-covers differ from each other in structure (Fig. 13), and the action cannot, as in the last family, be reversed. The left wing, which acts as the bow, lies over the right wing, which serves as the fiddle. One of the nervures (a) on the under surface of the former is finely serrated, and is scraped across the

"Ueber den Tonapparat der Locustiden, ein Beitrag zum Darwinismus,' “Zeitsch. für Wissensch. Zoolog.,” B. xxii., 1872, p. 100.

prominent nervures on the upper surface of the opposite or right wing. In our British Phasgonura viridissima it appeared to me that the serrated nervure is rubbed against the rounded hind-corner of the opposite wing, the edge of which is thickened, colored brown, and very sharp. In the right wing, but not in the left, there is a little plate, as transparent as talc, surrounded by nervures, and called the speculum. In Ephippiger vitium, a member of this same family, we have a curious subordinate modification; for the wing-covers are greatly reduced in size, but "the posterior part of the pro-thorax is elevated into a kind of dome over the wing-covers, and which has probably the effect of increasing the sound." "

We thus see that the musical apparatus is more differentiated or specialized in the Locustida (which include, I believe, the most powerful performers in the Order), than in the Achetidæ, in which both wing-covers have the same structure and the same function." Landois, however, detected in one of the Locustidæ, namely, in Decticus, a short and narrow row of small teeth, mere rudiments, on the inferior surface of the right wing-cover, which underlies the other and is never used as the bow. I observed the same rudimentary structure on the under side of the right wing-cover in Phasgonura viridissima. Hence we may infer with confidence that the Locustidæ are descended from a form in which, as in the existing Achetidæ, both wingcovers had serrated nervures on the under surface, and could be indifferently used as the bow; but that in the Locustidæ the two wing-covers gradually became differentiated and perfected, on the principle of the division of labor, the one to act exclusively as the bow, and the other as the fiddle. Dr. Gruber takes the same view, and has shown that rudimentary teeth are commonly found on the inferior surface of the right wing. By what steps the more simple apparatus in the Achetidæ originated, we do not

37 Westwood, "Modern Class. of Insects," vol. i. p. 453.

38 Landois, "Zeitsch. f. Wiss. Zoolog.," B. xvii., 1867, s. 121, 122.

know, but it is probable that the basal portions of the wing-covers originally overlapped each other as they do at present; and that the friction of the nervures produced a grating sound, as is now the case with the wing-covers of the females." A grating sound thus occasionally and accidentally made by the males, if it served them ever so little as a love-call to the females, might readily have been intensified through sexual selection, by variations in the roughness of the nervures having been continually preserved.

In the last and third Family, namely, the Acridiidæ,

or grasshoppers, the stridulation is produced in a very different manner, and, according to Dr. Scudder, is not so shrill as in the preceding Families. The inner surface of the femur (Fig. 14, r) is furnished with a longitudinal row of minute, elegant, lancetshaped, elastic teeth, from 85 to 93 in number;" and these are scraped across the sharp, propratorum; r, the stridulating ridge; jecting nervures on the winglower figure, the teeth forming the ridge, much magnified (from Landois). covers, which are thus made to vibrate and resound. Harris' says that when one of the males begins to play, he first "bends the shank of the hindleg beneath the thigh, where it is lodged in a furrow designed to receive it, and then draws the leg briskly up and down. He does not play both fiddles together, but alternately, first upon one and then on the other." In many species the base of the abdomen is hollowed out into a great cavity which is believed to act as a resounding board. In Pneumora (Fig. 15), a South African genus belonging

FIG. 14.-Hind-leg of Stenobothrus

39 Mr. Walsh also informs me that he has noticed that the female of the Platyphyllum concavum, "when captured, makes a feeble grating noise by shuffling her wing-covers together.

40 Landois, ibid., s. 113.

41 "Insects of New England," 1842, p. 133.

to the same family, we meet with a new and remarkable modification; in the males a small notched ridge projects obliquely from each side of the abdomen, against which the hind femora are rubbed. As the male is furnished with wings (the female being wingless), it is remarkable that the thighs are not rubbed in the usual manner against the wing

42

FIG. 15.-Pneumora (from specimens in the British Museum). Upper figure, male; lower figure, female.

covers; but this may perhaps be accounted for by the unusually small size of the hind-legs. I have not been able to examine the inner surface of the thighs, which, judging from analogy, would be finely serrated. The species of Pneumora have been more profoundly modified for the sake of stridulation than any other orthopterous insect; for in the

42 Westwood, "Modern Classification," vol. i. p. 46.

male the whole body has been converted into a musical instrument, being distended with air, like a great pellucid bladder, so as to increase the resonance. Mr. Trimen informs me that at the Cape of Good Hope these insects make a wonderful noise during the night.

In the three foregoing families the females are almost always destitute of an efficient musical apparatus. But there are a few exceptions to this rule, for Dr. Gruber has shown that both sexes of Ephippiger vitium are thus provided, though the organs differ in the male and female to a certain extent. Hence we cannot suppose that they have been transferred from the male to the female, as appears to have been the case with the secondary sexual characters of many other animals. They must have been independently developed in the two sexes, which no doubt mutually call to each other during the season of love. In most other Locustida (but not according to Landois in Decticus) the females have rudiments of the stridulatory organs proper to the male, from whom it is probable that these have been transferred. Landois also found such rudiments on the under surface of the wing-covers of the female Achetidæ, and on the femora of the female Acridiida. In the Homoptera, also, the females have the proper musical apparatus in a functionless state; and we shall hereafter meet in other divisions of the animal kingdom with many instances of structures proper to the male being present in a rudimenta y condition in the female.

Landois has observed another important fact, namely, that in the females of the Acridiidæ, the stridulating teeth on the femora remain throughout life in the same condition in which they first appear during the larval state in both sexes. In the males, on the other hand, they become further developed, and acquire their perfect structure at the îast moult, when the insect is mature and ready to breed.

From the facts now given, we see that the means by which the males of the Orthoptera produce their sounds are extremely diversified, and are altogether different from