POSITION OF AUDITORY RODS.

113

in other parts of the body; for instance, they have been discovered in the antennæ of a water-beetle (Dytiscus) and of Telephorus by Hicks, Leydig, and Graber, and in the body segments of several larvæ by Leydig, Weissmann, Graber, Grobben, and Bolles Lee. In the larva of Dytiscus, indeed, they have been observed in the body, antennæ, palpi, under lip, and legs. Moreover, while, as we have seen, in the tibia of Orthoptera and the halteres of flies they are numerous, in some of these cases they are few, sometimes, indeed, only a single rod being present, as discovered by Grobben in Ptychoptera. Nevertheless the evidence that they are really acoustic organs is, in the case of the Orthoptera, so strong, their structure is so peculiar, and the gradation of these organs from the most complex to the most simple is so complete, that it seems reasonable to attribute to them the same function.

*

Moreover, as regards the very simplest forms there is another consideration pointing to this conclusion. We have seen that in the Orthoptera the terminal filaments close up, and are attached to the skin. Now, it seems to be a very general rule, in reference to these organs, that they are attached to the skin at two points, between which is situated the attachment of the nerve. These points, moreover, are so selected as to be maintained at the same distance from one another, thus preserving an equable tension in the connecting filament.

Fig. 72, for instance, represents part of one segment of the body of the larva of a gnat (Corethra). This larva is as transparent as glass, and very common in ponds, a most beautiful and instructive microscopic object. EG is the ganglion; a is the nerve in question, which

*Sitz. der K. Akad. der Wiss. Wien, 1876.

I

114

CHORDOTONAL ORGAN OF GNAT-LARVA.

swells into a little triangular ganglion at g; from g the auditory organ runs straight to the skin at e, and contains two or three auditory rods (not, however, shown in the figure) at the point Chs; in the opposite direction, a fine ligament passes from g to the

Fig. 72.-Right half of eighth segment of the body of the larva of a gnat (Corethra plumicornis); after Graber. EG. Ganglia; N, nerve; g, auditory ganglion; gb, auditory ligament; Chs, auditory rods; a, auditory nerve; e, attachment of auditory organ to the skin; b, attachment of auditory ligament to the skin; hn, hn', termination of skin-nerve; tb, plumose tactile hair; h, simple hair; tg, ganglion of tactile hair; Im, longitudinal muscle.

skin at b. Hence the organ ge is suspended in a certain state of tension, and is favourably situated to receive even very fine vibrations.*

There are, as we have seen, a large number of observations which point to the antennæ as organs of hearing, and many more might have been given. When we come to consider, however, the anatomical provision which renders the perception of sound

Similar organs occur in other insects, as, for instance, in Ptychoptera.

AUDITORY HAIRS ON ANTENNE OF GNAT. 115 possible, we are met by great difficulties. The evidence is, I think, conclusive that the antennæ are olfactory as well as tactile organs, and I believe that they serve also as organs of hearing. There are, moreover, as shown in the last chapter, various remarkable structures in the antennæ, and I have given reasons for thinking some of them to be the seat of the sense of smell. Which, if any, of the remainder convey the sense of sound, it is not easy to determine. I have suggested that Hicks's bottles (Fig. 43) may act as microscopic stethoscopes; but they occur, so far as we at present know, only in ants and certain bees.

That some of the antennal hairs are auditory can, I think, no longer be doubted. Johnson, whose figure I give (Fig. 73), suggested† in 1855 that the hairs on the antennæ of gnats serve for hearing. Mayer also,‡

* I am glad to see that Leydig, who, however, does not appear to have read either Hicks's paper or mine, also regards these as chordotonal organs (Zool. Anz., 1886).

+ Quarterly Journal of Microscopical Science, 1855.

American Journal of Science and Arts, 1874.

116

SYMPATHETIC VIBRATIONS.

led by the observations of Hensen, has made similar experiments with the mosquito, the male of which has beautifully feathered antennæ. He fastened one down on a glass slide, and then sounded a series of tuningforks. With an Ut, fork of 512 vibrations per second he found that some of the hairs were thrown into vigorous movement, while others remained nearly stationary. The lower (Uts) and higher (Ut,) harmonics of Ut also caused more vibration than any intermediate notes. These hairs, then, are specially tuned so as to respond to vibrations numbering 512 per second. Other hairs vibrated to other notes, extending through the middle and next higher octave of the piano. Mayer then made large wooden models of these hairs, and, on counting the number of vibrations they made when they were clamped at one end and then drawn on one side, he found that it "coincided with the ratio existing between the numbers of vibrations of the forks to which co-vibrated the fibrils." It is interesting that the hum of the female gnat corresponds nearly to this note, and would consequently set the hairs in vibration.

Moreover, those auditory hairs are most affected which are at right angles to the direction from which the sound comes. Hence, from the position of the antennæ and the hairs, a sound will act most intensely if it is directly in front of the head. Suppose, then, a male gnat hears the hum of a female at some little distance. Perhaps the sound affects one antenna more than the other. He turns his head until the two antennæ are equally affected, and is thus able to direct his flight straight towards the female.

The auditory organs of insects, then, are situated in

ORGANS OF HEARING IN VARIOUS PARTS OF BODY. 117

different insects in different parts of the body, and there is strong reason to believe that even in the same animal the sensitiveness to sounds is not necessarily confined to one part. In the cricket, for instance, the sense of hearing appears to be seated partly in the antennæ, and partly in the anterior legs. In other cases, as in Corethra, the division appears to be carried still further, and a "chordotonal" organ occurs in each of several segments.

No doubt the multiplication of complex organs, like our ears, arranged as they are to appreciate a great variety of sounds, would be so great a waste that any theory implying such a state of things would be quite untenable; but with simple organs, such, for instance, as that of Corethra* (gnat; Fig. 72), the case is different, and there would seem to be an obvious advantage in such organs occurring in different parts of the body, ready to receive sound-waves coming from different directions. Moreover, the different organs exist; they do not appear to be organs of touch, yet they are clearly organs of sense, and that sense, whatever it be, whether hearing or any other, and though it may well be simple, and even perhaps confused, must be seated in various parts of the body. The fact of their being so distributed does not make it more improbable that they should be organs of hearing, than of any other sense.

At the same time, it is an interesting result of recent investigations that the auditory organs of insects are not only situated in various parts of the body, but are constructed on such different principles.

* Where, however, the number does not approach to that in certain Medusa (see ante, p. 84).