By means of a tuning-fork a spider may be made to eat what it would otherwise avoid. I took a fly that had been drowned in paraffin and put it into a spider's web, and then attracted the spider by touching the fly with a fork. When the spider had come to the conclusion that it was not suitable food, and was leaving it, I touched the fly again. This had the same effect as before, and as often as the spider began to leave the fly I again touched it, and by this means compelled the spider to eat a large portion of the fly. The few house-spiders that I have found do not seem to appreciate the tuning-fork, but retreat into their hiding-places as when frightened; yet the supposed fondness of spiders for music must surely have some connection with these observations; and when they come out to listen, is it not that they cannot tell which way to proceed? The few observations that I have made are necessarily imperfect, but I send them, as they afford a method which might lead a naturalist to notice habits otherwise difficult to observe, and so to arrive at conclusions which I in my ignorance of natural history must leave to others.[76] _General Habits._ Coming now to general habits, our attention is claimed by the only general habit that is of interest--namely, that of web-building. The instinct of constructing nets for the capture of prey occurs in no other class of animals, while in spiders it not only attains to an extraordinary degree of perfection (so that, in the opinion of some geometers, the instinct is not less wonderful in this respect than is that displayed by the hive-bee in the construction of its cells), but also ramifies into a number of diverse directions. Thus we have, in different species, wide open networks spread between the branches of bushes, &c., closely woven textures in the corners of buildings, earth tubes lined with silk, the strong muslin-like snare of the Mygale, which, as first noticed by Madame Merian,[77] and since confirmed by Bates,[78] is able to retain a struggling humming-bird while this most beautiful animal in creation is being devoured by the most repulsive; and many other varieties might be mentioned. It may at first sight appear somewhat remarkable that this instinct of spreading snares should on the one hand occur only in one class of the animal kingdom, while on the other hand, in the class where it does occur, it should attain such extreme perfection, and run into so much variety. But we must here remember that the development of the instinct obviously depends upon the presence of a web-secreting apparatus, which is a comparatively rare anatomical feature. In caterpillars, which are not predaceous, the web is used only for the purposes of protection and locomotion; and it is easy to see that the spreading of snares would here be of no use to the animals.