But in spiders, of course, the case is otherwise. Once granting the power of forming a web, and it is evident that there is much potential service to which this power may be put with reference to the voracious habits of the animal; and therefore it is not to be wondered that both the anatomical structures and their correlated instincts should attain to extreme perfection in sundry lines of development. The origin of the web-building structure was probably due to the use of the web for purposes of locomotion or of cocoon-spinning, as we see it still so used in the same way that it is used by caterpillars for descending from heights, and in the case of the gossamer spider for travelling immense distances through the air. As the anatomical structures in question differ very greatly in the case of spiders and in that of caterpillars, we may wonder why analogous if not homologous structures should never have been developed in the case of any other animal having predaceous habits--especially, perhaps, in that of the imago form of predaceous insects. It is easy to see how, if there were any original tendency to secrete a viscid substance in the neighbourhood of the anus, this might be utilised in descending from low elevations (as certain kinds of slugs use their viscid slime as threads whereby to let themselves down from low branches to the ground); and so we can understand how natural selection might thus have the material supplied out of which to develop such highly specialised organs as the spinnerets of a spider. But if we are inclined to wonder why this should not have happened among other animals, we must remember that any expectation that it should rests on negative grounds; we have no reason to suppose that in any other case the initial tendency to secrete a viscid substance was present. One inference, however, in the case of spiders seems perfectly valid. As this comparatively rare faculty of web-spinning occurs so generally throughout the class, it must have had its earliest origin very far back in the history of that class, though probably not so far back as to include the common progenitors of the spiders and the scorpions, seeing that the latter do not spin webs. I shall now give a few details on the manner in which spiders' webs are made. Without going into the anatomy of the subject further than to observe that a spider's 'thread' is a composite structure made up of a number of finer threads, which leave their respective spinneret-holes in an almost fluid condition, and immediately harden by exposure to the air, I shall begin at once to describe the method of construction.