No work could be carried on without their help; it is bacteria which, at every stage of decay, assist in breaking up leaves, twigs, insects' bodies, worm-casts, and other manures. The way in which they work is too difficult to explain here, but to get an idea of the romance of the underground world one must try to picture to oneself these swarms and myriads of germs and bacteria all incessantly and busily engaged at their several duties. In the uppermost layers there are probably in a single cubic inch of good soil from 54,000,000 to 400,000,000 of these microbes. Many are absolutely necessary to the harvest; a few may be of little importance, but there are sure to be some of those dangerous sorts which might devastate a continent with disease in a single summer. There are also quantities of other fungi. The fairy rings which one sees year after year in widening circles of bright, fresh green are the work, not of fairy footsteps, but of an underground fungus (_Marasmius oreades_ and others). Its threads are thin, white, and delicate; they attack the roots of grasses, etc., on the outer side of the ring. It is therefore on this outer side yellow, dry, and more or less withered. On the inner side, however, the grass is luxuriant and of a rich bright green. Here the fungus has died off, and its remains, as well as those of the plants which it destroyed, form a rich manure for the new grass following on its track. Every year the ring widens; at a certain time in summer one sees the irregular line of mushroom-like fungi which are formed by the destructive underground absorbing threads. This, however, is but one of the underground fungi. There are many kinds; some are useful, others are very destructive. Upon the upper surface of the soil there falls not only rain, but another sort of rain consisting of seeds, dead leaves, insects' bodies, fungus spores, bacteria, and dust. Every year when the ploughman turns the sod there is a revolution in the whole of these populations. So far nothing has been said about the roots themselves, which penetrate, explore, and exploit all these layers of dead leaves, soil, and subsoil. The length of roots produced is very much greater than any one would suppose. A one-year-old Scotch fir seedling when grown in sand produced in a season a total length (branches, etc.) of no less than thirty-six feet of root. The total surface of this root system was estimated to be about twenty-three square inches. This little Scotch fir after six months' growth was laying under contribution a cone of earth twenty to thirty inches deep and with a surface of 222 square inches. In certain kinds of corn the same author estimated the total length of the roots as from 1500 to 1800 feet. S. Clark estimated the length of the roots of a large cucumber plant as amounting to 25,000 yards (fifteen miles), and made out that it was occupying a whole cubic yard of ground.