Even more startling are the biological relations in the soil between the absorbing roots of the higher plants and some of these bacteria and fungi, for it has now been established beyond all doubt that certain fungi enter the living roots and there flourish not as mere destructive parasites, but as messmates not only tolerated by the plant, but even indispensable to its welfare. It is probable that nearly half the plants of our fields, moors, and forests entertain such fungi in their root-tissues. The curious, and long-known nodules on the roots of leguminous plants--peas, beans, clover, etc.--are filled with bacteria which enable these plants to avail themselves of the free nitrogen of the air, and so enrich the soil with nitrogenous substances. The roots of most forest trees, orchids, and plants of the moorlands, meadows and marshes are similarly occupied by fungi, which in some way convey salts--probably especially phosphates and potassium compounds--to the plant in return for the small tax of organic carbon-compounds it exacts from the latter. In some cases at any rate, as Bernard has lately shown, the very existence of the plant depends on its seedling roots obtaining this advantageous attachment and co-operation (symbiosis) of the fungus immediately on germination. These remarks must suffice to illustrate this part of my subject, and to emphasise the statement that the question whether a given plant can be grown in a given soil, is by no means one of simply the physical and chemical constitution of the latter. The plant will have to run the gauntlet of a long series of vicissitudes brought about by the presence or absence, relative proportions and vigour, and specific nature of the organisms in the soil at its roots, and it is easy to see that many cases of disease may be due to the absence of advantageous bacteria or fungi, or to circumstances which disfavour their life, as well as to the predominance of competing organisms. It will now be evident that the old points of view must be abandoned, and with them, especially, the widely prevalent notion that chemical analyses of the plant and soil can explain the real problems of agriculture.