BOOK in.] Introduction. 373

nutrition in plants with Saussure's results. Agricultural chemists were chiefly engaged till nearly 1860 with the questions, whether all or certain constituents of the ash of a plant are indispensable parts of its food, and whence these constituents are derived, and with cognate considerations on the exhaustion of the soil by cultivation and its remedy by suitable manuring. In France Boussingault had undertaken experimental and analytical investigations on these subjects before 1840, and it was he who in the course of the next twenty years made the most valuable physiological discoveries ; of these the most important was the fact that plants do not make use of free atmospheric nitrogen as food, but take up compounds of nitrogen for the purpose. In Germany the interest in such questions was increased by the instrumentality of Justus Liebig, who gathered from the knowledge that had been accumulated up to 1840 all that was fundamental and of real importance, and drew attention to the great practical value of the theory of the nutrition of plants in agriculture and in the management of woods and forests ; considerable state provision was soon made for investigations of the kind, but these often wandered from the right path for the reason, that being designed to promote practical interests they lost sight of the inner connection between all vital phenomena. Still a great mass of facts was accumulated, which careful sifting might afterwards render serviceable to pure science. Some of the best agricultural chemists deserve the credit of vindicat- ing purely scientific as well as practical points of view, and explained in comprehensive works the general subject of the nutrition of plants, so far as it was possible to do so without going deeply into their organisation ; among these were Bous- singault and the Germans Emil Wolff and Franz Schulze. But the questions of the nutrition of plants, which are con- nected with the chemical processes of assimilation and meta- bolism within them, remained still undecided, though some

valuable preliminary work on these points dates from this time.