the style therefore, as might be expected, is dry and unattractive; the writer seems to confine himself too anxiously within the
limits of what is given in experience, and there is no doubt that
many errors in later times might have been avoided if the
inductive proof of de Saussure's doctrines had been accompanied with a deductive exposition of them of a more didactic
character.
The processes of vegetation examined by de Saussure were, for the most part, the same as those which Ingen-Houss and Senebier had studied at length and correctly described in their general outlines. But de Saussure went beyond this, and by means of quantitative determinations struck a balance between the amount of matter taken up and given off by the plant, thereby showing what it retains. In this way he made two great discoveries: that the elements of water are fixed in the plant at the same time as the carbon, and that there is no normal nutrition of the plant without the introduction of nitrates and mineral matter. But we cannot form a due idea of de Saussure's services to physiology without going further into the detail of his work.
We will first consider his investigations respecting the assimilation of carbon in plants. Here we have the important result, that larger quantities of carbon dioxide in the atmosphere surrounding the plants are only favourable to vegetation if the latter are in a condition to decompose them, that is, if they are in sufficiently strong light; that every increase in the amount of carbon dioxide in the air in shade or in darkness is unfavourable to vegetation, and that if that increase is greater than eight times in the hundred it is absolutely injurious. On the other hand he found, that the decomposition of carbon dioxide by the green parts in light is an occupation that is necessary to them, that plants die when they are deprived of it. The first clear insight into the chemical processes which accompany the decomposition of carbon dioxide in the interior of the plant was obtained by perceiving, that plants by appro-
