Page:History of botany (Sachs; Garnsey).djvu/485

] earth, and all these principles together with those which are mixed with the water would weigh at least from two to three ounces, and this multiplied by four thousand, the number of the species of plants, would give a weight of five hundred pounds.

These arguments like those of Jung, and in the main also those of Malpighi, rested on facts which were on the whole as well known in ancient times as in the 17th century; but no one had before given heed to considerations, which were in themselves quite sufficient to do away with the Aristotelian teaching on the subject of the nutrition of plants.

In the second part of his letter Mariotte discusses the phenomena of vegetation which depend on nutrition; he compares the endosperm in the seed with the yolk of the egg in animals, and the entrance of the water into the roots with its rising in capillary tubes; he takes the milky juice to be the nutrient sap and compares it with arterial blood, the other watery juices answering to venous blood. He says something quite new about the pressure of the sap; he notices the high pressure at which the sap stands in plants, and concludes from it that there must be contrivances in them, which allow of the ingress of the water but not of its egress. The existence of the pressure is well demonstrated by the outflow from plants which contain milky juice when they are wounded, and is compared with the pressure on the blood in the veins. Equally striking is his further conclusion, that the pressure of the sap expands the roots, branches, and leaves, and so contributes to their growth. The sap, he adds, would not be able to remain at this pressure, if it did not enter by pores, which forbid its return. In these remarks lay the first germs of speculation on the growth of plants, such as we shall meet with in Hales also in a somewhat different form, but in the backward state in which phytotomy then was they could not at present be further developed; we shall recur to them further on, though in a different connection.