desired substances in the same amount as in the wild. When the problem of cultivating the guayule presented itself, the first question that arose, and most naturally, was the relation of secretion of rubber to the supply of soil water. Though there is no very good a priori ground for expecting one result or the other, it was known that the amount of rubber obtained from certain latex trees is influenced by the amount of soil water during the period of tapping. It was also found that in the guayule the occurrence of rubber in the new growths of the particular season intervenes only after a distinct interval of time, for which reason the collecting of the shrub at the close of the rainy season involves a great waste, since the increased weight is in no degree to be referred to a greater rubber content.
The careful study of the guayule plants which were under cultivation and had been freely irrigated developed, in addition to several other points of interest, the knowledge that the greater the amount of growth (itself dependent on the amount of water available), the slower the secretion of rubber and the smaller its total amount relative to the volume of the tissues. On the other hand, the rate of growth under irrigation is so much more rapid that the total volume of rubber-bearing tissue is very greatly in excess of that in field plants. Irrigated plants present other features of difference which must be taken into account by the manufacturer and which are of very great interest to students of the physiology of desert plants. I need mention only the more salient here, namely, the greater relative volume of the woody portion of the stem, its harder and more rigid character, the smaller volume of the cortex (bark) (Fig. 16) and the quantity of rubber in individual secreting cells.
The increased hardness of the "wood" in the irrigated plant is due, in part at least, to the reduced volume of the medullary rays. And as