28 THE JOURNAL OF INDIAN BOTANY
some places it was reported that Babul trees {Acacia arabica Willd.) had died. As a matter of fact many of the Banian trees (Ficus benga- lensis L) especially those which happened to be growing on road sides near dry hills showed an unusual leaf-fall and remained practically bare of leaves for some time. Even the very resistant prickly pear [Opuntia nigricans Haw.) showed signs of suffering, and in April 1919 had taken on a yellow colour and was drooping over extensive areas. It seemed a matter of some interest under these very exceptional conditions to ascertain what plants remained actually alive and grow- ing, and to find the amount of available moisture round the roots of such plants. The enquiry did not concern itself with trees, but only with herbaceous and sub-woody plants, and particular attention was devoted to (l) finding out to what degree the moisture conditions can be reduced, without preventing the wild indigenous weeds, mostly of arable land, from living and growing (2) *noting the general behaviour of the plants or the particular adaptation of the different plant- organs which seemed to enable the plants to tide over the drought, and (3) making special notes of any plants of agricultural or other economic value (especially of fodder value) that lived and grew through the drought, so that they might be tried as additions to the drought resisting plants actually grown in the country. This last aspect of the study was considered to be of very great importance, for shortness of fodder is now the great characteristic of famines in the Deccan.
A chain of stations was made across the Deccan, and working from these all the plants, whether likely to be of use or no, which were living and growing at the hottest and driest part of 1919 in the month of May were secured, and in many cases samples of the soil in which they were growing were collected for a moisture determina- tion. The samples of soil were taken, chiefly from the lower one third of the root region.
Moisture determination in the soil thus collected gave some difficulty. It was obvious that a mere determination of the loss on heating in a steam bath would not give the actual amount of "free" water in the soil, — and it is the "free" water that it was important to
• In doing this a comparison of the herbarium material of the difierent species collected from time to time in different situations as regards rainfall and soil, conditions was made, with the plants now collected to see what deviations from the normal type could be detected in each case. And although we cannot rely solely upon such a comparison in determining the behaviour of a plant in a particular situation under different conditions of soil moisture still in the absence of any direct comparison, that was the only course left open.
