INTERNAL TElVrPBRATURE OF LEAVES. 291 



The month of March is dry, with less than 4 inches rainfall, 

 and the rainfall increases in April and May. In April there is 

 still a good deal of sunshine, and a moderate amount in May. 

 In June the character of the weather changes. With the 

 breaking of the '• big monsoon," as it is called, not only is there 

 a large increase in the amount of rainfall reaching the large 

 total in June of 47 inches, but it becomes more continuous 

 with scarcely any sunny intervals. This type of weather con- 

 tinues through July and August. Though the rainfall is less 

 in these months, yet the rain is fairly continuous and only faUs 

 off in intensity. The most striking point in the curve is 

 the falling off of the yield of tea during the end of May and 

 through June. The lower value is continued through July 

 and August. The falling off coincides approximately with the 

 break of the monsoon about the end of May and the consequent 

 ushering in of almost sunless weatlier. If the figures of relative 

 humidity of the atmosphere were available, tliey would no 

 doubt be even more to the point than those of tlie rainfall. 

 But the correspondence between the falling off of the amount 

 of tea plucked in June and the increase in the rainfall suffii-: 

 oiently illustrates the point. In this case the factor limiting 

 the growth is not moisture, nor is it temperature, as the 

 average temperature during these months is slightly higher 

 than in the drier periods. It seems justifiable to suppose that 

 in this case the limiting factor is mineral food supply, as the 

 transpiration stream bringing the dissolved salts is almost 

 absent. If this is the case for a shrub-like tea, which under 

 suitable conditions is making some new growth at all periods 

 of the year, it would apply with special force to those species 

 in which the habit of rapid growth for a very limited period 

 has been evolved. 



Those species, whose growth period occurred in the wet 

 months, would have their growth checked by lack of mineral 

 food. Any variation toward a drier season would produce a 

 more vigorous growth and a larger tree . other conditions being 

 equal. Thus the habit of growth in the dry season would be 

 confirmed, and the habit of growth in the wet season checked, 

 until of the prevailing indigenous trees a majority grew in the 

 dry season. On this view we must suppose that those species 

 7(1)09 ' (38) 



