50 HORTICULTURAL SOCIETY OF NEW YORK. 



vided it in a normal way in the season of summer rains. The 

 power to absorb and transmit water was sufficient in all the 

 forms to support life (that is, to prevent wilting and desicca- 

 tion) during the drought period, but was adequate for growth 

 during this period only in castor bean and muskmelon. 



That the responses just described were due to the change 

 in the evaporating power of the air is hardly to be doubted. 

 There was not a sufficiently great change in temperature to ac- 

 count for it, as has been shown in preceding paragraphs. t It 

 might be suggested that the decrease in light intensity incident 

 to the oncoming of the summer season may have been the cause 

 of the response noted; but in the absence of reliable data as to 

 the effect of such variations in light intensity (we have as yet 

 no practicable photometer for such studies that measures the 

 energy of the light as a whole instead of measuring mainly the 

 less refrangible portion) and in the face of the a priori con- 

 sideration that it is fully as probable that such variations affect 

 the plant through changes in the evaporating rate as that the 

 light intensity is per se the active agent, the evaporation power 

 of the air seems by far the most probable climatic element to 

 which to attribute the plant responses here dealt with. 



Some of the native desert perennials respond to the change in 

 seasons in a marked way, as by losing their foliage in the dry sea- 

 son ; but it is difficult in these cases to distinguish between the ef- 

 fects of dry soil and those of high evaporating power of the air. 

 The deeper soil layers of the Desert Laboratory reservation con- 

 tain throughout the year a considerable water supply, so that it 

 is entirely possible that the more deeply rooting perennials do 

 not suffer directly from desiccation of the soil. A case where 

 the rate of evaporation appears to be the controlling factor in 

 the seasonal response is furnished by the experiment of Lloyd 

 with Foiiquieria splend'ens, in which he was able to cause the 

 local development of leaves during the dry season by simply 

 wrapping a portion of the leafless stem with a cloth kept wet by 

 means of a siphon. No water was allowed to reach the soil 

 about the plant, so that the response must have been due either 

 to the local check imposed upon transpiration or to water ab- 

 sorbed through the bark of the stem, bud scales, etc. From 



