122 SCIENCE AND FRUIT GROWING 



thermometer does not really measure the amount of radiation 

 occurring, but is affected chiefly by the temperature of the air 

 surrounding the instrument, and this temperature will be higher 

 on the high ground, though radiation is most active there, owing 

 to the flowing away of the air as it becomes cooled (see p. 127) . 

 For a thermometer to measure radiation properly, it should 

 be in vacua, with its bulb blackened, as in the case of a solar 

 radiation thermometer. 



The earth temperature is a better guide to the energy of 

 radiation than the earth radiation thermometer, and the indica- 

 tions here are very clear as to the greater radiation occurring 

 at the upper station. In the summer radiation will tell in the 

 direction of increasing the earth temperature, the heat absorbed 

 from the sun being greater than that lost by the ground for good 

 radiators, it must be remembered, are good absorbers (p. 126) ; 

 whereas in winter, with its feebler sun and long, cold nights, the 

 balance of radiative action will be in the direction of a loss of 

 heat by the soil. The earth temperatures show that in both 

 seasons there is greater radiation activity at the upper than at 

 the lower station : in summer the earth at the upper station 

 is o'86 hotter than at the lower, in winter it is 0-97 colder, 

 in spring and autumn intermediate values obtain; and the 

 combined difference of nearly 2, applying as it does to points 

 6 inches below the surface, must be considerably smaller than 

 the difference at the surface itself, which is the seat of radiation. 

 The graphic representation of the values in Fig. 17 show very 

 clearly the reversal of the conditions at the two stations in 

 winter and summer. 



The remaining feature measured at the two stations is the 

 relative humidity of the air, and here the results are rather 

 surprising, for the air at the upper station is found to be appreci- 

 ably moister than at the lower, at any rate in the summer and 

 autumn, the values at the two stations for the winter and spring 

 showing, with certain irregularities (see Fig. 17), no appreciable 

 difference. The explanation of this is not very evident. Moist 

 air, it is true, is lighter than dry air at the same temperature, 

 it consequently rises and expands, becoming cooled in so 

 doing, and its relative humidity is thereby increased, even to the 

 point of water separating from it in the form of mist one of 

 the modes of origin of clouds ; but it is difficult to imagine that 

 any such action can occur to an extent sufficient to produce 

 measurable results when a rise of only 15 feet is in question. 

 It is more likely that the differences observed are, in a sense, 



