CH. Ill, 6] TRANSPIRATION FROM PLANTS 45 



Little less surprising than the copiousness of transpiration 

 is the variability in its amount. Much depends upon the 

 character of the plant, for, in general, thick-leaved compact 

 kinds transpire less than thin-leaved open sorts, and hairy 

 less than smooth kinds, and slow-growing less than quick- 

 growing, though occasional surprising exceptions to these 

 rules occur. 

 But it also 

 varies greatly 

 at different 

 times in the 

 same plant, as 

 shows very 

 clearly when 

 a plant is 

 weighed fre- 

 quently, or 

 still better, is 

 made to write 

 upon a drum 

 of a transpiro- 

 graph (Figs. 

 20, 21) a con- 

 tinuous rec- FIG. 19. A plant prepared for weight-determinations 

 nrrl nf ltd nwn of the amount of transpiration ; X f . 



A thin aluminum shell covers the pot, and the roof is 

 transpiration rubber, which may be lifted at will for watering and 



day and night a6ratingthesoiL 



for a week or longer, the proper arrangements of course 

 being made to insure that all water loss shall take place from 

 the plant alone (as in Fig. 19). If simultaneously, whether 

 by personal observation or by use of recording meteorolog- 

 ical instruments, records are taken of the conditions of 

 weather, temperature, humidity, light, winds, the reason 

 for the fluctuations in transpiration is found. For thus it 

 becomes clear that the rate of transpiration is increased by 

 light, heat, dryness (of the air), and winds, and is lessened by 



