424 Mr. D. Thoday. Experimental Researches on [Mar. 1, 



25° C. ; minimum, 18° C. Experiment closed at 3 p.m. ; duration, five 

 hours. 



Table I. — Asymmetry error per sq. decim. per hour : average about 

 • + 1'6 milligrammes ; maximum, about + 3 milligrammes. 



a = control half -leaf ; h = experimental half -leaf. 



Leaf. 



Area, in 

 sq. cm. 



T)rv 



weight, 

 in 



grammes. 



Estimated 



change of 

 area of 

 experimental 

 half-leaf. 



Corrected 



dry 

 weight, 

 in 



grammes. 



Increase 

 of dry 

 weight, 

 in 



grammes. 



vJ clill IJci. 



hour per sq. 

 decim., in 

 milligrammes. 



of leaf 

 during 

 experiment. 



1 



{a) 

 (*) 



60 

 60 



0 -2653 

 0 -3130 



-t-1 -2 



0 -2653 

 0 -3168 



0 -0515 



17 -2 



Turgid. 



2 



(a) 

 (*) 



60 

 60 



0 ■2362 

 0 -2744 



-H -2 



0 -2362 

 0 -2777 



0 -0415 



13 -8 





3 



(a) 

 (S) 



70 

 70 



0 -2705 

 0 -3245 



-hO-9 



0 -2705 

 0 -3274 



0 -0569 



16 -3 





4 



(«) 

 (6) 



40 

 40 



0 -1652 

 0 -1832 



-Hi -0 



0 -1652 

 0 -1850 



0 -0198 



9-9 



Limp. 



5 



(a) 

 (*) 



75 

 75 



0 -3333 

 0 -3753 



-7-7 



0 -3333 

 0 -3463 



0 -0130 



3-5 



Wilted. 



The average of the rates of increase shown by the turgid leaves is 

 15'8 milligrammes, which approaches very nearly to Sachs' result, 16'5 milli- 

 grammes. It is to be noted that the latter was obtained during a period of 

 unl)roken sunshine, in the open air ; the conditions under which my experi- 

 ment was carried out were less favourable and less imiform. 



The other two leaves, 4 and 5, gave much smaller results. Leaf 4 had 

 been a little limp throughout the experiment, and only recovered its turgor 

 when it was taken into the laboratory. It showed a rate of increase of 

 10 milligrammes per square decimetre per hour. 



Leaf 5 was quite flaccid throughout the experiment and shrank in area to 

 the extent of al)out 8 per cent. Its stomata must have closed almost com- 

 pletely soon after it was taken into the open air, and it increased in weight 

 at the relatively small rate of 3'5 milligrammes per square decimetre 

 per hour. 



Subsequent sets of experiments all gave similar results. 



In these later experiments of 1909 tlie templet method employed in 

 Expt. 1 was superseded by the stamping method, whereby errors from 

 change of area are completely eliminated* and laborious corrections avoided. 



* Thoday, loc. ctt., j). 44. 



