344 smith : APPLICATION OF THE THEORY OF 



is in a straight line from the point of origin to the earth. 

 Their growth is surprisingly rapid. 7 mm. per hour having been 

 irded. 



Plate 25 shows the curve of growth in mm. per two hours 

 (three added) from July 6th, 10 a.m., to July 7th, 5 p.m., with 

 the curve of humidity, and below two growth curves for two 

 separate roots from January 9th. 6 a.m., to January 11th, 7 

 \.m. . with the curve of temperature for the same period. Tak- 

 ing the curves of earlier date first there is no doubt of the general 

 correspondence between the curve of growth and the curve 

 of temperature. The most rapid growth is always during 

 the day. though these were days of continuous sun and low 

 humidity, and the slowest growth coincides with the minimum 

 temperature, which it must again be pointed out was on these 

 dates much lower than is experienced at Peradeniya during 

 the greater part of the year. 



The curve for the same roots on July 6th to 7th shows a 

 striking contrast to that on January 9th to 11th. Here the 

 occurence of low humidity and sunshine during the day is 

 always followed by either slow growth, total cessation of 

 growth, or even in some cases a decrease in length. During 

 the night or during heavy rain in the daytime the growth is 

 rapid. There is no trace, even during the night when water 

 Bupply might be expected to be in excess, of any correspon- 

 dence between growth and temperature. Throughout the 

 whole period there is visible a close correspondence between 

 growth and humidity , and whether by night or by day an in- 

 crease of humidity is followed by an increase of growth. 



This coni rasi bel ween i he ourves of growl h for two different 

 tune- ol the year is puzzling, tor it will be noted thai one set 

 of temperatures is not uniformly either higher or Lower than 

 the other. If the January ourve had been Lower throughout 

 the case would have been easily explained on our theory, for 



at 1 he low t .■m pe] al u re we should e\ peel I he temperature to be 



limiting, while when the temperature rose the supply of water 

 would probably fall short oi that required for the more rapid 



