S6U SMITH : APPLICATION OF THE THEORY U) 



growth at Hakgala should on the whole be parallel with the 

 temperature curve. The following observations bear on this 

 point. 



On fourteen nights at Hakgala the rate of growth for the 

 first four hours of the night, i.e., from about 6.30 p.m. to 

 L0.30P.H., was measured and compared with the rate of 

 growth for the next eight hours, i.e. 10.30 p.m. to 6.30 A.M. 



During these fourteen nights varying weather conditions 

 prevailed, and the curves of air temperature through the night 

 varied from a considerable fall as night went on to a practically 

 uniform temperature through the night. 



If the temperature is the limiting factor in growth we should 

 expect to get a falling curve of growth on the nights when the 

 temperature falls and a steady curve on the nights when the 

 temperature is uniform. The average temperature for the 

 first four hours of the night was obtained by taking the mean 

 of the temperature readings at the beginning and at the end 

 of that period, and so also for the next eight hours. The 

 temperature at 6.30 a.m. is as a rule either the minimum tem- 

 perature for the night or very little removed from it. 



The mean temperature for the 8-hour period was subtracted 

 from the mean temperature for the 4-hour period. The 

 difference is a number which enables us to compare the different 

 nights as regards the fall of temperature. Similarly the rate 

 i if growth per two hours for the 8-hour period was subtracted 

 from the rate of growth for the 4-hour period and the difference 

 gives a measure of the fall of growth rate during the night. 

 'i 'he results are given in Plate 21 (below, on left) and constitute, 

 it seems to me, striking evidence for the truth of our hypoth- 

 th.it the two factors measured in this way should come bo 

 near to proportionality aa the ourve shows by mere coincidence 

 is unthinkable, and the general form of the curve must, it seem- 

 to me. reflect a real connection between the two factors. 



Onlythree of the points on the emve are seriously out of the 

 Straight line of direct proportionality, and the general conclu- 

 sion that when the temperature falls considerably the rats' 



