LIMITING FACTORS TO GROWTH MEASUREMENTS. 321 



1 4th to May 5th. During the greater part of this period its 

 growth was measured twicea day, in the morning about 7 a.m. 

 and in the evening about 6 p.m. A series of two-hourly 

 measurements was made through the night, April 9th to 10th, 

 which are not represented on the plate, and on several occa- 

 sions other measurements were made at special times when 

 the weather conditions were in any way unusual. When these 

 observations resulted in striking evidence as to the growth 

 under different conditions they will be quoted in discussing 

 the results. 



The culm of which the growth is represented in the upper 

 part of Plate 23 measured on July 10th to 12th was the subject 

 of observations from July 9th at 5 p.m. to July 12th at 6 a.m., 

 and of no others. 



Results. 



The following facts seem to me to stand out clearly on a 

 comparative examination of the curves of growth given fer 

 the three places. 



1. Growth and Rainfall. 



The rate of growth is in no way directly dependent on the 

 rainfall. This is most clearly brought out by the measure- 

 ments made at Anuradhapura in the dry country. The culms 

 begin their growth here as a rule in June in the dry season, and 

 throughout their growth, which is completed or nearly so by 

 the end of September, there is as a rule little rain. 



A clump with young culms in different stages of growth was 

 under observation from June loth to 25th, and during this 

 time the total rainfall was "03 inch falling on two days, *02 

 being the highest amount per day. The average rainfall for 

 the month of June is 1 • 4 inch and for July 1 ■ inch as has 

 before been stated. 



Notwithstanding this the highest rate of growth per 2 hours 

 ever recorded in the whole of my observations was observed 

 here , and the average rate of growth during the night was 

 higher than at either of the other places. 



9(8)06 (29) 



