LIMITING FACTORS TO GROWTH MEASUREMENTS. 335 



in tact be equal to the growth during a larger number of hours 

 favourable (as regards water supply) at night when the tem- 

 perature is low. 



It is to be noted that at Anuradhapura, where the high 

 temperature allowed of very rapid growth at night and where 

 the day was as a rule dry and sunny, there is no approach 

 to equality between the growth-rates by night and by 



day. 



Another Theory : Periods of rest necessary. 



Our theory, it will be seen, provides an explanation which 

 covers the facts of the case fairly completely. Another suppo- 

 sition which would account for some of the facts, is that which 

 is sometimes held that a growing organ is not capable of con- 

 tinuous rapid growth, but must have intermittent periods 

 of rest. 



In this connection a paper by W. E. Kellicott (8) is of some 

 importance. He shows that in the root of Allium the periods 

 of slowest growth are the periods of most active cell division. 

 Thus it seems that there are two operations going on in the 

 growing organ: (1) cell-division, which takes place when growth 

 is slow ; (2) increase in size of the cells thus formed causing the 

 more rapid elongation of the organ. If slow growth or cessa- 

 tion of growth is necessary for active cell-division, we get the 

 idea that periods of rest are necessary for subsequent rapid 

 growth. 



On this theory we should expect that growth would be more 

 rapid if favourable conditions occurred immediately after a 

 period of rest than if similarly favourable conditions occurred 

 after a period of activity. The very rapid growth during 

 2-hourly periods of rain during the day, when such periods 

 follow bright sunshine (a time of rest for the growing culm), 

 would fall in with tins idea. It would also account for the 

 rapidity of growth at the beginning of the night at Hakgala. 

 The return of favourable conditions after the resting or 

 almost resting period during the day causes a greater growth 

 than the continuance of those conditions later in the night is 



