Winifred K. I5renchley 227 



In interpreting these formulae it should be remembered that as C, D 

 and E are closely associated the alteration of any one of them without 

 the other is to some extent an unnatural process. Thus the fact that 

 in both formulae the coefficient of D is negative, in the first to an 

 insignificant extent but in the second significantly, shows that the 

 greater growth has in general been made with the lower night tempera- 

 tures after allowance has been made for the cooler days and less sunshine 

 which are in fact associated with cooler nights. In the same way sun- 

 shine is detrimental to the seedlings after allowance has been made for 

 its beneficial effect upon the day temperatures. Once the average effect 

 of the environmental conditions is ascertained it is possible to obtain 

 a truer representation of the relative rate of growth at different ages. 

 For this purpose the average of the relative rates of increase for any 

 age is corrected by means of the regression formula to its probable value 

 under standard environmental conditions. The result of so doing is 

 shown by comparing Figs. 7 and 8. The falling off of growth with age is 

 more gradual after allowance has been made for changing environment; 

 the points lie closer to the regression line for the older series, but the 

 four first series are distinctly less linear. The whole series so corrected 

 is likely to give a much truer picture of the normal history of the plant 

 than when no correction is made. On the other hand, it is clear that 

 irregularities are still present though they are smaller, and the three 

 measures which have been used evidently do not give a complete picture 

 of the plant's environmental experience during the week to which they 

 refer. Among the other factors of environment that cause these irregu- 

 larities are probably the distribution of the hours of sunshine over the 

 day and week, the variation in the actual intensity of light apart from 

 bright sunshine and the humidity of the surrounding air. 



The abrupt rise in the relative rate of increase for the mean 9th 

 period may possibly be mere coincidence but is more likely to have a 

 physiological significance. Examination of the actual curves of effi- 

 ciency indices show that somewhere in the neighbourhood of this period 

 a real or secondary maximum efficiency index occurs, and in several 

 cases the data show that just at this time the plants were in their first 

 flush of flowering. This rise may therefore be real and connected with the 

 initiation of sexual reproduction. If this be so it affords an interesting 

 parallel to the increase in growth occurring at the time of puberty in 

 human beings^. 



1 Hall, G. S. (1908). Adolescence, Vol. i, ^jp. 59, 84, 93. 99. 

 Ann. Biol, vi 16 



