542 PLANT GROWTH AND PLANT COMMUNITIES 



along these lines were first carried out at Oxford with H. annuus, where 

 between May and September in two years consecutive experiments at 

 weekly intervals were conducted with plants which at the start of each 

 experiment possessed three pairs of true leaves, the third of which was 

 just visible. At the beginning and end of each experiment the leaf area 

 and the dry weights of the roots, leaves, and stem were measured, and 

 the diurnal variations in light level and air temperature were continu- 

 ously recorded. Multiple regressions linking the net assimilation rate, 

 the leaf-area ratio, and the relative growth rate with the seasonal varia- 

 tions in light and temperature were then fitted to the data. In the statis- 

 tical treatment it was found possible to allow for the ontogenetic drift 

 due to variations in the environment in the interval between sowing and 

 the initial sampling (see Blackman, Black, and Kemp, 1955). Subse- 

 quently Black (1955) undertook a similar investigation, following the 

 weekly growth of TrifoUum subterraneum throughout the year under 

 the conditions of Adelaide, Australia. More recently G. L. Hodgson and 

 M. R. Sampford repeated the Oxford experiment on H. annuus under 

 the cooler conditions of Scotland, and they added a second species, 

 Vicia faba. I am indebted to them for their permission to quote some of 

 their as yet unpublished results. 



The main conclusions reached on the basis of statistical analysis 

 are summarized in Table II. Under all conditions there is a positive 

 efiFect of solar radiation on the net assimilation rate and the relative 

 growth rate and a negative efiFect on the leaf-area ratio. The influence 

 of temperature is clearly dependent both on the environmental condi- 

 tions and on the species. Take first H. annuus: the net assimilation rate 

 is temperature-dependent at Invergowrie but not at Oxford, while for 

 the leaf-area ratio the position is reversed. The sensitivity of the net 

 assimilation rate to temperature is shared by V. faba, but, in contrast to 

 H. annuus, its leaf-area ratio is positively linked with temperature. The 

 main climatic diflFerences between the two centers were that at Inver- 

 gowrie the days were longer and the temperatures lower, and it can be 

 suggested that at Oxford the temperature range was above a threshold 

 value which restricted assimilation at Invergowrie. 



Although only the influence of mean daily temperature is cited in 

 Table II, the statistical analysis included an examination of possible 

 efiFects of the diurnal fluctuations in temperature. In only two instances 

 did these fluctuations prove to be significant. The net assimilation rate 

 of T. subterraneum was positively linked with the mean daily range, 

 while the leaf-area ratio of V. faba was correlated with both the mean 

 and minimum temperatures. 



One of the most striking features of these studies was the high 

 proportion of total variance that could be accounted for in the individ- 

 ual multiple regressions. Thus there were good grounds for concluding 



