226 NATURALISATION OF FOREST TREES. 



The relative humidity in winter must also not be neglected. 

 Even if growth is dormant, a low relative humidity with warm 

 days and especially if accompanied by wind, will cause rapid 

 transpiration from evergreens, which is likely to result in injury 

 by drought and which also accentuates the danger from frost. 

 A species accustomed to a dry winter climate is less likely to 

 suffer from these causes than one from a moist winter climate. 



Hence care must be taken with regard to both summer and 

 winter humidity in the selection of species, but from the general 

 point of view of the possibility of forest growth, a large part 

 of South Africa, including the high veld, is in a favourable position, 

 even if we accept Professor Mayr's figure and make no allowance 

 for the influence of the forest itself. 



Evaporation is also a most important factor in connection 

 both with the process of transpiration and with the loss of moisture 

 generally from the soil. It must always be taken into consider- 

 ation together with the rainfall when determining the moisture 

 demands of a species and the available moisture of any locality. 

 In a paper on the zonal distribution of forests in Eastern America, 

 Mr. E. N. Transeau takes as the main basis of distribution, not 

 temperature, but the combined factor of the ratio of rainfall to 

 evaporation and is supported in this by B. E. Fernow. It 

 is interesting that he finds that the borderland between the 

 deciduous Eastern forest and the prairie has a rainfall of about 

 80 per cent, of the evaporation, but here again this cannot be 

 accepted as a world-wide rule. Probably the percentage is 

 much lower, for instance, in the case of the borderland between 

 coniferous forest and grassland in the South-West of America. 



Season of Rainfall. The season in which the rainfall occurs, 

 though of much importance, has perhaps received greater em- 

 phasis than it is entitled to, as compared with other factors, 

 such as temperatures of the growing season, and atmospheric 

 humidity, although it does largely control these. At any rate, 

 it does not seem to have been recognised sufficiently that conifers 

 are much less likely to be affected by the season of rainfall than 

 are broad-leaved species. 



It is difhcult to say whether, with a given annual rainfall, the 

 amount of moisture available for forest growth is affected by the 

 season in which the rain falls. On the one hand, a winter rainfall 

 occurs when growth is dormant and a greater proportion of it is 

 lost by percolation deep into the soil before growth begins ; while, 

 on the other hand, a summer rainfall occurs in the growing season 

 but when evaporation is most active, so that a greater proportion 

 of it is lost by this means. A given rainfall may produce woody 

 growth if it is a winter one, but not if it is a summer one, but the 

 explanation of this is probably to be found rather in the competition 

 from grass in the latter case, for evidently a summer rainfall is 

 much more favourable than a winter one for shallow-rooted plants. 

 By the establishment of a forest and its proper sylvicultural treat- 

 ment, this competition with grass is removed. 



Hence, on the whole, it seems probable that as regards the total 

 moisture requirements of a spieces, it does not make much differ- 

 ence whether a given rainfall occurs in winter or summer. 



