BRANCH-GROWTH OF DOUGLAS FIR. 1 47 



years. In the wood of more rapid growth the branches die 

 oflf in five years as opposed to seven or eight years in the 

 slower-growing wood. Nevertheless in these five years the 

 branches of the former wood reach a much larger size than 

 those in the latter wood reach in seven. 



To state generally, therefore, that branch development varies 

 with the planting distance is somewhat misleading. What is 

 probably more correct is that within a given quality-class, 

 i.e. where the rate of growth is similar, the closer the planting 

 distance, the shorter the life of the branches, and consequently 

 the smaller their development. 



Such a distinction may seem trivial, but in any experimental 

 investigation into branch development, by means of a series 

 of plots, such as Brig. -General Stirling of Keir suggested in 

 the first part of this volume, it is very essential that the area 

 selected should be of uniform quality throughout, or that the 

 various plots should be carefully duplicated over the area, so 

 as to average out any differences which are not obvious on 

 examination. Also, each quality-class must have its series 

 of plots. 



Experience shows that ground in Scotland is remarkably 

 variable, and it is extremely difficult to obtain even one acre 

 of quite uniform quality anywhere. A carefully carried out 

 experiment would probably show that the rate of growth 

 remains constant throughout the whole area for all distances, 

 but that there will be differences in size, due to the different 

 lengths of life of the branches in the various plots. The length 

 of life is itself dependent upon the time when the canopy closes, 

 which is in turn influenced by the rate of growth of the 

 branches. There are thus complications arising, and the need 

 for definite information is all the greater. 



