9 G TKAXSACTIOXS AND PPvOCEEDINGS OF THE [Sess. lxii. 



doincr this is to o-et the direct influence of these conditions 

 while the trees are actually growing. 



The mean temperature and rainfall for the year do not 

 represent the means for the summer months ; at the same 

 time the means for the winter and spring mouths, and also 

 for the year, have been before me in studying the curves 

 of increment. 



The relative humidity of the air during the six growing 

 months has been represented by another curve. 



I am indebted to Dr. Alexander Buchan, Secretary, Scot. 

 Met. Society, for kindly placing at my disposal his MS. 

 tables of sunshine for the whole of Scotland, and from 

 these tables curves have been drawn ; these curves, how- 

 ever, are only approximately true for Braemar, but they 

 are a great help to the present investigation, as light is an 

 important factor in the growth of plants. 



Since the meteorological observations at Braemar began, 

 in 1856, I had to make that year the starting-point for 

 my curves; the period under consideration is, therefore, 38 

 years. 



The following are the specimens now under considera- 

 tion : — Specimen A, common larch, 66 years old at 

 section near the ground ; on N. slope of hill ; S.W. 

 exposure. Specimen B, Xorway spruce, 68 years old at 

 section near the ground ; on N. slope of hill ; N. exposure. 

 These two trees grew about 2 miles west of Braemar, and 

 about 100 feet above the river Dee. Specimen E, Scots 

 pine, 100 years old; on N. slope of hill, one mile E. of 

 Braemar, and 2 feet above the Dee ; S.W. exposure. 

 Specimen F, common larch, 74 years old, but section 

 at 32 feet above ground, so that tree must have been 

 considerably over 100 years old ; 1 mile N. of Braemar, 

 on S.E. slope of hill N. of the Dee ; exposure uncertain. 

 Specimen G, common larch, 64 years old at section near 

 the ground ; on level alluvial soil, with gravel subsoil, 

 ^ mile N. of Braemar ; exposure N.E. 



From these notes and the first part of the paper it will 

 be seen that conditions and aspects of the trees varied 

 considerably, and these may account for some of the minor 

 differences found in the curves. 



A more accurate method of investigation would be to 



