THE GIRTH-INCREASE OF TREES. 43 
To compensate for this interruption in the deciduous group, 
I selected in 1892 a fresh set of twenty young trees, each of a 
different species, almost all growing in the North and West 
borders of the Arboretum, subject to the same objection of want 
of preparation of the ground before planting as the former set, 
but favoured by a rather better soil. Once more, however, the 
continuity of observation was interrupted, in 1896, by a close 
pruning of branches and roots, in preparation for transplanta- 
tion, which at once reduced the aggregate girth-increase by 
nearly one-half. As to the Evergreens, discouraged as I was 
by the comparative failure of the first set, I made no effort to 
increase their number, although the observations on the original 
set were continued. 
The introduction of Chesterman’s steel tapes insured an 
accuracy of measurement unattainable in the original experi- 
ments, and enabled me to initiate new inquiries, such as the 
determination, within narrow limits, of the seasonal beginning of 
girth-increase in the different species, the weekly rate of growth, 
the relation of girth-increase to the development of the leaves 
and twigs, &c.; and the various results were communicated to the 
Royal Society of Edinburgh in 1883, and to the Botanical Society 
in 1887, ’88, 89, and ’92. In the present Report I shall confine 
myself to the annual and monthly results,—and in the deciduous 
_ trees only,—bringing them down to 1897, which completes a 
period of twenty years. The observations for 1892-97, both 
annually and monthly, and the annual observations of the original 
set for the second decade, which have not yet been published, 
will be given in detail, but only the general results for the first 
decade are reproduced from my former papers. In these papers 
were incorporated observations made on trees at Craigiehall, 
near Cramond, but these have long been discontinued, and the 
results will only be incidentally used here. Thus, the present 
Paper becomes peculiarly a record of the life-history, as indicated 
by girth-increase, of a considerable number of deciduous trees 
in the Botanic Garden and Arboretum, over periods of from six 
to twenty years. - 
Doubts have been ee as to the possibility of measuring 
the girth of trees to the twentieth of an inch with accuracy, and 
unquestionably in stems of great size, and in all stems of irregular 
