198 BOTANICAL GAZETTE [MARCH 
rules of growth for the species. Inasmuch as a description of all 
the trees used would be confusing, short silvicultural notes on each 
specimen will be given in the text where it seems necessary. 
Previous investigations of growth in forest trees 
Before entering upon a description of the methods employed in 
this work, a brief résumé of those of other investigators will perhaps 
add interest to the present study. 
Von Mout (25) sought to determine the growth of trees by 
making measurements of the circumference at a definite place on 
the bole. From these the radius was computed and the increase 
in thickness noted. CHRISTISON (5) pursued the same method and 
computed data for a large number of species, including both hard 
and soft woods. The data of Jost (13) were based in part on the 
methods given above and in part on measurements which he 
obtained by the use of a “Fuhlhebel.”’? Any data securedsthrough 
bark measurement are unreliable because of continual changes 
going on in the older parts of the secondary cortex and changes 
which bear no relation to the newly forming rings. As a result, 
only broad generalizations can be drawn from data based on such 
methods. 
T. Hartic (12) sought to determine the growth of tree species 
in a different manner. Choosing even-aged, pure stands, where 
growth conditions appeared to be similar, he felled typical speci- 
mens from these at different periods and made comparative studies. 
He assumed that in such stands all individuals exhibited similar 
characteristics of growth, a view that is untenable in the light of 
our present knowledge. Hartic’s method is open to criticism 
in that it was extremely inaccurate and could therefore never give 
reliable results. Growth varies markedly not only in different 
individuals in a stand, but also in different parts of the ring at 4 
given height. 
MIscHKE (24) took the first steps in the direction of securing 
accurate results. Using an increment borer, he studied the annual 
ring at different periods in its development and obtained in this 
way the first results which were in any way accurate. WIELER (39) 
2 For description see reference. 
