230 BOTANICAL GAZETTE [MARCH 
of this sort, because the transition from early to late wood is always 
very gradual, and it is difficult to distinguish the first formation 
of late wood tracheids. Larch should prove much more satis- 
factory for this study. But in spite of the difficulties mentioned 
above, it was obvious that late wood formation was in evidence in 
tree I on August 5, 1912, and on August 13 of the following year. 
On each of these dates all four cuttings showed some traces of it, 
and, further, it appeared to be slightly more advanced in cuttings I 
and II than in the ones taken lower on the bole. This was to be 
expected from what has been previously said; late wood forma- 
tion begins first in the upper portions of the bole. 
Primary growth in aerial parts 
Some attempt was made in the investigations to secure reliable 
data concerning the elongation of aerial parts. A sample plot of 
0.05 acre was laid off on May 3, 1913, in the vigorous young white 
pine stand mentioned previously. It included 115 trees which 
ranged from 4 to 11 years. The soil was of medium thickness, 
underlaid by sandstone and shale; exposure was open. All the 
trees were seemingly vigorous. 
The trees were first examined on May 3 as to elongation of 
aerial parts. At that date elongation had already begun in all the 
trees on the plot, which varied from 0.5 to 2.5 inches. Greatest 
‘elongation had occurred in the terminal leader, while it was less 
pronounced in the slower growing lateral branches. The leaf 
fascicles were in evidence, but had as yet attained no appreciable 
length. . 
Observations corresponding to the above were subsequently 
made on May 30, June 17, and July 4 of the same season. Accurate 
measurements of the terminal leader and of the preceding year’s 
growth for 50 trees were made in each case and the results collected 
in table L. The average growth of the preceding season is con- 
sidered as the mean of the average preceding year’s growth as found 
on May 30, June 17, and July 4. It is to be observed that the last 
vary slightly, as no attempt was made to select the same 50 trees 
on each date. The measurements are given in inches and fractions — 
of inches. | 
