226 BOTANICAL GAZETTE [MARCH 
tracheid occurred between a cambial cell and a fully formed 
(as to size) tracheid, and occasionally here and there in a cutting 
this was lacking entirely. Again, in the cuttings of September 19 
of the following year the same condition of affairs existed. While 
growth as to relative amount had to all purposes ceased, still all 
indications pointed to the fact that in all four cuttings it was going 
on, though very slowly. In both cases growth appeared to be most 
sluggish in cutting III, but no reason can be assigned to account 
for this fact. 
The data from the preceding paragraph lead to the following 
conclusions. Xylem formation in tree I continued during two 
successive years until the last half of September, possibly even as 
late as October 1. Further, it was in evidence throughout a large 
part of the bole, as cutting I was 38 feet above the ground and the 
terminal leader extended only 17 feet beyond. Whether it still 
continued in the terminal leader cannot be concluded from the 
present investigation. If we correlate these deductions with those 
previously reached in the paper, the following points are evident: 
(a) growth intensity falls off first in the upper parts of normal white 
pine trees, more tardily below; (6) cessation of xylem formation 
does not follow the same law, but persists sluggishly in all parts 
of the bole (with the possible exception of the terminal leader) 
until the latter part of September; (c) the exact time of the end of 
xylem formation was not determined in the present investigation, 
but it is safe to conclude that it was practically complete by 
October 1. 
The results of the present study are contrary to those of ROBT. 
Hartic (9), who says, “Der Abschluss der Zuwachsthatigkeiter- 
folgt oben entsprechend friiher, als unten.’’ Too much emphasis 
must not be placed upon this statement, because (1) Hartic made 
external measurements only, and (2) his results may have been 
influenced by subsequent phloem formation after xylem formation 
had ceased. Rost. HARTIG avoids the issue in part when he states 
that cambial activity occurs in the tops of trees in Pinus silvestris, 
Picea excelsa, and Larix decidua during the months of May, June, 
and July, and at the base during June, July, and August. While 
he implies also a cessation of growth, he does not say it in so many 
