634 
Harper . — Defoliation : its Effects upon the 
of the thin-walled spring wood, suppose that its formation is regulated by 
the requirements of the transpiring crown ; when these needs are adequately 
provided for only thick-walled autumn wood cells are formed until growth 
ceases for the year. Pfeffer 1 cautiously admits that nothing definite is 
known of the causes of the seasonal change in structure of the woody ring. 
In the particular case of the uppermost of these two sections it is 
especially noteworthy that the ring fails altogether on the south side instead 
of being intermediate in condition between the east and west, as it should 
have been if differences of temperature or wind-pressure had been the 
cause of the inequality shown by Figs, n and 12. It is hard to suppose 
that the transpiration of the symmetrically developed uppermost few feet 
of the crown above the section should have been so negligible on the east 
side if it were responsible for all the spring wood on the west ; but there 
might have been a difference in the nutrition of the opposite sides of the 
section at that point. Now the distribution of growth in a thriving and 
well-nourished tree lends weight to the view taken by Sachs 2 and Pfeffer 3 
that growth in the different parts of a plant is not governed in the first 
place by the quantity of food available in each, but rather that food- 
materials are translocated especially to those parts where cell-division is 
proceeding most energetically. But after several defoliations have reduced 
the reserves the quantity of food at hand in any part might conceivably 
become a limiting factor to the growth there, and, indeed, in the particular 
tree in which this remarkable ring occurred growth had quite failed at the 
base a year or two before, which suggests that the upper parts even might 
have become by this time very sensitive to slight local variations in 
food supply. Numerous ‘ringing’ experiments seem to show that the 
elaborated food-materials tend to pass vertically down the trunk without 
diffusing around in all directions, so that in a very impoverished tree 
growth of the main stem might be taking place only just below the point 
of junction of each branch . 4 Supposing in such a case that just above the 
section there were a branch on the east side, but that on the west the 
nearest considerable branch were a little higher up, it is easy to see that 
the east side of the section would get the more food, while on the west the 
descending nutriment, having farther to go, might not last out so far down 
as the point where the section was taken. A similar result would follow if 
the west side of the crown were more severely defoliated than the east. 
It is unlikely that the formation of autumn wood in general can be the 
result of a single cause, but in this particular case differences in the amount 
of food-material available on either side of the section seem to afford the 
explanation most fitting to the facts described. This matter cannot be 
settled until the whole of the upper part of a starving tree has been 
2 Sachs (2), p. 439. s Pfeffer (2), pp. 584, 592, 600, 601. 
4 Cp. Nordlinger, p. 3; Hartig (2), p. 23. 
1 Pfeffer (1), p. 215. 
