638 
Harper. — Defoliation : its Effects tip on the 
so constant a difference as was found between the thick-walled autumn 
wood of the Japanese Larches and the abnormal thin-walled zone outside 
it is not easily explained merely by the youth of the trees. Whatever 
inferences can be drawn from these cases are all in favour of the view that 
diminished nutrition may be a cause of the occurrence of unthickened cells 
at the boundary of the autumn wood. 
For the view that in the autumn wood of the defoliated larches such an 
abnormal formation of thin-walled cells outside the thickened ones is really 
a starved attempt at autumn wood, there is yet a further argument in the 
steadily decreasing size of the. cells from the beginning of the thickened 
zone right outwards to the limit of the ring. There are no published 
figures of the ‘ double-ring ’ described by Strasburger in a larch to which 
reference has already been made, but Fig. 14, from a section taken at 
random from a species of Abies, shows that the thinner-walled cells between 
the first and second thick-walled zones are actually larger than those 
on either side of them. S. J. Record’s illustration of Juniperus virginiana 
shows more clearly a similar state of things, which would appear to be the 
normal result of a renewal of cambial activity towards the end of the grow- 
ing season. But in the abnormal rings of the defoliated larches, as Figs. 4 
and 13 show, the cells become ever smaller up to the end of the year’s 
growth, even though they are without any characteristic autumn thickening. 
In the two Japanese Larches from Oxford the ring of 1911 was bounded by 
cells that appeared as if checked in development before they had properly 
outgrown the flattened prismatic form and thin-walled condition of the 
cambial cells from which they were derived. But almost to the last the cells 
of the abnormal rings in Figs. 4 and 13 are rounded though unthickened, 
showing that they had been growing since they were cut off from the cambium. 
Their relatively small size corresponds to that of the outermost cells of 
normally thickened autumn wood (which are not always markedly flattened), 
and it may be supposed that they are not thick-walled for the simple 
reason that there was not sufficient food-material at hand, although the 
cambial cells still retained their normal capacity for repeated division. The 
mere formation of new cells and the thickening of their membranes must 
be regarded as two distinct phenomena, governed independently of each 
other by stimuli at present insufficiently understood. 1 A case like that of 
Fig. 3 seems to represent a genuine attempt at forming autumn wood, 
in which, however, the extra deposits of thickening upon the wall had to be 
omitted. Only in the extreme case shown in Fig. 9, between A and B, where 
the cells at the boundary are no smaller than those of the succeeding 
1 The thickening and lignification of the elements often proceeds very slowly in young normal 
larches, so that in August there may be found a zone of thin-walled cells between the cambium and 
the completed tracheides, which last may still give the cellulose reaction in the thick secondary layers 
of their walls. 
