542 
ALBAN STEWART 
which occur near the bases of such. There is often a parting of the 
fibers to right and left tangentially in such regions, a condition which 
is still apparent for some distance above and below a chamber. Maule 
(15, pp. 19-20, pi. I, fig. 8) describes and figures a somewhat similar 
arrangement of fibers in the wood formed in a direct line with longi- 
tudinal wounds on stems of Cornus sibirica. The structure of the 
masses of lignified parenchyma, which occur around the bases of the 
larval chambers (fig. 3) are very complex and no regular arrangement 
of the cells can be distinguished. The cells in such places are mostly 
short and are not dissimilar to the short-celled elements which are 
formed immediately after wounding. Fibers are sometimes mixed 
with the parenchyma cells in which case they pursue a very irregular 
course. Maule (15, p. 7) mentions conditions somewhat similar to 
this near wounds on stems of Caragana, Abies, Aesculus, Salix, and 
Tilia. 
The course of the cells in the wood appears to be nearly normal 
in most places in cross sections of galls as the greater part of them are 
not so greatly bent from their upright course as to be readily apparent 
in such sections. There are places, however, in which the course of 
the cells is very much altered, and small segments of the wood may 
be turned over nearly at right angles so that the long axes of the cells 
are presented. Such an arrangement is shown in the lower right hand 
side of fig. 14. In places of this kind the shortness of the cells which 
are turned over is very apparent. The change in direction of the cells 
from vertical to horizontal or nearly so, is usually sudden and there 
are seldom pronounced transitional stages between. Cross sections 
also show a great reduction in the number of vessels, and where the 
disturbance has been greatest, an entire absence of them. They 
occur more commonly in the gall wood first formed, from which 
narrow vessel-containing segments may extend outward into wood 
from which they are otherwise absent (fig. 12). Sometimes the narrow 
vessel-containing segments extend nearly to the cambium. In trau- 
matic wood of these species of oak, formed soon after wounding, there 
is a great reduction in the number or an entire lack of vessels. 
In cross sections of many of the older galls a layer of wood has been 
formed just inside the cambium which is normal in many respects 
(figs. 5 and 14). This layer sometimes extends through nearly the 
entire circumference of the gall, but more often only segments of it 
occur. The cells in this layer usually have nearly their normal 
