1910] EAMES—THE BROAD RAY IN QUERCUS 163 
huge, homogeneous ray. Fig. 3 shows this process in a seedling 
of Q. velutina, the partial formation of three rays being evident. The 
ultimate shape and structure is brought about by the loss of the 
separating rows of fibers, tracheids, or wood parenchyma, or rather 
by the transformation of the elements included by the rays-as they 
approach one another. This change is gradual; more and more 
parenchyma is formed, with corresponding decrease in fibers, until the 
cambial cells that lay down these radial rows cut off only ray 
cells. Ray formation of this sort exists in these species only as a 
passing phase in the seedling. In the wood of the mature plants 
large rays, of course, are continually formed, but their origin is 
nearly always abrupt. The transition from lignified elements to 
parenchyma occurs generally within the width of a single annual ring, 
or sometimes two or three. Later there is usually an increase in width 
which is due, however, not to the uniting of any neighboring small 
Tays and the transformation of adjacent elements, but to simple 
growth in size with the increase of the stem in diameter. The seedlings 
of the black oaks, then, give evidence of the mode of origin of the 
broad rays so characteristic of the mature wood of the group. In 
So doing they repeat the ray structure that probably occurred through- 
out the wood of their rather recent ancestors. 
Among the white oaks, seedlings of Q. alba L., Q. bicolor Willd., 
and Q. prinoides Willd. were examined. In all three species the 
wood of the stem for a distance of several inches, or even one or 
two feet, above the root possesses only uniseriate rays. Moreover, 
until the plant has attained a considerable size no large rays nor 
any signs of grouping of the small rays appear. Cases were seen in 
Q. alba and Q. bicolor, where in seedlings 15 to 20 years old only 
the linear rays existed. This condition would suggest that these 
white oaks, perhaps, are somewhat more primitive than the black 
oaks, for the seedlings of the former revert in structure to an ancestral 
type in which only uniseriate rays existed, and had not as yet begun 
to aggregate. Fig. 7 shows a transverse view of this condition in 
Q. alba in about the fifteenth to twentieth annual rings) When the 
large rays do appear in these species, they are formed as in the black 
oaks, rather abruptly, the phase of compounding being confined to 
only one or two annual rings. Often nearly the whole ray arises at 
