1910] EAMES—THE BROAD RAY IN QUERCUS 165 
of resemblance. A prominent feature of the broad ray of Quercus 
is the strong “dipping in” of the annual ring wherever these radial 
masses of parenchyma occur. This is well shown in jig. 7. In the 
cases among the lower Cupuliferae, and in fossil species of Quercus 
where rays of the compound type exist, the same feature is noticeable. 
In Alnus, as appears in fig. 17, this condition is prominent; in the 
fossil oak it can be seen, but only faintly, for the annual rings are not 
strongly marked, probably as a result of the milder climate of the 
Miocene. This peculiarity serves thus further to connect the two 
kinds of rays. 
In view, then, of the apparent phylogenetic relation of the two 
sorts of medullary rays found in Quercus, the designation of the 
broad ray as primary and of the small one as secondary, a use that 
is to some extent prevalent, seems inadmissible. The terms would 
certainly be more applicable if interchanged, but the adjectives 
small and broad or large, or small and compounded may perhaps be 
more appropriately applied. 
Summary 
The medullary rays of the oak are of two sorts: small, linear sheets 
of parenchyma, only a single row of cells wide, as seen in transverse 
and tangential sections; and large, generally fusiform masses, many 
cellsin width. The former have been to some extent called secondary 
rays, the latter primary rays. 
Fossil evidence points to the probable derivation of the existent 
oaks from ancestors which possessed only the linear type of ray. 
The large ray has apparently originated by the aggregation and 
fusion of many of these small rays, through the loss of the separating 
elements, or the transformation of the latter into ray cells. A fossil 
oak from the Miocene gold gravels of California shows an inter- 
mediate condition, a stage where all the large rays consist merely of 
aggregations of small rays, with separating fibers. The seedlings 
of some, at least, of the black oaks form rays of this nature for 
a brief period, usually during the first ten to fifteen years. There 
* is during this time a recapitulation of the history of the develop- 
ment of the ray; the small, uniseriate rays, which are at first formed, 
gradually approach each other in groups, and fusion slowly occurs 
