1093 
Relationships of the Gnetales. 
type. Bailey 1 emphasized the transitions in the compounding process 
furnished by the false rays of the Cupuliferae which, as we have seen, like- 
wise have their counterpart in Ephedra. He further showed that the com- 
pounding took place primarily around the leaf-trace. The development of 
broad rays is obviously an adaptation for the storage of assimilates manu- 
factured in the leaf, and the most convenient place for their storage is in the 
stem adjacent to the leaf-trace. From their original position around the 
trace he considers that the broad rays have spread throughout the wood. 
Not only in the possession of broad rays, therefore, but also in their 
origin does Ephedra resemble the lower Dicotyledones. It should be noted, 
however, that the compounding process bears no obvious relation to the 
leaf-traces, but may give rise to rays which do not touch a trace at any 
point. In fact, it seems to be quite independent of the leaf-trace, and is 
likely to result in the formation of a broad ray on any radius of the stem. 
It is true that the trace is accompanied by a broad ray, as may be seen in 
PI. XCIV, Fig. 7, but this ray is neither so broad nor so high as the usual type 
of compound ray. Moreover, it rarely shows the fusion process. In fact, it 
represents rather the leaf-gap of other woods. The reason for its poor de- 
velopment is probably to be found in the non-functional character of the leaves 
In practically all cases the leaves are merely small scales which have little 
or no chlorophyll, and have given over their function of assimilation to the 
cortex of the young plant which is abundantly supplied with chlorophyll 
and stomata. Not the restricted area around the leaf-trace, therefore, 
but the whole circumference of the branch would feel the need for storage. 
Of course it is quite possible that the original position of ray development 
was in the vicinity of the leaf-traces, and that with the reduction of the 
leaves and the functioning of the cortex in assimilation the rays may have 
lost their original connexions. Indeed, the occasional occurrence of com- 
pounding around the trace and the constant presence of a ray in that 
position would justify such an inference. In any case the resemblance 
both in regard to the history of development and adult form between the 
rays of Ephedra and those of the lower Dicotyledones is remarkably 
close. 
Just as in many oaks the compounding process is visible only as 
a passing phase in the seedling, the broad rays of other regions extending 
right to the pith, so in certain regions of Ephedra the broad ray has 
worked back almost or quite to the pith. This is particularly true of those 
regions in which storage is greatest, namely, in roots and underground 
stems. In these places the broad ray develops at a very short distance 
from the pith, and may indeed be quite continuous with the pith. PI. XCVI, 
Fig- 33, which represents a transverse section of a subterranean stem of 
1 Bailey, I. W. : Relation of the Leaf-trace to the Origin of Compound Rays in the Lower 
Dicotyledons. Ann. Bot., Jan., 1911. 
