Jeffrey and Torrey . — Transitional Herbaceous Dicotyledons. 237 
by the same faults as many of the others, namely, they are not taken 
properly in the nodal region or the stems are too slender to reveal the 
condition of transition of fundamental importance in the present connexion. 
These shortcomings are all the more serious because the two illustrations 
in question are all that the authors have presented for the very important 
herbaceous group, the Compositae. Fig. 18, a further example of the 
Ranunculaceae, is not from a nodal region of Clematis , and as a consequence 
naturally throws no light whatever on the typical foliar ray, which in woody 
herbs like the species of Clematis illustrated is of shallow vertical length 
and confined as a consequence to the region immediately below the node 
and the trace to which it is related. 
The elucidation of the results reached by the examination of trans- 
verse and longitudinal sections of transitional herbs is naturally best 
accomplished by means of diagrams. Text-fig. 1 shows at once the tan- 
gential and transverse aspects of three different genera of the Compositae. 
In the centre appears a representation of the stem of the southern woody 
genus Baccharis , in the region of the node. The leaf-trace is diagram - 
inatically indicated by elements with spiral markings. It will be noted in 
the tangential aspect which faces the observer that there is only a slight 
concentration and enlargement of the rays of the wood in proximity to the 
foliar trace. On the opposite side of the transverse aspect of the central 
region of the diagram is shown the transverse section of a leaf-segment of 
a higher node. Obviously we have here to do with an aggregate ray 
related to the leaf-trace, since no vessels are present in the region of the 
cylinder radially subtending the vascular supply to the leaf. To the right 
in the diagram appears the nodal tangential aspect of Bidens sp. Here 
there is a considerable fusion and enlargement of the normal rays of the 
wood in relation to the foliar trace, and as a consequence the storage 
conditions are farther advanced than they are in the central segment of the 
diagram. No transverse section of the foliar ray is represented in the case 
of Bidens. To the left in the diagram is shown the tangential aspect of the 
nodal situation in the genus Helianthus. The storage devices here have 
become better developed than in Bidens , since a well-marked compound 
and as a consequence purely parenchymatous ray has made its appearance 
in relation to the foliar trace. A notable feature in this instance is the 
presence of a central tongue of unmodified wood in the lower region of the 
foliar ray. This structure has the effect of subdividing the leaf-ray into two 
in the lower part of its course. The two narrow forks of the foliar ray in 
many herbs run for a long distance down the stem, and as a consequence 
through a number of internodes. The vertical length of the foliar ray 
other things being equal, has a direct relation to the degree of herbaceous- 
ness of the axis. It may vary in the same stem, for in the lower aerial 
region, where the cylinder is thick and woody, the vertical extension of the 
