236 Jeffrey and Torrey . — Transitional Herbaceous Dicotyledons 
who peruse these lines, it is hoped, that the foliar ray, practically universally 
characteristic of Dicotyledonous herbs, is a compound structure derived 
from the aggregation of storage tissues in the region of the foliar trace, as it 
enters the stem at the node. 
Before passing on to the diagrammatic illustrations of the derivations 
of transitional herbs from woody ancestors, it will be well to pass in review 1 
the figures of the article (1) which it has been necessary to discuss in the 
present pages. 
Their first illustration of the Compositaceous genus Arctotis is an 
excellent example of the peculiarly inconsistent reasoning which charac- 
terizes the anatomical statements of our critics. After objecting strongly 
to the use of the decumbent stem of Potentilla palustris by Eames (1) in his 
article proceeding from these laboratories, they use the ‘ somewhat de- 
cumbent aerial stem of Arctotis grandis 9 in support of their own conception 
of the origin of the herbaceous type. Fig. 2 of their article represents the 
stem of the American Beech and has no bearing on the present discussion. 
Fig- 3> of Hypericum aureum , is not taken in the region where the leaf- 
traces enter the stem at the node, and consequently has no bearing on the 
matter in dispute, namely, the presence of foliar rays in relation to foliar 
traces. We have examined several species of Hypericum and have 
invariably found foliar rays related at the node to leaf-traces. The following 
figure of another species of Hypericum is open to the same criticism as 
Fig. 3. The next figures (5 and 6 ) illustrate the stem of the Solanaceae, || 
but not at the node. Here, as in the Hypericaceae, well-marked foliar 
rays are to be seen where the leaf-traces enter the stem. Fig. 7, of Salvia 
sp., may be taken as illustrative of the Labiatae, which are as well 
characterized by the presence of foliar rays as are other herbaceous orders. 
Figs. 8 and 9, illustrating the Malvales, have been sufficiently criticized in 
the foregoing pages. Fig. 10, of the stem of Rosa rugosa , has little bearing 
on the discussion. The authors do not figure any herbaceous representa- 
tives of the Rosaceae, with the exception of Sanguisorba , in which foliar 
rays are clearly present even in the aerial stem, although not shown in their 
illustration. The futility of their statements regarding the Ranunculaceous 
genus Xanthorhiza has been revealed in earlier paragraphs. The other 
illustration of the Ranunculaceae, Delphinium , is through too slender a stem 
to throw any light on the question of the origin of herbs from woody 
ancestors. Acanthopanax shows some progress towards the herbaceous I 
condition and the formation of foliar rays, although this is not manifest in 
the illustration supplied by our critics. Fig. 15, illustrating the Umbelli- 
ferae, is made through the internode of an extremely herbaceous axis, and 
as a consequence throws no light whatever on the origin of the herbaceous 
type from the woody, which is naturally best elucidated by herbs of 
intermediate and transitional organization. Figs. 16 and 17 are characterized 
