1 8 edmund w. sinnott and irving w. bailey 
Phylogenetic Conclusions 
If the hypothesis which has here been presented is a sound one 
and the trilacunar palmate leaf type is indeed the most ancient 
among Angiosperms, the theory which derives these dominant seed 
plants from Cycad-like Gymnosperms meets with grave difficulty. 
The Mesozoic Bennettitales, from which many phylogenists believe 
the Angiosperms to have descended, were unilacunar and possessed, 
like all Cycads, pinnately compound leaves. The difficulty of deriv- 
ing from such a leaf the foliar types which prevail among living Angio- 
sperms has been noted by many writers. Wieland^^ suggests that the 
gap may have been bridged by a fusion of the leaflets of the com- 
pound leaf, with the eventual production of a simple pinnate one. 
But if a palmate leaf was indeed the primitive Angiosperm con- 
dition it is difficult to see how it could have arisen from 
any cycadaceous type. The Conifers, on the other hand, 
are invariably palmate in their venation. Although the common 
leaf type is a narrow one with but one or two strands, there are num- 
erous cases in the Araucarineae and Podocarpineae where the leaf 
becomes broader. In accommodation to this change the vascular tissue 
increases in amount, but always by a basal multiplication of the 
strands in palmate fashion rather than by the origin of branches 
from a midrib (figs. 74 and 75). It is possible to imagine how such a 
broad-leaved form, developing a trilacunar type of insertion (Agathis 
is occasionally bilacunar) might easily give rise to the palmate leaf 
which we have regarded as ancient. Evidence from the leaf certainly 
favors a coniferous rather than a cycadean stock as ancestral for the 
Angiosperms. 
Information from this source is also important in connection 
with the ancestry of the Monocotyledons. These plants are almost 
invariably palmate in their venation, and in many cases are three- 
veined. In the broad-leaved members of the Potamogetonaceae, 
Alismaceae, Araceae, Liliaceae, Dioscoreaceae and others we often 
meet with leaves which are essentially like those of the Piperaceae, 
Melastomaceae and other palmate Dicotyledons (figs. 76 and 77). 
The nodal conditions in such forms often suggest the dicotyledonous 
type, three traces departing to each leaf (fig. 78), and the petiole is 
10 Was the Pterophyllum Foliage Transformed into the Leafy Blades of 
Dicotyls? Wieland, G. R., Am. Jour. Science 38: 451-460. 1914. 
