HISTOLOGY OF PHLOEM IN WOODY ANGIOSPERMS 37 1 
shown in Robinia Pseudo- Acacia (fig. 30) and indicated in literature for 
Cytisus and Halimodendron (16). 
In contrast with the famihes that show striking variation of sieve- 
tube type there are also those which show unusual uniformity. Thus 
the Salicaceae, Juglandaceae, and Betulaceae show almost universally 
the presence of sieve tubes of type i, as shown both in the present 
research and in literature. The Urticaceae also, as indicated in 
tables I and 2, show the third type of sieve tube as a constant character. 
If sieve-tube type has any phylogenetic significance, it might well be 
argued that this family belongs higher in the scale of evolution than 
it now stands. 
Sieve-tube type within the genus seems to be constant in the great 
majority of cases. Certain exceptions do occur, however, even in the 
rather small number of plants that have been examined. Perhaps 
the most glowing example of variation within a genus is shown in 
Fraxinus. Here Fraxinus americana has, beyond question, sieve tubes 
with oblique end walls as in Tilia, but Fraxinus nigra, as shown in this 
study, and Fraxinus excelsior, as shown by Russow (16), have sieve 
tubes with transverse end walls. In the genus Prunus, Prunus serotina 
has sieve tubes of the second type but Prunus Persica has those of type 
3. This latter case is perhaps not so much to be wondered at because 
the two species are not closely related within the genus. An indica- 
tion of this condition is shown in literature by Lecomte (13), who lists 
Vitis canescens as belonging to type 3, when it is well known that the 
"vine type" of phloem, as in V. vinifera, V. lahrusca, and V. vulpina, 
has the oblique end wall. So far as has been determined, sieve-tube 
type is constant within a species. In the present study several col- 
lections of a species have been made in many cases, the material being 
chosen from different individuals and from various habitats, and in 
no instance has any essential discrepancy been apparent. 
In view of the evidence brought out in tables i and 2, there seems 
to be no basis for the statement of Hemenway (6) that the study of 
the sieve tube adds further evidence to the theory that herbaceous 
plants are more advanced in evolutionary development than is the 
woody type. This theory regarding the position of herbaceous plants 
is doubtless well grounded; the point emphasized here, however, is 
that the so-called "herbaceous type" of sieve tube is of such common 
occurrence among woody plants that on the basis of our present knowl- 
edge there seems to be little foundation for the statement that the 
