bo 
DR. А J. EWART ON HECTORELLA CSPITOSA. 
There is nothing in this description to connect Hectorella any more definitely 
with the Portulacacese than with the Alsinoidee-Polycarpes. The chief 
distinction lies in the fact that in Lyallia the pair of leaves just beneath the 
base of the perianth are considered to be prophylls, and in Hectorella sepals. 
The latter is, however, an error. Of the two supposed “sepals” one can 
usually be seen to arise at a slightly higher level than the other as in Lyallia, 
while the vascular bundles are inserted obliquely and distinctly below the 
origin of those which run to the perianth and stamens. In the figure here 
given the line a а gives the point at which the perianth-segments usually 
separate, b b the level of the thalamus. The “sepals” are inserted below b 0, 
and at or above с с. The vascular bundles of the stamens and petals vary 
as regards their exaet origin in different flowers, but always diverge abruptly 
at the apex of the flattened flower-stalk, forming a distinct group of bifurca- 
tions well above the point of origin of the vascular bundles of the sepals, and 
splitting up at once or after a slight preliminary branching in the fused base 
of the perianth. In the figure of the flower given by Pax (l. e. p. 53, fig. 20), 
the close relation of the prophylls to the flower is somewhat exaggerated, 
and the bluntly-pointed tips of the petals and their slight concavity are not 
shown, so that the flower appears more widely dissimilar from that of Lyallia 
than it really is. 
In regard to the stamens, Pax doubts whether they really alternate with 
the perianth-segments, as stated by Hooker. The latter author is, however, 
correct, as the term is ordinarily understood, although when the vascular 
bundles of stamen and perianth-segment have a short common branch, that 
from the stamen may be seen to twist laterally and join the common branch 
from the inside, as though antipetalous, in cleared and mounted preparations 
showing no signs of lateral distortion. Out of the large number of flowers 
examined all but two had 5 stamens, the exceptions having 6 and 7 respec- 
tively, and the extra stamens being opposite to the “ petals” at their junction 
with the perianth-tube. In regard to the stigmas, Hooker and also Pax 
give the number as from 1 to 3. Out of several dozen flowers examined, all 
but 1 had two stigmas only, the exceptional flower having one of its stigmatic 
