AND ANATOMY OF THE GENUS RESTIO. 217 
for ensuring fertilization, but upon which points no evidence yet 
exists. 
Inflorescence.—This consists in both the male and female plants 
of spikes or spikelets of various forms and sizes, according to the 
Species or to the age of the individual plant. The spikelets are 
sometimes placed singly at the ends of the culms or branches, 
more frequently they are disposed in cymes, the cymes being more 
or less dense, sometimes spike-like, or more frequently paniculate. 
In some cases the inflorescence of the male and female plants is 
alike, while in others it differs; in this latter case the male plant 
generally has numerous spikelets arranged in panicled cymes, 
while the female spikelets are less numerous and spicate in 
their arrangement. Occasionally monccious plants are met with 
wherein both male and female spikelets occur in the same inflo- 
rescence. Kunth mentions this in R. intermedius, and I have met 
with it in R. ferruginosus. The individual spikelets are erect or 
spreading, very rarely pendent ; some are sessile (terminal), others 
pedicellate, the pedicel being compressed from side to side, 
grooved lengthwise to accommodate the spikelet, and increasing 
in breadth from below upwards. In R. triticeus, as observed by 
Rottboell, the back of the spikelets is parallel to the rachis—in 
other words, the spikelets are attached to the rachis in a trans- 
verse manner as in Tritieum. Inthe great majority of the species 
the spikelets have their edges (not their backs) turned towards 
the rachis, as happens also in Lolium among Grasses. From the 
cylindrical or somewhat globose form of the spikelets in some 
instances, it becomes difficult to ascertain the real position of the 
spikelets with reference to the axis. If the spikelets be crowded, 
they are somewhat curved so as to get out of one another’s way ; 
but if the spikelets are placed at some distance one from another, 
no such curvature is necessary. 
In R. graminifolius a peculiar circular cushion-like pad of 
cellular tissue is to be seen at the angles formed by the diverging 
pedicels. The purport of this little cushion is unknown to me; 
its existence has not, so far as I am aware, been previously noted, 
nor have I yet seen anything similar in any other species. 
Each spikelet consists of a number of bracts or scales, in the 
axils of some or all of which are placed the flowers. In the male 
plants all the bracts, with the exception perhaps of the outermost 
or lowest and the one next above it, are fertile, while in the female 
spikelets it not unfrequently happens that one or two only of the 
upper scales have flowers in their axils. The scales are arranged 
