128 BOTANICAL GAZETTE [FEBRUARY 
tance below the inflorescence (51), which terminates the culm of the 
main stem; they are exactly alternate, and so are the glumes of the 
median spikelet (S‘); in the two lateral the same position is notice- 
able, but these have been turned so as to form an angle of 90° with 
the median, the terminal spikelet. In the axil of L' a complex of 
shoots may be seen, beginning with a prophyllon (P*) with its back 
toward the main axis, but alternating with the leaf Lt. The subse- 
quent leaves (L3-L°) of this axillary shoot, on the other hand, are 
turned go° to the side of the leaf Z', but are otherwise alternate. 
The first of these leaves (L3) supports a bud with a prophyllon and 
two small leaves, and similar buds are also observable in the axils 
of the other leaves (L4—L°). The apex of the whole shoot terminates 
in a single spikelet (S?), the arrangement of whose glumes, alternat- 
ing with the green leaves of the same axis, shows that it is a median 
and not one of the lateral spikelets; these have become suppressed 
entirely in this instance. 
There is still another shoot observable, however, namely the one 
whose prophyllon is marked P?, and which has only two green leaves. 
This little shoot comes from the axil of the large prophyllon P’, 
Munroa thus illustrating a case, not unusual among Gramineae, in 
which prophylla may subtend shoots. It is furthermore noticeable 
that this prophyllon (P?) and the leaves of its shoot lie distinctly in 
the same plane as the first leaves (L' and L?) and the first prophyllon 
(P). In this way the spikelet S* terminates the main culm, while the 
other (S?) terminates the shoot developed from the axil of the leaf L.* 
The succession of these various leaves (L and P) seems invariably 
the same, inasmuch as no instance was observed where the inflores- 
cence was directly preceded by a prophyllon alone. But the number 
of vegetative shoots in Munroa is sometimes much larger than that 
of the floral, and the reason seems to be, at least judging from this 
particular instance (jig. 9), that the shoots were so crowded within 
the leaf axil that the development of flowers became arrested. This 
explanation may seem justifiable when we examine the other diagra™ 
(fig. 10), which shows a more balanced growth of the whole syste™ 
of axes, instead of the one-sided development of shoots shown in the 
former diagram (fig. 9), where no axillary branch was developed from 
the leaf L?, but only from L’. 
