56 THE LONG-JOINTED BAMBOO FOR BLOW-PIPES. 
decreasing. ‘These, like all sterile spiculz, bear in the axils of most of 
their glumes more or less reduced branchlets in the shape of small buds, 
and there is no prolonged internode in their rachilla. As all ramifications 
within the inflorescence begin with a bicarinate bract (‘‘ adossirtes vor- 
blatt,” Eichler), and those bracts are all of the same shape and, as also 
the bracts of the other series to which the gemmiparous and the fertile 
glumes belong, are like each other, and only differ in size, it is often 
nearly impossible to distinguish the single spicule. 
‘So far the composition of the spicula is almost the same as in 
B. Griffithii, and it approaches very much that of Nastus. The only 
deviation from the structure in the latter genus is the presence of 
gemme. The anthers, styles, and ovary are as in B&B. Griffith. But 
whilst the lodicule in B. Grifithit are, according to Munro, unequal, as 
in Nastus, I find them equal, or almost so, in &. Wrayz. On the other 
hand, both B. Grifithii and B. Wray: differ from Nastus in the shape of 
the ovary, which is oblong, gradually tapering into the short style, which 
persists in the fruit as a cylindric beak, instead of being widened above. 
The glumes are coriaceous, but not so prominently nerved as in NVastus, 
and more like those of most Bambuse. In habit and leaves B. Wrayz 
differs very much from &. Grifithiz, approaching more nearly B. 
Arundinacea, at least so far as herbarium specimens are concerned. If 
we imagine the “gemmez’’ of the fertile spiculae as being all entirely 
suppressed, we should have exactly the same structure as we have in 
Nastus. The step to this is hardly greater than from the many-flowered 
spicula of the majority of Bamdbusx to the one-flowered spicule of 
B. Grifithi’, and if we retain this in Bambusa we ought, I think, to 
reduce Nastus also to Bambusa, where it might stand as a section or as 
a sub-genus, distinguished by the more distinct reduction of the structure 
of the spiculz, but linked very closely to the true Bumbusxe by B. Wray 
and B. Grifithii. From a geographical point of view, it is very interest- 
ing to note that the centre of Mastus lies quite at the periphery of the 
common area of the other Bambusz, in Madagascar and in the island of 
Bourbon, and that it just reaches Western Sumatra (Angkola, Junghuhn, 
according to Munro). Here in the most western part of the Malayan 
flora it is joined by its nearest ally, B. Wrayz, whilst B. Grifithit was 
found in the extreme north of Burma and in Manipur by Mr. C. B. Clarke. 
“Mr. L. Wray, Jr., states that this Bamboo was only found in the 
two localities mentioned above. According to him it dies off after having 
fruited. The Semangs call it Buloh bersumpitan, and they ‘use the 
large canes for the outer case of their arrow blow-pipes; the small ones 
for the inner tube. They straighten them over a fire when green, and 
hang them up in the smoke of their cooking-places to dry.’ In numerous 
sterile spicula the 4th or 5th glume was deformed by insect action. The 
basal portion was much enlarged and adnate to the lkewise enlarged 
next internode of the rachilla. This portion, a kind of oblong gall, 
contained the larva of the insect in the cavity. 
‘In regard to the plant described above (Bambusa Wrayr), the 
following interesting particulars respecting it are contained in a letter 
received at Kew from Mr. Wray, dated Perak, June 14th, 1892 :— 
