IMP 
` 
| 
i 
! 
| 
FRUIT OF MELOCANNA BAMBUSOIDES. 407 
produced mainly on the gynodynamous inflorescences, slender spikes from a few to 
30 em. long. Where they occur in androdynamous inflorescences which represent much 
decompound panicles, they are always few and usually terminal or subterminal on the 
branches or branchlets bearing them. They spring from sessile, 1-2-flowered spikelets, 
the bracts (glumes, valves, pales) of which persist after flowering in an unchanged though 
withered condition. Nor does the axis of the spikelet which aets as a very short fruit- 
stalk outwardly suffer any transformation, the necessary strengthening being provided 
for, here as well as in the axes of the inflorescences, by internal changes resulting in the 
development of sufficient mechanical tissue. "Thus the fruits hang, by means of very 
short and slender stalks of great tensile strength, from the flexible and wiry axes of 
the inflorescences. 
III. DURATION or Lire (“ FLOWERING PERIOD ”). 
-Metocanna bambusoides is, like many other bamboos, so far as we know, monocarpic. 
It is a gregarious plant forming extensive forests, and flowers and fruits simultaneously 
over large tracts at long periods, each period corresponding to one generation. Kurz (in 
‘Ind. Forester,’ i. p. 257) says it flowers and fruits in Arakan every 30 to 35 years. The 
dates of flowering and fruiting which I have been able to ascertain are: 1801 for 
Chittagong (Pierard in Roxburgh, Hort. Beng. p. 25), 1849 for Tenasserim (Falconer in 
Hb. Kew.) *, 1863-1866 for the Botanic Garden, Calcutta (Gamble, Ind. Bamb. p. 120), 
1864-1865 for Arakan (Kurz, J. c. p. 258; this is evidently the fruiting to which Theobald 
refers in Mason, ‘Burmah, 2nd ed.ii. p. 101), 1889 for the Garo Hills (specimen in Gamble's 
herbarium), 1892 for “ Assam” (Gamble, Z. c.), and 1900 to 1902 for the Garo Hills, 
Cachar, and Chittagong (specimens received by Sir Dietrich Brandis). It is clear from 
these dates that we cannot speak of a general * flowering period " of 30-35 years in the 
case of Melocanna bambusoides, although it may be quite true that the normal duration 
of life of this monocarpic bamboo is 30-35 years. We can hardly assume o priori that 
the flowering should take place simultaneously (or, what is about the same, that all the 
individuals of this species should be of the same age) over the whole area with its varied 
conditions of soil and climate, and in fact we know that this is not the case with other 
gregariously flowering bamboos (see Brandis in ‘Indian Forester,’ xxv. 1899, p.10). The 
dates for the Garo Hills—to which that for * Assam” may be added, as by “ Assam” 
no doubt the south-western corner of the Kamrup District is meant—are particularly 
instructive, as we learn from them that flowering took place at least three times during 
a period of about 12 years within a limited area. There is, however, no evidence to show ` 
what portions of the area were affected each time, or how far the flowering was general 
in the affected portions. More numerous observations, extending at the "e time over 
the whole area, are necessary, in this as in the case of other “ periodically " flowering 
bamboos, before we can understand this remarkable " periodieity." 
* A flowering specimen, collected by Falconer in “ Moulmein," was formerly in en Kew Volpe. EE 
to a note it was sorted out because it was identical with another flowering specimen ent by Kurz in the Botenie 
Garden, Calcutta. Falconer was in Moulmein and the Attaran District early in 1849, and I — the Koen 
was then collected, although he does not mention it especially in his " Report on the Tenasserim Teak Forests, 
