IX, ©, 6 Gates: Pioneer Vegetation of Taal Volcano 409 
of the line of ordinary storm-wave action, although it has not 
yet acquired sufficient density to eliminate Ipomoea. Meager 
evidence of further successional relations indicates the develop- 
ment of the Themeda consocies of the cogon association, or of the 
Acacia consocies of the parang. The extreme intolerance of 
Sesbania makes this association very easy to replace by shading. 
The presence of two plants of Pandanus tectorius in the back 
strand of the southeastern region and one of Hrythrina indica 
near Pirapiraso is all that there is at present to indicate the 
development of the Pandanus strand association, one which is 
quite common on the seacoast strand elsewhere in the Philip- 
pines, but no other strand associations were indicated. 
THE COGONAL OR GRASSLAND FORMATION 
THE COGON ASSOCIATION 
This association of grasses in one or another of its consocies 
is the most widespread association on the island, although almost 
nowhere does it reach its normal density. As this association 
is characterized by certain species of grasses, nearly any one of 
which may develop to the exclusion of the others in a given 
area, the whole association is easily divided into consocies, based 
on the specific identity of the grass which dominates. On Taal 
Island three of the consocies of this association are definitely 
represented, namely, the Saccharum spontaneum, the Themeda 
gigantea and the Imperata cylindrica consocies, while there is 
a suggestion of a fourth, the Miscanthus consocies, in a very few 
places. On the devastated slopes the first two of these have 
appeared, the first in greatest abundance, while the third has 
made its appearance in the ridges and valleys of the northeast 
cape, where the devastation was not so severe. 
The Saccharum spontaneum consocies.—The light, silky- 
haired, wind-distributed seeds abundantly produced by Saccha- 
rum spontaneum were widely distributed over Taal Island by 
the northeast monsoon. The sides of ridges, particularly those 
facing to the northeast, were first vegetated, the sweep of the 
wind and the lack of water preventing seeds from lodging on 
the backs or the crests of ridges and the wash of water after 
rains preventing them from remaining in the drainage channels. 
From these many centers of dispersal, vegetative reproduction, 
together with the plants from the excessively abundant crop 
of seeds produced on the island in 19138, is fast obliterating the 
striking relation of seeding to exposure to wind. 
From ash-buried rhizomes one would have expected the ap- 
