TAISAKA HOUSES 263 



made of woven cloth, indeed they seem to have little, if any, 

 knowledge of spinning or weaving. On the other hand, they 

 are clever in straw-work and in manufacturing mats and 

 baskets. 



Their houses are very small, made of a slight framework and 

 filled in with the midrib of the leaves of the traveller's tree 

 in the same way that the zozoro (papyrus) is used in Imerina, 

 and looking almost exactly like zozdro. These leaf-stalks, 

 which are called faldfa, are fixed together on long fine twigs so 

 as to make a kind of stiff mat, the triangular stems easily fitting 

 in alternately. These mats are the ordinary mattress, and are 

 used in various other ways. One of them forms the door on 

 either side of the house, being shifted to one side or another as 

 required, and is kept from falling by sliding within a pole hung 

 from the framework. The flooring, which is always raised above 

 the ground, is made of the bark of the traveller's tree, pressed 

 flat so as to form a rough kind of boarding ; while the thatch of 

 every house is the leaves of the same tree, which forms a neat 

 and fairly durable covering. Here also, as among the other 

 coast tribes which we have seen, the traveller's tree might be 

 called with equal or greater propriety, " the builder's tree." 

 The hearth is at one end of the house, in the centre, with a 

 strong square framework above it, having two or three rows of 

 shelves. The trdno-dmbo, or elevated house for storing rice, 

 seems common to every tribe we have visited since leaving the 

 Betsileo province. The villages here are arranged in groups of 

 from two to half-a-dozen in a line, and with only a small space 

 between each group. 



The rice-fields in this flat swampy district have a very 

 different appearance to those in Imerina or Betsileo ; they are 

 like immense pits, in some places dug out to some depth in the 

 sides of the low elevations. The people do not transplant their 

 rice, as do those of the central provinces, but reap it where it 

 has been sown. We continually came across traces of volcanic 

 action; ancient streams of lava, conical-shaped hills and, on 

 the coast, reefs of basalt rock, gradually being broken up by 

 the action of the waves. All this showed that the great groups 

 of extinct volcanoes in the central provinces had their counter- 

 part in these southern regions of the island. Another inter- 

 esting fact was, that we found unmistakable signs also of 



