395 



riar (Snake yam) and the T'biuni Dracnnnm or T^hi Boaya (Crooo- 

 tlilo yam). He tells us that he grow thorn in ricli clay soil free of 

 all stones in order to demonstrate to hiniself that it was not 

 ohstaeles in the soil which produced the unusual direction, and 

 " many of them thrust their tails out of the soil." 



The (ork-screw apiiearancc which Kum|)hius caused to he 

 drawn, hai)pens to he in excess of anything' produced in Sing'a|)ore, 

 where the longer M'ere gently sinuous. 



Tamil gardeners in Singapore, were in charge of the yam heds, 

 and when the yams, as Rumphius ejjpresses it, thrust their tails 

 out of the soil, these men having never seen anything similar in 

 fndia, were at a loss what to do: hut by earthing up as may 

 he seen in Plate 6, the tip was covered whenever it extruded, and 

 its elongation maintained. Experience in 1914 and 191 o had 

 shown that if not earthed up the apex dies. 



On plate C^ the reader will find figured the two most striking 

 of the races of this type which were grown. Tn ])reparation for the 

 figures there given, a rough hamhoo scaffolding was made above 

 the beds and to cross poles the tu1)ers were tied by strings before 

 the earth was removed from around them : and by this means the 

 tubers were retained in their natural position. The race of yam 

 shown in the upper block of plate fi is shown again in plate 5, lower 

 block: and the upper block of nhite .") exhibits a transition stage 

 between what is usual and this peculiar development wherein the 

 yams curve in the soil but not so as to extrude. Although the 

 extreme conditions such as are shoM'n on plate 6 appear to be con- 

 fined to eastern Malaya, the condition seen in the u]iper block of 

 Plate 5 is rather widely distributed : and being for instance a mark 

 of the " B\dl's born yam " of the Shans. 



The races with upgrowing tubers will now be taken seriatim 

 commencing with the one which forms the transition. 



Race Xo. 38 has already been figured in this Bulletin as Xo. 1 

 on page 301 from a tid:)er grown in 1914. Tn 1915 it showed a 

 tendency to branch, hut it had bad soil. In 1916 it produced yains 

 as figured on plate o. The flesli is white hut there is magenta 

 sap Just under tlie ^kin. The skin is not rooty. Tlie average re- 

 turn was 7 Ills. T oz. (3374 grammes) which is per acre 7 tons 9| 

 cwt. 



This is the last yam for which a rate per acre can he given, 

 as the others need to tlie cultivated further apart that they may 

 be earthed up easily; and the nnmlier of hills to an acre would thus 

 be reduced below 2350. 



Race Xo. 38 is that whicli is sliown in two plates, l)oth plate o 

 and plate fi. From plate 5 the slightly sinuous curving of the 

 tubers may be realised: but it hai)pened to be greater in one tuber 

 too elongated for convenientlv arranging in that grou]). It will be 

 observed that in this race tlic direction nf the tuber at first has been 

 downwards but that at a \ovy eai-ly stage a strong curve takes 

 place by which the point begins to ascend more or less vertically. 

 The age at which this change of direction comes in varies, and 

 in one of the tubers figured the tulier did not grow down at all 



