742 



THE TROPICAL AGRICULTURIST, 



[March i, 1883. 



so high as '216, while the Ceylon figures are 'OTS to 

 •173. Tliere can be little doubt that the superiority 

 of the Mlgiri soil in potash is due to its freer, more 

 disintegrated condition. Mr. Hughes' opinion was that 

 there %\'as uo necessity to apply potasli directly to our 

 soils, an opinion which was challenged by a gentle- 

 man interested in South of India coffee properties 

 Apart from the potash contained in superphosphate 

 of lime and castoroake. our silicates, even those which 

 arc described as "insoluble," but which gradually 

 yield to the influence of our hot moist climate, con- 

 tain abundance of this ingredient. In soda, again, 

 the Nilgiri soils have somewhat the advantage of 

 Ceylon, the range being from '030 to '164, while in 

 Ceylon we go from "a trace" in the old estate to 

 only '079 as our highest. We now come to the most 

 important element of all, phosphoric acid, and here 

 we would seem to have the superiority, in consequence, 

 we suspect, of the old estate, which gave 1'14, having 

 been recently manured with superphosphate. A ques- 

 tion was also raised about another estate, the soil 

 specinieu from which gave '211. Leaving out of count 

 the old estate, our range of phosphoric acid is from 

 •051 to •211, while in the case of the Nilgiri pro- 

 perties the range is from 'lOS up to •339. The su- 

 periority of tiie Nilgiri soils, therefore, in this, the 

 leading element of fertility is great and decided. 

 There is, however, a suflicient quantity of phosphoric 

 acid in most of our soils to enable them to be classed 

 as fertile, if we could only discover some cheap 

 method of converting the clay substratum from a stiff 

 to a free coudition. Sulphuric acid is present in but 

 small proportion in Ceylon soils, but the Nilgiri soils 

 are still more detioieut in this element, there being 

 only "a trace" in four of the specimens analyzed, 

 the ligures in the other four cases ranging from ■0'27 

 to ■113. The Ceylon figures range from '003, by far 

 the lowest, to •109. An average would shew de- 

 cidedly in favour of Ceylon, but the soil in both cases 

 vould be improved by the application of fresh suptr- 

 pliosiihate. We now come to chlorine, the compar- 

 ative abundance of which in the Dimbula soils has 

 a clobC bearing on our suggestion that the disease 

 which has leoently withered the leaves and "morti- 

 fied " the stems of eucalypts and cinchonas may be 

 due t(^ excess of salt spray carried on the wings of 

 the monsoon winds. In the case of the Nilgiri soils, 

 the lange for chlorine is so low as between ^002 and 

 004. With us, in Diinbula, the range is from '002 

 to so high as '017. Salt in the atmosphere, driven 

 against vegetation with the force of strong and con- 

 tinuous wmd, is a very different thing to salt in soil. 

 Had the new leaf-disease appeared in its virulent form 

 when Mr. Hughes was here, and had its saline origin 

 been then suggested, he would have helped us with 

 analyses of monsoon rainwater. What he did say in 

 1879 was ;— 



Chlorine, as in all inland soils, is present only in 

 smuU quantity. No. 7 being the only exception, and 

 in this soil it is not prestnt in any abnormal quantity, 

 considering its exposure to the north-east monsoon 

 and a somewhat high rainfall, which, being im- 

 pregnated with saline matter, would account for the 

 larger quantity of chlorine probably present as com- 

 mon salt. 



It 15 obvious that a quantity of chloride of sodium, 

 mixed with rain and driven by fierce winds for weeks 

 and months continuously against vegetation of foreign 

 origin, might do an amount of mischief, which niight 

 uevtr be suspected from analyses of soil on which 

 the rain was deposited. This by the way. Of carb- 

 onic iicid there is only "a trace" in any of the 

 Dimbula soils, as might be expected from Ihs paucity 

 of liine. In the Nilgiri soils, on the othi>r hand, 

 carbonic acid shewed in every case, ranging from 

 .OdO up to 'JWO. Here is a real difference, therefore, 



whatever its signifieance may be. In regard to the 

 Ceylon soils Mr. Hughes remarked : — 



Lime in all these soils is very low, the only excep- 

 tion boing No. 9, which is taken from a epot situated 

 at the base of a limtstoue rock. 1 was unable to de- 

 termine any appreciable amount of carbonic acid in 

 this, or any of the soils, so that doubtless the liiiie 

 originally present as carbonate (or bicarbonate if in any 

 kind of solution) has been decomposed by the humic 

 acid of the organic matter in the soil. 



But the presence of what tliere is of carbonic acid 

 iu the Nilgiri soils, Mr. Hughes does not consider 

 worthy of special notice. It is possible, however, 

 that the presence of this acid may have aided humic 

 acid in the reduction of the mineral constituents of 

 the soil ? We now, finally, come to the quarlzy 

 matter or silicates, and here, at the end as at the 

 beginning, Mr, Hughes has rendered exact comparison 

 impossible by adopting a different distinction iu the 

 Nilgiri analyses to that which he made when dealing 

 with the specimens from Dimbula, In our case the 

 usual formula was adopted of " silica soluble in alkali" 

 and "insoluble silicates." Of the former, the per- 

 centages ranged from 540 to 3 '765. The latter were 

 so high as from 43069 up to 74'798, the projmrtions 

 of quartz in those insoluble silicates b^ing from 13'120 

 np to 26'SO. In the case of the Nilgiri soils, Mr. 

 Hughes has not shewn " silicates soluble in alkali," 

 or the proportion of quartz in the " insoluble silicates." 

 What he does is to distinguish " pure quartz sand," 

 on which the freeness of the soil no doubt largely 

 depends, from " insoluble silicates." We suppose a 

 good deal of the quartz sand is also insoluble (?), in 

 which case, we suppose, the "insoluble silicates" are 

 really not so much lower in the case of the Nilgiri 

 soils, as the figures seem to indicate. As the Nilgiri 

 figures stand, they shew "pure quartz sand," rang- 

 ing from 7'2S0 to 449'f)0, while " insoluble silicates" 

 range from 10 51G up to 46 312. Here our comparison 

 ends, and, so far as it goes, we submit, it supports 

 the proposition we advanced at the commencement of 

 our remark", that the superiority of the Nilgiri soils 

 is not so much due to the presence in greatly larger 

 quantity of the main elements of fertility — nitrogen, 

 phosphori'j acid, potash and lime, as to better mechan- 

 ical condition, indicated by the description, " a 

 rich surface mould two fiet deep" In this respect 

 even large portions of the Nilgiri gr.'iss lands seem 

 superior to ours in Ceylon. We feel bound to notice 

 that none of the Dimbula analyses refer to grass land, 

 while in the case of the Nilgiri specimens Nos. 1 and 4, 

 are described as formerly grass land. In proportions 

 of nitrogen, phosphoric acid, lime and potash, these 

 soils do not compare very unfavourably with the 

 others, while their mechanical condition, judging from 

 proportions of alumina and quartz sand, does not seem 

 worse. On No. 4, hnwever, succirubras are represented 

 as growing very badly, while condauiineas are doing 

 " fairly well." W'e should have liked to have known 

 the age of the plants, because we can quite imagine 

 cinchonas growing "fairly well" on grass land at 

 first and then " dying out." We have ^rass laud 

 and grass land iu Ceylon. Any traveller through D< m- 

 bula can see cinchonas doing well on portions of " the 

 Devon patana," while a carefully conducted experi- 

 ment further up has resulted iu failure. With good 

 grass land to begin with and thorough digging to 

 a depth of at least IS inches, no doubt success can 

 be attained. We said that the deep rich soil of the 

 Nilgirr plantations enabled ihem to resist drought as 

 well as wind, but we have often asked ourselves the 

 que.stion whether conditions of climate in Southern 

 India and British S.khim which involve very heavy 

 falls of rain in from one-third to one-half of the year 

 and drought tempered by dews during the remaining 

 two-thirds or ouc-half, uiay not, after all, be, on 



