April 28, 1910] 



NATURE 



249 



correctly forecasting the monsoon rains either as to time 

 or intensity is still far from complete solution. They are 

 so elusive that the forecasting of rainfall even for three 

 days in this part of the world is not yet conspicuous for 

 its infallibility. 



The importance of the laws of rainfall is so incalculably 

 great that it is not extravagant to say that their discovery 

 is the ultimate object of, and excuse for, the millions of 

 meteorological figures that are published annually in all 

 civilised nations. Hence the discussion of rainfall in all 

 its aspects is worthy of encouragement, and in this con- 

 nection the centre of gravity of annual rainfall may 

 possibly, after all, be of some value. J. Cook. 



Edinburgh, April i8. 



I AM glad to read Mr. Cook's reply to my remarks, but 

 believe that my criticism cannot be dismissed as a mere 

 a priori one, and that it goes to the root of the matter. 

 It is true that Mr. Cook illustrated his proposal in a 

 most exhaustive manner, and that he did not suggest that 

 his method might be of service in comparing the rain- 

 falls of places in quite different climatic regions. But the 

 general reasoning in the first paragraph of my former 

 letter cannot be both correct and incorrect. Assuming it 

 to be correct, it follows directly that even if we confine 

 our attention to the records for a single station we might 

 have the same C.G. for two years which differed greatly 

 from one another as regards the monthly distribution of 

 rainfall. In such a case, what possible significance could 

 attach to the position of the C.G. ? 



I am heartily in sympathy with Mr. Cook's feeling that 

 the discussion of rainfall in all its aspects is worthy of 

 attention, but note that he himself does not maintain that 

 his method is, but only that it may possibly be, of 

 some value. It is certainly at first sight surprising that 

 the calculated C.G. of rainfall for a large number of 

 places for a given jear in, say, Scotland should be very 

 nearly the same, for the monthly rainfalls as ordinarily 

 tabulated exhibit a bewildering complexity ; but if the 

 monthly values for the various stations are expressed as 

 percentages of the year's total, the resulting picture is 

 usually of a very simple and symmetrical character, which 

 would lead one to anticipate that the C.G.s for the 

 \ arious stations would approximate closely. — 



Andrew Watt. 



Scottish Meteorological Society, 122 George Street, 

 Edinburgh, April 22. 



The Fertilising Influence of Sunlight. 



The beneficial effect of heat on soil is recorded by Virgil 

 in the following passage, to which Mr. F. B. Smith has 

 directed my attention : — 



" Often too, 'tis eood 

 To bum the stubbles and » ith crackling flames 

 Consume the empty staiks ; whether from thence 

 The earth derives a hidden store of strength 

 And fattening food, or whether 'tis that fire 

 Rakes out the subtlest vice and sweats away 

 Exces-sive damp, or whether by the heat 

 New pores are opened and the choked are cleared," &c. 



(" Georgics," Bk. i., hnes 100 et seq.). 



It is interesting to learn from Mr. Fletcher (April 7) 

 that the natives of Bombay, in certain circumstances, sub- 

 ject their soils to heat. Mr. Fletcher regards the explana- 

 tion given by Dr. Hutchinson and myself as incorrect, 

 and suggests that the effect is due to the destruction of 

 some toxin. This was the first hypothesis we examined, 

 but was found to be insufficient. 



(i) Toluened soil {i.e. soil treated with a small quantity 

 of toluene, which is subsequently allowed to evaporate 

 in situ without washing anything from the soil) is more 

 fertile and more favourable to bacterial activity than the 

 original untreated soil. 



(2) When an aqueous extract of untreated soil is added 

 to the toluened soil, there is a still further increase in 

 fertility and in bacterial activity. The same result follows 

 when a minute amount of the untreated soli itself is added 

 instead of the aqueous extract. 



(3) When a larger quantity (5 per cent.) of the untreated 

 soil is added a similar effect is produced for a time, then 

 the bacterial activity begins to be depressed. This action 

 increases, and finally the depression, both in bacterial 



NO. 2 113, VOL. 83] 



activity and in fertility, is out of all proportion to the 

 5 per cent, of soil originally added. 



Experiment (2) is conclusive against the hypothesis that 

 a soluble toxin exists in the untreated soil which can be 

 put out of action by toluene. For such a toxin should 

 cause a decrease, and not an increase, in productiveness. 

 Experiment (3) is equally conclusive against a relatively 

 insoluble toxin ; had this been present the depression should 

 have shown itself at once, and should have been pro- 

 portional to the amount of toxin, i.e. of untreated soil, 

 added. 



The growth of the injurious factor in experiment (3) 

 seems to necessitate a biological hypothesis. Considering 

 these and our other experiments in detail. Dr. HtKchinson 

 and I see no way out of the conclusion that organisms are 

 present in soil inhibiting the development of bacteria, and 

 therefore of plant food. The organisms, whatever they 

 are, must be larger than bacteria, or they would occur 

 in the extract of experiment (2) along with the numerous 

 bacteria there present — indeed, the beneficial effect of this 

 extract was traced to the unweakened races of bacteria 

 present, partial sterilisation having somewhat weakened 

 the soil bacteria. Further, they develop more slowly than 

 bacteria. As similar phenomena have been observed in all 

 the soils examined, we are justified in supposing that the 

 organisms are widely distributed, and constitute an 

 important factor in soil fertilit}-. 



Mr. Fletcher's water-culture experiment is not germane 

 to the point. A toxic body that occurred there would not 

 necessarily come direct from the plant or be found in the 

 soil. It is extraordinarily difficult to keep prolonged water 

 cultures sterile, and until some attention is paid to the 

 bacterial changes going on it is impossible to regard the 

 results as proof of the presence of toxins in soils. Indeed, 

 I know of no satisfactorj- evidence of their existence in 

 normal soils. E. J. Russell. 



Rothamsted Experiment Station, Harpenden. 



Pneumatolysis. 



It is thirtj--nine years this month since Nature, over 

 diffident initials, published my first scientific communica- 

 tion that ever saw print. For more than thirtj' of those 

 years I have been much interested in the physics of 

 plutonic rocks. Quite recently an event has occurred 

 which must be almost without precedent in science. The 

 petrologists have apparently repudiated, with unanimitj-, 

 what is an axiom beyond dispute with chemists. 



For some years past the petrology of plutonic rocks has 

 been based on the new doctrine of " pneumatoljsis, " or 

 the solvent powers of gases over solids. 



Perhaps the last published important work on chemistry- 

 is the English version of Ostwald's " Fundamental Prin- 

 ciples of Chemistrj," 1909. Referring in that work to a 

 certain diagram, representing the behaviour of one solid 

 and one gas, the author writes : — 



" From this point the liquid phase exists in the presence 

 of the gaseous phase to the end of the diagram, because 

 solid substances do not form solutions with gases " (pp. 

 1S6— 7). Italics mine. 



I believe that everv^ chemist will assent to the above 

 statement. If a gas is to mix with a solid, as a solution, 

 the solid must first be vapourised ; but if this be so the 

 greater part of twentieth-centurj' petrology breaks down, 

 because it is everywhere relying on the truth of pneumato- 

 lysis. 



From the student's point of view the situation is as 

 paralysing as it is stupefying, and there seems nothing 

 to be done but to put away the microscope. It is no work 

 for students to discuss first principles. 



Southwood, Torquay, .April 18. Arthutj R. Hunt. 



Anomalous Reading of Hygrometer. 



May not the observation referred to in Nature of 

 April 7 (p. 165) be a very simple case of latent heat evolu- 

 tion by condensation when the atmosphere is supersaturated 

 with vapour? I think I have seen the wet bulb register- 

 ing a temperature higher than the dry bulb ; but this 

 explanation seemed at the time so obvious that I made no 

 careful verification of the apparent phenomenon. 



Hugh Richardson. 



Bootham School, York, April 11. 



