January i6, 19 13] 



NATURE 



541 



theoretical calculations is more than satisfactory ; 

 especially in view of the difficulty of making, in a 

 material soft and easily distorted, the cuts called for 

 by the theory. 



It would certainly be preferable to make use of 

 celluloid, as Prof. Coker now does. I did indeed 

 attempt it at the time of these experiments, but ex- 

 perienced some little difficulty in glueing the celluloid 

 after cutting it, inasmuch as, in accordance with 

 Volterra's theory, pressure is exerted on some regions 

 of the faces in contact, tension on others. 



The optical method, which permits the investlga- 



Without having experimented on the effects of com- 

 plete sterilisation of a sick soil, Messrs. Russell and 

 Petherbridge state {},oc. cit., p. 90), "our experiments 

 thus lead to the conclusion that at least two factors 

 are concerned in soil sickness : a falling off in bac- 

 terial activity and an accumulation of plant parasites 

 and disease organisms." 



.A.S a matter of fact, all that is proved is, as in Drs. 

 Russell and Darbishire's paper quoted, that partial 

 sterilisation produces both increased crop and increased 

 bacterial activity. The illogical conclusion is then 

 drawn from this that increased crop is due to increased 



Fic. 3. Fig. 



tion of the internal distribution of tensions, has thus 

 given us the means of verifying the theory of elas- 

 ticity in a salient point, namely Volterra's theory of 

 distortions, just as it has given Prof. Coker the means 

 of supplying, by the use of experimental models, cer- 

 tain deficiencies of the theory — a theory which cannot 

 always submit to rigorous calculation the complex 

 conditions of internal strains to which the materials 

 of practical construction are subject. 



O. M. CORBINO. 



Rome, Physical Institute of the Royal University. 



The Bacterial Theory of Soil Fertility. 



In vol. v., part i. (October, 1912) of The Journal oj 

 Agricultural Science. Messrs. Russell and Pether- 

 bridge state that " partial sterilisation appears to be 

 ■the proper method of dealing with ' sick soils ' " 

 (p. 91). I venture to think that the theory of partial 

 sterilisation — which is indeed very attractive — will not 

 stand examination. .According to this theory the fer- 

 tility of a soil depends largely on its bacterial popu- 

 lation, the enemies of which are destroyed by partial 

 sterilisation, which the bacterial spores survive. 



Now if this theory is correct, it should follow that 

 complete sterilisation must diminish the fertility of a 

 soil, since all bacterial spores will have been destroyed. 

 This, however, is not the case, as Dr. Russell is 

 apparently aware, for Drs. Darbishire and Russell in 

 the same journal (vol. ii., part iii., December, 1907, 

 p 319) state: "a few experiments have been made 

 with soils heated to 120° C. The same kind of 

 results are obtained as at the lower temperatures, but 

 thev are somewhat intensified." 



In other words, complete sterilisation gave an in- 

 creased crop over partial sterilisation. The 

 corr'parative effects of complete and partial steri- 

 lisation on a soil were shown by the present writer 

 (Cairo Sci. Jour., vol. iv.. No. 4:;, .'\pril, 1910), maize 

 in soil untreated, soil heated to 95° C, and soil 

 heated to 170° C. yielding green weights in the 

 proportion of i45'5. iS''?- i"5S"6 (see Fig. i). 

 NO. 2255, VOL. 90] 



bacterial activity. This erroneous deduction would not 

 have been arrived at had a few parallel experiments 

 been conducted with completely sterilised soils. 



.Again, the authors find that plants grow as well 

 in extracts from the "sick" soil as in extracts from 

 partially sterilised soil. They conclude that the 

 "sickness" is therefore not due to a soluble toxic 

 substance. But does this prove it? The phenomenon 

 of absorption (or adsorption) of soluble salts by soils 

 appears to have been overlooked. Further, these 

 results are in direct contradiction to the very elaborate 

 experiments carried out by the U.S.A. Department of 

 Agriculture. 



With regard to the growth of seedlings in water 

 extracts of soils, very little detail of the method of the 

 experiments is given. For instance, it is not stated 

 at what stage the seedling was planted in the water 

 extract. That certain precautions may have been 

 overlooked would appear possible from the statement 

 (used as an argument against the toxic theory) that 

 " cucumber seeds are very sensitive to unfavourable 

 conditions, but they germinate fully as well in ' sick ' 

 soil as in partially sterilised soil." 



Now Pickering (Journ. Agric. Sci., vol. ii., part iv., 

 and vol. iii., parts i. and iii.) pointed out that ger- 

 mination is delayed in heated soils — a fact long known 

 to farmers — and supposed that this was due to the 

 production bv heat of a toxic substance. The present 

 writer (loc. cit.) proved that this delay is due entirely 

 to a physical cause, viz. the increased osmotic pressure 

 in the water contents (and water extract) of a heated 

 soil ; this causes imbibition by the seeds to be checked, 

 in some cases to such an extent that they rot before 

 they have absorbed sufficient water to cause germina- 

 tion. 



Now if the seeds in Russell and Petherbridge's 

 experiments were germinated in the soil extracts, 

 after five days' growth (plate iii.. Fig. 3a), we should 

 scarcely expect the seedling in the extract from the 

 heated soil to have made up for time lost in ger- 

 mination. Even if the seeds were all germinated 

 under the same conditions, e.g. in water, and seed- 



