November 19, 19 14] 



NATURE 



309 



oxidation and productiveness, a set of pot experi- 

 ments was put up, and these showed that the 

 \ rtihty of the soil was also increased after partial 

 -ierilisation. Typical results are illustrated in 

 Fig-. I : the upper part gives experiments with rye, 

 the lower part with tomatoes. 



At this stage it was found that the fact had 

 been discovered before by practical growers. Soil 

 that had been heated, e.g. by a bonfire, or treated 

 with the antiseptic (carbon disulphide) used for 

 destroying- certain pests, gave larg-er crops than 

 before. The fact had even been reported to one 

 or two of the German experiment stations, but it 

 had not attracted very much attention. One im- 

 portant discovery had been made, however : it was 

 found that partial sterilisation caused increases in 

 the numbers of bacteria, and this was explained 

 by supposing that the weaker races had been 



^^0 



S4 



^'i.iV.X'i 



ji/L ()_. ^5/ C.Oi. 



Fig. 2. — Lime as an agent for soil sterilisatiun. Pot 73, control ; 76-87, succe-^sively increasing 

 doses; between 79 and 80 lies. the partial bterilisatioa point ; 87 contains an excessive 

 amuuiit. (H. B. Uuichin;on.) 



killed, leaving the moVe vigorous in clear posses- 

 sion of the field. 



Systematic investigations at Rothamsted soon 

 sliowed that the increased productiveness of par- 

 tially sterilised soils was due to an increased 

 formation of ammonia, and this in turn was 

 traced to the increased numbers of bacteria. The 

 new flora, .however, was not composed of more 

 vigorous races as had been supposed; on the con- 

 trary, it was' less vigorous than the old, and it 

 wed its extra decomposing power to its numeri- 

 al superiority. This point was established by 

 umerous experiments because of its great im- 

 portance in explaining the phenomena. The 

 reason for the increased bacterial numbers was 

 traced to a detrimental factor present in normal 

 soils and keeping down the bacterial numbers ; 

 this factor was put out of action by partial sterili- 



NO. 2351, VOL. 94] 



sation, whereupon the bacterial numbers began to 

 rise. Owing to its highly complex nature soil 

 forms very unpromising material for a direct 

 attack ; investigations have nearly always to be 

 indirect. In the present instance the method 

 adopted for discovering the detrimental factor was 

 to ascertain what agencies put it out of action and 

 what did not, and thus to draw up a list of its 

 properties. It was found, for example, that the 

 detrimental factor was sharply put out of action 

 by heating to 55° C, or by any poison, and that 

 it never set up again so long as the soil was kept 

 free from dust ; it did, however, appear when a 

 little untreated soil was added, but the reappear- 

 ance was slow and somewhat erratic. These and 

 other experiments all indicated that the factor was 

 biological ; but definite evidence was obtained that 

 it was not bacterial. Suspicion therefore fell on 

 the protozoa, which were found to 

 occur in all the natural soils ex- 

 amined but to be killed by the 

 partial sterilisation. A very large 

 number of experiments showed that 

 the destruction of the soil protozoa 

 always went hand in hand with the 

 increase in bacterial numbers ; up to 

 the present no exception has been 

 found. This is strong presumptive 

 evidence, but unexpected difficulties 

 arose when attempts were made to 

 clinch the evidence by inoculating 

 mass cultures of protozoa into par- 

 tially sterilised soils, and up to the 

 present these have not been over- 

 come. We are, therefore, thrown 

 back on indirect methods. 



The first difficulty that was raised 

 against the identification of the de- 

 trimental factor with the soil proto- 

 zoa was that the protozoa were 

 probably present in the soil as cysts, 

 and not in the trophic state. It was 

 argued that the soil conditions must 

 be unfavourable for the development 

 of animals like protozoa, which 

 stand in constant need of air and 

 water. The first investigations 

 support this view. The copious de- 

 of protozoa when untreated soil is 

 into hay infusion, of course, proves 

 nothing ; and when attempts are made to entice 

 anything out of the soil by thermo- or galvano- 

 tactic methods nothing happened. The test or- 

 ganism was Colpoda ciicuUus, a ciliate occurring 

 abundantly in the culture and easy to pick out 

 under the microscope, and no evidence could be 

 obtained that it occurred in the trophic state in 

 the soil. 



However, the experiments were continued. Mr. 

 Martin found that a film was formed when soil 

 was vigorously shaken with concentrated picric 

 acid solution which, on examination, was seen to 

 contain protozoa. The significance of the obser- 

 vation is, of course, that the acid instantly kills 

 the organisms in the soil, so that the state in 



Si 



_!/ CJOl 



seemed to 

 velopment 

 inoculated 



