218 EXPERIMENT STATION RECORD. 



From the results obtained it is concluded that the bacterial activity in soil 

 depends on the size of soil grains, the water content, and the concentration of 

 food. The speed and the final stage of decomposition vary with these three 

 factors. 



In order to obtain physiologically comparable results it is shown that the 

 soil solution must contain the same food concentration in all experiments, which 

 does not correspond with conditions in cultivated soil. With the same food con- 

 centration in the soil the decomix)sition of some substances shows approximately 

 under all conditions the same final point of cleavage, so that only the speed of 

 decomposition and not the final stage is influenced by the size of grain and 

 water content. This was found to be the case in the decomposition of peptone 

 by B. mycoides. However, with other bacteria in previous experiments the 

 speed of decomposition and also the point of final decomposition were influenced 

 by the physical properties of the soil. 



Coarse spongy organic substances, such as decomposed cellulose, acted in dry 

 soils as water absorbers and diminished the bacterial action, but in moist soils 

 they increased ventilation and thereby increased the activity of aerobic bacteria. 



Occurrence of Azotobacter in tropical soils, J. Groenewege {Aleded. Proef- 

 stat. Java-Suikerindus., 4 {1913), No. 13, pp. 241-2U; Arch. Suikerindus. Neder- 

 land. Indie, 21 {1913), No. 26, pp. 790-798; ahs. in Gliem. Ahs., 7 {1913), No. 

 21, p. 3635). — Contrary to the conclusions of De Kruyff, the author found 

 Azotobacter chroococcum in all but one of a series of Java soils, and in this the 

 chlorin content was 3.S6 i^er cent, indicating suSicient sodium chlorid to kill the 

 bacteria, BaciUus radiohacter was also found to be generally present in Java 

 soils. 



Methods in soil bacteriolog'y. — VI, Arnnionification in soil and in solution, 

 F. LoHNis and H. H. Greex {CentN. Bakt. [e^c], 2. Aht., 37 {1913), No. 22-25, 

 pp. 534-562; als. in Jour. Chem. 8oc. [London^, 104 {1913), No. 609, I, pp. 797, 

 798; Chem. Abs., 7 {1913), No. 19, pp. 3380, 3381).— This is a critical study of 

 the factors affecting ammonification and nitrification of blood meal, flesh meal, 

 and horn meal under laboratory conditions. It is considered that in determin- 

 ing the relative value of laboratory tests in soil and solution media the specific 

 mode of application of methods must be clearly defined, since widely varying 

 results may be obtained in both cases according to variation in the experi- 

 mental conditions. *' The most significant cause of variation appears to be 

 that of aeration." 



Ammonification, as well as nitrification, proceeded much more rapidly un er 

 aerobic than under anaerobic conditions. It is believed that aerobic condi- 

 tions especially favor the later stages of breakdown resulting in the formation 

 of ammonia itself. Under conditions of insufficient aeration, increasing the 

 quantity of nitrogenous material used retarded the ammonification, but with 

 adequate aeration it had little influence. In general smaller quantities of 

 material experienced more complete ammonification although the chemical 

 nature of the material used seemed to be the dominant factor. The duration 

 of the experiment affected the apparent extent of ammonification, the rate 

 of ammonification being in general greatest during the first ten days in both 

 soil and solution. The three materials used showed very different rates of 

 decomposition. 



In solution tests the losses of ammonia by evaporation were highest and the 

 relative extent of ammonia assimilation lowest with the larger quantities of 

 material. In these tests the use of magnesium hydrogen phosphate effected 

 marked conservation of ammonia only in the longer periods of decomposition. 

 However, the ammonia not lost through evaporation was assimilated by the 



