1917] AGBICULTUBAL CHEMISTRY AGEOTECHNY. 413 



The determination of nitrogen in the form of calcium cyanamid in lime 

 nitrogen, A. D. Bekkhout, K. D. Hekdriksz, and J. Wind {VersUuj. Landbouwk. 

 Onderzoek. Rijkshnulbomvproefstat. [Netherhnids'l, No. 20 (1917), pp. 43-52).— 

 The following method has been found to yield most reliable results : 



Two gm. of the finely pulverized sample is rubbed up with repeated small 

 amounts of distilled water and transferred to a 1,000-cc. measuring flask by 

 decantation, using about 900 cc. of water. The flask is then placed in the 

 shaking machine and thoroughly shaken for about one hour. The solution, 

 after being made up to volume, is thoroughly shaken, and treated by either of 

 the following methods : 



(1) Fifty or 100 cc. of the solution is neuti'alized in a flask with 10 per cent 

 nitric acid and then treated with from 2.5 to 5 cc. of 2.5 per cent ammonia. 

 Tenth-normal silver nitrate solution is immediately added in excess from a 

 burette while the solution is being thoroughly shaken to precipitate the cyanamid 

 as silver cyanamid. The material in the flask is made up to volume and after 

 thorough agitation is filtered. After fliis precipitation the solution should be 

 decidedly alkaline. From 50 to 100 cc. of the filtrate, after being acidified with 

 from 5 to 10 cc. of 10 per cent nitric acid, is titrated with tenth-normal ammo- 

 nium thiocyanate solution, using from 2.5 to 5 cc. of a cold saturated ferric 

 ammonium alum solution as indicator. If chlorids are present they should 

 be previously determined and the proper correction made. 



(2) One hundred cc. of the solution is transferred to a 250-cc. beaker, 

 neutralized with 10 per cent nitric acid, and after the addition of 2 cc. of 10 

 per cent ammonia a slight excess of tenth-normal silver nitrate is added with 

 thorough agitation. The precipitated silver salt is quantitatively filtered, 

 thoroughly washed, and the wet filter and contents subjected to a Kjeldahl nitro 

 gen determination, using 10 cc. of 50 per cent sulphuric acid, 0.6 gm. of mer 

 cury, and 20 cc. of a mixture of 1 liter of sulphuric acid (specific gravity 1.84) 

 and 200 gm. phosphorus pentoxid. 



On the Duclaux method for the estimation of the volatile fatty acids, F. 

 W. Upson, H. M. Plum, and J. E. Schott (Joiir. Amer. Chem. Soe., 39 {1917), 

 No. 4< PP- 731-742). — From the results of a critical study the authors conclude 

 that small, unavoidable variations in the experimental results such as ai"e 

 within the limits of error of the method may cause such wide variation in the 

 calculated results as to make them of no quantitative value. When more than 

 two acids are present in a mixture practically identical series may be calcu- 

 lated from mixtures of different acids in varying proportions. " Results which 

 indicate the presence of one acid may just as well be calculated in terms of 

 three or more acids. Small amounts of acids may be distributed just as well 

 between the acids next higher and lower in the series." 



The theory of the method is deemed to be unsound, and the procedure there- 

 fore not to be applicable for either quantitative or qualitative determinations 

 of the composition of unknown mixtures of fatty acids. 



The experimental data are submitted and discussed in detail. 



A note on the Duclaux method for volatile fatty acids, A. R. Lamb {Jour. 

 Amer. Chem. Soc., 39 {1917), No. 4, pp. 746, 7^7).— From a critical study of 

 the method in connection with some investigations on the fermentation of corn 

 silage the author, at the Iowa Experiment Station, considers it to be impracti- 

 cable to attempt to determine more than two or three acids in the same fraction. 

 " By proper fractional distillation the acids in the mixture may be quite well 

 separated. This obviates the necessity of calculating as many as four acids 

 from a single distillation, which is objected to by Upson et al." (as noted above). 



The lack of agreement between different sets of Duclaux's constants is con- 

 sidered to be due to impurities in the acids used and to avoidable experimental 



