AGRICULTUEAL CHEMISTRY. 309 



tenth-normal sodium bydroxid and after the addition of an equivalent amount 

 of sulphuric acid is used for the eolorimetric determination of salicylic acid by 

 tlie use of iron alum. 



Befraction of nonvolatile fatty acids in butter, W. Lttdwig (Ztschr. rnter- 

 -iich. yahr. //. Gcniissnith. I'f {1001), Xo. 3. pp. 20S-213). — The indices of refrac- 

 tion of 111 samples of butter ranged from -in to 43.6 and of the nonvolatile fatty 

 :;(ius from 2S.0 to 31.7. 



Refraction of nonvolatile fatty acids in butter, H. Sprixkmeykb and A. 

 FiRSTEXBEBG (Ztschr. rnter.^iich. Xahr. ii. Gfiiiissmtl.. I'f {1907),\o. 3. pp. 

 U3-21')^. — The authors determined the indices of refraction of butter, cocoanut 

 oil. lard, and other fats, and of the nonvolatile fatty acids contained therein. 

 Ten samples of butter showed indices ranging from 42.S to 4.5. and the nonvola- 

 tile fatty acids of the butter indices ranging from .30.8 to 3.3.4. Similar results 

 were obtained with other fats. The authors do not believe that the refrac- 

 tometer numbers of the nonvolatile fatty acids show less variations thau those 

 of the fats. 



Refraction of the nonvolatile fatty acids, T. Sudexdorf {Ztschr. Untersuch. 

 Xahr. u. GenussmtL. 14 (1907), Xo. 3. pp. 216-220). — The refractometer numbers 

 of 56 samples of butter from various sources ranged between 40.1 and 46.3 and 

 the numl>ers of the nonvolatile fatty acids of the butter between 28.7 and 34.9. 

 Determinations were also made of the indices of refraction of cocoanut oil, 

 lard, and tallow, and of the nonvolatile fatty acids of these fats. As a means 

 of detec-ting adulterations, the author finds no advantage in determiuiug the 

 index of refraction of the nonvolatile fatty acids over that of the fats themselves. 



A new method for the detection of cocoanut oil in butter, R. Cohx (Ztschr. 

 uffvntl. Chcni.. 13 {,1907). Xo. 16. pp. 30S-311 i. — The fat is s:ipouified as in the 

 Reichert-Meissl method, the alcohol is evaporated, and the soap is dissolved in 

 warm water, c-ooled. and treated with a saturated solution of sodium chlorid. 

 To the clear filtrate from this is added hydrochloric acid, which causes a cloudi- 

 ness if cocoanut oil was present in the original sample. The amount of cocoa- 

 nut oil is indicated by the degree of cloudiness. 



The new "'international*' method of tannin analysis {CoUefjiuw. 1907. 

 Xo. 266. pp. 2-}P-2-5-J). — This gives the final recommendations of the international 

 commission appointed at the 1906 conference of the International Association of 

 Leather Trades Chemists to investigate and make comparative tests of various 

 methods of tannin analysis with a view of selecting the most satisfactory. 



An apparatus for the determination of carbon dioxid in carbonates, P. 

 Malherbe {Aiiit. Chi))i. Anal.. 12 {1907). Xo. 7. pp. 26t-263. fi<7s. 3). — An appa- 

 ratus so c-oustructed as to obviate danger of suction of the sulphuric acid into 

 the flask where the carbon dioxid is being disengagetl is described. 



The rate of combustion and pressure developed in a calorimetric bomb, 

 F. G. Benedict and F. P. Fletcher (Jour. Anier. Chctn. Soc. 29 (1907). Xo. -5. 

 pp. 739-757). — The Berthelot-Atwater bomb calorimeter was used. In this 

 study of the effects of different amounts of material and different initial pres- 

 sures upon c-ombustion pressure, the effect of different sized capsules and of 

 loose and compressed material, and related factors. The authors* summary of 

 their results follows: 



" With an initial pressure of 3tK> lbs., the pressures in the calorimetric bomb 

 rarely exceedetl 7<Xt lbs. The greater the quantity of material the greater the 

 pressure, although the pressure is not proportional to the weight of the sub- 

 stance. The greater the initial pressure the greater the maximum pressuiv, 

 although regularitj- and proixtrtionality are not observed. 



