114 EXPERIMENT STATION RECORD. 



ducted as follows: "The sample of butter, taken with a trier or otherwise, is 

 first warmed to about 100° F. and thoroughly stirred to insure the mass being 

 uniform. About 10 cc. of the sample is placed in a sedimentation tube and 

 whirled in a [hand] centrifuge for a few seconds. The sample should be suffi- 

 ciently liquid in order to insure a good reading after being whirled in the 

 centrifuge. After reading the amount of the sample on the tube scale, about 

 5 cc. of gasoline is added and the tube carefully inverted 2 or 3 times, holding 

 the thumb or finger over the top of the tube. Let the solution of fat and 

 gasoline drain a few^ seconds before removing the finger. Place the tube in the 

 centrifuge and whirl again for 15 or 20 seconds. 



" The gasoline dissolves the fat, forming a clear layer on the top. The non- 

 fats, that is the water, salt, and curd, being immiscible with the gasoline and 

 also heavier, form the lower layer. The second whirling drives the nonfats to 

 the lower end of the tube almost completely, at the same time forming a sharp 

 line of division between the 2 layers. The amount of nonfats is then carefully 

 read on the tube scale. . . . 



" Care should be taken that this test is made at a fairly uniform temperature 

 in order to eliminate as far as possible the changes in relative volumes due to 

 variations in temperature. In case the sample when first placed in the sedi- 

 mentation tube is not sufficiently liquid to insure a good reading on being 

 whirled, it may be warmed by placing it in water or in an oven for a few 

 minutes at a temperature not over 110°." 



The method yields slightly higher results than the official method. 



A simplification of the method for determining the Beichert-Meissl and 

 Polenske numbers, A. Goske {Ztschr. Untersuch. Nahr, u. Genussmtl., 24 

 {1912), No. 4, pp. 274-276, fig. 1). — ^The apparatus consists of a boiling flask, a 

 distilling tube (1 bulb), a Liebig condenser with a flared upper end, a funnel, 

 holding a piece of filter paper, attached with a cork to the lower end of the 

 Liebig condenser, and a 110 cc. receiving flask divided into 10 cc. divisions. 

 The advantages claimed for this apparatus are that (1) filtration after dis- 

 tillation is unnecessary; (2) titration is done directly in the 110 cc. obtained; 

 (3) rinsing of the apparatus is eliminated, and in this way the losses observed 

 in the usual procedure are avoided; and (4) no special preparation of the 

 filter is necessary for the second determination. 



Estimation of essential oil in mustard, D. Raquet {Ann. Chim. Analyt., 

 17 {1912), yo. 5, pp. 174-178; ahs. in Analyst, 37 {1912), No. 436, p. 309).— It 

 is pointed out that mixing mustard with water previous to distillation and 

 allowing it to stand for some time is often followed by inconcordant results. 

 Micro-organisms develop and exert their activity under these conditions, which 

 results in the loss of oil. " If, however, dilute alcohol be used in place of 

 water, the digestion may be allowed to proceed for even 24 hours without loss 

 of essential oil. Having regard to these conditions, the following method is 

 recommended for the estimation of the oil : 



" Five gm. of the mustard flour is mixed in a 250 cc. flask with 100 cc. of 

 water and 20 cc. of 90 per cent alcohol ; the flask is now closed and set aside 

 for 6 hours, or heated to a temperature of 35° C. for 1 hour. The contents are 

 then distilled, and 50 cc. of the distillate is collected in a 100 cc. flask in which 

 10 cc. of ammonia have been placed previously; 20 cc. of tenth-normal silver 

 nitrate solution is now added, the distillation is continued until the 100 cc. 

 flask is filled nearly to the mark, and after the flask has been closed with a 

 stopper carrying a long glass tube, the contents are heated to 85° for 1 hour. 

 When cold, the mixture is diluted to 100 cc, filtered, and the excess of silver 

 is titrated in 50 cc. of the filtrate by means of tenth-normal thiocyanate solu- 

 tion after the addition of nitric acid. The number of cubic centimeters of 



