284 Journal of Agricultural Research voi. x, no.6 



in moisture content between the silage from wilted alfalfa and from wilted 

 alfalfa plus water is not as great as would be expected. This is no doubt 

 due to the fact that during the first two weeks' fermentation moisture was 

 lost because the stoppers could not be kept in place. 



Acidity content. — Acidity was determined by the method of extract- 

 ing in 50 per cent alcohol. Investigations made at this laboratory ^ had 

 shown that the method of water extraction did not give as high a per- 

 centage of acidity as the extraction with alcohol. The method used was 

 as follows: Two hundred gm. of carefully mixed silage are passed 

 through the clover cutter and are then placed in a 2-quart Mason jar. 

 For well-mixed alfalfa silage where only total acidity is to be determined 

 half of this amount is sufficient. For mixed silage from a silo or for 

 ordinary com silage it is best to use this size of sample, as there is a suffi- 

 ciency of extract for other determinations, such as volatile acids and 

 nitrogen in amino form. To the silage in the Mason jar are added 500 c. c. 

 of 95 per cent neutral alcohol. The advantage of this procedure is that 

 when the determination can not be finished at once the alcohol will stop 

 bacterial and enzymic action, and the sample can be kept as long as 

 desired. The amount of water present in the sample, as known from the 

 moisture determination, is calculated, and water equivalent to the differ- 

 ence between this and 500 c. c. is added to the jar, making the total 

 quantity of liquid in contact with the silage i ,000 c. c. The jar is then 

 placed on a shaking machine for about six hours, after which the contents 

 are thrown on a 32 cm. folded filter. Triplicate portions of 25 c. c. of this 

 filtrate are titrated with NJ20 sodium hydroxid, using phenolphthalein as 

 the indicator. 



The figures for acidity are probably large. The titration was made on a 

 portion of extract representing 5 gm. This means that the analytical 

 error in computing to percentage is multiplied 20 times. The extract is 

 highly colored, and the end point is not easily read. To compensate for 

 this, all the acidity titrations, as well as the formol titrations, were made 

 by the same person during the last two years of work. This gives the 

 figures greater comparative value. 



The results for acidity titration are expressed as cubic centimeters of 

 Nl2o sodium hydroxid for the extract from 5 gm. of silage, and also as 

 percentages of lactic acid. This gives a higher percentage of acidity 

 than the reality; but as only comparative values are needed here, this 

 method of calculation is sufficiently accurate. Farther on attention 

 will be called to the relative amount of volatile and nonvolatile acids in 

 alfalfa silage made with various supplements. The results for acidity, as 

 well as for moisture, total nitrogen, and amino nitrogen, were averaged 

 for each set of four bottles; but only those of approximately equal 

 quality on the basis of percentage grade were included in this average. 



1 Sw ANSON, C. O., Calvin, J. W., and Hungkrford, Edwin, acidity in silage; method op deter- 

 mination, /n Jour. Axner. Chem. Soc., V. 35, no. 4, p. 476-483. 1913. 



