876 EXPERIMENT STATION RECORD. [Vol. 35 



when the butter was stored at a temperature of 0°, and this decrease was still 

 more marked when it was allowed to remain at a temperature of 32°. This 

 butter, stored at 0°, scored 93 when made, 88 at three months, and 84 at six 

 months, and had a pronounced unclean flavor at the end of this time. 



From these analyses it is concluded that the decomposition caused by cream 

 acidity progresses at a temperature of 0° in a package of butter and can be 

 measured by an analysis of the gas extracted therefrom. 



A series of experiments was then undertaken to determine whether this 

 measurable decomposition occurs in the fat of the butter itself, in the butter- 

 milk, or in both. Fat from butter made from pasteurized cream was so pre- 

 pared as to exclude, as far as possible, by melting, filtering, and washing, all 

 ingredients of the butter other than fat. The butter fat so prepared was found 

 to contain 0.05 per cent of protein (NX 6.38). Samples of this fat kept in 

 storage at 0° showed no physical alteration of any kind after six months or 

 even after 1.5 years. Analyses showed that no chemical changes had occurred 

 in fat kept in storage under these conditions for months, and an analysis 

 of the extracted air showed that no appreciable oxidation had occurred during 

 a storage interval of five months. A sample of this purified butter fat was 

 exposed to the action of a large quantity of air and stored for about three and 

 a half months at 32°. Under these conditions very little oxidation took place. 



To determine whether oxidation takes place in one or more of the nonfatty 

 constituents, butter was made from pasteurized cream ripened with a pure 

 culture. A part of this butter was washed until the wash water was barely 

 clear; another part was given additional copious washing in four changes of 

 water ; and with the third part the buttermilk was drawn off and the butter 

 allowed to remain unwashed. Each of these samples was worked on a table 

 worker to the extent of 40 revolutions. During the storage period the exces- 

 sively washed butter and the normally washed butter were found to have prac- 

 tically the same keeping qualities as shown by scoring, and the chemical con- 

 stants of the fat showed that there was very little or no chemical change. 

 However, analyses of the confined air in these samples showed that the total 

 amount of oxygen originally present underwent a markedly progressive decrease 

 during storage at 0°. The sample of unwashed butter scored 92 at one month 

 and 86 at six months, at which time it was stale, fishy, and sour. No note- 

 worthy chemical change had occurred in the fat at six months, but striking 

 changes had occurred in the composition of the confined air. The maximum 

 amount of carbon dioxid (31.67 per cent) was found after a storage of three 

 months, at about which time the characteristic " off-flavor " became distinctly 

 noticeable. Very little oxygen was found in the sample after two months' 

 storage. With another sample of unwashed butter 19.89 per cent of carbon 

 dioxid was found in the butter 1.5 hours after manufacturing. This amount 

 increased to 22.91 per cent at the end of two weeks when the butter was held 

 in the dark at room temperature. Under these conditions the oxygen content 

 decreased from 20.42 per cent at 1.5 hours to 10.61 per cent at the end of two 

 weeks. 



To further test the effect of air on the nonfatty ingredients of butter, tubes 

 filled with broken pumice were impregnated with buttermilk from butter made 

 from pasteurized cream acidified to 1 per cent with lactic acid before churning. 

 This acid buttermilk was held at 32°. When the buttermilk was 4.5 days old 

 it was found to contain 17.67 per cent of oxygen. At 26 days the oxygen 

 content had decreased to zero. The carbon dioxid content of the buttermilk 

 rose from 2.37 per cent at 4.5 days to 34.37 per cent at 26 days and decreased 

 to 31.76 per cent at 62 days. In a control experiment with a 1 per cent solu- 

 tion of lactic acid the oxygen content remained practically constant and at 



