EXPERIMENT STATION BULLETINS. 563 



used as a comparator the convenience of having the standards at hand is 

 valuable. The three middle holes may be used for storing compensating 

 blanks. An extra hole at the end holds the comparator clamps when the 

 cover is adjusted. 



EXPERIMENTAL WORK 



The experimental work which has been completed for this paper is in- 

 tended to show briefly the application of the colorimetric hydrogen ion de- 

 termination to the subjects considered. Studies which are under way and 

 projected will take up in detail the various phases of the work. 



MILK PLANT SUPPLY 



A method which could be used to judge the keeping quality of milk as it 

 is delivered to the city milk plant, creamerj^, ice cream plant, and condensery 

 has long been needed. Methods which are in use at the present time are 

 faulty in that they do not give a true index of the keeping quality of the 

 milk, are subject to considerable error, require a trained technician, are 

 expensive of material and labor, and require too long a time before results 

 are available. 



The results taljulated in Table I are designed to show how the score of 

 milks tested by the pH method compares with the bacterial count and with 

 the actual keeping quality as judged by tasting after the milk had been 

 stored for 24 hours at 21 °C. The samples are scored as to pH change as 

 shown in Table III taken from a previous paper. 



A summary of the results in Table I rearranged in Table II shows that 

 the average bacterial count checks up quite closely with the score based 

 upon rate of change in pH. As shown in Table I there are many individual 

 cases in which a high bacterial count does not indicate a sample of milk vnth 

 poor keeping quality, nor does a low bacterial count always indicate good 

 keeping quality. 



In Table I it will be seen that samples with counts of 100,000, 120,000, and 

 185,000 bacteria per cubic centimeter were sweet at the end of 24 hours at 

 room temperature while samples with counts of 11,000, 30,000 and 36,000 

 were sour or curdled. This was undoubtedlj^ due to the fact that the first 

 three samples were contaminated largely with inert or slow acting bacteria 

 while the last three samples contained very active varieties which quickly 

 soured the milk. In the actual practice of grading milk by the bacterial 

 count alone the first three samples might have been condemned while the 

 last three samples with low counts and poor keeping quality would have 

 passed as a very good grade of milk. In this way an injustice is often done 

 when milk is graded upon the bacterial count alone. 



