1360 Wisconsin Academy of Sciences, Arts, and Letters. 
As shown in the above tables (XI. and XII.) all of the seven¬ 
teen samples of raw milk examined curdled in less than 18 hours 
and in nearly all cases the curd formed was of the lactic acid 
type. In the pasteurized milks, two out of twelve, remained fluid 
for 18 hours but otherwise the character of the curds was not 
much different from that of the raw milks. In the case of the in¬ 
spected milks, nineteen samples, eight remained fluid for eighteen 
hours or over. The character of the curd varied more than in the 
previously considered classes with more P. and G. than J. curds. 
All of the certified milks remained fluid for more than 18 hours 
and the P. and G. curds were considerably in excess of the J. 
curds. 
IV. EXPERIMENTAL PASTEURIZATION. 
From the results which have just been discussed, especially in 
testing for B. coli in pasteurized milk, it seemed necessary to un¬ 
dertake sonte pasteurization experiments where the conditions 
could be accurately controlled. For this work a specially con¬ 
structed water-bath has been used. This consists of a bath 
containing about 10 liters of water, fitted with a Roux thermo- 
regulator, and a standardized thermometer, and also provided 
with a mechanical stirrer in the shape of a fan or paddle wheel 
run by a small electric motor. The milk was pasteurized either 
in test-tubes or in bottles. When the test-tubes were used, 
some of them were stoppered with cotton plugs, but in these 
cases duplicates were always made in test-tubes closed with rub¬ 
ber stoppers. The bottles were either pint milk bottles, closed 
with the ordinary paper cap, eight ounce bottles fitted with a per¬ 
forated lubber cork containing a thermometer, or oil sample- 
bottles which were fitted with a screw cap. In the test-tube ex¬ 
periments only a few centimeters of milk were placed in the bot¬ 
tom of the tube, and care was taken to immerse the test-tube in 
the water-bath as low as possible. Some of the bottles were im¬ 
mersed entirely; in other cases only up to the caps, but not over 
them. 
The results obtained are shown in the following table (XIII). 
