CO:\[PAKISOX OF RESULTS WITH WATKU OVKN, 



21 



mately 8 inches) hig^li and is graduated in fifths, the one with a read- 

 ing capacity of '20 cubic centimeters and the other with a read- 

 ing capacity of 2;") cubic centimeters. For samples of \ erv wet grain 

 hirger measuring cylinders will be necessary. With the graduations 

 in fifths it is very easy to make the reading in tenths of a per cent, 

 which is a sufficiently close percentage for all conunercial grading 

 of grain, as samples taken on ditl'erent days will show a much witler 

 variation. The graduations are in cubic centimeters, so that when 

 100 grams of grain are used for the test the percentage of water 

 can be seen at once, 1 cubic centimeter of water representing 1 per 

 cent. The cylinders should be of a uniform height, so that they can 

 be used indiscriminately beneath any of the condenser tube<. 



In expelling the water from the corn 

 a small quantity of oil (less than one- 

 half a cubic centimeter) is carried over 

 into the graduated cylinders, Avhich 

 prevents them from drying rapidly 

 after the readings have been made and 

 the contents emptied at the close of 

 the test. However, preparatory to 

 their being used again the cylinders 

 must be cleaned and dried, which can 

 best be done with a test-tube cleaner 

 having a small piece of sponge 

 attached to the end. 



-20 



4-16 

 -=■15 



4 13 

 4-12 

 -r II 

 4-10 

 4-9 



4-8 



4- 7 

 4-6 

 ^5 

 4-4 

 -^3 

 ^ 2 

 -E- I 

 -r / 



t25 

 424 

 -5-23 

 4 22 



4 2' 

 -f 20 

 -^19 



4 '8 

 -f-17 

 4 16 

 4 15 

 -E-I4 

 413 

 4 12 



4ll 

 410 



49 



4-8 

 4 7 



4^ 



4-5 

 -f 4 



43 

 4-2 



COMPARISON OF RESULTS WITH DE- 

 TERMINATIONS MADE IN A WATER 

 OVEN. 



The method and the apparatus for 

 making moisture determinations of — < 

 corn as described in the foregoing fig. i2.-Graducued cyiindtr-. iur mcusur- 



,. ,, 1 • /Ti c ing the water expelled from the grain. 



pages is so radically diiierent from 



ihat commonly used in chemical laboratories that a comparison of 



results seems advisable. 



Duplicate tests were made according to the rapid method and 

 likewise in a water oven, the average percentage of moisture obtained 

 in each case being shown in Table I. 



9» 



