106 BURTON EDWARD LIVINGSTON 



may be more thorough. A tooth-brush serves very well for 

 the latter but the softer cups are to be treated more carefully. 



In moist situations moulds and algae frequently appear on 

 the cups after a time. These may be removed by washing with 

 strong alcohol and a sponge or cloth. As soon as a colony is 

 detected, it should thus be destroyed, but great care should be 

 exercised to prevent the alcohol touching the shellacked surface, 

 shellac being readily soluble in alcohol ; with cups having a glazed 

 base no danger need be feared in this connection. 



Since accidents of various kinds are not to be entirely avoided, 

 it is well, where possible, to operate the instruments in duplicate 

 or in triplicate, the cups being so placed as to have as nearly the 

 same exposure as possible." Ordinarily the average of the two 

 or three corrected readings is taken as the measure for the 

 period, but a reading is discarded when it is obvious that some 

 misfortune has befallen the corresponding cup. 



Restandardization (see the next section) may be avoided by 

 using two instruments exposed about a meter apart and replac- 

 ing one of them at a time with an unused cup of known co- 

 efficient. The first one or two periods of operation after such 

 replacement (if not too long) may be considered as restandardi- 

 zation periods for the cup longest in operation. 



Restandardization. Even with all possible precautions, the 

 coefficients of correction are apt to become altered through 

 changes in the effective evaporating surface of the cup. It is 

 therefore requisite that each cup be frequently restandardized. 

 By this means changes in the magnitude of the coefficient may 

 be detected and allowed for. It has usually been found ade- 

 quate to restandardize each cup after from two to four weeks 

 of operation in the open. Since it requires some time for 

 this restandardization, the record is kept continuous by re- 

 placing the cup which is to be restandardized by another stand- 

 ardized cup. During the following period of from two to four 

 weeks the first cup is restandardized and is ready to return to 

 operation when the second comes away. 



It is to be borne in mind that the purpose of the restandardi- 

 zation is to determine the coefficient of correction of the cup at 



