214 The Plant World. 



were made to derive probable coefficients for the different 

 weeks intervening between calibrations, it being borne in mind 

 that where two cups are in operation side by side, their weekly 

 readings constitute a continual series of measurements of the 

 fluctuations in their coefficients with reference to each other. 

 After the derivation of the coefficients, all readings were corrected 

 and the resulting pair of records (where two cups had operated 

 simultaneously at the same place) were averaged to give the 

 corrected rate for each week and station. 



Thirty-eight co-operators in the United States and Canada 

 were provided with instruments. Accidents and interruptions 

 of various sorts occurred, so that the final records are not nearly 

 all complete, but an excellent series of weekly rates was obtained, 

 thanks to the careful foresight and persistence of Dr. Shreve 

 and to the general efficiency of the co-operators at the several 

 stations. There was consideiable variation in the date of ori- 

 ginal installation at the various stations, and some in the date 

 of discontinuance. For the majority of the stations, the avail- 

 able series of readings extends over a period of fifteen weeks, 

 from May 25 to Sept. 7, 1908. In order to indicate the general 

 change in evaporation intensity as the season advances, and not 

 to complicate matters by too much detail, the corrected weekly 

 rates have been averaged in three groups of five weeks each, the 

 averages thus obtained roughly corresponding to the three sum- 

 mer months. The three five-weeks averages have again been 

 averaged to obtain the mean weeklv rate throughout the season. 

 These averages, together with the names of the various stations 

 and of the co-operators, are given in Table I. The superscript 

 figures after certain averages denote the number of weeks, if 

 not five, and the number of five-weeks periods, if not three, from 

 which the average is derived. 



Since measurements of evaporation have heretofore usu- 

 ally been made in terms of linear xmits of depth, the averages 

 from the porous cup atmometers have been reduced to centimeters 

 and inches, and these are also given in the table. In this con- 

 nection it cannot be too strongly emphasized that no unit of 

 evaporation is of any value unless the surface from which evap- 

 oration occurs is accurately defined, (see Plant World 13: 115 

 1910). In making the reductions here presented, our standard 



