242 



Journal of Agricultural Research 



Vol. X, No. s 



TabIv^XIV. — Analysis of relative humidity from hygrograph record for the week of Sep- 

 tember II to i8, igi6. {See fig. 7) 



Day. 



Relative humidity (per cent). 



Noon 

 read- 

 ing. 



7 a. m. 

 read- 

 ing. 



7 P- in- 

 read- 

 ing. 



Mean from 7 

 o'clock read- 

 ings. 



Day. Night. 



Mean obtained 



by planimeter 



from graph. 



Day. Night. 



Monday. . . 

 Tuesday. . . 

 Wednesday . 

 Thursday . . 



Friday 



Saturday 



Sunday . . . . 

 Monday 



17 

 10 



71 

 54 

 45 

 45 



50 



67 



85 



100 



79 

 81 

 80 



82 

 36 

 43 

 73 

 58 

 58 

 70 



Weekly mean from noon readings ... 47 



Weekly mean from 7 a. m. readings . . j | 79 



Weekly mean from 7 p. m. readings . 

 Weekly mean from all 7 o'clock read- 

 ings 



Weekly mean from the graph 



60 



43 

 55 

 79 

 79 

 68 

 76 



66 

 52 

 64 

 86 

 68 

 70 

 75 



70 



80 



69 

 56 



45 



47 



65 



76 

 68 

 78 



95 

 87 

 82 

 87 



BAROMETRIC PRESSURE 



Theoretically variations in atmospheric pressure as shown by the 

 barometer should have an effect upon the rate of evaporation from water 

 surfaces. With the variation of i inch at the Denver laboratory this 

 effect could not be determined or even detected. It is quite probable 

 that within the limits of variation of barometric pressure at any one 

 station the effect of the variation is not material at the temperatures of 

 the water bodies under investigation. 



PART II.— EVAPORATION FROM THE SURFACES OF STREAM-BED 



MATERIALS 



In many places the North Platte River, the South Platte River, the 

 Rio Grande, and other western streams the water of which is used 

 for irrigation have smooth and nearly level beds from ^ to i mile and 

 over in width. Part of the year these may be covered with water; at 

 other times the water table is below the surface of the sand. When they 

 are covered, there is a loss going on from evaporation, but usually these 

 floods occur at a season when evaporation rates are relatively low. Much 

 of the time during the months when evaporation rates from a water 

 surface are high the water table in the rivers is below the top of the sand. 

 That there is a loss from this sand surface has not been questioned, but 

 there have been no figures from which to estimate this loss. 



