356 CARNEGIE INSTITUTION OF WASHINGTON. 



The investigations of the behavior of the gravitation needle in a partial 

 vacuum were pursued through the summer of 1922 and a complete set of night 

 observations was recorded. In addition to the static deflections of the needle, 

 periods and logarithmic decrements were also tabulated, all under a variety of 

 experimental conditions. All are fundamentally subject to discrepancies, 

 owing to the uniform presence of radiant forces. The endeavor to disen- 

 tangle these complications was but partially successful and led to no trust- 

 worthy results. The apparatus has since been taken apart and the period in 

 question determined indirectly. Using the night observations of 1922, which 

 as a whole were remarkably uniform, the final outcome was nevertheless 

 disappointing. It does not appear that, under ordinary reasonable laboratory 

 conditions in the summer and a quartz fiber apparatus, the gravitation con- 

 stant can be obtained with a degree of assurance exceeding 1 per cent of 

 uncertainty. 



Hayford, John F., Northwestern University, Evanston, Illinois. Investiga- 

 tion of the laws of evaporation and stream- flow. (For previous reports see 

 Year Books Nos. 12-16, 19-21.) 



One year ago this investigation was concentrated on the study of evapora- 

 tion from the Great Lakes, using observations on Lake Superior and on Lake 

 Michigan-Huron. For Lake Superior one least-square solution (involving a 

 single month of observations on Lake Superior) had been completed and four 

 such solutions for Lake Michigan-Huron, each involving either one or six 

 months of observations of the elevation of the lake-surface. The evaporation 

 investigation rested directly upon the investigation of effects of winds and of 

 barometric pressures on the Great Lakes which had been completed during 

 the preceding year. 



During the year ending September 5, 1923, five more least-square solutions 

 for determining the evaporation from Lake Superior were completed. The 

 last four of these five were each based upon 6 months of observations of the 

 elevation of the water-surface. At the close of the year a solution based upon 

 28 months of observations was in progress. 



During the year four more least-square solutions for determining the 

 evaporation from Lake Michigan-Huron were completed. Three of these 

 were based upon 6 months of observations of the elevation of the water- 

 surface of Lake Michigan-Huron, and the last one was based upon 28 months 

 of observations, the warmer months of 5 years. 



The investigation has now progressed sufficiently to make it certain that 

 when the evidence from 28 months of observations on each of the two lakes, 

 Superior and Michigan-Huron, has been fully analyzed by the methods now 

 in use, the constants in the formula expressing the evaporation for each day 

 from these lakes in terms of the temperature, vapor pressure, and wind 

 velocity, as observed by the Weather Bureau, will have been determined with 

 a fair degree of accuracy. It will then be possible to compute, with reasonable 

 confidence, the evaporation on any day from any such water-surface anywhere 

 in the world in terms of the meteorological elements observed in the vicinity of 

 that lake, reservoir, river, gulf, or sea. It is reasonably certain that the 

 analysis of the evidence from 28 months of observations on each of these two 

 lakes will be completed during the year 1924. 



