300 CARNEGIE INSTITUTION OF WASHINGTON. 



gated as thoroughly as the data permitted. Since especial instructions 

 had been issued for Cruise VI for obtaining data regarding the dimen- 

 sions of waves, it was expected that the investigation might explain 

 a large part of the discordance in results heretofore noted. A relation 

 was found between the height of waves and the observed dip-of- 

 horizon, but it will account for only a small part of the discrepancies. 

 Dr. Slocum confirmed our earlier conclusions to the effect that the 

 instruments at present in use at sea, while sufficient for navigational 

 purposes, do not yield results of the precision required for investigating 

 changes in the coefficient of refraction. 



Dynamic deviations. — Automatic records of the rolling and pitching 

 are being received from time to time with the usual observations 

 from the Carnegie. The records were taken and have been preserved 

 under difficult conditions of heat and moisture, but nevertheless they 

 are very satisfactory. It is regretted that more were not obtainable, 

 but power for running the gyrostat was not available daily, as had 

 been hoped. With the assistance of the Magnetic Survey Division, 

 spools and winding-machine have been provided to facilitate readings, 

 scalings, etc. An examination of the records shows three important 

 points in connection with dynamic deviations, viz : 



(1) The period of rolling is not constant, but varies during the observation 

 or experiment between comparatively wide limits. For example, the time of 

 a double roll during declination observations on June 10, 1920, varies between 

 6 seconds and 12 seconds. 



(2) The motions of rolling and pitcliing are so complex that in curves 

 extending over intervals of 50 to 60 minutes there are comparatively few and 

 sometimes no exact repetitions. 



(3) Any estimation of the average amplitude of roll made without consider- 

 ing the frequency might be very misleading. 



Investigation was begun by Dr. Slocum to examine how far these 

 conditions would modify the formula for simple harmonic motion 

 used, for example, by Bidlingmaier (Erdmagnetische See-Beobach- 

 tungen, I Toil, pp. 271-292). After examining the work already done, 

 Dr. Slocum deduced formulae for the deviations of a compass, subject 

 to the condition that the point of support of the needle or card is 

 moving in a plane with simple harmonic motion. He subsequently 

 investigated the same problem for damped and forced oscillations. 

 By confining the motion to a plane, it is expected that tilting deviations 

 will be eliminated to a large extent, if not wholly. 



A preliminary motion table has been planned with Mr. Fleming's 

 help and constructed in the workshop, for experimental purposes in 

 connection with Dr. Slocum's investigation. This table does not take 

 the place of the more elaborate one proposed for reproducing the actual 

 motion of the Carnegie from automatic records. Besides experimental 

 purposes, the table would be useful in balancing the special compass 

 variometers in case of need. 



