ATMOSPHERIC REFRACTION AT LOW TEMPERATURES. 691 



The difference in the times corrected for the personal equation 

 of the observer by taking repeated observations will be the time 

 Sound takes to travel the distance. 



On January 28, 1870, the velocity of sound at a temperature 

 of —21-2° C. or —6-2° F. was 321-58 metres, and on February 

 20, at a temperature of —33-9° C. or —29^ F., the velocity of 

 sound was 317*52 metres per second. 



Assuming the ordinary law for the reduction of the observa- 

 tions to 0° C. to be true, the mean of the two results at 0° C. or 

 32° F. is 337 metres or 1,105 feet per second. 



YII.— TERRESTEIAL MAGNETISM. 



1. Observations in Davis Strait and Baffin's Bay, by Sir 

 James Ross and Captain Sabine. Phil. Trans. 1819, and 

 Voyage of Sir John Ross in 1818. 



The observations on the dip and magnetic intensity were made 

 by Sir James Ross and Captain Sabine at Waygate Islands, and 

 at several points on ice in Baffin's Bay, by observing the time of 

 100 vibrations of the dipping-needle both in and perpendicular to 

 the magnetic meridian. There was no opportunity for using 

 instruments which required fixing during the voyage. 



The dipping needle used was one by Nairne and Blunt, and the 

 balance of the needle was truly adjusted by screws on the cross- 

 wires attached to its axis, so that no alteration took place in the 

 indication of the dip on reversing the poles. 



The intensity of the magnetic force was determined with the 

 dipping needle, the needle being drawn to a horizontal position by 

 a magnet and then allowed to oscillate freely. Observations of the 

 time of 100 vibrations, and the extent of arc traversed at every 

 tenth vibration, were made both in and perpendicular to the mag- 

 netic meridian. 



The observations in the magnetic meridian, as well as those for 

 the dip, were repeated with the face of the instrument towards the 

 east and towards the west. 



These observations were compared with observations made with 

 the same instruments at Regent's Park in April 1818 by Captain 

 Kater, and in March 1819 at the same spot and also at Brassa 

 Island, Shetland, by Sabine. 



The magnetic declination was determined by means of Captain 

 Kater's azimuth compasses, described in " Instructions for the use 

 *' of instruments furnished to the Northern Expeditions." The 

 compass was placed on a copper-fastened stool, and carefully levelled 

 by means of a spirit level. Each altitude and azimuth is a mean 

 of several observations, the compass being removed and levelled 

 afresh between every one, thus making each faithfully distinct. 



The latitudes and longitudes are deduced by the ship's log 

 from the nearest observed. " The influence of the ship's iron on 

 " her compasses, increasing as the directive power of magnetism 



X X 2 



