transit instrument. During the second year, this box was replaced by one much heavier 

 and more solid a petrol case about 2 feet by 2 feet by 3 feet which was frozen to the 

 ground in exactly the same way. The stand was placed a few feet to the north-west 

 of the hut in order to get as much shelter as possible from the prevailing south-easterly 

 winds. (Plate 3.) 



The fixed mark for the transit instrument consisted of a wooden post 2 inches by 

 3 inches in section, placed upright on the frozen ground and surrounded by a heap of 

 stones and sand cemented together and to the ground by frozen water. The post was 

 perfectly rigid and acted quite satisfactorily during the cold winter months. Through 

 the post, about 3| feet above the ground, was bored a hole f inch in diameter. Behind 

 this hole was nailed to the post a tin box cut away at the back, and with another much 

 smaller hole (-| inch) bored in the tin to correspond approximately with the centre of 

 the hole in the wood. In the tin box was placed a small oil bicycle lamp which shone 

 through the hole and furnished the fixed mark used throughout the first series of observa- 

 tions. Owing to the necessity for sheltering the transit instrument in the lee of the 

 hut, and to the fact that a few feet north of the instrument was the shore of McMurdo 

 Sound, it was found necessary to put the fixed mark to the south of the transit instru- 

 ment. Unfortunately, the furthest distance visible in that direction was only about 

 125 yards away. The fixed mark was therefore set up close to the magnetic hut and 

 at an altitude of 4 -5' from the transit instrument. Owing to the small distance 

 separating the fixed mark from the transit instrument, the light was not properly focussed 

 and appeared as a bright spot of appreciable diameter surrounded by a set of concentric 

 circles of light, of decreasing brightness as the distance from the centre increased. It 

 was, however, quite easy to bisect the circles. Of greater importance was the fact 

 that the standards of the transit instrument allowed a lateral movement of - s inch. 

 This was of importance because of the closeness of the fixed mark, and the difficulty of 

 always keeping the instrument in the same relative position. The angular value of 

 - a - 5 - inch in 125 yards is l"-8, giving a possible error in time from this cause of 0-12 sec.* 



This was a serious matter, but there seemed no way of avoiding the difficulty. The 

 only other instruments that might have served for the observations for rate were the 

 small 4-inch sledging theodolites and a 5-inch altazimuth instrument which had only 

 a single cross wire and in which the telescope was transited by reversing in the wyes. 

 Neither of these was furnished with a striding level. 



As might be expected from the preceding account, the observations with different 

 stars showed very considerable variations, though in general the observations of the 

 passage of a single star over the different wires were in fair agreement. 



A preliminary calculation of the time observations was made in the south, but, 

 since our return to England, Mr. W. E. Curtis, lately of the Imperial College of Science 

 and Technology, South Kensington, has checked the whole series, and it is his compu- 

 tations that are here used. 



Several times during the course of the winter, preparations were made to start 

 * 0-12 sec. for equatorial stars. For high-declination stars this error would be much greater. 



34 



