Appendix B. 



On li'th May, at Entelihe (noon at Mombasa), we obtained, by the above- 

 mentioned telegraphic comparison, as correction on the mean Greenwich time : 



Ki = + 3" 16" or- 

 Ko = + 3 25 52-0 

 Ivj = + 3 2 10 -5 



From these elements we get the following diurnal corrections for the three 



chronometers : 



Bi = - 2^-762 



B. = + 2 ■ 215 



B., = + 3 • -137 



The elements respecting the chronometer No. -t have been omitted because 

 on 7th May, when the party reached Entebbe, that chronometer was stolen. 



The start for Euwenzori was made at Entebbe, capital of the I'^ganda 

 Protectorate, and on the march some astronomical observations were made in 

 order to fix the position of some points which were generallj' those of encamp- 

 ments. Obviously it was not a ease in which too much reliance could be placed 

 on the Greenwich time, as indicated by the chronometers during the period of 

 one month of rough travelling, that being about the time occupied in reaching 

 Bujongolo, last point where were obtained astronomical observations, and where 

 were begiui the topographic operations for the survey of the Ruwenzori 

 highlands. In order, however, to secure the greater or less efficiency of the 

 chronometric observations, it was found expedient to observe, with the 

 determination of the local time at Bujongolo, the emersion of B A C 81 from 

 the lunar disk, with a view to calculating the hour of Greenwich time at the 

 moment when the phenomenon was observed. 



During the march the chronometers were carried on his person l)y H.R.H., 

 who kept them carefully wrapped up. This expedient should have reduced to 

 a minimum the influence of the changes of temperature, if, during the hours of 

 rest, when being replaced in their own boxes, they had not had to feel the 

 effects of the temperature inside the tent. Such effects, though little different 

 from those of the atmosphere, always differed greatly from those due to contact 

 with the human l:)ody. Still, when we consider that the period of repose was 

 daily repeated for aJjout the same length of time, it may he inferred that the 

 daily recurring correction cannot have been affected by .serious error due to 

 this cause. 



In any case it is to be regretted that of the three chronometers one alone 

 displayed a sufficiently regular action, maintaining a fairly slight daily 

 correction. This may easily be seen from the record of the chronometers. 



310 



