4. Observations of Compass-Variations

For finding the pole-star easily when the sky was not very dark or was partially obscured by clouds, and also for obtaining the approximate bearings of time-stars for their subsequent identification, the theodolite was invariably set up in the magnetic meridian by its trough-compass, and the magnetic bearing of the pole-star was read on the horizontal circle after each latitude-observation and the time noted. In this way approximate compass-variation was determined at each camp, with the following results:

Variation of the Compass

°
SollumDec., 19223obs.234W.
SiwaJan., 19231242
JaghbubFeb., 19235225
Near JaloMarch, 19231412
Jalo (El Erg) “ “ 145
Buttafal “ “ 1
El Harrash “ “ 1348
TajApril, 19236332
Arkenu “ “ 2325
Ouenat “ “ 1332
ErdiMay, 19231357
Agah “ “ 140
Enebah “ “ 1421
Bao “ “ 1459
FurawiaJune, 19232432
Um Buru “ “ 2325
Kuttum “ “ 1426
El Fasher “ “ 2251

The method of estimating the compass-variation with the theodolite is, of course, only rough; but the values found are probably correct at most places within half a degree, and they serve to show that there is no likelihood of any serious errors of dead-reckoning owing to local irregularities of the compass-variation. They have accordingly been utilized in reduction of the compass-bearings of the traverse to true bearings over the major portion of the route, where no previous determinations exist, and where in consequence the distribution of the isogonic lines is not known with any exactitude.