76 REPORTS ON THE STATE OP SCIENCE. 



In all, sixteen main stations were occupied. The chain of triangu- 

 lation extended from 1° 10' N. to 1° 10' S., and consisted of five figures, 

 namely: a northern complex figure, including the base net; three 

 quadrilaterals; and a southern tetragon, or quadrilateral with an 

 additional centre-point. The width of the chain is about 30 miles; the 

 longest ray was 47 miles. 



Every station was permanently marked with an iron or brass peg, 

 surmounted by a large cairn of stones, in the centre of which was fixed 

 nn iron beacon. The Uganda Government was informed of the position 

 of all stations. 



The work in the field took eleven months, from [March 1908 to 

 February 1909, of which a month and a half was occupied in base 

 measurement. The atmospheric conditions for observing were bad, 

 and delayed the work considerably. 



The actual cost of the British Section was 3,750Z. A good deal of 

 expense was saved by the fact that the majority of the officers and men 

 had been on the Anglo-Congolese Boundary Commission, and the 

 expense of their journeys, camp equipment, &c, were thus saved to 

 the Arc Survey. 



Preliminary computations were carried out in the field as much as 

 possible, and found useful in the subsequent precise calculations. The 

 latter were undertaken on the return to England, and are now practically 

 completed. 



The absolute probable error of the base, taking into account all 

 possible sources of error, was found to be 



± 14.92 mm., or 1 in 1,108,000, 

 a result which is considered very satisfactory. 



The average error of closure of the triangles was 



+ 0".812, 

 and the probable error of an observed angle 



± 0".390. 



The final report, which includes a full description of the work of 

 the survey, the methods adopted, the determination of the errors of 

 the instruments, the measurement of the base, the adjustment of the 

 triangulation, a discussion of the height of the datum point on Lake 

 Albert, and the adjustment of the vertical observations, &c, is almost 

 complete. There remains to be done the section dealing with the 

 geodetic positions of the points. The computation of these positions 

 from the triangulation is necessarily bound up with the results of the 

 latitude observations, as an adjustment has to be made between the two. 

 Unfortunately the latitude results have not yet been received from 

 M. Dehalu ; but as soon as they are, the final computation of the geodetic 

 positions will be made and the report published. 



It may be mentioned that, in addition to the main work of the 

 survey, observations were taken with a view to determining the height 

 of Buwenzori and of the Mufumbiro volcanoes; and a connection was 

 made with the survey of Uganda. 



M. Dehalu also carried out a large number of magnetic observations. 



