All detail visible along lines of march from camp to camp was recorded on plane-table sheets on a scale of 1:100,000. The usual process was as follows: The plane-table sheet was first provided with a graticule at 10′ intervals of latitude and longitude, and all the triangulation points previously fixed within the area covered by the sheet were marked in their computed places. Stations were chosen along the route at an average distance apart of two or three kilometres, the most commanding hills being selected, and the positions of these were found by plane-table re-section from three or more triangulation points. The compass, being frequently disturbed by magnetic rocks,[76] was only used to get a first approximation to the true orientation of the table. The plane-table station having been fixed on the map, tacheometric readings were taken to all conspicuous points easy of access within a radius of about two kilometres, and plotted at their measured distances along the directions given by the alidade. In the telemetric measurements a 5-inch tacheometer was used side by side with the plane-table, and two staff-men were employed. As the scale of the map was small, the sights were much longer than is usual in tacheometry, and the maximum distance of 800 metres directly readable by the four-metre staves employed was generally exceeded. For the long distance readings, where the distance between two cross-wires subtended more than the length of the staff, I devised the following process. Bringing the centre wire to the base of the staff, a reading of the vertical circle was taken; next, by the tangent-screw, the wire was brought to the top of the staff, and a second reading of the vertical circle was taken, the difference giving the angle subtended by the four-metre staff. It is clear that the distance is as many times greater than 800 metres, as the angle subtended is less than 17′ of arc, and the distance is thus found by simple proportion.