In setting a straight row of stakes across the glacier, our mode of proceeding was in all cases this:--The theodolite was placed on the mountain-side flanking the glacier, quite clear of the ice; and having determined the direction of a line perpendicular to the axis of the glacier, a well-defined object was sought at the opposite side of the valley as close as possible to this direction; the object being, in some cases, the sharp edge of a cliff; in others, a projecting corner of rock; and, in others, a well-defined mark on the face of the rock. This object and those around it were carefully sketched, so that on returning to the place it could be instantly recognized. On commencing a line the point of intersection of the two spiders' threads within the telescope was first fixed accurately upon the point thus chosen, and an assistant carrying a straight baton was sent upon the ice. By rough signalling he first stood near the place where the first stake was to be driven in; and the object end of the telescope was then lowered until he came within the field of view. He held his staff upright upon the ice, and, in obedience to signals, moved upwards or downwards until the point of intersection of the spiders-threads exactly hit the bottom of the baton; a concerted signal was then made, the ice was pierced with an auger to a depth of about sixteen inches, and a stake about two feet long was firmly driven into it. The assistant then advanced for some distance across the glacier; the end of the telescope was now gently raised until he and his upright staff again appeared in the field of view. He then moved as before until the bottom of his staff was struck by the point of intersection, and here a second stake was fixed in the ice. In this way the process was continued until the line of stakes was completed. Before quitting the station, a plummet was suspended from a hook directly underneath the centre of the theodolite, and the place where the point touched the ground was distinctly marked. To measure the motion of the line of stakes, we returned to the place a day or two afterwards, and by means of the plummet were able to make the theodolite occupy the exact position which it occupied when the line was set out. The telescope being directed upon the point at the opposite side of the valley, and gradually lowered, it was found that no single stake along the line preserved its first position: they had all shifted downwards. The assistant was sent to the first stake; the point which it had first occupied was again determined, and its present distance from that point accurately measured. The same thing was done in the case of each stake, and thus the displacement of the whole row of stakes was ascertained.[A] The time at which the stake was fixed, and at which its displacement was measured, being carefully noted, a simple calculation determined _the daily motion_ of the stake. [Sidenote: THE FIRST LINE.]