On the Motion of Glaciers. 275 



which it presented for the setting-up of the instrument (a con- 

 dition not everywhere fulfilled), and also because at this spot 

 the glacier formed a tolerably continuous mass, without any 

 longitudinal, transverse, or marginal fissures worth mention- 

 ing ; here, too, was the best prospect of being free from acci- 

 dental anomalies of the motion. 



In setting up the scales we proceeded in the following 

 manner. The stake on which the scales were fixed was sunk 

 into the ice to the depth of half a metre, and a mound of ice 

 30 centims. in height raised around it, in order as much as 

 possible to prevent melting away of the ice at this spot. The 

 mound was covered with earth and stones ; so that the whole 

 resembled one of the well-known earth-mounds. A stake 

 thus planted was perfectly stable and could be employed for 

 the observations for about four days. The reading-telescopes 

 were set up firmly on the bank, and sufficiently sheltered from 

 the action of the sun. The readings were nearly always exe- 

 cuted by both of us, each independently of the other. The 

 scales were graduated in half-centims.; millimetres were esti- 

 mated. The uncertainty of the estimation amounted, with 

 unfavourable illumination or with trembling of the image in 

 consequence of unequal heating of the intervening air, to about 

 0*1 centim.; but it sometimes rose to 0'2 centim.; in normal 

 circumstances, however, the readings of both observers were 

 perfectly concordant. 



In order to convince ourselves that the motions observed on 

 the scales were not motions of the scales themselves, but really 

 indicated the motion of the point in question of the glacier, 

 near one of the signals a second was fixed in the ice, and so 

 close to the first that both appeared in the field of the telescope. 

 With this arrangement the observations showed that both pairs 

 of scales moved in close conformity to one another. 



We select as evidence for our above-expressed assertion the 

 observations of the 3rd and 4th of September, because on those 

 days the motions of the glacier took place under different con- 

 ditions : the 3rd was all bright and cloudless, while the sky 

 during the forenoon of the 4th (the time of the greatest mo- 

 tion) was covered with cumulus clouds and the glacier was 

 only at times exposed to the sunshine. 



The times in the following Table are reckoned from midday 

 to midday. The numbers give the horizontal and vertical 

 motions in centimetres. + denotes a downward motion, — a 

 motion upward (uphill). The readings took place halthourly 

 during the period from 6 a.m. to 6 p.m. 



