SUMMER ASCENTS. 



283 



On the morning of December 25th, 1911, it was obvious from the smoke of Erebus that 

 the lower atmosphere was remarkably calm, and therefore an attempt was made to "et a 

 very high ascent. A fuse burning for over 40 minutes was used, and silver paper was 

 attached to the instrument to make it more visible when far away. The ascent has 

 already been described (page 217) and it was so successful that the instrument was 

 recovered after it had been nearly 7,000 metres high. The record revealed four regions in the 

 atmosphere : 



(a) From th's ground to 2,000 metres the gradient was moderately high with a mean 



value of -71 C. per 100 metres. 



(b) From 2,000 to 3,000 metres the gradient was irregular, indicating various layers. 



In this region the gradient was small, being on the average -SO^C. per 100 

 metres, but from 2,-500 to 3,000 metres only SS^C. 



(c) From 3,000 to 6,000 metres the gradient was again higher, being on the avera"-e 



70°C. per 100 metres. 



(d) Above 6,000 metres the gradient fell to ■Si'C. per 100 metres; the ascent, however, 



ceased at 6,743 metres. 



In (a), (b) and (c) we recognise the three layers which have been such a marked feature 

 in all the previous ascents in the summer. The decrease of the gradient at 6,000 metres is 

 significant, but the ascent was not sufficiently high to make sure that this was caused by 

 an approach to the stratosphere. It is exceedingly disappointing that this ascent did not 

 continue for another 1,000 metres higher, as it proved to be the last one which we were 

 able to make. The mean giadient from sea-level to 6,750 metres shown by this ascent was 

 •64" C. per 100 metres. 



Combined Summer Results. 



Ascents were niade on six days during the summer months of November and December. 

 It has been shown that in each one of these ascents a region was met with in which the 

 gradient was smaller than in the regions above and below. 



The following table gives the approximate height of this region : — 



Table 147. 

 Approximate height of the beginning and end of the region of small gradient. 



