Heights of the principal Hills of Swaledale, Yorkshire. 121 



minution of the depression of Shunnor Fell must be accounted 

 for, notwithstanding the unvarying value of that of Ingle- 

 borough, by admitting the existence of local refraction. On 

 a calm day, when the temperature reaches its maximum about 

 noon and declines rapidly towards sunset, the thin stratum of 

 air based on an extensive plateau, has a tendency to become 

 rarer about midday than the one immediately above it, but 

 acquires in return a proportionate augmentation of density at 

 nightfall ; — the refraction, during the interval, passing in con- 

 sequence from its least to its greatest value. This has evi- 

 dently been the case in the observation of Shunnor Fell, where 

 the ray, under circumstances of weather peculiarly disposed 

 to develop local refraction, literally grazed for more than a 

 mile the broad and level summit of Great Whernside. In the 

 direction of Ingleborough, the ground, on the contrary, abso- 

 lutely precipitous at first, forms a steep declivity to the very 

 base of the mountain. In confirmation of the explanation pro- 

 posed, it may be added, that the height of Shunnor Fell, cal- 

 culated from the mean of the observed depressions, and with 

 the usual estimate for refraction, comes out within half a foot 

 of the truth. The measurements at Ingleborough, it is pro- 

 per to mention, were not made by the horizontal wire, but 

 by a new and infinitely superior method hereafter to be de- 

 scribed. 



The large levels of the sector had preserved their adjust- 

 ments so nearly unaltered during the survey of Wensleydale, 

 that it was deemed advisable on the present occasion to ascer- 

 tain their reversing points from time to time, leisurely and 

 under favourable circumstances in the valley; rather than, as 

 had hitherto been the plan, from hurried and generally unre- 

 peated experiments with the instrument resting on such sup- 

 ports (sometimes of questionable stability) as the summit of 

 the fell presented. The satisfactory determination of the two 

 points of the scale between which the bubble of the level will 

 come to rest, in two opposite directions of the telescope within 

 its Ys, is a task requiring extreme precaution and address. 

 The support, based on solid ground, must be perfectly firm ; 

 its surface sufficiently ample not to require any part of the 

 instrument to project over it; and without being smooth, 

 which is highly objectionable, it should be selected so nearly 

 a true plane that its contact with the under surface of the stand 

 may extend throughout its length. It is also of importance 

 that the surface should not yield, from the friable texture of 

 the material, to increased pressure. At Reeth, the operation 

 was undertaken with great prospect of success on a pile of se- 

 veral tons of pig lead, but the results, probably from the soft- 



N. S. Vol. 8. No. 44. Aug. 1830. R ness 



