TRANSACTIONS OF SECTION A, 417 
Table II. gives the values obtained at different heights at hourly in- 
tervals. The results itidicate a very considerable diurnal variation of 
temperature even at a height of 10 kin., but the irregularities are too great 
to allow very accurate deductions to be drawn. ‘The fact that the variation 
is similar at all heights up to 10 km., and that solar radiation has no 
appreciable effect on the highly polished instrument at low altitudes, seems 
to point to a real variation, as shown by the curves. On the other hand, 
the increase of the amplitude of the variation with increasing height appears 
to indicate that it is due to insolation. 
During the ascents the region was under the influence of an anticyclonic 
system of moderate intensity, whose centre was situated N.W. of Ireland. 
The pressure distribution changed but slightly during the period considered. 
4. A Balloon Spectrograph. By Professor W. J. Humpnreys. 
5. The Effect of Atmospheric Pressure upon the Harth’s Surface. 
By F. Navisr Denison, F'.R.Met.Soc. 
Since the year 1899 the author has taken up the study of slow-period 
movements of the Milne horizontal pendulum, which has been in constant 
operation here from 1899 to the present time, 
In order to make a thorough investigation of this phenomenon he has 
measured the photographic records from this instrument with a millimetre 
scale, and has noted the amounts and times of occurrence of all changes, 
including the diurnal and longer-period deflections. 
These observations have been entered into a specially designed register, 
and, as they are often of sufficient amplitude to necessitate the resetting of 
the boom by altering the levelling adjustment, they have been corrected to 
form a continuous record. 
By studying the earlier observations with the Victoria daily weather 
charts of the Pacific slope the author became convinced that most of these 
movements were due to meteorological causes. 
A daily curve of the pendulum’s movements for 1899 was plotted, and 
formed the material for a paper, entitled ‘The Seismograph as a Sensitive 
Barometer.’ This was read before the Royal Meteorological Society in 
June 1901. Later in the same year the author personally presented at the 
Glasgow Meeting of the British Association an illustrated paper upon the 
same subject derived from the plotted curves for 1899 and 1900 in con- 
junction with the weather charts of the Pacific slope. 
Acting upon the advice of Sir George Darwin, to obtain more data 
bearing upon this subject, and to install another horizontal pendulum to 
swing N-S, in order to study this direction of tilting in conjunction with 
the ‘Milne’ E-W pendulum, the author has succeeded in keeping a con 
stant record of the K-W movements, and in January 1907 he personally 
constructed a simple form of horizontal pendulum. This is set up on the 
solid rock in the basement of the Victoria Post Office; it swings N-S, and 
is about 500 feet from the E-W instrument. 
The accompanying diagram, which was part of a paper published by 
the Royal Astronomical Society of Canada in 1908, shows the daily move- 
ment of both pendulums, and the mean daily temperature from February, 
1907, to January 31, 1908. 
The upper curve represents the movements of the E-W pendulum, 
whose period of vibration is fifteen seconds and its angular value 0°76. The 
intermediate curve shows the N-S movement, which, having a ten-minute 
period of vibration, is not so sensitive as the former instrument. The lower 
curve gives the mean daily temperature. 
There is a remarkable correspondence between the two pendulum curves— 
that is, when the E-W pendulum swings towards the west the N-S one 
1909. EB 
