I20 Report S.A.A. Advancement of Science. 



used for the purpose of reducing the coast observations, but with 

 how much success is not known. From observations made during 

 the six years 1841-6 he determined a series of corrections to be 

 appUed to observations made at the Royal Observatory at any hour 

 to obtain the true mean. These have been published in some of 

 the earlier Annual Reports of the Meteorological Commission. He 

 further deduced the formula to the second harmonic term : — 



d = 40-58 sin (i5h + 580- •43')+i°-3o sin (3oh + 62°- -29'), 

 where d is the deviation from the mean temperature, and h the number 

 of hours from Cape mean noon. This is of historic interest, as 

 being the only result ever evolved by a Cape Observatory official 

 from the mountainous piles of routine meteorological observations 

 accumulated there. 



No suitable material exists for the reduction of cloud, although 

 it is not improbable that the observations at VIII. may 

 give a very fair approximation to the mean. And, finally, it is 

 opportune to mention, though the matter is somewhat outside the 

 limits of this paper, that observations of wind-direction at VIII. 

 may only be used in comparing one month with another ; they give 

 otherwise no information of any value. There is a very strong 

 diurnal variation in the surface wind-movement over the table-land, 

 and pretty well every item of infallible weather-wisdom, involving a 

 wind factor, in circulation, is fundamentally unsound through not 

 taking it into account. 



In Table i will be found the annual values of Pressure, Tem- 

 perature, Dew-point, and Humidity, derived from observations made 

 hourly during the five years 1898- 1902. The dew-points and 

 humidity-ratios are determined by means of the Greenwich factors, 

 and are therefore liable, particularly in the dry hours about mid-day, 

 to whatever inaccuracy Glaisher may be supposed to have introduced. 

 They are also in error to a certain extent, because thev are derived 

 from the indications of a wet bulb in ordinar)- air, which is known 

 to read higher on the whole than when it is placed in a forced 

 draught. It is important to observe that the dew-points and 

 humidity-ratios are computed for each observation, and not, as is 

 sometimes done, by applying the factors to the monthly means of 

 dry and wet bulbs : the latter process makes the dew-point a little, 

 and the humidity-ratios a great deal too small. The hours of ob- 

 servation are reckoned in the old hours of Cape Colony civil time 

 (22-^°E.), from midnight to XXIII. Kimberley being some 2°. 10' 

 to the east of this, each consecutive mean value properlv belongs 



to the local times oh. 9m.. ih. 9m., 2h. 9m " This is 



important \n appl)ing the results to stations in different longitudes. 



When we compare the mean values at VIII. with the means 

 for the (lay, we see at once how wide of the mark our appreciation 

 of the climate is likely to be if we take onlv the results of second- 

 order stations to define it. The pressure is 0-4 inch, or nearlv one- 

 half Its whole daily range in excess ; the temperature is 4° too low • 

 the dew-point is nearlv a degree, and the humiditv nearlv seven per 

 cent. tr,o high. Also the mean of the registered maximum and 



