168 REPORT — 1839. 



derived from Dr. Apjohn's formula, applied to an extensive se- 

 lies of hourly observations. 



When we consider of how great importance it is to Meteoro- 

 logy to perfect the method of hygrometric observations by the 

 moistened thermometer, I cannot but hope that those approxi- 

 mations may lead to future and more accurate results. Having 

 found the formula in very many instances give close approxima- 

 tions to the dew point, as determined by the direct method of 

 Daniell, I cannot but think that it will eventually become an 

 important acquisition to this department of science. 



The very recent completion of the numerical results of these 

 first years' hourly observations on the barometer and dew point 

 necessarily limits this Report to some of the more immediate de- 

 ductions which they present. It is further to be observed that 

 the weather of these years has been of a disturbed and unset- 

 tled character ; especially the last two, which have abounded 

 in heavy rain and hard gales. Comparing the easterly and 

 westerly winds (leaving the due north and south out of the ac- 

 count), I find that the west winds have blown for one-half the 

 whole period, and the easterly for one-quarter only. The ratio 

 being nearly 2:1. Comparing the northerly with the south- 

 erly winds (leaving the due east and west out of the account), 

 the latter have blown for one-fourth of these three years, and 

 the former for one-eighth ; the ratio being also about 2:\. The 

 great prevalence of the southerly and westerly over the north- 

 erly and easterly winds has been accompanied by a mean press- 

 ure probably below what will appear in the mean of a greater 

 number of years. In these years there occurred twenty-eight 

 hard gales, principally from S.E. to N.W. 



Horary Observations on Temperature. 



Since the discussion of the observations on temperature, two 

 valuable communications have been placed in my hands ; the 

 first by Professor Bache, which contains an hourly register of 

 the thermometer from June 1837 to June 1838, at Frankford 

 Arsenal, by Captain Alfred Mordecai, of the United Service 

 Corps of Ordnance. The second contains three hourly regis- 

 ters for the year 1837, by Major Ord, of the Royal Engineers, 

 at the three most important meteorological and geographical 

 stations in the island of Ceylon, viz., Colombo, Kandy, and 

 Trincomalee. 



I am enabled from these registers to extend the table given 

 in page 199 of the Fifth Report of the Association, so as to in- 

 clude these places. 



