176 LIEUT.-COLONEL SYKES ON THE ATMOSPHERIC TIDES 



of observations made in India. My own observations are also exceptions to the law. 

 In latitude 18° 30', at 1823 feet, with a mean temperature of 78°, the mean diurnal 

 oscillation for one year, at the limit hour of the tide, was -1166, whilst in Calcutta, 

 latitude 22° 33', mean temperature 78°* 13, the mean of three years' oscillations (1829, 

 1830, and 1831,) give only "1100; and as the observer was Mr. Prinsep, his name 

 is a guarantee for his accuracy. 



But the exceptions are not confined to the tropics, for we find that the mean of 

 five years' observations at Marseilles, latitude 43° 16', mean temperature 60°*8, gives 

 a less oscillation (-0326*) at a few toises above the sea than at Berne, latitude 46° 57', 

 mean temperature 53°'6, at 532 toises above the sea, where the mean of ten years' 

 observations gives an oscillation of •0354-}-. Here, therefore, the oscillation unques- 

 tionably should have been less, because the latitude is higher, the temperature lower, 

 and the height above the sea greater. But these discrepancies may be attributed to 

 the observations not having been taken at the exact limit hours of the tides, and do 

 not therefore give the true oscillation ; nor will satisfactory light be thrown upon 

 the irregularities of the tides until hourly observations are made for lengthened pe- 

 riods in various parts of the earth. 



On the subject of decrement in oscillation, consequent on elevation above the sea, 

 I have collected such data as were available, and have thrown them into the form of 

 a table. Humboldt found that at the Caraccas, at 936 toises above the sea, the 

 oscillation was greater ('1063 J, mean temperature 69°*8,) than at Cumana at 10 toises 

 above the sea, where it was '1004, mean temperature 78°'8. My own careful obser- 

 vations at Poona furnish a similar anomaly. At 1823 feet above the sea the mean 

 oscillation for a year was greater ("1166) than at Bombay, where for nine months the 

 mean was -0765 at the Engineer Institution ; and in my occasional visits I found it 

 respectively -0836 in April 1827, '1123 in March 1828, and '1141 in December 1828. 

 At Madras, in a lower latitude than Poona, at the level of the sea, I have shown it 

 to be only '0790 ; whilst at Calcutta, in a higher latitude than Poona, the means of 

 three years make it '1100. Proceeding to higher levels, however, we find a marked 

 diminution in the extent of the diurnal tide. At Mahabuleshwur, at 4500 feet, the 

 means of eight months reduce the oscillation to '0694 ; at Hurreechundurghur, at 

 3900 feet, the oscillation for the three hottest months was "0969 ; whilst at Kota- 

 gherry, at 6407 feet, it was for five months from noon to sunset only '0498. The oscil- 

 lation at Mahabuleshwur, at 4500 feet, was in fact less than Humboldt's oscillation 

 at Mexico of -0708 at nearly 7000 feet. 



When we pass to the other tides we find the same puzzling anomalies. The mean 

 rise from sunrise to 9 — 10 a.m., whether at Hurreechundurghur, at Mahabuleshwur, 

 or Kotagherry, instead of being less than at Poona, is in fact greater. The mean of 

 three years on the level of Poona gives -0445, whilst the first place gives -0488, the 

 second place -0476, and the last '0490. The maximum night tide, on the contrary, 

 * 0™°»'83, t 0""n-90. X 2«"°'7o. 



