94 Proceedings of the British Association. 



and December were the greatest, November and February the least, 

 January and September the two nearest the mean. 



Mr. Harris now proceeded to discuss the supposed Influence of the 

 Moon on the Barometer, and with this view had reduced about 4,000 of 

 the observations, so as to show the pressure at the time of the moon's 

 southing, and for each hour before and after ; but he could not discover 

 any differences which could be supposed to arise from the moon's 

 influence. He was therefore disposed to agree with the conclusion 

 lately arrived at by Mr. Lubbock, from a discussion of the Barometric 

 Observations at the Royal Society — viz. that no lunar irregularity is 

 observable from this method of discussing the observations — that, if at 

 any time established, it must prove extremely small. He could not, 

 however, avoid mentioning, as a singular coincidence in the results of 

 the two years, that taking the mean pressures about the four periods of 

 the lunar changes, it appeared that the pressure was less at the new 

 moon, and that it increased up to the last quarter, when it was the 

 greatest. The first object being to arrive at certain great periodical 

 variations, those had been principally kept in view; hence, mere ac- 

 cidental disturbances remained as yet unconsidered. Mr. Harris, how- 

 ever, had observed, as a very general result, that when the pressure 

 decreased at night, whilst the temperature increased, the succeeding 

 weather was always disturbed and uncertain — in winter, gales of wind 

 from the S.E. and S.W., with rain ; whilst, on the contrary, a decreasing 

 temperature, with an increasing pressure, was generally followed by 

 fair weather, with winds varying from N.W. to N.E. The observations 

 hitherto made with the dry and wet bulb thermometer had not yet been 

 reduced. Of the ordinary thermometer, more than 50,000 hourly obser- 

 vations were now completed. Mr. Harris had received two very inter- 

 esting communications on the Hourly Changes of Temperature, which 

 enabled him to contrast the curves of Plymouth and Leith with those of 

 Frankfort Arsenal, near Philadelphia, and three places in Ceylon. The 

 Association was indebted to Major Ord, R.E., for the latter, and to 

 Capt. Mordecai, of the United States' Corps of Ordnance, for the former. 

 Hourly observations had been obtained by these gentlemen, similar to 

 those which had already appeared in the Transactions of the Associa- 

 tion, and which fully confirmed the results arrived at by Sir D. Brewster, 

 to whom the scientific world is indebted for the first perfect series of 

 hourly observations of the thermometer, and also the results of those 

 arrived at by Mr. Harris in the discussion of similar observations carri- 

 ed on at Plymouth, at the request of the Association. Mr. Harris here 

 exhibited, under the form of curves, the mean hourly progress of the 

 temperature at these different places. It appeared, from these observa- 

 tions, that the line of mean temperature at the three stations in Ceylon, 

 between 6° and 8° N. lat., was crossed between 9 and 10 a.m., and at 9 

 p.m. The mean temperature at these stations being 74° at Kandy, and 

 from 80° to 81° of Fahrenheit at the others, which did not materially 



