134 Mr. J. Glaiblier's Remarks on the Weather 



inches; and atGreenwicli on December 7 at S"^ a.m., and it was, 

 as stated above, 28'383 inches. A reading so low as tliis is 

 of rare occurrence. The previous instances at Greenwicli are 

 as follows: — In 1783, on March 6, the reading was 28'12 

 inches; in 1809, on December 17, the reading was 28*20 

 inches ; in 1821, on December 25, the reading was 27'89 inch- 

 es; in 1824, on November 23, the reading was 28*37 inches; 

 and in 1843, on January 13, the reading was 28*10 inches. 



During the quarter there were eightexhibitions of the aurora 

 borealis, which occurred on the following days: — Oct. 15, 23 

 and 24 ; Nov. 1,2,19; Dec. 2 and 19. Tiiat oil Oct. 24 was one 

 of the finest I have seen. (See the Philosophical Magazine 

 for November, and the Athenaeum for November.) At every 

 one of these times the magnets were much disturbed. The 

 magnetic disturbance connected with the aurora of October 24 

 exhibited a greater amount of consecutive disturbance than 

 had been before experienced at Greenwich since the establish- 

 ment of the magnetic observatory in 1840. (See the Pliiloso- 

 phical Magazine for January 1848 ; and a forthcoming account 

 of the aurora seen in Cambridge, by Mr. Morgan, of the Cam- 

 bridge Observatory.) 



From the prece(ling remarks, it will be seen that the weather 

 during the past quarter has been very unusual indeed. I have 

 searched all meteorological records at my command, which 

 have been made in the previous fifty years, and I have failed 

 to find any season of similar character. In the year 1806 the 

 average temperature for the last quarter of the year was 50°*1 

 (see Philosophical Transactions for that year) ; and this result 

 nearly agrees with that found by Luke Howard, Esq. (see his 

 Climate of London), which was 50°*3 ; but although this value 

 is greater than that of the past three months, yet 1 am inclined 

 to think that the temperature of this period in 1806 did not 

 really exceed that of 1847, as at this time all mean tempera- 

 tures depended solely on uncompared self- registering thermo- 

 meters ; and it is found that even with good self-registering 

 thermometers, a subtractive correction is always required to 

 deduce from them the true mean for the month. The only 

 October in this century whose temperature seems really to 

 have exceeded that of the past October, is that of 1811 ; so 

 that whether we compare the weather of the past quarter by 

 longer or shorter periods with that of similar periods in past 

 years, it has evidently been of a very remarkable character, 

 and of rare occurrence. 



To the report of the Registrar-General are appended the 

 monthly values at every station, from which the average values 

 for the quarter have been determined, and which are contained 

 in the following table : — 



