1838.] On the Revolution of the Seasons. 1§3 



rity. In this I have been so far disappointed, that the lists received 

 have been very few, and not above two of them extending more than 

 25 years back. That the data which they, together with those already 

 in my possession, afforded, should be insufficient to give results regular, 

 or nearly regular, was no more than experience with the most recent 

 period (from 1835 to 1806) led me to believe. But there are some 

 farther impediments to extending the investigation beyond 1806 which 

 must be particularly noticed. 1st. Many of the lists do not extend 

 beyond 1812 ; only 12 out of 30 reach to 1792, and only 5 to 1750. 

 Each of these lists has a particular mean dependent upon local circum- 

 stances, and the cessation of it affects the general average. The only 

 way to obviate this source of error, would be to reduce each list to a 

 common mean, the labour of which would be very great. 2ndly. No 

 registers of the prices of corn for this period have been kept with a 

 view of ensuring accuracy ; consequently the lists have been filled up 

 from the best information that could be procured, such as the private 

 memoranda of individuals, merchants, and others : of course the more 

 distant the time the more scanty and uncertain those data would become, 

 and we can hardly believe but that many of the results they furnish must 

 be, in a degree, erroneous. 3rdly. Previous to the year 1806 great part 

 of the country was still in its age of chivalry ; at one time subject to the 

 passage of numerous armies, sacking, plundering and devastating; at 

 another to the forays of bands of knights-errant. Besides this, the 

 different governments often took the corn dealer under their paternal 

 care, and he was made to sell as the caprice of a tyrant, or the clamours 

 of an unthinking people, obliged him. 



All this being premised we have now to state what the actual result 

 of an examination of the lists has been. The lists which formed the 

 N. W. line in a previous paper were not increased in number, but sepa- 

 rated into two divisions, according as the places from which they came 

 were situated nearer, either to Dehli, or to Lodiana. — The average 

 was taken of each division, and, as in the former case, the difference 

 between the maximum and minimum from 1750 to 1835 was divided 

 in* 1000 parts. The Benares line remained the same as before. — * 

 The Bengal line was increased by a list from Calcutta taken from the 

 12th vol. of the Asiatic Researches. The last, or southern line (Jub~ 

 bulpore, Indore, and BhopalJ, was increased by the addition of a line 

 from Dundivala in the Panjdb. The average of the five lines was 

 then taken and the general line traced out in the manner described in the 

 former paper. I have before mentioned the notion that a complete revolu- 



