232 M. Arago on the Thermometrical State 



may be remarked, irj the first place, that the partial congelation 

 of a river can never alone characterize a climate. There are 

 several circumstances connected with atmospheric vicissitudes, 

 which may cause incidentally to descend, upon any given point 

 of the globe, very cold and dry strata from the higher regions 

 of the air : and it would appear that the natural cold of these 

 strata, with that from the abundant evaporation which their dry- 

 ness produces, together with that which the night originates, 

 from radiation, more especially if the atmosphere be perfectly 

 serene, is perfectly sufficient to occasion the freezing of rivers in 

 every region of the earth.* And thus, within these few years, 

 we have learnt, if not without surprise, at least without esteem- 

 ing the circumstance as altogether inexplicable, that in Africa, 

 the water in the leathern bottles of Captain Clapperton was one 

 night frozen, not far from Mourzouck, and in a plain but little 

 elevated above the level of the sea. On the same accounts, 

 meteorologists have not placed among the assertions unworthy 

 of examination, what Abd-Allatif (See M. Silvestre de Lacy's 

 translation, p. 505,) reports, viz. that in the year 829, when the 

 Patriarch, James of Antioch, and Denys of Telmacher, went 

 with the Caliph Mamoun into Egypt, they found the Nile 

 frozen. 



But previous to engaging in this discussion, we ought to have 

 inquired if it be well established that the rivers in the south of 

 Italy are never frozen in our times: Be this however as it may, 

 to the testimony of Virgil might have been opposed a very de- 

 cided passage from Theophrastus, whence we learn that at that 

 remote period, the dwarf-palm (Chamaerops humilis) covered a 

 great extent of ground in Calabria. The vegetation of this 

 shrub is quite compatible with a degree of cold, as in the king- 

 dom of Valencia, where there are occasional incidental frosts of 



• These considerations serve to explain, 1^ how, in 1 709, the Seine was 

 not entirely frozen over at Paris, even between the bridges, whilst at Tou- 

 louse, the inhabitants were promenading on the Garonne, and in Languedoc 

 they went on the ice from Cette to Boussigny and Balaruc ; 2dly, how the 

 maximum of cold occurred at Paris two days later than at Montpellier; 3c?/y, 

 how, after a partial thaw, the cold returned to Montpellier sooner than to 

 Paris ; and, 4<%, how, in 1 829, the cold was above a degree more intense at 

 Toulouse than at Paris, which is situated 5|° more to the north. 



