775 



cording to Shuckburg, to be 761.18 mm (at the 

 temperature of zero), I naturally concluded 

 from that comparison, that the barometric mean 

 at the level of the sea in the torrid zone, was a 

 little less than in the temperate # . Uncertain 

 with respect to the capillarity of the barometer 

 I had employed, I estimated that difference at 

 two millimeters in my View of the Equinoxial 

 Regions, and which I attributed to the ascend- 

 ing movement of the tropical atmosphere, 

 which bears the layer of air strongly heated, 

 towards the polar regions. Having made, with 

 my instruments, long journies by land from 



p. 107. Schumacher Jstr. Nachr. Beit, Tom. ii, No. 65 j 

 Hertha, n. 3, p. 246. On the almost constant depression 

 which the barometer undergoes near Cape Horn, where the 

 western winds blow impetuously, see Krusenstem, Rec. de 

 Mem, hydr., Tom. i, p. 29; Leopold de Buck; in Gilbert, 

 Ann der Physik., Tom. xxv, p. 230 ; Id. Barometrische 

 Windrose, p. 4. 



* See my Essay on the Geography of Plants, p. 90. In 

 the first half of the 18th century, Richer, Bouguer, La Con- 

 damine, Ulloa, and Don Jorge Juan, believed that the baro- 

 meter at the level of the equinoxial seas, was 27 In y 

 28 in 0°, or 28 in I 1 '. The instruments used by those travel- 

 lers had no doubt the air but very imperfectly expelled, for no 

 correction being employed for the temperature, the barome- 

 tric heights must have been found too great. If the mean 

 barometric heights at the level of the seas of Europe, have 

 been recently a little exaggerated, it is no doubt on account 

 of the uncertainty that envelops the effect of capillarity. 



