NOTES. XCV 



comprised 933 stations, an error of three feet would not be at all surprising ; 

 and we may consider this result as rather tending to confirm the equilibrium 

 of the waters which communicate round Cape Horn (Arago, Annuaire du 

 Bureau des Longitudes pour 1831, p. 319.) I had inferred, from my baro- 

 metric observations in 1799 and 1804, that, if there were any difference of 

 level between the two oceans, it could not exceed 3 metres (Rel. hist. T. iii. 

 pp. 555 557 ; and Annales de Chimie, T. i. pp. 55 64.) The measure- 

 ments which seem to establish a higher level for the waters of the gulph of 

 Mexico, and for those of the Adriatic, (in the later case, by the combination 

 of the trigonometrical operations of Delcros and Choppin with those of the 

 Swiss and Austrian engineers,) appear to me to be subject to many doubts. 

 Notwithstanding the form of the Adriatic, it is scarcely probable that the 

 level of its waters at the upper end should be almost 26 French feet higher 

 than that of the Mediterranean at Marseilles, or 2 3 '4 French feet higher 

 than the level of the Atlantic. (See my Asie centr. T. ii. p. 332). 



( 7 ) p. 298 Bessel iiber Fluth und Ebbe, in Schumacher's Jahrbuch fur 

 1838, S. 225. 



t 368 ) p. 299. The density of sea water depends concurrently on the tem- 

 perature and on the degree of saltness a consideration which has not been 

 sufficiently attended to in investigations into the cause of currents. The 

 submarine current which brings towards the Equator the cold water of the 

 poles, would follow an exactly opposite course, if difference in respect to 

 saltness were alone concerned. In this point of view, the geographical dis- 

 tribution of temperature and of density in the waters of the ocean is of great 

 importance. The numerous observations obtained by Lenz (Poggendorif s 

 Annalen, Bd. xx. 1830, S. 129,) and by Captain Beechey (Voyage to the 

 Pacific, Vol. ii. p. 727), are deserving of particular attention. Compare also 

 Humboldt, Relat. hist. T. i. p. 74 ; and Asie centrale, T. iii. p. 356. 



(^ p. 300. Humboldt, Relat. hist. T. i. p. 64; Nouvelles Annales dcs 

 Voyages, 1839, p. 255. 



( 3 ?) p. 300. Humboldt, Examen crit. de 1'hist. de la Ge'ogr. T. iii. p. 100. 

 Columbus adds shortly after (Navarrete, Coleccion de los viages y descubri- 

 mientos de los Espanoles, T. i. p. 260), " that the movement is strongest in 

 the Caribbean Sea." In fact, Rennell terms this region, " not a current, 

 but a sea in motion." 



( 371 ) p. 300. Humboldt, Examen crit. T. ii. p. 250 j Relat. hist. T. i. pp. 

 6674. 



C 372 ) p. 300. Petrus Martyr de Angleria, De Rebus Oceanicis et Orbe 



