68 Introduction. 



stationary. The number of observations is arbi- 

 trary, but three or four at least should be made 

 daily ; for instance, at six or eight o'clock in the 

 morning, at noon, and about four or six ; and, if it 

 can be, also at ten in the evening. The observ- 

 ation at noon is the most important. It must not 

 be forgotten to compare, before or after the 

 voyage, the height of the marine barometer with a 

 good portable barometer, because, without this 

 comparison, a chief result of the barometrical ob- 

 servations, the determination of the absolute height 

 of the barometer in high latitudes would be lost. 

 It is likewise requisite to notice, at every observ- 

 ation, the warmth of the barometer, by a thermo- 

 meter fixed to it, or at least hanging near it in the 

 room, and also the temperature of the external air. 

 With regard to the latter, it is to be observed, that 

 the thermometer must be suspended in a place 

 perfectly accessible to the external air, but not ex- 

 posed to either the direct or the reflected rays of 

 the sun, and in case these conditions cannot be 

 combined, it might be better to have two hanging 

 in different places. On such voyages it is proper 

 to take several thermometers, not only in case of 

 accidents by breaking, but also to have an oppor- 

 tunity of putting into the hands of a diligent ob- 

 server in remote places, one of these instruments, 

 the observation of which is so easy, and yet at- 

 tended with such profit to climatology. 



Connected with the changes of the atmosphere, 

 is the moving power of all navigation, the winds. 



