284 METEOROLOGY, 



were never corrected for temperature. During 1856 and 1857 Green's 

 Smithsonian barometer was employed, and its readings reduced to 32° 

 Fahrenheit. No correction for altitude was ever made, as the cisterns 

 of the various instruments employed were at so small an elevation 

 above the level of the sea. Neither was the elastic force of vapor 

 applied at the time of the record. This force has been calculated only 

 during the past year, according to the rules established by Regnault 

 for deriving every degree of it exhibited in the atmosphere from the 

 readings of the wet and dry-bulb thermometers. It will be seen that 

 it increases directly with the temperature, and amounted during 1857 

 to nearly half an inch during midsummer, or one sixty-seventh of the 

 entire atmospheric weight. 



The absence of either abrupt or great changes gives indication of 

 the tropical feature which the climate possesses. As a general rule 

 the atmospheric pressure varies but little, and that through slow and 

 long continued movements, rather than in the sudden manner charac- 

 teristic of the latitude on the Atlantic coast and elsewhere. Never- 

 theless, although the mercurial column rises and falls within very 

 restricted limits, yet there are changes, represented it is true by small 

 measurements, which occur with wonderful regularity and certainty, 

 diurnal movements at fixed hours, as well as annual ones, having 

 reference to the position of the sun in the ecliptic. The former, or 

 horary oscillations, as revealed on the chart of diurnal barometrical 

 curves, present, in a marked degree, the two diurnal maxima and 

 minima observed within the tropics; the ante-meridian maximum, at 

 about 9 to 10 a. m., being more constant than that at the same period 

 post meridian. Without a single exception the pressure is always 

 less at 3 p. m., and this has no reference to whether the column stands 

 high, as in the cold, or low, as in the hot season. 



The following table, calculated from the horary observations, taken 

 once a month during 1857, gives the mean successive hourly range for 

 the year. The signs -|- and — denote the range of each hour above 

 or below the mean of the twenty-four hours. 



