56 DIURNAL OSCILLATIONS—ABNORMAL CHANGES. 
the appearance of the profile of the line. In conformity, however, with the practice of meteo- 
rologists, this element of variable pressure has not been taken into account. All the data upon 
which it depends are submitted ; and when scientific investigations, that are now in progress, 
have sufficiently tested the subject, the error due to the variations in tension of vapor may be 
determined, and the altitude of each station properly reduced. 
The corrections applied to the original record of the barometric observations are as follows: 
ist. For relative expansion of column of mercury and scale due to variations of heat, in order 
to reduce the observations to an equivalent, at a temperature of 32? Fahrenheit. For this 
purpose, Professor Guiot’s excellent tables were used. ! 
2d. For constant error of barometer, or the difference between the zero point of its scale and 
that of the standard with which it was compared. The errors of the various barometers used 
are shown in the Appendix. 
3d. For oscillations of the ra column, due to diurnal variations of atmospheric pres- 
sure, and determined by hourly observations, taken for several consecutive days, at various points 
upon the route. The results obtained for this correction are also shown in the appendix. 
All barometric oscillations seem to be due primarily to variations of temperature. Let the 
atmosphere be placed in equilibrio, and everywhere, at equal elevations, the columns of mercury 
supported would be the same. But as soon as one portion of the ærial fluid is warmed by the 
sun it expands, rises, and overflows the regions contiguous, producing currents of air more or 
less violent. The heated column becomes lighter, and the colder heavier, in consequence of the 
superincumbent mass. The effect upon the barometer is, however, greatly modified by the 
vapor of water which is at the same time generated by heat, and by its elastic force tends to 
increase the pressure. By the daily revolution of the earth, consecutive portions of its atmo- 
sphere are warmed, causing horary oscillations, which amount to about 2.5 millimetres within the 
tropics, and become almost inappreciable as we approach the frigid zone. Two points of maxi- 
mum and two of minimum pressure occur during the twenty-four hours,* and their periods are 
determined by the relative value of the influences produced by heat upon dry air and upon moisture. 
4th. For abnormal changes of atmospheric pressure, sometimes termed barometric waves. 
Contact of the air with an extensive surface of land heated by the sun's rays, and currents of 
wind charged with a greater or less amount of moisture, create another system of atmospheric 
dilatations, the effects of which may, in some respects, be compared to tides upon the ocean. 
Within large areas they produce simultaneous oscillations of the barometric column, called 
abnormal.t They recur at irregular periods, and are variable in their character. Their magni- 
tude is less within the tropics than in the temperate zones. Within the limits of our western 
9 The remarks submitted above were entirely suggested by the reduction of the observations made upon the Pacific rail- 
road explorations, € were informally communicated to the department some time since. Lately, through the kindness 
of Professor Henry, the distinguished secretary of the Smithsonian Institution, I have received a copy of a lecture on me- 
teorology, ps lies before that institution by Robert Russell, Esq., with an appendix by the secretary, giving much in- 
formation upon this subject. The whole perenne is highly interesting. With regard to the horary oscillations of the 
barometer, Ser Henry remarks as follow 
the barometer exhibits ita maxima and two minima in the course of twenty-four hours has given rise to 
much mix as to its ca The most common explanation is, that it is due to the joint action of the variation of 
the temperature and fn de the aci the moisture rising into the atmosphere increases the weight of the air 
more than the heat diminishes it by expansion; while at about ten o'clock a. m. the effect of heat overbalances that of 
vapor, and again as the sun begins to decline La weight of the latter predominates. Mr. Espy attributes the daily oscil- 
afte of the barometer to the upward and downward motion of the particles of air and moisture as they are separated or 
approximated by the change of temperature. The particles weigh more while the rate of separation is increasing, and less 
while it is diminishing. This is a true cause, but we are not certain that it is a sufficient one. Whatever may be the cause 
of the daily tine of the barometer, we know the effect is nearly the same in parts of the earth most widely separated.” 
According to the analysis of Mr. Dove, (see Kaemtz's Meteorology,) the pressure of dry air, when freed from the tension 
of vapor, pond but one diurnal maximum and minimum. aem have occurred where the readings of the barometer, 
uncorrected for the force of vapor, have given a similar res 
+ Abnormal oscillations include the residual errors of T oeil results, when corrected for all the pee 
changes that may be measured. 
EE he ir 
