38 



REPORT 1869. 



tion, but of a progressively different form, being the most inclined the furthest 

 from, and the least so, the nearest to, the time of sunset. 



By reading the temperatm-es from these curves at every 100 ft. of elevation up 

 to 1000 ft. the next Table was foi-med. 



Table showing the decrease of Temperature with increasing elevation at every 



100 ft, up to 1000 ft. 



The numbers in these Tables prove that which was indicated by the several free 

 ascents, viz. that the decrease of temperature -with increase of elevation has a 

 diurnal range, and different at diflereut hours of the day ; the changes being the 

 greatest at midday and the early part of the afternoon, decreasing to (at or about) 

 sunset, when with a clear slcy there is little or no change of temperature through 

 several hundred feet fmni the earth; whilst with a cloudy sky it decreases from 

 the midday hours at a loss ra]iid rale to (at fir about) sunset, when the decrease. is 

 nearly uniform at the rate of 1° in 200 ft. I was not able t(i take any observations 

 after sunset ; but such observations are greatly needed, as there seems to be a very 

 great probability that the temperature at the height of 1000 ft. may not undergo a 

 greater range of temperature during the night-hours than during the day-hours ; 

 and if this be the case, then the temperature at night must increase from the 

 gromid with elevation : this inference seems to be conlirmed by the after-sunset 

 observations of Oct. 2nd, 1805, but it is desirable and very important that the facts 

 slioidd be determined by direct experiments. The law witli a clear sky may 1)e 

 thus represented : — Take the lieights as ordiuates of a curve of which the corre- 

 sponding changes of temperature are tlie corresponding abscissee (considered posi- 

 tive when tlie temperature decreases, i. c. so that a decrease of 10° at 1000 ft. would 

 correspond to a point on the curve whose positive abscissa is 10 and ordinate 1000, 

 then the curve thus formed will be somewhat hyperbolic, for the changes are 

 greatest near the earth), the concavity being turned towards an ordinate through 

 tlie origin or axis. The concavity will be greatest when the curve represents the 

 decline of temperature at a time soon after midday ; but as the afternoon advances 

 the curve gradually closes up to and coincides with the axis at or about svmset, 

 becoming then rectilinear : after passing this critical position, in which the tem- 

 perature is uniform and equal to that on the earth for the first 1000 ft., the curve 

 probably becomes hyperbolic again, its concavity still being turned towards the 

 axis, so that an increase of temperature corresponds to an increase of height, and 

 t-lie extreme position is readied probably at or soon after midnight, when the cm-ve 



