ON NINE BALLOON ASCENTS IN 1863 AND 1864. 265 



wet-bulb thermometer, have been adopted, without combination with similar 

 results otherwise derived, excepting only when the wet-bulb failed to act 

 either at times when the temperature of the air liad just descended below the 

 freezing-point, or just ascended above it, and when I have had occasion to 

 apply water to the wet-bulb at such times the dew-point as found from 

 either DanieU's or Kcgnault's hygrometer has been used. 



§ 5. Vaeiation- of Tempeeature of the Ate with Height. 



Every reading of temperature in the preceding Tables, or the means of 

 smaU groups of readings when observations have been taken in quick succes- 

 sion, at about the same altitude, was laid down on a diagram ; all these 

 points were joined, and a curved line was drawn passing through or near 

 them, giving an e(]ual weight to every point, and such that the area of the 

 spaces between the original and adopted lines on one side of the adopted line 

 was equal to that of the spaces on the other side of the line. The curves 

 thus formed, for the most part, in the previous experiments have shown a 

 gradual decrease of temperature with increase of elevation, and a gradual in- 

 crease of temperature with decrease of elevation ; biit this was not the case 

 this year, and I have not been able to adopt any curves for January 12 and 

 April 6. 



On the other days of experiments, a curve of assumed normal temperatui-e 

 has been adopted, and by comparison between the results as read from this 

 curve, and the observations at the same elevations, the places and amount of 

 disturbances are shown in the following Tables. 



The numbers in the first column show the height in feet, beginning at the 

 ground and increasing upwards ; the numbers in the second column show the 

 interval of time in ascending to the highest point; the notes in the third column 

 show the circumstances of the observations ; the niimbers in the fourth and 

 fifth columns the observations and the approximate normal temperatures of the 

 air ; and those in the next column the difference between the two preceding 

 columns, or the most probable efl^'ect of the presence of cloud or mist on the 

 temperature, or of other disturbing causes in operation. 



The next group of columns is arranged similarly for the descent, and the 

 other groups for succeeding ascents and descents. 



