ON NINE BALLOON ASCENTS IN 18G3 AND 1864. 273 



It is very clear from the numbers in the following Table, that they differ 

 very much from those in previous years, and that they cannot be combined, or 

 all used in deducing general laws. The ascent on August 31 was made in 

 the evening with a partially clear sky, and the results are somewhat abnormal. 

 Some of the numbers in January 12 and April 6 are affected by + signs, 

 circumstances never before met with ; on June 13, at sunset, no difference was 

 found within 2000 feet of the earth ; and those of Juue 20, made a Uttle before 

 sunset, and those of June 27, made a little before and after sunset, aU seem 

 to differ from the general laws as found by experiments made when the sun 

 is situated at a good altitude. 



The only experiments this year which can be combined with previous re- 

 sults are those of September 29 with a cloudy sky, those of October 9 and 

 August 29, mostly with clear skies. 



It is certain from the numbers in this Table that there are at times, in the 

 higher regions of the earth's atmosphere, spaces subjected to great cold, and 

 others to considerable heat ; and from the notes made at the time of passing 

 through clouds, that there exist some clouds of very low temperature, and 

 some, as those of January 12, of high temperature. 



The presence of such either cold or hot currents passing over the country 

 must play an important part in aU our meteorological phenomena, and must 

 exert a great influence upon our climate. . 



The numbers in columns 24 and 26 show the mean results from the experi- 

 ments of the year, the former when the sky was cloudy, September 29, and 

 the latter when clear or mostly clear, on October 9 and August 29. 



The numbers in column 25 show the number of experiments upon which 

 each result in column 24 is based ; at heights up to 5000 feet these vary from 

 13 to 22, at 6000 feet and 7000 feet to 5 and 7, and to heights exceeding 

 7000 feet to 4, these having been made on two days only, viz. June 26 and 

 September 29, 1863, on which days the balloon was frequently enveloped in 

 fog and clouds to the height of 3 and 4 miles. 



The numbers in column 32 show the total number of experiments upon 

 which the numbers in column 31 are based ; they vary from 7 to 17 up to the 

 height of 23,000 feet, and there can be but little doubt that the numbers in 

 column 31 are very nearly true, and approximate closely to the general law. 

 Above 24,000 feet the number of experiments are too few to speak confi- 

 dently upon them, but they are in accordance with the series below this ele- 

 vation. 



The numbers in column 28 show the decrease of temperature at each 1000 

 feet increase of elevation with a cloudy sky, they differ from those in column 

 31, showing the decrease for the same space with a clear sky, the former being 

 smaller, the latter up to the usual height of the cloud plane, and are nearly 

 alike above that plane, but the observations of cloudy skies at heights exceed- 

 ing 5000 and 6000 feet, are too few to place great confidence in them. 



In forming the last six columns of the following Table, no use has been 

 made of the observations taken on July 17, 1862, Aug-ust 31, 1863, January 

 12, April 6, June 13, 20, and 27 of the year 1864. The results on all those 

 days are abnormal ; I defer studying the results on these days till I can com- 

 pare them with some confidence with the general laws, which may be found 

 to hold good at times of the year at which they were made ; their study then 

 will be profitable, and probably fruitful in yielding valuable results. 



The numbei-s in column 30 show the average increase of elevation at every 

 1000 feet for a decrease of 1° with a cloudy sky ; from these we see that 

 up to 1000 feet the average space is 223, the space for each 1000 feet in- 

 creasing, till at 22,000 feet it requires more than 1000 feet for a change of 

 1° of temperature. 

 1864. 



