The Meteor of July, 1S60. 375 



sition of the orbit, both in respect to the earth's surface and in 

 space. 



(3.) The close approximation to parallelism to the earth's sur- 

 face of the eastern portion of the observed path leaves it a matter 

 of doubt, considering the imperfection of the observations, whether 

 the meteor finally passed out of the atmosphere and went on its 

 way in a disturbed orbit, or descended gradually into the Atlantic. 

 The former supposition is perhaps the more probable, especially if 

 the path was curved, as above suggested, instead of a straight 

 line. 



(4.) The meteor exhibited different appearances in different 

 parts of its course. It seems to have been observed first as a 

 single body, more or less elongated, gradually increasing in bril- 

 liancy, throwing off occasionally sparks and flakes of light, until 

 it reached the neighbourhood of Elmira, N. Y. Here something 

 like an explosion occurred, and the meteor separated into two 

 principal portions with many subordinate fragments all continuing 

 on their course in a line behind each other, and still scattering 

 luminous sparks along their track, until a point was reached about 

 south of Nantucket, when a second considerable explosion took 

 place, and afterwards the principal fragments passed on till lost 

 to view in the distance. The most trustworthy observations 

 represent the meteor as disappearing while yet several degrees 

 above the horizon, (generally from 3° to 6° or 8°). Besides the 

 actual changes of form which the body successively underwent, 

 apparent changes would present themselves to each observer aris- 

 ing from chanQ:e of direction in which the meteor was seen. 



(5.) It is not easy, from the observations in hand, to determine 

 with much accuracy the volocity of the meteor while passing 

 through our atmosphere. The time of flight is doubtless largely 

 over-estimated by most observers, especially those unaccustomed 

 to measure intervals of a few seconds. Timing with a watch, a 

 repetition of the acts performed during the flight of the meteor 

 usually reduces the interval to not more than one third, or even 

 one fifth, of the observer's own estimate. From 15 to 30 seconds 

 is a fair range for good observation?^, and probably to no observer 

 was the meteor in sight over 45 seconds or a minute, although a 

 minute and a half and two minutes are very common estimates 

 A comparison of the most probable estimatee of time with the 

 length of path observed, gives a velocity ranging from eight to 

 fifteen miles a second. Probably 12 or 13 miles is a tolerable 



