OP A METEORIC FIRE-BALL. 37 



satisfactory way of computing it from physical causes, various arbitrary values vv^ere 

 assumed for it, by way of experiment, ranging from 2 to 160. The latter afforded 

 an orbit which satisfied the observations tolerably well, but the assumption seemed so 

 extravagant that this line of research was abandoned, and a change being introduced 

 in the elements near longitude 74°, as already remarked, an orbit was computed, 

 using only the unresolved value of R, and so taking no account of the difference 

 of resistance in the horizontal and vertical directions. The path thus divides itself 

 into three sections, the 1st and 3d of indefinite length, over only a small portion of 

 which the meteor was visible, and the second an intermediate one, 160 miles long, 

 where it was most brilliant. The calculations for the resistance of the atmosphere 

 were commenced at the point where the first change in the elements was made, the 

 height of the meteor being there about 56 miles ; but as the effect of this resistance 

 was appreciable, Avhile the meteor was receding, up to the height of not less than 

 64 miles, within the limits of accuracy to which the computations extended, allow- 

 ance should properly have been made for it through the latter part of the 1st sec- 

 tion, though at that height it was but very trifling. Indeed, the entire modification 

 of the path occasioned by it was so small, compared with that which the considera- 

 tion of the spheroidal form of the earth would occasion, that it seemed to savor of 

 useless refinement to allow for the former, while the latter was omitted. And with 

 the hope that the subject might hereafter receive, at the hands of others, a more 

 thorough discussion, in which both might be taken into account, I concluded to 

 slightly modify the elements so as to afford an undisturbed orbit that would differ 

 so little from the disturbed one just described, that the azimuths and altitudes in 

 Table II, which had been already computed for the one, might answer for the other. 

 This undisturbed orbit is given in Table I. The computed changes in the values 

 of semi-axis major, eccentricity, velocity, longitude of perigee and perigeal dis- 

 tance, caused by the unresolved resistance of the atmosphere, in the 2d and 3d 

 sections of path were as foUows :— 



