ATM 



lowed it. It muR be added, that in this calculation, the 

 dire-fl and rcflefted rays are fuppofcd to be ri^rht lines ; 

 whereas in faft they are curves, formed by tiie pe'rpetual re- 

 fraction which the rays undergo in pafling through a feries 

 of different denfities of air. Compntin,^ then upon them, 

 as two fimilar curves, or rather as a fnigle curve, one ex- 

 treme whereof is a tangent to the earth ; its vertex, equally 

 diftant from both the extremes, determines the height of 

 the atmofphere ; which therefore will be found fomewhat 

 lower than in the former cafe ; the point of concuirence 

 of two right lines, which are here only tangents to the 

 curve, the one at one end, and the other at the other, 

 being higher than the vertex of the curve. In this way, 

 M. de la Hire finds the atmofphere ■>fii<';i fathoms, or 16 

 leagues. Hiil. de I'Acad. Roy. de Scien. an. 17 13. p. 71. 



The nature of the curve, which is defcribed by a ray of 

 light in palling through the atmofphere, has been the fub- 

 jeA of afiiduous inveftigation. M. De la Hire took "-reat 

 pains to demoilrate, that, fuppofing the denfity of the at- 

 mofphere proportional to its w-eight, this curve is a cycloid : 

 and he fays, that if the ray be a tangent to the atmofphere, 

 the diameter of its generating circle will be the height of 

 the atmofphere ; and that this diameter iacreafes, till at laft, 

 when the rays are perpendicular, it becomes infinite, or the 

 circle degenerates into a right line. This reafoning fup- 

 pofes that the furface of the atmofphere is a plane ; but fince 

 it is a cun'e, he obfcrves that thefe cycloids become in faft 

 epicycloids. Htrmannus, in hi; " Phoronomia," has detefted 

 the error of M. De la Hire, and fliewn that this curve is 

 infinitely extended, and has an afymptote : and Dr. Brook 

 Tsylor obferves, " Method. Incrcm." p. 16S. that it is 

 one of the m.oft intricate and perplexed that can well be 

 propofed. This ingenious author computes the refradlive 

 power of the air, to be to the force of gravity at the 

 furface of the earth, as 320,000.000 to 1. 



The extreme rarity of the atmofphere at conCderable 

 altitudes, fuch as thofe of forty or fifty miles, bound- 

 ing the prcdudlion of twilight, has perplexed philofo- 

 phers in accounting for meteors, whicli, whatever be 

 their origin, whether cleftrical or otherwife, are ob- 

 fcrved at a much greater elevation than that to which the 

 refractive power of the atmofpherical air extends. A very 

 remarkable one of this kind was obfcrved by Dr. Hal- 

 ley in the month of March 1719 ; the altitude of which 

 he computed to have been between 69 and 734 Enghfli 

 miles ; its diameter being 2800 yards or more than a mile 

 and a half, and its velocity about 3J0 miles in a mivaite. 

 Others of a umilar kind, but of a greater altitude and velo- 

 city, have been obferved by oth.ers ; and particularly one 

 feen in Auguil 1783, whofe height above the earth could 

 not be Icfs than ninety miles, and its diameter v/as not lefs 

 than the former, whilil its velocity was certainly not lefs 

 than 1000 miles in a minute. From analogy and reafoning 

 it is very probable, that fuch meteors are not eflfentially 

 different from thofe that are feen near the furface of the 

 earth. Neverthelefs in the high regions where they are 

 obferved, the atmofphere, according to our computation, 

 ought not to have denfity fiifBcient to fupport flame and to 

 propagate found ; and yet fuch meteors are commonly 

 fucceeded by one or more explofions, and are accompanied, 

 as it has been reported, with a hi.Tmg noifi; as they pafs over 

 our heads. The meteor of 17 19 was not only very bright, 

 fo that for fome time it changed the night into day, but 

 was attended with an exploilon that was heard over all the 

 illand of Britain, occalioning a violent concuffion of the 

 atmofphere and feeming to (hake the earth itfclf. And yet, 

 ift the regions in which this mrieor moved, the air ought to 



ATM 



have been 300 thoufand times rarer than the air we breathe, 

 or 1000 times rarer thnn the vacuum commonly mnde by a 

 good air-pump. Dr. Halley conjectures, that the immcnfe 

 magnitude of fach bodies may comptnfate for the rarity 

 of the m.cdium in which they mcne. Allowing them to be 

 eleflrical phenomena, difficulties occur in explaining feveral 

 circumflancts attending them ; and particularly the fplendor 

 of their appearance, which requires a circumambient fluid 

 capable of confining and condenling the electric matter of 

 which they are compofed. From late experiments, it has 

 been inferred, that the electric fluid cannot pervade a perfect 

 vacuum. See Mi-teor. 



Atmosphere, Rtfraa'ton and RefleP.ion of the. That 

 the atmofphere has a rcfraftive power, which is the caufe 

 of many plienomena, is unqueftionable. This power is af- 

 certained by the production of twilight above noticed, and 

 by many other facts and experiments. Alhazen the Ara- 

 bian, who lived about A. D. I 100, ftems to have been 

 more inquifilive into the nature of refraction than the pre- 

 ceding writers. But neither Alha/en, nor his follower 

 Vilellio, knew any thing of its jull quantity, which was 

 not known to any tolerable degree of exaCtnefs, till Tycho 

 Brahe, with incredible diligence, fettled it. But neither 

 Tycho, nor Kepler, difcovered in what manner the rays of 

 light were refracted by the atmofphere. Tycho thought 

 the refraftion was chiefly caufed by dcnfe vapours, very 

 near the earth's furface. Kepler placed the caufe wholly 

 in the higher regions of the atmofphere, which he took to 

 be uniformly denfe ; and thence he determined its altitude 

 to be little more than that of the higheft mountains. But 

 the true conllitution of the atmofphere, deduced afterwards 

 from the Torricellian experiment, afforded a jufter idea of 

 thefe refraftions, efpecially after it appeared by a repetition 

 of Mr. Lowthorp's experiment, that the air's refraftivc 

 power is proportionable to its denfity. By thi? variation 

 of the air's denfity, a ray of hght, in paffing through the 

 atmofphere, is continually refrafted at ever)' point, and 

 thereby defcribes a curve, and not a ftraight line, as it 

 would have done were there no atmofphere, or were its 

 denfity uniform. See Rffraction. 



The atmofphere, or air, lias alfo a rcfleftive power; and 

 this power is the caufe that enlightens objefts lo unifnrmly 

 on all lides. The abftnce of this power would cccafion a 

 fti-ange alteration ia the appearance of things ; their fliadows 

 would be fo very dark, and their fides enlightened by the 

 fun fo very bright, that probably we could fee no more of 

 them than their bright halves ; fo that, for a view of the 

 other halves, we muit turn them half round, or, if immove- 

 able, mull wait till the fun could come round upon thein. 

 Such a pellucid unrefledtive atmofphere wouldjndeed have 

 been very commodious for allroHomical obfcr\atio: s upoiv 

 the courfe of the fun and planets among the fixed liars, 

 vifible by day as well as by night ; but then fuch a fuddcn 

 tranlition from darknefs to liglit, and from light to darknel'i 

 immediately, upon the rifing and fetting of the fun, without 

 anv twilight, and even upon turning from or to the fun at 

 noon day, would have been very inconvenient and ofieniive 

 to our eyes. 



However, though the atmofphere is greatly aflillant to 

 the illumination of objects, yet it mult alfo be obferved 

 that it flops a great deal of light. By M. Bouguer's expe- 

 riments, it feems that the light of the moon is frequently 

 2000 times weaker in the horizon, than at the altitude of 

 66 degrees ; and that the proportion of her light at the 

 altitudes of 66 and 19 degrees, is about 3 to 2. The 

 lights of the fun mull bear the fame proportion to each 

 other at thofe heights, which M. Bouguer made choice ot, 



as. 



