148 Remarks on Dr. Enjleldh Institutes 



demonstration. — After all, we should have been much bet- 

 ter pleased to see the proposition entirely omitted, than any 

 attempt made to amend it. The hypothesis that the rays 

 which come to the eye at the end of twilight are brought by 

 a single reflection, is a very questionable one. The power 

 of reflecting light possessed by the atmosphere, must de- 

 pend on one or both of two causes : 1. It may reflect some 

 of the rays which pass through it in consequence of a defect 

 of transparency. 2. It may reflect in the same manner as 

 light is ordinarily bent back into a denser medium. This 

 last mode of reflection, if it ever takes place without an ab- 

 rupt change of density, is evidently more likely to take 

 place, in proportion as the variation of density is more rapid. 

 Now whichever of these causes operates to produce twi- 

 light, it must evidently exist in a far higher degree in the 

 lower, than in the higher regions of the atmosphere. Hence 

 instead of a single reflection at the height of forty-two 

 miles, two or more successive reflections may quite as pro- 

 bably transmit to the eye the light with which twilight clo- 

 ses.* — But even admitting the correctness of the assump- 

 tion that twilight is produced by a single reflection, it is 

 most obvious that no inference can be deduced concerning 

 " the height of the atmosphere," or even the height at 

 which it ceases to .reflect light. The only legitimate con- 

 clusion is, that forty-two miles is the limit beyond which 

 light is not reflected in sufficient quantity to affect the or- 

 gans of vision. If, instead of this vague proposition, the 

 law of atmospheric density at different altitudes had been 

 inserted in its proper place in the Pneumatics, the subject 

 would have been exhibited in a far more interesting and in- 

 structive form. 



The subject of the moon's librations, in props. 78 — 82, 

 is nianaged with singular infelicity. The introductory pro- 

 position should be, that " the time of the moon's rotation 

 on its axis is equal to the mean time of its revolution round 

 the earth," — instead of beginning with the fact that "the 

 moon always has nearly the same side toward the earth," 

 and drawing the strange' inference that " if the moon re- 

 volves about its axis, its periodical time must be equal to 

 that of its revolution round the earth." The librations 



* See Vince's Ast. I. Art. 206. 



