VOL. XXXI.J PHILOSOPHICAL TRANSACTIONS. 517 



which the moose will wade far and deep, and by the noise they make in the 

 water, the hunters often discover them. In the winter they live on browse, or 

 the tops of bushes and young trees, and being very tall and strong, they will 

 bend down a tree as thick as a man's leg ; and where the browse fails them, they 

 will eat off the bark of some sorts of trees, as high as they can reach. They 

 generally feed in the night, and lie still in the day. 



The skin of the moose, when well dressed, makes excellent buff; the Indians 

 make their snow-shoes of it : their way of dressing it is thus : after they have 

 haired and grained the hide, they make a lather of the moose's brains in warm 

 water ; and after soaking the hide for some time, they stretch and supple it. 



Some Remarks on the Allowances to he made in Astronomical Observations for 

 the Refraction of the Air. By Dr. Edm. Halley, R. S. S. Astronomer Royal. 

 With an accurate Table of Refractions. N° 368, p. ] 69. 



Were the medium of our air much more in quantity, or the force of gravity 

 much greater than it is, or in a word, were the refractive power of the air much 

 more sensible than we find it, nothing could have been a greater impediment to 

 discoveries in astronomy : for all objects appearing by refraction higher than 

 really they are, till such time as the laws and quantity of that refraction had 

 been ascertained, it would have been impossible to have been secure of the true 

 observed place of any celestial object. But as it is so little, that none but nice 

 instruments can perceive its effects, it was not discovered to be any at all, till 

 Bernard Walther's time, about the year 1500 ; nor brought to any sort of rule 

 till Tycho Brahe ; nor ascertained, till our worthy president [Sir I. Newton] 

 made the first accurate table of it. The curve which a beam of light describes, 

 as it approaches the earth, being one of the most perplexed and intricate that 

 can well be proposed, as Dr. Brook Taylor has shown in the last proposition of 

 his Methodus Incrementorum. 



By this table it follows, that the ratio of the sine of the angle of incidence, 

 to that of the refracted angle, increasing as the beam approaches, makes a very 

 notable difference in the place of an object near the horizon: but in objects 

 that are much elevated, the refractions become small, and their differences 

 scarcely exceed a second per degree ; so that they are much the same, as if the 

 incident and refracted angles were on the surface of a sphere of air of the same 

 uniform density, close adjoining to the eye. 



When therefore the stars are 20° or more elevated above the horizon, we 

 may take it for granted, without sensible error, that the sines of the true 

 and apparent distances from the vertex, are in the same constant ratio. Hence 



