140 Remarks on Dr. KiiJiehVs Institutes 



true aniplilude is 8'^ towards the sa^tne side, then the varia- 

 tion of the needle is 2'^ east." 



Prop. 10. SchoL 2. The late and accurate observations 

 of Mr. Gilpin* and Col. Beaufoy on the diurnal variation 

 of the needle in different months of the year, present wide 

 deviations from the results here given from Mr. Canton. Both 

 the observers just mentioned, make the extremes of the 

 mean diurnal variation in different months, about 11' and 

 4' ; and both place the time of the maximum earlier in the 

 year than was done by Mr. Canton. Col. Beaufoy (Thom- 

 son's Annals, 1819,) places it in April. 



Prop. 11. Schol. 1. The dip of the needle is here rep- 

 resented as probably " unalterable at the same place." — 

 Whatever be the cause of the dip, this supposition is ex- 

 tremely improbable, while the declination is known to be 

 variable, and to be, in common with the dip, the result of 

 the tendency by which the needle places itself in the mag- 

 netical line. Nor do the observations made at London du- 

 ring the last century, warrant the inference made in this 

 Scholium. As measured by Whiston in 1724, it was 75'^' 

 10' : and nearly accordant with this result is that of Gra- 

 ham, obtained in the following year. Cavendish, in 1775, 

 found it to be 72° 30', and Gilpin, in 1805, 70° 21'. These 

 observations, after every allowance is made for the imper- 

 fection of the instruments employed, leave no doubt that 

 the inclination of the needle has undergone a gradual dimi- 

 nution in London, during the last century. According to 

 M. Humboldt, (see Biot— Traite de Physique, IIL 136,) a 

 similar diminution has taken place, during the same period, 

 in France. 



Book VL Prop. 13. Schol. 1. The statements concern- 

 ing the ratio of the sine of the angle of incidence to that 

 of deviation, in passing to and from water and glass, are 

 true only under a limitation which is not distinctly pointed 

 out, — namely, that the angle of incidence is indefinitely 

 small. 



Schol. 2. The partial reflection of light by the second 

 surface of transparent media, when the angle is within the 

 limit for light to be refracted, is erroneously ascribed to 

 " inequalities" of the surface. If this were the true cause, 



" Philos. Trans. 1816. 



