572 



NA TURE 



[October 8, 1908 



The full circles show the apparent positions of the 

 sun due to refraction, at different horizons, if we apply 

 the refraction correction and consider the sun visible 



degree of apparent change of place brought about in 

 this way; and how the difference between the true 

 and apparent places rapidly diminishes as the true 



SUHNIER :.S0LST1'CE . 



48 - 

 47 - 



It 



_ : . AZIMUIHS,.- ::._: 



:L ■ -J: ■ Jr--rtf%5F^ 



G. 21.— The Azimuths of the Sunrise (upper limb) at th= Summer Solstice. The values given in the table have been plotteH, and the effect of the height 

 of hills on the azimuth is shown. The range of latitude given enables the diagram to De used in connectuii with the solstitial alignments at Cariiac, Le 

 IMcnac, and other monuments in Brittany. 



with 2' of its diameter showing, whilst the dotted 

 circles show the real position of the sun at the same 

 moment. Thus, considering the lowest full circle. 



horizon is left behind. Thus at the sea horizon the 

 true and apparent suns are just separated; with the 

 horizon i° high thev interlock. 



Fig. 22. — Azimuth of Suiiriie (upper limb) at Winter Sol&tice. 



in Fig. 19 we see that the azimuth of apparent sun- 

 rise, with a sea-horizon, is N. 37° 1' E. 



A comparison of the full-line circles with the lower 

 dotted circles in the diagram will give an idea of the 



XO. 2032, VOL. 78] 



The next diagram gives the conditions for lat. 50°. 

 In this latitude, while the sun appears to rise at 

 the present time over the true sea horizon at azimuth 

 N. 5oJ° E., instead of X. 37° E., as at Stenness, 



