26 



PROCEEDINGS OF THE AMERICAN ACADEMY 



A B 



G into two pencils, one of which 

 passes without deviation through the 

 eyepiece and the hole in the eyestop 

 to the eye. The other pencil is 

 thrown to one side by C, and is cut 

 off by the eyestop. The light passing 

 through B is deviated about G'* by 

 the difference in inclination of its 

 two inclined sides. This lisht is 

 also divided into two pencils by (7, 

 one retaining the deviation imparted 

 by B, and being cut off by the eye- 

 stop. The other is deviated by C, 

 but in such a manner as to counteract the inclination imj^arted to 

 it by B. It therefore passes centrally through the hole in the eye- 

 stop to the eye of the observer. The latter accordingly receives two 

 pencils of light formed by the same object-glass, one receiving the 

 liglit from A, the other that from B. These two pencils are polarized 

 by G in planes at right angles, and their relative brightness may 

 accordingly be varied at will by turning the Nicol E. The instrument 

 in principle closely resembles the meridian photometer for some years 

 in use at this Observatory. The same device is employed to secure 

 two equal pencils polarized in perpendicular planes, but in that instru- 

 ment two equal object-glasses are employed, instead of two images ot 

 the same objective. 



The eyepiece is focused on the front surface of the prisms A, B, 

 so that their adjacent edges appear as a line dividing the field into two 

 equal parts. By turning the Nicol the brightness of either part of the 

 field may be reduced indefinitely, so that the brighter may always be 

 brought to equality with the fainter. Placing the whole instrument 

 at the principal focus of the telescope, we see side by side in the two 

 halves of the field images of objects really about 16' apart. 



The observations were made by placing the edge of the prism par- 

 allel to the sun's limb at the point nearest Venus, and bringing Venus 

 into one half of the field. A portion of the sun's disk near its centre will 

 be seen in the other half of this field, and may be compared directly 

 with Venus by turning the Nicol. Settings were made in the four posi- 

 tions of the Nicol in which the images appeared equal, and the posi- 

 tions read to tenths of a degi'ee. The observation was then repeated, 

 moving the telescope so that the portion of the sky close to the sun's 

 limb should be measured in the same manner. Eight settings taken 



