1895. ] NEW YORK ACADEMY OF SCIENCES. 171 
Accordingly a careful series of observations was made with 
the polarimeter. The experiments were all conducted in a well 
darkened room, and care was taken not to allow any light to 
enter the instrument except that emitted by the uranium glass. 
In order to make the determination of the angles of emission 
convenient, the light from the lantern was thrown Vertically 
down upon the surface of the uranium glass by means of total 
reflection in a right angled prism. The cube of glass was care- 
fully levelled so that the emitting surface was always horizontal. 
The arrangement of apparatus is shown in the figure. 
Ten readings were taken for every angle of emergence. The 
results are given in full. 
Bea 85° 80° 
Left.» Right. Left. Right. Left. Right. 
40.5 29.8 42.3 30.7 43.0 32.0 
39.5 29.9 43.0 BiEo 43.3 S225 
39.8 28.5 42.5 30.5 43.8 32.3 
40.0 29.0 41.7 30.9 44.0 32.6 
39.0 28.9 40.9 31.0 43.3 BIS 
39.76 29.22 42.1 30.86 43.48 32.34 
——— — — Oe 
2 w = 68.98° 2 w = 72:86° 2 w = 75.82° 
p= .358 p = .293 i— 240 
15°. 70° 65° 50° 
Left. Right. Left. Right. Left. Right. Left. Right. 
44.5 34.5 45.5 35.0 47.0 37 49.1 38.2 
45.0 35.0 46.4 36.4 47.2 36.3 48:9°5 38.9 
44.3 35.0 46.5 36.5 47.5 37.9 49.0 38.9 
44.5 33.5 47.0 35.0 48.0 317i 50.0 39.3 
44.5 34.4 46.3 36.2 47.4 36 49.3 39.0 
44.56 34.5 46.34 35.8 47.62 36.8 49.296 38.52 
——s>—— —— ——_ — 
2 w —79.06° PY nea=lewal 2 w = 84.4° 2 w = 87.782 
— Ok j= et p= .098 p = .039 
The chief difticulty encountered in making these determina- 
tions was the lack of perfect’ uniformity in the emitting sur- 
face. The uranium glass, being rendered self luminous by the beam 
from the lantern, could not have entire uniformity over its surface 
unless the illuminating beam was uniform in intensity, which was 
not the case. The images corresponded -to points on the sur- 
face not more than 2 mm. apart, and yet it was found that the 
equality of the images could be sometimes disturbed by directing 
the instrument toward anew portion of the field. As great care 
as possible was taken to direct the polarimeter toward such por. 
