PHYSICS: C. BARUS 
361 
the fixed horizontal shadow in the spectrum due to the fine wire across 
the slit. It does not occur when rays retrace their path. The follow- 
ing experiment was made to coordinate the vertical displacement of the 
component rays and centers of elKptic fringes. A glass plate d = 0.1QS 
cm. thick was placed nearly normally in the beam hd, figure 1, and pro- 
vided with a horizontal axis and graduated arc. The amount (i) of 
rotation of the plate, corresponding to the vertical displacement of one 
central fringe in the telescope (i. e., the passage of the ring a, into h and 
then into in the spectrum, S, figure 3) was then found to be, if i is 
the angle of incidence, 
i h Ah 
No fringes, c fig. 3 3.5° 0.0149 cm 
One fringe, & fig. 3 5.0"* 0.0214 0.0065 cm. 
Two fringes, a fig. 3 6.5° 0.0281 0.0067 
where h is the corresponding vertical displacement of the rays bd (see 
figure 2) and computed from (m index, r angle of refraction) 
h = d (sin i — cos i tan r) 
Thus the vertical displacement of rays corresponding to the vertical 
semiaxes of the central ellipse or one fringe is between 0.0065 to 0.0067 
cm., i. e., a superior limit would be about 7 X 10~^ cm. Hence h = 
N 0.007 for N such central fringes is an excessive estimate. 
The question is now suggested, in how far such an arrangement 
would fall short of being able to exhibit the drag of the ether in a 
rapidly rotating body, should such drag occur. In figure 1, let a jS be 
a cylinder of glass with plane parallel ends, capable of rotating on the 
axle yd. If / is the length of the cylinder fi its index of refraction, and 
and r the distance of either component ray {ac, bd) from the axis y d,n 
the number of turns per second and V the velocity of light, we may 
write using the above estimate, N being the number of fringes displaced 
h = i N X 0.007 = lirnrl fi/V 
since ac rises while bd falls. li n = 200, r = 10 cm., 1 = 100 cm, V 
= 3XlOi«, M = 1.5 
N = 0.018, nearly. 
It would thus be necessary to estimate about 1 /60 of a fringe, which is 
just beyond the limit of certainty even if n, r can be increased, and / 
multiplied by reflection. The device suggested is nevertheless of inter- 
est and deserves further consideration. In fact I have succeeded in in- 
creasing the sensitiveness about four times. 
