164 TRANSACTIONS OF THE [aPR. 1, 
This is an instrument simple enough in principle, but difficult in 
construction. Moreover it does not possess a very high degree 
of accuracy, owing to the fact that its use depends upon the de- 
tection, by means of a bi-quartz polariscope, of the exact point 
at which all polarization disappears from a beam of light. The 
beam of light to be examined for polarization is made to pass. 
through a pile of glass plates set in such a manner with refer- 
ence to the beam as to produce, by refraction, polarization in a. 
plane at right angles to the plane in which the beam is already 
partially polarized. The inclination of the plates is altered until 
the halves of the bi-quartz plate show no traces of color. The 
instrument having been previously graduated by means of a 
beam of known polarization, the amount of polarization in the 
beam in question can be immediately read off upon the graduated 
circle. 
The graduation is effected by passing a plane polarized beam 
through a thick crystal of Iceland spar. Then, by the Law of 
Malus, the polarization in the transmitted beam is given by the 
cosine of twice the angle included between the plane of polariza- 
tion of the incident beam and the principal plane of the crystal. 
Both because of the difficulty of construction of this instru- 
ment, and because my own experiments with the bi-quartz polari- 
scope made me distrustful of the accuracy with which the point of 
no polarization could be determined, another form of instrument 
was devised for these experiments which is greatly superior to 
the Arago polarimeter in simplicity, and is probably more than 
equal to it in accuracy. The instrument had been constructed 
and used for several months before the discovery was made that 
the credit of the first conception and use of this method of 
measuring the constants of partially polarized light is due to. 
Cornu. 
In view of the exceeding naturalness and simplicity, as well as 
the accuracy of the method, it is surprising that it was not earlier 
discovered and has not been more generally employed. Cornu’s 
description of his instrument was published in ’82 in the “ Ass’n 
Francaise pour L’Avancement des Sciences, Compete Rendu;” 
but so far as I can discover, no reference was made to it at the 
time in any of the scientific journals, nor has it taken its place 
among other polarimeters in any of the text books on optics. 
The instrument as constructed and used for the purposes of 
these experiments was as follows. <A rectangular opening O.,-1 
mm. in width and 2.5 mm. in length, was made in a diaphragm H. 
which stood a short distance in front of the double Wollaston 
prism A. The prism was rotated until the extraordinary image 
of the opening was to the left of the ordinary ; the distance of 
