Chemistry and Physics. 263 
Of which the last five possess the property in a very slight degree only. 
The test used in every case was the dislocation of the rings of a 
plate of calc spar; of which a very good specimen was used, capable 
of exhibiting eight or nine red rings: and all the experiments were 
made by candle-light, which is indispensable. It will secure greater 
confidence in these results to say, that all the specimens which I submit- 
ted to Prof. Poweli’s examination, in a different instrument, were found 
by him to produce the above effect; and from his published observa- 
tions several more cases may be quoted in confirmation of the general 
result: such are—chromate of lead, litharge, plumbago, and Indian 
ink. The natural conclusion from these facts appears to be, that in a 
perfect mathematical theory of reflexion, both cases should be embra- 
ced in one set of formule, of which some terms or coefficients should 
be insensibly small, except when the refractive index was very large ; 
that, strictly speaking, no substances completely polarize common light 
at any angle, but that the residue of unaltered light is too feeble to af- 
fect the eye, when the refractive index is below a certain limit ;-—-and 
that plane polarized light always becomes elliptically polarized, but that 
the virtual difference of paths of the two compact vibrations ~-parallel 
pencils which have traversed it. It is to be provided with a 
frame and screw, capable of compressing it in the middle. (A similar 
apparatus has already been employed by Brewster, Ling an ouillet, 
to show that glass under pressure sses double refraction.) We 
may now proceed to the experiment itself. Let us suppose, then, that 
the arrangements have been made in a darkened room for producing 
the interference of two pencils of light, which are to be polarized in 
