94 ON DEPARTURES FROM FRESNEI/S LAWS OF REFLEXION
calico, stretched over a glass plate, soon reduced the angle to 4° or 3e further progress seemed more difficult. Comparisons were rendered s what-uncertain by the fact that different parts of the surface gave va: numbers. After a good deal of rubbing, a' - a was reduced to such fij as 2°, on one occasion apparently to 1£°. Sometimes the readings taken without touching the surface after removal from the rouge, at o the face was breathed upon and wiped. In general, the latter treat seemed to increase the angle. Strong sulphuric acid was also tried without advantage, as also putty-powder in place of or in addition to r The behaviour did not appear to be sensitive to moisture, or to appreciably when the surface stood for a few days after treatment.
Thinking that possibly changes due to atmospheric influences mig nearly half a century have penetrated somewhat deeply into the ( I re-ground and polished (sufficiently for the purpose) one of the. origi unpolished faces of the prism, but failed even with this surface to re a-a below 2°. As in the case of the diamond, it is impossible to ; absolutely that o! - a. cannot be reduced to zero, but after repeated I had to despair of doing so. It may be well to record that the refrii index of the glass for yellow rays is 1'6 80.
These results, in which k (presumably positive) remained large in of all treatment, contrast remarkably with those formerly obtained 01 refractive glasses, one of which, however, appears to contain load. Il then found that by re-polishing it was possible to carry k down to '/AM to the negative side, somewhat as in the observations upon wafco.r i( possible to convert the negative k of ordinary (greasy) water into one M small positive value, when the surface was purified to the utmost.
There is another departure from Fresnel's laws which is observed wl piece of plate glass is immersed in a liquid of equal index*. Under circumstances the reflexion ought to vanish.
The liquid may consist of benzole and bisulphide of carbon, of whic first is less and the second more refractive than the glass. If the IK inent is for the yellow, more benzole or a higher temperature will tak ray of equal index towards the blue and vice versd. "For a closer o: nation the plate was roughened behind (to destroy the second ruHc.xiorf was mounted in a bottle prism in such a manner that the incidence be rendered grazing. When the adjustment of indices was for the yi the appearances observed were as follows: if the incidence is pretty ob the reflexion is total for the violet and blue; scanty, but riot evauescer the yellow; more copious again in the red. As the incidence becomes and more nearly grazing, the region of total reflexion advances from the
* "On the Existence of Reflexion when the relative Refractive Index is Unity," lirit. Report, p. 585 (1887); Scientific Papers, Vol. m. p. 15.as was to be expected, the utilization of the whole number (ten)