W. LeConte Stevens—Lecent Progress in Optics. 281 
indirect method just described, but does not. surpass it in 
accuracy. Everyone knows that the meter is not an exact sub- 
multiple of the earth’s circumference, and that the determina- 
tion of its exact value from the seconds pendulum is full of 
‘difficulty. It may perhaps be said that the optical method is 
no more absolute than the pendulum method, for no human 
measurements can be free from error; that there is no possi- 
bility of the destruction of the original meter and all certified 
copies of it; and that there is no proof or probability that 
molecular changes are gradually producing modifications in 
standards of length. Even if we should grant that for all 
practical purposes the labor of determining the meter in terms 
of an unchanging optical standard has been unnecessary, the 
achievement is a signal scientific triumph that ranks with the 
brilliant work of Arago, Fresnel and Regnault. In prepara- 
tion for it much new truth has been elicited, and light waves 
have been shown to carry possibilities of application that 
Fresnel never suspected. 
The physicist is nearly powerless without the aid of those 
who possess the highest order of mechanical skill. The inter- 
ferential comparer could never have been utilized for such 
work as Michelson has done with it had not Brashear made its 
optical parts with such an approach to perfection that no error 
so great as one-twentieth of a wave length could be found 
upon the reflecting surfaces.* In the conception, mechanical 
design, and execution, the entire work has been distinctively 
American. 
The interferential refractometer has been used with much 
skill by Hallwachst for comparing the variation of refractive 
index of dilute solutions with variation of concentration. The 
fact of solution brings about a change of molecular constitu- 
tion, affecting both the electric conductivity and the refractive 
index; and the changes in optical density are measurable in 
terms of the number of interference fringes which cross the 
field of view for a given variation of dilution. 
Luminescence. 
While all work on the visible spectrum is confessedly optical, 
we can no longer make an arbitrary division point and declare 
that one part of the spectrum belongs to the domain of optics 
and the other not. Since the days of Brewster and the elder 
Becquerel fluorescent solutions have enabled us to bring within 
the domain of optics many wave lengths that were previously 
invisible. Stokes’s explanation of this as a degradation of 
* Travaux et Mémoires du Bureau internationale des Poids et Mésures, vol. xi, 
ed, 1895. 
+ Wiedemann’s Annalen, xlvii, p. 380, and liii, p. 1. 
