PHYSICS: C. BARUS 
669 
corresponding rays d and d' were obtained by cleaving the white beam 
from a collimator symmetrically, by a knife edged prism with silvered faces. 
4. Interferences from Rough Surfaces. — The question now at issue is 
whether these interferences can be retained when the collimator is 
removed and the Hght comes directly from a ground glass surface or a 
Nernst filament. The spectrum fringes go at once when the slit is 
widened. Not so the achromatic sets. After obtaining this result 
with sunlight and ground glass, I replaced both by the light from a 
Nernst filament, under the impression that ground glass might, to a 
small degree, behave like plate glass. Having produced the achromat- 
ics as usual with the collimator, its objective was first removed and it 
was then seen that the two washed slit images were no longer super- 
posed. Bringing the images together (by rotation of the mirror n 
around a vertical axis), a position was soon found in which the achro- 
matic fringes appeared brilHantly in a white field, quite out of focus. 
The slit could now be widened or removed altogether, but the fringes 
persisted though with less brilliancy. It is thus possible to obtain these 
fringes directly from a Nernst filament or a narrow vertical strip of 
sunlight. They are so mobile with changes of A N and A a, that to 
find them it is necessary first to produce the spectrum fringes with col- 
limator and spectro-telescope, then to find the achromatics on removing 
the spectroscope, next to remove the objective of the collimator and 
adjust for superposed images and finally to remove the slit. They 
practically cover the whole width of the washed slit image with streamers 
extending laterally into the glare some five times further. SHt images 
may even be slightly separated while each alone retains the achromatic 
fringes, a rather puzzling phenomenon. One may note that the slit 
images here are not reversed. 
Experiments made as to the nature of these achromatic fringes 
showed that they are probably Fresnellian interferences. To prove 
this the objective of the collimator was removed and strong fringes 
obtained by passing the two vague images of the slit gradually over 
each other, horizontally. The fringes in this motion passed from hori- 
zontal maxima of size gradually to vertical hair lines as the images 
slid from contact of their nearer edges to contact of their further edges. 
The coarse fringes were even strongly present in the narrow dark gap 
between slit images prior to contact. The telescope was now focussed 
on the slit so that sharp linear images appeared. The fringe then 
vanished, but it appeared that the coarse fringes corresponded to coin- 
cident sharp slit images when observed out of focus, and the fine hair 
lines to sharp slit images far apart also seen out of focus. The whole 
