Prof, Knoblauch on the Interference of Heat. 127 
. In order to diminish to the utmost the absorption in the tra- 
versed media, and thus to increase the intensity of the effects, 
a grate and lens of rock-salt were employed. With such a 
grate, containing 600 lines in an inch, the deflection for the 
eentre was 31°-0, and for the first spectrum 1°-5 ; these two lumi- 
nous bands were separated by a colder one corresponding to a 
deflection of 0°°3. With a finer grate the deflection was observed 
to be 17°-25 in the centre, 3°°5 in the first spectrum, and only 
0°-5 between the two. The deflections which with rock-salt 
remain on the dark bands are due to diffusion of the rays, which, 
with this substance, cannot be avoided. 
The above effects can certainly not be ascribed to accidental 
sécondary actions; for the several differences of temperature 
were still observable when the pile exposed the same surface 
during its displacements, or even when it presented a greater 
aperture to the rays in the dark bands. 
At the same time a new proof of the divergence of the rays 
of heat through inflexion may be deduced from these experi- 
ments. For whilst, without the grate, the enclosing limits of 
these rays at the place of measurement were, say, 2°5 millims. 
apart, after replacing the grate the extreme limits of heat were 
not even reached at a distance of 300 millims. on each side of 
the centre; consequently at places which were 600 millims. 
distant from each other. 
2. Difference of phase with the same length of path in consequence 
of the passage of rays through a body of unequal thickness. 
After interference-bands had been produced by introducing 
into the path of the solar rays an interference-prism in place of 
the grate, and a cylindrical glass lens in place of the achromatic 
one or the rock-salt lens, and after the thermo-electric pile had in 
this case, too, distinguished in the most unmistakeable manner 
the dark bands from the neighbouring light ones by a deflection 
of from 0°25 to 1°25 at the multiplier, a somewhat conical 
strip of glass was interposed behind the interference-prism, in 
such a manner that the rays of heat, in order to traverse one- 
half of the same, had to pass through a greater thickness of 
glass than was necessary in order to traverse the other half. By 
this means a displacement of the interference-bands was pro- 
duced ; for the thermo-multiplier indicated a decrease of tempe- 
rature, on the introduction of the glass, when the thermic pile 
was at a place of original maximum of heat, and an increase of 
temperature when the pile was situated in one of the former cold 
bands: this latter fact is the more significant, since the action 
was there opposed by the absorption of the interposed glass. With 
respect to interference, therefore, the influence of unequal thick- 
