478 
DR. WALTER NOEL HARTLEY ON THE ABSORPTION SPECTRA 
connections being of metal or of glass. It must not be overlooked that the time 
occupied in making the analyses was much longer than in taking the photographs, 
and as the vapour had to be transferred from one vessel to another, this admitted of 
changes occurring in the composition of the vapour, by condensation at one period and 
saturation at another. Moreover, at low pressures there was a difficulty experienced 
in extracting all the vapour from the tube and submitting it to analysis. 
The Absorption Spectrum of Benzene Vapour at Constant Temperature 
and Reduced Pressures. 
In the first experiments at reduced pressures a glass tube was used with glass taps, 
the same arrangement being used for vaporising the benzene, but the outlet tube, 
instead of being connected with an aspirator, was connected by a glass T-piece with 
a filter-pump and a pressure gauge. Air was drawn through the benzene and 
regulated by the flow from the pump until the mercury in the gauge was steady; the 
exhaustion still proceeding, the exposure of the plate was made, after which the 
exhaustion was continued until a steady and further reduction in the pressure .was 
obtained, when another exposure was made, and so on. In this way the vapour was 
continually being removed at one end of the tube and renewed at the other, but 
always at the same pressure for each exposure. Photographs were taken at tempera¬ 
ture 1T J, 5 C., and 778 mm., 483 mm., 253 mm. and 21 mm. pressure. It was observed 
that in the first series of experiments the photographs taken at reduced pressures 
were all somewhat weak, and the tube for working at low pressures would probably 
be improved by being lengthened to render the absorption bands more distinct. The 
spectra were photographed on Cadett and Neale’s spectrum plates, developed with 
ferrous oxalate, and cleared with a solution of alum acidified with nitric acid. Some 
few photographs were taken on “ Rainbow ” Warwick plates, the period of exposure 
being the same in each case, namely, two minutes. 
The Measurement of the Spectra. 
The plates were measured with a micrometer made many years ago by the late 
Mr. A. Hilger, the screw of which, according to my instructions, was made of 
platinoid, because this metal is rustless. The pitch of the screw is 100 turns to the 
inch. It carries a microscope made by Zeiss, which has fine definition and a flat 
field. I’he instrument is fitted on a stand tilted at an angle of 45° C. to facilitate 
prolonged observation through the microscope. Measurements to the l/l0,000th of 
an inch are easily made; but in the case of absorption bands this cannot always be 
accomplished, because it is difficult to determine, under the necessary magnification, 
where the bands begin and end. The wave-length at the edges on each side of a 
band were measured whenever they were broad enough and well-defined. In some 
