470 MR. A. E. TUTTON ON THE THERMAL DEFORMATION OF THE CRYSTALLISED 
twisting round the centre or any other irregularity, throughout the whole of the 
interval of time. Frequently slight cracking of the crystal is accompanied by 
widening or narrowing of the bands during one part of the observation and movement 
in the inverse direction during the other, or possibly by twisting for a whole revo¬ 
lution, and very frequently l)y merely jumping several bands, the appearance at the 
end Ijeing much the same as at the beginning. Such an observation is, of course, 
valueless, but the Abbe method would not detect this fact. Undoubtedly, the 
author’s method, although very fatiguing, is the only one which is trustworthy 
when fragile substances are under investigation. 
The counting of the hands was achieved precisely as described in the dilatometer 
memoir (p. 348) with the aid of the tape-puncturing recorder, the induction coil 
which illuminated the hydrogen Geissler tube being actuated at sufficiently rapidly 
succeeding intervals to enable the author to observe the passage of at least every 
f[uarter of a band. Timing the transit with the watch is an excellent aid, as, if the 
observation is trustworthy, there should be no sudden changes of rapidity in the move¬ 
ment of the bands. When the Fletcher ring-gas-burner below the double air-bath is 
first ignited, the bands move very slowly, the rapidity then growing with a regular 
increment until it reaches, in the case of large expansions where at least forty bands 
pass during the interval of 45° of temperature, a maximum of two bands per minute ; 
the rapidity then as gradually diminishes until, with the attainment of constancy at 
the higher limit for that particular interval, the bands cease to move altogether. 
Moreover, if the temperature recorded by the inner bent thermometer, whose bulb 
is tied to and in contact with the platinum-iridium tripod, shows any slight tendency 
to descend a fraction of a degree, the bands should immediately begin to retrace their 
steps to a corresponding extent. No observation has been accepted during which 
these conditions were not fulfilled. 
The temperature limits em|)loyed were respectively the ordinary temperature, 
o])tained as low as j)0ssible by commencing work about 7 A.M., the neighbourhood 
of 56°, and that of 96°. The determinations of the positions at these temperatures, 
of the two bands nearest to the reference point, were made precisely as described in 
the previous memoir {p. 346). The monochromatic light employed throughout was 
red hydrogen light, corresponding to the C line of the solar and hydrogen spectrum, 
separated from all other radiations by a train of prisms in the manner described in 
the dilatometer memoir (pp. 322 and 342). The wave-length of this radiation 
employed in the calculations was 0'0006562 millim. 
When adequate time for complete cooling had elapsed, after the second series of 
observations, the measurement of the thickness of the crystal and the length of the 
tripod screws was made, l:)y means of the thickness measurer described on p. 337 of 
the former memoir. For this purpose the interference chamber was carefidly raised 
out of the bath by means of the rackwork on the pedestal, and the tripod, together 
with the supported crystal and compensator, after cutting the thread binding the 
