320 
DR. E. H. GRIFFITHS AND MR. EZER GRIFFITHS ON THE 
Introduction. 
In two previous papers* accounts have been given of experimental investigations 
into the variation with temperature of the specific heats of certain metals, the range 
of temperature covered being, in general, from 0° C. to 100° C., or, in the case of 
sodium, to nearly 140° C. 
The present paper is a further contribution to the same subject, the temperature 
range being continued down to about —160° C. 
As pointed out in the previous papers, determinations of the true specific heats 
throughout this range are needed to bridge the large gap between Nernst’s observa¬ 
tions at liquid hydrogen and liquid air temperatures and the observations above 
referred to at higher temperatures. 
For determinations in the vicinity of fixed points at low temperatures the method 
employed by Nernst is an admirable one and his experimental values carry great 
weight. 
He has given, in the cases of silver and lead, the results of investigations at 
higher temperatures ( e.g ., at —78° C.), but these determinations cannot, for several 
reasons, be regarded as satisfactory. The validity of this conclusion is indicated by 
the fact that individual values at the same temperature differ by about 4 per cent. 
PART I. 
Experimental Methods and Results. 
(l) Description of Apparatus. 
Our method of investigation demanded a constant temperature enclosure, and the 
apparatus described in succeeding pages was designed for this purpose. 
We abandoned the use of liquid air on account of the uneconomical nature of the 
processes that would be involved, and utilised the Joule-Thomson cooling effect in a 
direct manner. 
Fig. 1 is a sectional view of the apparatus. Compressed airt at a pressure of 2000 to 
3000 lbs. per square inch entered the interchanger by the pipe A. This interchanger 
BB was constructed of ordinary solid drawn copper tubing, bore, coiled in the form 
of flat spirals. Successive layers of the coil were separated by strips of cardboard and 
the entire coil was inserted into a length of motor tyre tubing which fitted it closely. 
From the inter changer coils the air was carried to the valve C, by means of which an 
observer controlled the flow, excess of air being discharged at the safety valve on 
the compressor. Although not shown in the figure, a Hook’s joint was inserted in 
the spindle controlling the valve. 
* ‘ Phil. Trans.,’ A, 500, vol. 213, p. 119, 1913;' ‘Roy. Soc. Proc.,' A, vol. 89, p. 561, 1914. 
t Freed from moisture and CO^. 
