at the Royal Institution, 1 900-1 907. o95 



by the oscillatory uioveiueiit at each junction ; as the temperature 

 falls, the flanges come more nearly into apposition and ultimately 

 coincide. Conductiyity is then at its maximum, the rate of flo\y 

 depending alone on the sectional area of the tube and the skin 

 friction, being no longer checked by the oscillations of the tubes. 



Thermometry at Low Temperatures. 



Owing to the fact that the rate of change of resistance of 

 platinum becomes gradually smaller at yery low temperatures, the 

 accurate determination of low degrees of temperature cannot well be 

 effected by means of resistance thermometers. 



An elaborate inyestigation of a number of constant pressure gas 

 thermometers has been made from which it appears, that either a 

 simple gas such as helium, hydrogen or oxygen or a compound gas 

 such as carbon dioxide, at an initial pressure somewhat less than one 

 atmosphere, may be made use of as the thermometric substance in 

 determining temperatures near to but above that at which it boils. 



The average value of the boiling point given by these experi- 

 ments in the case of oxygen is - 182° '5 and in the case of 

 hydrogen -252° '5 or 20° '5* absolute. The value for oxygen 

 is in agreement with the mean results obtained by Wroblewki, 

 Olzewski and others. 



The melting-point of liydrogen, determined by a helium thermo- 

 meter, is 15° absolute. 



A careful study of a large number of resistance thermometers 

 has shown that at low temperatures these give yaria])le results and 

 that no thermometer of the kind will afford accurate values up to 

 and below the boiling-point of hydrogen. At the boiling-point of 



* lu the interest of historical accuracy, it should be pointed out that this 

 value was first given in the Bakerian lecture delivered to the Royal Society 

 on February 7, 1901, published in their Proceedings, vol. Ixviii,, pp. 44-54. 



If reference be made to the third English edition of Ernst von Meyer's 

 History of Chemistry from the earliest times to the present day (Macmillan), 

 an authoritative work published in 1906, the following passage will be found 

 (p. 523) :— 



" Dewar was the first to succeed in obtaining a measurable amount of 

 liquid hydrogen (about 50 c.c. at one time) and he has since then been able to 

 solidify it. An apparatus has also been designed by Travers by means of 

 which liquid hydrogen can be obtained in quantity. 



Liquid hydrogen is clear and colourless ; it shows no absorption spectrum 

 and the meniscus is as well defined as in the case of liquid air. The boiling- 

 point was first determined by Dewar with a platinum resistance thermometer 

 to be — 238° C. but more recent determinations by Travers, using a helium 

 thermometer, have given —252*5° C, a number now accepted by Dewar." 



As a matter of fact, Travers and Jaquerod make the following statement 

 ^p. 489) in a communication read to the Royal Society on June 19, 1902 

 Proceedings, vol. Ixx., pp. 484) : — 



