( 170 ) 



which is to be calculated from the isotherms, and the Joule-Kelvin 

 effect has therefore a sufficient value to yield an appreciable quantity 

 of liquid in a given apparatus in a delinite time. 



Three years ago I had so far advanced with the investigations 

 which led to the isotherms of helium, 1 ) that these determinations 

 themselves could be taken up with a reasonable chance of success. 



At first the great difficulty was how to obtain sufficient quantities 

 of this gas. Fortunately the Office of Commercial Intelligence at 

 Amsterdam under the direction of my brother, Mr. 0. Kameklingh 

 Onnes, to whom I here express my thanks, succeeded in finding in 

 the monazite sand the most suitable commercial article as material 

 for the preparation, and in affording me an opportunity to procure 

 large quantities on favourable terms. The monazite sand being 

 inexpensive, the preparation of pure helium in large quantities 

 became chiefly a matter of perseverance and care. *) 



The determination of isotherms of helium was not accomplished 

 before 1907. 



The results of the determinations of the isotherms were very 

 surprising. They rendered it very probable that the Joule-Kelvin 

 effect might not only give a decided cooling at the melting point 

 of hydrogen, but that this would even be considerable enough to 

 make a Linde-Hampson process succeed. 



Before the determinations of the isotherms had been performed 

 I had held a perfectly different opinion in consequence of the failure 

 of Olzewski's and Dewar's attempts to make helium liquid, and had 

 even seriously considered the possibility that the critical temperature 

 of helium, might lie if not at the absolute zero-point, yet exceedingly 

 low. In order to obtain also in this case the lower temperatures, 

 which among others are necessary for continuing the determi- 

 nations of isotherms below the temperatures obtainable with solid 

 hydrogen, I had e.g. been engaged in designing a helium motor (cf. 

 Comm. N°. 23) in which a vacuumglass was to move to and fro 

 as a piston in another as a cylindre. And when compressed helium 

 was observed to sink in liquid hydrogen (Comm. N°. 96, Nov. 1906) 

 I have again easily suffered myself to be led astray to the erroneous 

 supposition of a very low critical temperature. 



In the meantime I had remained convinced that onlv the deter- 



l ) [Clomp, for cryostats : Comm Ns. 14, 51, 83, 94, for thermometry: Comm. 

 Ns. 27, 60, 77, 93, 95, 99, 101, 102, for manometers, piezometers and deter- 

 mination of isotherms : Comm. Ns 44, 50, 69, 70, 71, 78, 97 99, 100J. 



~) [Even the quantity of 200 liters (and 160 liters of reserve) of the extreme 

 purity required, though requiring a great deal of labour, was not out of reach]. 



