( 115 ) 



in all possible dirci'tioiis would give tlie fullest pmniise, notwilli- 

 standing- the great difüculties which are encountered when woi-king 

 with liquid helium. It is therefore all the nioi-e necessary to esta- 

 blish beyond all possibility of doubt the property of which advantage 

 would be taken in such experiments. With this end in view niodifed 

 measurements are being made. 



It is further worth noting that just as the resistance of constantin 

 changed but little when the temperature fell from ordinary to li(inid 

 hydrogen temperatures so too the change is slight as the tem|)eratiire 

 sinks further to those of liquid helium. This proi)erty was utilised to 

 obtain rough confirmation of the value of the latent heat of vapori- 

 zation of helium, which can be calculated from Clapeyron's formula 

 using the data which have already been published ccmcerning its 

 vapour pressure and vapour density. (Conq^are the above remarks 

 as to the ratio between the Joule heat development to the latent 

 heat of the liquid which is used for cooling). 



Physics. — ''Researches on Magnetism. 111. On Para- and 

 Dia-magnetism at very low temperatures.'' By H. Kamereingii 

 Onnes and Albert Perrier. Communication N". 122'^ from the 

 Physical Laboratory at Leiden. 



§ 1. Tntroduction. In an earlier research (Comm. N". JK), April 

 1910) w^e investigated the magnetisation of oxygen dow^n to lem])e- 

 ratures close to the freezing point of hydrogen, and we foimd 

 deviations from Curie's law which seemed to us to be connected 

 with the problem as to how far the electrons which occasion 

 magnetic phenomena are frozen fast to the atoms when the substance 

 is cooled to very low teiuperatures^). This made it very desirable 

 to extend the research, especially as far as hydrogen temperatures 

 were concerned, to other paramagnetic substances which follow 

 Curie's law at ordinary temperatures. If, as we found to l)e the 

 case, it were found that deviations of the same character as those 

 for oxygen were encountered with these substances also: it might 

 well be assumed that such deviations, and also, should they be 

 connected with the freezing of the electrons, this freezing itself 



V This idea of which repealed use has been made in former Communications 

 (cf. Weiss and Kamerlingh Onnks, Comm. N'. 114, Febr. lülO p. f) note 1) did 

 not appear to be applicable to the case of ferromagnetic substances (loc. cit. p. 9) 

 as the temperature sank to the freezing point of hydrogen. Our experiments give 

 rise to the further question as to how these substances would bebave when tiie 

 temperature was lowered still further. 



