Conditions of Chemical Change. 177 



cules * consisting of an unknown and probably great number 

 of potassium atoms," even though the determination of its 

 molecular weight, alike by the methods of vaporization, the 

 depression of freezing-points, and diminution of vapour- 

 pressure, would point to a molecular number 39*1, if that of 

 hydrogen be taken as 2. Thus the only point of differentiation 

 is the assumed enormous electric charge. But even if it be 

 granted that such a charge would alter the properties of 

 a kind of matter (of which, be it remarked, there is no expe- 

 rimental evidence) precisely as yellow phosphorus is converted 

 by the process of heating into red phosphorus, yet the dif- 

 ference between these two last modifications is for the most 

 part one of degree and not one of kind. But we are asked to 

 believe that ionic potassium would not decompose water, and 

 that ionic chlorine would possess no odour. 



The propounders of this hypothesis take away attributes 

 hitherto regarded as essential to the various elementary kinds 

 of matter, and substitute for them mere mental abstractions. 

 As such they might be useful in the same sense as circles and 

 triangles, and we might possibly deal with them as implied 

 postulates ; but to predicate properties of kinds of matter of 

 which we can have no cognition, and to differentiate them 

 from kinds of matter of which we have cognition, is to carry 

 scientific imagination beyond its legitimate bounds. The 

 following sentences taken from Ostwald's Lehrbuch illustrate 

 the relations given in the text : — (i) " Chlor is ein stark 

 riechendes Gas ; " (ii) " Das molekulare Chlor, Cl 2 , gelit in 

 Chlorionen iiber;" (iii) " Ebenso muss in Gefasse B ein 

 Ueberschuss von Chlorionen vorhanden sein ... in keinem 

 Gefasse nimmt man etwas besonderes wahr — weder . . . noch 

 Chlor geruch in B." 



Though the element chlorine when combined with potassium 

 in potassium chloride or with hydrogen and carbon (in 

 chloroform, for example) has none of the properties of ele- 

 mentary chlorine by itself, yet this would not affect the 

 question ; for by the ionic dissociation theory the chlorine in 



* It is time to protest against such a sentence as this in a book (Nernst, 

 cf. supra) written presumably, as our English phrase hath it, " for the use 

 of schools and colleges": — "Wasserstoff und Chlor sind uns beide bei 

 gewohnlicher Temperatur nur als H 2 und Cl 2 hekannt." We know the 

 properties of hydrogen — or at least we have a rational expectation, 

 amounting almost to a certainty, that the properties of a specimen 

 of hydrogen collected to-day are the same as those of a specimen of 

 hydrogen collected fifty years ago ; but we do not know hydrogen as II 2 . 

 All that we can say is this, that, adopting certain hypotheses of the con- 

 stitution of matter, we believe, or we imagine, or at best we conventionally 

 represent the molecule of hydrogen to be H a . 



