481 



required f'oi' dixisioii, diluiioii, .solulion, melting and evaporation 

 remain of subordinate imporlance '). Hence notwithstanding the 

 complicated character, we have to do with : 



1. comparatively characteristic reaction temperatures; 



2. with the decisive influence of melal-formations. 



We can now divide I again into the following more elementary 

 processes : 



a. MCO, = MO + CO, c. + C = CO 



b. MO = M -f (I. C + CO, = 2 CO 



in which c and d follow the same reaction equation for all metal 

 cyanide-syntheses, whereas h is first of all the reaction which claims 

 the lion's share of the energy-supply. 



c. The primary reaction. From W. Kossel's *) point of view the 

 course of l> means only this, that under the influence of the supplied 

 energy in the metal oxide the oxygen cedes again the negative elec- 

 tron taken from the metal. In our case this view entails the diffi- 

 culty that this restitution would have to take place more easily 

 wiih metal atoms with relati\ely great aflinity to the leturning 

 electron than to more electro-positive elements. In reality, however, 

 the reaction appears to set in more easily with increasing electi'o- 

 positivity of the elements. 



It is, however, not necessary to assnme that in every metal- 

 metalloid com|)onnd one or more of the metal-eleclrons has qnite 

 gone over to the metalloid. In many cases it may he more a question 

 of partial transition, i. e. conditions will be found in which only 

 partial separation ("Lockernng", dislocation) of the metal- (valency-) 

 electron with regard to the metal atom must be assumed. In con- 

 nection with the spectral interpretations by Hohr this may also be 

 expressed thus, that the electron in qnestion will on an average be 

 at the disposal both of the oxygen atom and of the metal atom in 

 a characteristic "abnormal" path. 



The decreasing reaction lemperatnre with the increasing electro- 

 positivity of the metal leads us further to the assnmption that pri^ 



1) In the same way the dependence of the reaction-energy on the temperature 

 may be neglected within a wide margin. The possibility of all the.se approximations 

 stamps the cyanizrngreaction as one of the few chemical conversions, which like 

 the rare ideal photo-chemical reactions are able to give experimentally demon- 

 strable indications in favour of the theories which at the same time bear a funda- 

 mental and idealizing character [of. e. g. F. Weigert, Zeitschr. f. Phys. 14, 

 383 (1928)]. 



") W, KossEL, Ann, d. Phys. 49, 229 (1919). 



