46 



PROCEEDINGS OF THE AMERICAN ACADEMY 



Co., 



CI 



NIT— X IT..— CI 

 Nil,— XH3— Nil— CI 

 ) Nil,— NH— NIL— CI 

 NII-NII3-CI ' 

 CI 



and Co, < 



XH.,— CI 

 js^H.'- NTT— CI 

 Nil"— nit'— NH..— CI 

 NH.!— NH.!— NTT.;— CI 

 NH.!— NH.— CI " 

 NH.,— Cl 



As I have adopted this view throughout, further illustrations will be 

 unnecessary. The advantages of the theory are, I think, first that it 

 regards cobalt as tetratomic, and reduces the whole series virtually to 

 the type of C02O3, or Co^Cl,;, and secondly that i; enables us to explain 

 the different cases of isomerism without arbitrary assumptions as to 

 the existence of allotropic forms of cobalt. Thus it is easy to see that 

 there may well be a difierence between compounds having respectively 

 the symmetrical structural formulas, — 



f NH3— Cl 

 I NH.,— NH3— Cl 

 jr.] NH.!— NH.,— Cl 



I- '^o, s ]srH _xii^_ci 



NH 



r CI 



NH3— NH,— CI 



II. Co, i 



NH3-CI 



NH3-(;i 



N H,— N H„— N TL— CI 

 N h!!— N 1 1 J— N H3— Cl 

 NH3— NIT..— Cl 

 Cl 



f Cl 

 NH„ 



nh!! 



"I- c^^ s;;:-Sn:: 



Cl 



NIT..- NIL 



-NIT.,— CI 



NH3— CI 

 Cl 



-NH3— NTT3— Cl 



IV. Co., y 



Cl 

 Cl 

 NH3— NH3— NH,— NTT. 



NH., 



Cl 

 Cl 



-NH, 



-XH._CI 

 -NH.,— NTI3— NII3— Cl 



and that supposing that the six units of affinity of Co^, are qu;ditatively 

 exactly e(iual, the complex, Co^,(NIT.,),„Cl,;, admits of being arranged 

 in a great variety of ways. Now since the salts of roseocobalt and ]uu- 

 pureocobalt resemble each other as regards the types to which they 

 belong and as eitlier of these types may be fornnilated in a great 

 variety of ways, it becomes vci-y dillicult to di'tcrmine which strucliiial 

 fornnda to adopt for the com|)oun(ls of either series. 



Thus the salts of roseocobalt and of jxirpureocobalt. nsually at least, 

 belong eitlier to the types RcX^ and I^•X^., to RcX^Y^ and rcX^\\„ 



