W. Gibbs on the theory of Atomicities. 415 
"Priv a 
pur} Cle ®,(8NH) 
Devi 
« © No.2, PtpClO,Hn®2N, mpriy Cl, e(8NH)” 
Raewsky’s base No. 1, Pt,Cl,0,Hy,0O,N, 
If it be objected to the assumption that potassium, sodium, &e., 
are triatomic in certain combinations, that these elements have 
not yet been observed to form compounds with chlorine and 
Cl 
other usually monatomic elements of the form K 1 Cl, Ireply first 
Cl 
that thallium, which unquestionably belongs to the alkaline 
group, is distinctly triatomic in TICI,, and secondly that if chlo- 
rine itself be sometimes triatomic, the compound KCl may be 
represented as K™Cl" as well as K'Cl', each element being satu- 
rated upon either view. With respect to the elements belonging 
to the nitrogen group I will simply call attention to the fact that 
nitrogen is heptatomic in the hyperiodids of certain complex 
ammoniums and enneatomic in others. Thus we have 
w{ CH), ma wf CMs 
3 5 
so that nitrogen furnishes the most complete series of atomici- 
ties, being 1, 3, 5, 7 and 9-atomic. ; 
t is easily seen that there is a marked difference between 
the elements both of odd and of even atomicities as regards the 
maximum intensity of combining power. Thus pam Be. sodi- 
um, hydrogen, chlorine and some other elements have their maxi- 
when they become penta- 
tomic, and becomes nearly insensible when they are hepta- 
tomic. Finally, tantal d niobium have their maximum 
inally, tantalum and nl sity of five. (Ear 
H.)”. But we have, as I think, a right to in- 
