92 
Proceedings of the Royal Irish Academy. 
If ^^e use the atomic weights and valencies of these nine elements 
(taking scandium = 44*3), we can find the equation of the cubic 
passing through these points, the form of which I have plotted in 
Plate VIII., and which has one real asymptote like the former cubic, but 
has a much more complex shape. ^ The dyad lines, both positive and 
negative, have a remarkable relation to the cubic curve. They are 
both very near the position of the horizontal tangent. 
In fact, the positive dyad line intersects the cubic in three real 
points, viz. : — 
79, 6.5-064, 65. 
The first of these is the element selenium, and the third the element 
zinc, with another element almost identical with zinc in atomic weight, 
both points being nearly on the horizontal tangent. 
The negative dyad line intersects the cubic in three real points 
also, viz. : — 
40, 52, 56 ; 
which represent exactly the atomic weights of the three elements — 
calcium, chromium, and iron. 
Now, we must remember that the cubic curve was constructed 
without any reference to iron, or its atomic weight, and it is very 
remarkable to find it taking its place on the curve after chromium, 
and near nickel and cobalt, which are not far ofi the dyad line. 
In fact, the cubic curve points out, by clinging to the dyad line, 
from 52 to 59, the possibility of forming, in that interval, a number of 
elements similar to each other in physical and chemical properties. 
These elements have been actually formed, and are — 
Chromium, Iron, Mckel, Cobalt. 
Excepting these four elements formed on the cubic branch of the curve, 
the distribution of the remaining elements, between the lineal and 
<jubic branches, is similar to that found in the curve described in 
Part I. 
The first three elements are on the cubic, viz. — 
K I Ca I Sc. 
The last three elements are also on the cubic branch, viz. — 
As I Se I Br. 
I venture to call it a serpentine 'whiplash' cubic. 
