Haughton — Notes on Newtonian Chemistry. 639 
The general shape of the curve is shown in figure 4, where + w', 
— is the axis of x, and -f n^, + n^, the axis of y. The axis of w' is 
an asymptote, and there are two other asymptotes AB and CD 
parallel to the axis of y, and lying at distances from the origin 
+ a and - h corresponding to the real roots of the quadratic 
in o)'. 
There are three real branches of the curve, one lying between AE 
and 0, w'; the second between CD and 0, - oi)'; and the third lying 
between AB, CD, and the axis of x. Of these branches the first two 
are to be rejected, because they correspond to negative values 
of n^. 
Of the remaining third branch of the cubic curve, the right-hand 
side corresponding to positive values of oi' must be rejected, because 
is never greater than unity, and the point of the curve which de- 
termines w' must lie very high up on the left-hand branch, because 
must be an enormously great number. 
Fig. 4. 
5. Mydroflmric Acid (HF). — The probabilities are very strong 
that the molecular volumes of fluorine and hydrofluoric acid are the 
same as those of hydrogen, chlorine, and hydrochloric acid. 
Assuming this to be the case, I proceed to determine the rotations 
of fluorine and hydrofluoric acid by means of equations (4)' and (5)', 
making ^ = 19 and writing for 22000 the constant 37600, which 
