Commentary on Proposition VIII.21¶
Back to Proposition VIII.21.
The converse of VIII. 19. If $a, c, d, b$ are in geometrical progression, $a, b$ are “similar solid numbers.”
Let $\alpha, \beta, \gamma$ be the least numbers in the ratio of $a, c, d$ (and therefore also of $c, d, b$). [VII. 33 or VIII. 2]
Therefore $\alpha, \gamma$ are prime to one another. [VIII. 3]
They are also “similar plane numbers.” [VIII. 20]
Let
$$\alpha = mn, \quad \gamma = pq,$$
where
$$m : n = p : q.$$
Then, by the proof of VIII. 20,
$$\alpha : \beta = m : p = n : q.$$
Now, ex aequali,
$$a : d = \alpha : \gamma, \qquad \text{[VII. 14]}$$
and, since $\alpha, \gamma$ are prime to one another,
$$a = r\alpha, \quad d = r\gamma, \quad \text{where } r \text{ is an integer.}$$
But $\alpha = mn$: therefore $a = rmn$, and therefore $a$ is “solid.” Again, ex aequali,
$$c : b = \alpha : \gamma,$$
and therefore
$$c = s\alpha, \quad b = s\gamma, \quad \text{where } s \text{ is an integer.}$$
Thus $b = spq$, and $b$ is therefore “solid.” Now
$$\begin{aligned} \alpha : \beta = a : c &= r\alpha : s\alpha \\ &= r : s. \end{aligned} \qquad \text{[VII. 18]}$$
And, from above,
$$\alpha : \beta = m : p = n : q.$$
Therefore
$$r : s = m : p = n : q,$$
and hence $a, b$ are similar solid numbers.