Steps / Historical AI

Leonhard Euler · 1770 (1828 English ed.) · Internet Archive

Elements of Algebra

All 627 leaves · leaf 493

Transcription

gives $2a^3 = y^3 - x^3$; it is likewise impossible to find two cubes having their difference equal to the double of another cube; and the same impossibility extends to the sum of two cubes, as is evident from the following demonstration. 247. *Theorem*. Neither the sum nor the difference of two cubes can become equal to the double of another cube; or, in other words, the formula $x^3 \pm y^3 = 2z^3$ is always impossible, except in the evident case of $y=x$. We may here also consider $x$ and $y$ as prime to each other; for if these numbers had a common divisor, it would be necessary for $z$ to have the same divisor; and, consequently, for the whole equation to be divisible by the cube of that divisor. This being laid down, as $x^3 \pm y^3$ must be an even number, the numbers $x$ and $y$ must both be odd, in consequence of which both their sum and their difference must be even. Making, therefore, $\frac{x+y}{2} = p$, and $\frac{x-y}{2} = q$, we shall have $x=p+q$ and $y=p-q$; and of the two numbers $p$ and $q$, the one must be even and the other odd. Now, from this, we obtain $x^3+y^3 = 2p^3 + 6pq^2 = 2p(p^2 + 3q^2)$, and $x^3-y^3 = 6p^2q + 2q^3 = 2q(3p^2 + q^2)$, which are two formulæ perfectly similar. It will therefore be sufficient to prove that the formula $2p(p^2+3q^2)$ cannot become the double of a cube, or that $p(p^2+3q^2)$ cannot become a cube: which may be demonstrated in the following manner. 1. Two different cases again present themselves to our consideration: the one, in which the two factors $p$, and $p^2+3q^2$, have no common divisor, and must separately be a cube; the other in which these factors have a common divisor, which divisor, however, as we have seen (Art. 243), can be no other than 3. 2. *Case* 1. Supposing, therefore, that $p$ is not divisible by 3, and that thus the two factors are prime to each other, we shall first reduce $p^2+3q^2$ to a cube by making $p=t(t^2-9u^2)$, and $q=3u(t^2-9u^2)$; by which means it will only be farther necessary for $p$ to become a cube. Now, $t$ not being divisible by 3, since otherwise $p$ would also be divisible by 3, the two factors $t$, and $t^2-9u^2$, are prime to one another, and, consequently, each must separately be a cube. 3. But the last factor has also two factors, namely $t+3u$, and $t-3u$, which are prime to each other, first because $t$ is not divisible by 3, and, in the second place, because one of
confidence 0.89 33 expressions source scan at the Internet Archive

This text was produced by machine and has not been corrected by hand. Where it matters, read the scan. Confidence is the model's own, not an external check.