Steps / Historical AI

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

Elements of Algebra

All 627 leaves · leaf 435

Transcription

398 ELEMENTS PART II. This new solution is also obtained from the last method, in the following manner: If we make $x + y\sqrt{-ac} = (p\sqrt{a} + q\sqrt{-c})^2$, and $x - y\sqrt{-ac} = (p\sqrt{a} - q\sqrt{-c})^2$, we shall have $x^2 + acy^2 = (ap^2 + cq^2)^2$, and, consequently, equal to a square. Farther, because $x + y\sqrt{-ac} = ap^2 + 2pq\sqrt{-ac} - cq^2$, and $x - y\sqrt{-ac} = ap^2 - 2pq\sqrt{-ac} - cq^2$, we find $x = ap^2 - cq^2$, and $y = 2pq$. It is farther evident, that if the number $ac$ be resolvible into two factors, in a greater number of ways, we may also find a greater number of solutions. 185. Let us illustrate this by means of some determinate formulæ; and, first, if the formula $x^2 + y^2$ must become a square, we have $ac = 1$; so that $x = p^2 - q^2$, and $y = 2pq$; whence it follows that $x^2 + y^2 = (p^2 + q^2)^2$. If we would have $x^2 - y^2 = \square$; we have $ac = -1$; so that we shall take $x = p^2 + q^2$, and $y = 2pq$, and there will result $x^2 - y^2 = (p^2 - q^2)^2 = \square$. If we would have the formula $x^2 + 2y^2 = \square$, we have $ac = 2$; let us therefore take $x = p^2 - 2q^2$, or $x = 2p^2 - q^2$, and $y = 2pq$, and we shall have $x^2 + 2y^2 = (p^2 + q^2)^2$, or $x^2 + 2y^2 = (2p^2 + q^2)^2$. If, in the fourth place, we would have $x^2 - 2y^2 = \square$, in which $ac = -2$, we shall have $x = p^2 + 2q^2$, and $y = 2pq$; therefore $x^2 - 2y^2 = (p^2 - 2q^2)^2$. Lastly, let us make $x^2 + 6y^2 = \square$. Here we shall have $ac = 6$; and, consequently, either $a = 1$, and $c = 6$, or $a = 2$, and $c = 3$. In the first case, $x = p^2 - 6q^2$, and $y = 2pq$; so that $x^2 + 6y^2 = (p^2 + 6q^2)^2$; in the second, $x = 2p^2 - 3q^2$, and $y = 2pq$; whence $x^2 + 6y^2 = (2p^2 + 3q^2)^2$. 186. But let the formula $ax^2 + cy^2$ be proposed to be transformed into a square. We know beforehand, that this cannot be done, except we already know a case, in which this formula really becomes a square; but we shall find this given case to be, when $x = f$, and $y = g$; so that $af^2 + cg^2 = h^2$; and we may observe, that this formula can be transformed into another of the form $t^2 + acu^2$, by making $t = \frac{afx+cgy}{h}$, and $u = \frac{gx-fy}{h}$; for if $t^2 = \frac{a^2f^2x^2 + 2acfgxy + c^2g^2y^2}{h^2}$, and
confidence 0.95 50 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.