Steps / Historical AI

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

Elements of Algebra

All 627 leaves · leaf 544

Transcription

CHAP. II. ADDITIONS. 507 If we had, therefore, to resolve the equation $Ap^m + Bp^{m-1}q + cp^{m-2}q^2 + \text{&c.} + vq^m = \pm 1$, in whole numbers, we must seek for the values of $p$ and $q$ by the method of the preceding problem, except in the case where the equation $Ax^m + Bx^{m-1} + cx^{m-2} + \text{&c.} + v = 0$, had roots, or any divisors commensurable; for then, it is evident, that the quantity $Ap^m + Bp^{m-1}q + cp^{m-2}q^2 + \text{&c.}$ might be decomposed into two or more similar quantities of less degrees; so that it would be necessary for each of these partial formulæ to be separately equal to unity, which would give at least two equations that would serve to determine $p$ and $q$. We have elsewhere given a solution of this last problem (Memoires pour l'Academie de Berlin pour l'Année 1768); but the one we are going to explain is much more simple and direct, although both depend on the same theory of con- tinued fractions *. 31. Problem 3. Required the values of $p$ and $q$, which will render the quantity $Ap^2 + Bpq + Cq^2$ the least possible, supposing that whole numbers only are admitted for $p$ and $q$. This problem evidently is only a particular case of the preceding; but it may be proper to consider it separately, because it is capable of a very simple and elegant solution; and, besides, we shall have occasion afterwards to make use of it, in resolving quadratic equations for two unknown quantities in whole numbers. According to the general method, we must begin, there- fore, by seeking the roots of the equation $Ax^2 + Bx + C = 0$, which we know to be, $\frac{-B \pm \sqrt{(B^2 - 4AC)}}{2A}$. 1st, If $B^2 - 4AC$ be a square number, the two roots will be commensurable, and there will properly be no minimum, because the quantity $Ap^2 + Bpq + Cq^2$ will become $0$. 2d, If $B^2 - 4AC$ be not a square, then the two roots will be irrational, or imaginary, according as $B^2 - 4AC$ will be $> 0$, or $< 0$, which makes two cases that must be considered separately; we shall begin with the latter, which it is most easy to resolve. First case, when $B^2 - 4AC < 0$. 32. The two roots being in this case imaginary, we shall * See also Le Gendre's Essai sur la Theorie des Nombres, page 169.
confidence 0.97 22 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.