Ferrari method
A method for reducing the solution of an equation of degree 4 to the solution of one cubic and two quadratic equations; it was discovered by L. Ferrari (published in 1545).
The Ferrari method for the equation consists in the following. By the substitution y=x-a/4 the given equation can be reduced to x^4+px^2 + qx +r = 0,\label{1} which contains no term in x^3. If one introduces an auxiliary parameter \def\a{\alpha}\a, the left-hand side of (1) can be written as x^4+px^2+qx+r =
=(x^2+\frac{p}{2}+\a)^2-\big[2\a x^2 - qx +\big(\a^2+p\a+\frac{p^2}{4}-r\big)\big].\label{2} One then chooses a value of \a such that the expression in square brackets is a perfect square. For this the discriminant of the quadratic trinomial must vanish. This gives a cubic equation for \a, q^2-42\a(\a^2+p\a+\frac{p^2}{4}-r)=0. Let \a_0 be one of the roots of this equation. For \a=\a_0 the polynomial in square brackets in (2) has one double root, x_0 = \frac{q}{4\a_0}, which leads to the equation (x^2+\frac{p}{2}+\a_0)^2 - 2\a_0(x-x_0)^2 = 0. This equation of degree 4 splits into two quadratic equations. The roots of these equations are also the roots of (1).
References
[1] | A.G. Kurosh, "Higher algebra" , MIR (1972) (Translated from Russian) |
Ferrari method. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Ferrari_method&oldid=14615