Skip to content

Mathematics · Ch 7 — Conic Sections

Tangents from a Point to the Ellipse

7.2.6

Tangents from a Point to the Ellipse

From a point P(x1,y1)P(x_1,y_1) in the plane, using the slope-form tangent y=mx±a2m2+b2y=mx\pm\sqrt{a^2m^2+b^2} from section 7.2.5, and forcing it through PP:

y1−mx1=±a2m2+b2.y_1-mx_1 = \pm\sqrt{a^2m^2+b^2}.

Squaring both sides to remove the ±\pm and the square root, then collecting terms in mm:

(x12−a2)m2−2x1y1m+(y12−b2)=0.(x_1^2-a^2)m^2-2x_1y_1m+(y_1^2-b^2)=0.

This is a quadratic in mm, so (as for the parabola) it has two roots m1,m2m_1,m_2 in general — confirming that exactly two tangents can be drawn to an ellipse from any external point.

By the standard relations between the roots and coefficients of a quadratic:

m1+m2=2x1y1x12−a2,m1m2=y12−b2x12−a2.m_1+m_2=\dfrac{2x_1y_1}{x_1^2-a^2}, \qquad m_1m_2=\dfrac{y_1^2-b^2}{x_1^2-a^2}. …