y=11+x2y = \dfrac{1}{1+x^2}

The Witch of Agnesi y=11+x2y = \dfrac{1}{1+x^2}

y=11+x2y = \dfrac{1}{1+x^2} is a rational function known as the Witch of Agnesi, and its graph is a bell shape closely resembling the Gaussian. Because its denominator is a polynomial rather than an exponential, its tails fall off more gently than the Gaussian's.

Domain and range The denominator 1+x21 + x^2 is always at least 11 and never zero, so the function is defined for all real xx. Since 0<11+x210 < \dfrac{1}{1+x^2} \le 1, the range is (0,1](0, 1].

Symmetry Because f(x)=11+(x)2=f(x)f(-x) = \dfrac{1}{1+(-x)^2} = f(x), it is an even function, symmetric about the yy-axis.

Monotonicity and extrema The derivative is f(x)=2x(1+x2)2f'(x) = \dfrac{-2x}{(1+x^2)^2}. It is positive for x<0x < 0 and negative for x>0x > 0, so the curve reaches its maximum at the peak (0,1)(0, 1) and decreases on either side.

Asymptotes and limits As x±x \to \pm\infty the denominator grows without bound and y0y \to 0, so the xx-axis is a horizontal asymptote. The decay is slow, of order 1/x21/x^2; for example y=0.1y = 0.1 already at x=3x = 3.

Inflection points The second derivative f(x)=6x22(1+x2)3f''(x) = \dfrac{6x^2 - 2}{(1+x^2)^3} changes sign at x=±13±0.577x = \pm \dfrac{1}{\sqrt{3}} \approx \pm 0.577, where y=34y = \dfrac{3}{4}. These shoulders mark the change from concave down to concave up.

Relation to other functions This function is exactly the derivative of the arctangent: ddxarctanx=11+x2\dfrac{d}{dx}\arctan x = \dfrac{1}{1+x^2}. Its total integral is dx1+x2=π\displaystyle\int_{-\infty}^{\infty}\dfrac{dx}{1+x^2} = \pi, and dividing by π\pi gives 1π(1+x2)\dfrac{1}{\pi(1+x^2)}, the probability density of the Cauchy distribution.

Applications and history The curve is named after the 18th-century Italian mathematician Maria Gaetana Agnesi (1718–1799). When it appeared in her 1748 textbook, the Italian word "versiera" (a turning curve) was confused with "avversiera" (she-devil), leading to the famous English mistranslation "witch." Today it arises as the Cauchy distribution in probability and the Lorentzian function describing resonance.