# §16.3(i) Differentiation Formulas

 16.3.1 $\frac{{d}^{n}}{{dz}^{n}}\mathop{{{}_{p}F_{q}}\/}\nolimits\!\left({a_{1},\dots,% a_{p}\atop b_{1},\dots,b_{q}};z\right)=\frac{\left(\mathbf{a}\right)_{n}}{% \left(\mathbf{b}\right)_{n}}\mathop{{{}_{p}F_{q}}\/}\nolimits\!\left({a_{1}+n,% \dots,a_{p}+n\atop b_{1}+n,\dots,b_{q}+n};z\right),$
 16.3.2 $\frac{{d}^{n}}{{dz}^{n}}\left(z^{\gamma}\mathop{{{}_{p}F_{q}}\/}\nolimits\!% \left({a_{1},\dots,a_{p}\atop b_{1},\dots,b_{q}};z\right)\right)=\left(\gamma-% n+1\right)_{n}z^{\gamma-n}\mathop{{{}_{p+1}F_{q+1}}\/}\nolimits\!\left({\gamma% +1,a_{1},\dots,a_{p}\atop\gamma+1-n,b_{1},\dots,b_{q}};z\right),$
 16.3.3 $\left(z\frac{d}{dz}z\right)^{n}\left(z^{\gamma-1}\mathop{{{}_{p+1}F_{q}}\/}% \nolimits\!\left({\gamma,a_{1},\dots,a_{p}\atop b_{1},\dots,b_{q}};z\right)% \right)=\left(\gamma\right)_{n}z^{\gamma+n-1}\mathop{{{}_{p+1}F_{q}}\/}% \nolimits\!\left({\gamma+n,a_{1},\dots,a_{p}\atop b_{1},\dots,b_{q}};z\right),$
 16.3.4 $\frac{{d}^{n}}{{dz}^{n}}\left(z^{\gamma-1}\mathop{{{}_{p}F_{q+1}}\/}\nolimits% \!\left({a_{1},\dots,a_{p}\atop\gamma,b_{1},\dots,b_{q}};z\right)\right)=\left% (\gamma-n\right)_{n}z^{\gamma-n-1}\mathop{{{}_{p}F_{q+1}}\/}\nolimits\!\left({% a_{1},\dots,a_{p}\atop\gamma-n,b_{1},\dots,b_{q}};z\right).$

Other versions of these identities can be constructed with the aid of the operator identity

 16.3.5 $\left(z\frac{d}{dz}z\right)^{n}=z^{n}\frac{{d}^{n}}{{dz}^{n}}z^{n},$ $n=1,2,\dots$.

# §16.3(ii) Contiguous Functions

Two generalized hypergeometric functions $\mathop{{{}_{p}F_{q}}\/}\nolimits\!\left(\mathbf{a};\mathbf{b};z\right)$ are (generalized) contiguous if they have the same pair of values of $p$ and $q$, and corresponding parameters differ by integers. If $p\leq q+1$, then any $q+2$ distinct contiguous functions are linearly related. Examples are provided by the following recurrence relations:

 16.3.6 $z\mathop{{{}_{0}F_{1}}\/}\nolimits\!\left(-;b+1;z\right)+b(b-1)\mathop{{{}_{0}% F_{1}}\/}\nolimits\!\left(-;b;z\right)-b(b-1)\mathop{{{}_{0}F_{1}}\/}\nolimits% \!\left(-;b-1;z\right)=0,$
 16.3.7 $\mathop{{{}_{3}F_{2}}\/}\nolimits\!\left({a_{1}+2,a_{2},a_{3}\atop b_{1},b_{2}% };z\right)a_{1}(a_{1}+1)(1-z)+\mathop{{{}_{3}F_{2}}\/}\nolimits\!\left({a_{1}+% 1,a_{2},a_{3}\atop b_{1},b_{2}};z\right)a_{1}\left(b_{1}+b_{2}-3a_{1}-2+z(2a_{% 1}-a_{2}-a_{3}+1)\right)+\mathop{{{}_{3}F_{2}}\/}\nolimits\!\left({a_{1},a_{2}% ,a_{3}\atop b_{1},b_{2}};z\right)\left((2a_{1}-b_{1})(2a_{1}-b_{2})+a_{1}-a_{1% }^{2}-z(a_{1}-a_{2})(a_{1}-a_{3})\right)-\mathop{{{}_{3}F_{2}}\/}\nolimits\!% \left({a_{1}-1,a_{2},a_{3}\atop b_{1},b_{2}};z\right)(a_{1}-b_{1})(a_{1}-b_{2}% )=0.$

For further examples see §§13.3(i), 15.5(ii), and the following references: Rainville (1960, §48), Wimp (1968), and Luke (1975, §5.13).