A note on the Dixon formula for a finite hypergeometric series (Q2493971)

In [\textit{B. M. Brown} and \textit{M. S. Eastham}, Proc., R. Soc. Lond., Ser. A, Math. Phys. Eng. Sci. 459, No. 2038, 2431--2448 (2003; Zbl 1047.33001)] the polynomial \(\Phi_{N}(x,\nu)\) of even degree was introduced as a truncated Bessel series for \(J_{-\nu}(i\sqrt{(x)})\). This function \(J_{-\nu}(i\sqrt{(x)})\) is connected to the Dirichlet Jost function for a certain potential for Schrödinger's equation. The formula can be written in an equivalent form as Kummer's \(\;_{2}F_{1}(-1)\) formula \[ _{2}F_{1}(a,b;1+a-b;-1)=2\cos \left(\frac{1}{2}\pi{a} \right) \Gamma\left[\begin{matrix} -a,1+a-b \\-\frac{a}{2},1+\frac{a}{2}-b\end{matrix}\right]. \] The second case in the article can be rewritten as a new form of Dixon's formula. This form is \[ _{3}F_{2}(a,b,c;1+a-b,1+a-c;1)= 2\cos\left(\frac{1}{2}\pi{a}\right) \Gamma\left[\begin{matrix} 1+a-c,-a\\ -\frac{a}{2},1+\frac{a}{2}-c\end{matrix}\right]\;_{2}F_{1} \left(\frac{a}{2},c;1+a-b;1\right). \] See also \textit{B. M. Brown} and \textit{M. S. P. Eastham}, J. Comp. Appl. Math. 194, No. 1, 131--140 (2006; Zbl 1093.33002).