19.15 Advantages of Symmetry19.17 Graphics

§19.16 Definitions

Contents

§19.16(i) Symmetric Integrals

19.16.1 \mathop{R_{F}\/}\nolimits\!\left(x,y,z\right)=\frac{1}{2}\int _{0}^{{\infty}}\frac{dt}{s(t)},
19.16.2 \mathop{R_{J}\/}\nolimits\!\left(x,y,z,p\right)=\frac{3}{2}\int _{0}^{{\infty}}\frac{dt}{s(t)(t+p)},
19.16.3 \mathop{R_{G}\/}\nolimits\!\left(x,y,z\right)=\frac{1}{4\pi}\int _{0}^{{2\pi}}\!\!\!\!\int _{0}^{{\pi}}\left(x{\mathop{\sin\/}\nolimits^{{2}}}\theta{\mathop{\cos\/}\nolimits^{{2}}}\phi+y{\mathop{\sin\/}\nolimits^{{2}}}\theta{\mathop{\sin\/}\nolimits^{{2}}}\phi+z{\mathop{\cos\/}\nolimits^{{2}}}\theta\right)^{{\frac{1}{2}}}\mathop{\sin\/}\nolimits\theta d\theta d\phi,

where p (\neq 0) is a real or complex constant, and

19.16.4 s(t)=\sqrt{t+x}\sqrt{t+y}\sqrt{t+z}.

In (19.16.1) and (19.16.2), x,y,z\in\Complex\setminus(-\infty,0] except that one or more of x,y,z may be 0 when the corresponding integral converges. In (19.16.2) the Cauchy principal value is taken when p is real and negative. In (19.16.3) \realpart{x}, \realpart{y}, \realpart{z}\geq 0. It should be noted that the integrals (19.16.1)–(19.16.3) have been normalized so that \mathop{R_{F}\/}\nolimits\!\left(1,1,1\right)=\mathop{R_{J}\/}\nolimits\!\left(1,1,1,1\right)=\mathop{R_{G}\/}\nolimits\!\left(1,1,1\right)=1.

A fourth integral that is symmetric in only two variables is defined by

19.16.5 \mathop{R_{D}\/}\nolimits\!\left(x,y,z\right)=\mathop{R_{J}\/}\nolimits\!\left(x,y,z,z\right)=\frac{3}{2}\int _{0}^{{\infty}}\frac{dt}{s(t)(t+z)},

with the same conditions on x, y, z as for (19.16.1), but now z\neq 0.

Just as the elementary function \mathop{R_{C}\/}\nolimits\!\left(x,y\right)19.2(iv)) is the degenerate case

19.16.6 \mathop{R_{C}\/}\nolimits\!\left(x,y\right)=\mathop{R_{F}\/}\nolimits\!\left(x,y,y\right),

and \mathop{R_{D}\/}\nolimits is a degenerate case of \mathop{R_{J}\/}\nolimits, so is \mathop{R_{J}\/}\nolimits a degenerate case of the hyperelliptic integral,

19.16.7 \frac{3}{2}\int _{0}^{{\infty}}\frac{dt}{\prod _{{j=1}}^{5}\sqrt{t+x_{j}}}.

§19.16(ii) \mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};\mathbf{z}\right)

All elliptic integrals of the form (19.2.3) and many multiple integrals, including (19.16.3), are special cases of a multivariate hypergeometric function

19.16.8 \mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};\mathbf{z}\right)=\mathop{R_{{-a}}\/}\nolimits\!\left(b_{1},\dots,b_{n};z_{1},\dots,z_{n}\right),

which is homogeneous and of degree -a in the z’s, and unchanged when the same permutation is applied to both sets of subscripts 1,\dots,n. Thus \mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};\mathbf{z}\right) is symmetric in the variables z_{j} and z_{{\ell}} if the parameters b_{j} and b_{{\ell}} are equal. The R-function is often used to make a unified statement of a property of several elliptic integrals. Before 1969 \mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};\mathbf{z}\right) was denoted by R(a;\mathbf{b};\mathbf{z}).

19.16.9 \mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};\mathbf{z}\right)=\frac{1}{\mathop{\mathrm{B}\/}\nolimits\!\left(a,a^{{\prime}}\right)}\int _{0}^{{\infty}}t^{{a^{{\prime}}-1}}\prod^{n}_{{j=1}}(t+z_{j})^{{-b_{j}}}dt=\frac{1}{\mathop{\mathrm{B}\/}\nolimits\!\left(a,a^{{\prime}}\right)}\int _{0}^{{\infty}}t^{{a-1}}\prod^{n}_{{j=1}}(1+tz_{j})^{{-b_{j}}}dt, a,a^{{\prime}}>0, z_{j}\in\Complex\setminus(-\infty,0],

where \mathop{\mathrm{B}\/}\nolimits\!\left(x,y\right) is the beta function (§5.12) and

19.16.10 a^{{\prime}}=-a+\sum _{{j=1}}^{n}b_{j}.
19.16.11
\mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};\lambda\mathbf{z}\right)=\lambda^{{-a}}\mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};\mathbf{z}\right),
\mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};x\boldsymbol{{1}}\right)=x^{{-a}}, \boldsymbol{{1}}=(1,\dots,1).

For further information, especially representation of the R-function as a Dirichlet average, see Carlson (1977b).

§19.16(iii) Various Cases of \mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};\mathbf{z}\right)

\mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};\mathbf{z}\right) is an elliptic integral iff the z’s are distinct and exactly four of the parameters a,a^{{\prime}},b_{1},\dots,b_{n} are half-odd-integers, the rest are integers, and none of a, a^{{\prime}}, a+a^{{\prime}} is zero or a negative integer. The only cases that are integrals of the first kind are the four in which each of a and a^{{\prime}} is either \frac{1}{2} or 1 and each b_{j} is \frac{1}{2}. The only cases that are integrals of the third kind are those in which at least one b_{j} is a positive integer. All other elliptic cases are integrals of the second kind.

(Note that \mathop{R_{C}\/}\nolimits\!\left(x,y\right) is not an elliptic integral.)

Each of the four complete integrals (19.16.20)–(19.16.23) can be integrated to recover the incomplete integral:

19.16.24 \mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};\mathbf{z}\right)=\frac{z_{1}^{{a^{{\prime}}-b_{1}}}}{\mathop{\mathrm{B}\/}\nolimits\!\left(b_{1},a^{{\prime}}-b_{1}\right)}\int _{0}^{{\infty}}t^{{b_{1}-1}}(t+z_{1})^{{-a^{{\prime}}}}\*\mathop{R_{{-a}}\/}\nolimits\!\left(\mathbf{b};0,t+z_{2},\dots,t+z_{n}\right)dt, a^{{\prime}}>b_{1}, a+a^{{\prime}}>b_{1}>0.