Let
then
This function is defined for all values of
except those lying between
and
,
and is seen to be many-valued. To make the function single-valued,
is taken as the
arc of smallest numerical value whose cosecant is
. This means that if
is positive,
we confine ourselves to points on the arc
(Figure 5.11),
taking
on values between 0 and
(
may be included);
and if
is negative, we confine ourselves to points on the arc
,
taking on
values between
and
(
may be included).
Figure 5.11:
The inverse secant function
using SAGE.
|
Figure 5.12:
The standard branch of
using SAGE.
|
Differentiating with respect to
by XVI and following the method of
the last section, we get
(equation (XXIII) in §5.1 above).
david joyner
2008-08-11