  plconf(z, nx, ny, kx, lx, ky, ly, clevel, nlevel, pltr, xg, yg)
  ***************************************************************

  plconf

  Draws a contour plot of the data in z[nx][ny], using the nlevel contour
  levels specified by clevel. Only the region of the array from kx to lx and
  from ky to ly is plotted out. A transformation routine pltr is used to map
  indicies within the array to the world coordinates, using the
  transformation arrays xg and yg. Contour plots for more information.

  z (PLFLT **, input): Pointer to a vectored two-dimensional array
  containing data to be contoured.

  nx, ny (PLINT, input): Physical dimensions of array z.

  kx, lx (PLINT, input): Range of x indicies to consider.

  ky, ly (PLINT, input): Range of y indicies to consider.

  clevel (PLFLT *, input): Pointer to array specifying levels at which to
  draw contours.

  nlevel (PLINT, input): Number of contour levels to draw.

  pltr (void *, input): Pointer to function that defines transformation
  between indicies in array z and the world coordinates. The function should
  have the form  pltr(x, y, tx, ty, xg, yg, nx, ny)

  x, y (PLFLT, input):  Specifies the position in the array through which
  the contour runs in terms of the array indicies.

  tx, ty (PLFLT *, output): Pointers to the world coordinates
  corresponding to the point (x, y).

  xg, yg (PLFLT *, input): Pointers to transformation arrays.

  nx, ny (PLINT, input): Array dimensions in x and y, respectively.

  xg, yg (PLFLT *, input): Pointers to arrays which define the
  transformation from (x, y) to (tx, ty), used as input to the transformation
  function pltr.

  Two transformation functions are provided: pltr1, for one dimensional
  mappings (i.e. tx = f(x), ty = f(y)), and pltr2, for two dimensional
  mappings (i.e. tx = f(x, y), ty = f(x, y)). Function values at locations
  between grid points are obtained via linear interpolation.

  NOTE: this routine cannot be called directly from Fortran. Use plcon0,
  plcon1, or plcon2 instead.

