A Plummer model in virial units (E=-1/4, M=1, G=1) can be generated by using
the default keywords **sigma=0.53192304** (sigma^2 = 8/(9pi)) and **r0=0.58904862**
(r0 = 3pi/16).

A King model...*doesn workk yet*.

A de Vaucouleurs model *has not
been implemented yet*.

A Jaffe model *has not been implemented yet*.

A Hernquist
model *has not been implemented yet*.

The method used is to compute, from given density as a function of radius, the density as a function of potential, after which Eddingtons inversion formula (an Abel integral equation, see also eq. (4-140a) in Binney & Tremaine, 1987) is used to compute the distribution function at the energies at which the radii were given. The density/energy has been modified to incoorporate anisotropic models a la Osipkov-Merrit (for details see pp.239 in Binney & Tremaine, 1987).

**out=***out_file*- Write output table in binary form
to
*out_file*, which will contain an**OsipkovMerrittModel**data set. A utility program*tablst(1NEMO)*can be used to get a table for plotting purposes. [no default]. **model=***name*- Which model to use. Possible options are
**plummer, king, devauc, jaffe**. [default:**plummer**, but see also keyword**in=**below)]. **rmax=***r_max*- Radius at which to cutoff the model. The potential value at this
edge is assumed zero. Some models may ignore this parameter. [default:
**100**]. **b=***b*- Anisotropy parameter (inverse anisotropy radius) [default:
**0**]. **nrad=***n_rad*- Number of radii to use for numerical integration [default:
**512**]. **njump=***n_jump*- Stride through integration tables to generate the output table. The size
of output table will be
*n_rad/n_stride*. [default:**1**]. **ngauss=***n_gauss*- Number
of points used for Gauss-Legendre integrations [default:
**48**] **sigma=***sigma*- (
**plummer**only) Something surely (velocity dispersion?) [default:**0.53192304**]. **r0=***r0*- (
**plummer**only) Model scaling radius [default:**0.58904862**]. **w0=***w0*- (
**king**only) [default:**1**]. **emtot=***emtot*- (
**king**only) [default:**-1**]. **rc=***rc*- (
**king**only) core radius [default:**1**] **in=***in-file*- (Ascii table ) file from which a table
of radius versus density is used as a model instead. [default: not used,
see keyword
**model**above]. **radcol=***rad_col_nr*- Column number in Iin-file where
radius is stored [default:
**1**]. **denscol=***dens_col_nr*- Column number in
*in-file*where density is stored [default:**2**]. **masscol=***mass_col_nr*- Column number
in
*in-file*where cumulative mass is stored. Currently not used, but future expansion allows usage of a**masscol**instead of a**denscol**[default:**0**].

The absolute value of the distribution
function *f* in Plummer models is wrong. I don’t know why this is yet. Doesn’t
matter for mkanmod though.

4-jun-88V1.0 createdPJT 28-oct-88V1.1 in= added + bug removedPJT 9-sep-90V1.2 got rid of fortranish stuffPJT