A parametrized model of the mass distribution within the Milky Way is
fitted to the available observational constraints. The most important
single parameter is the ratio of the scalelength R-d,R- Of the stella
r disc to R-0. The disc and bulge dominate upsilon(c)(R) at R less tha
n or similar to R-0 only for R-d,R-/R-0 less than or similar to 0.3.
Since the only knowledge we have of the halo derives from studies like
the present one, we allow it to contribute to the density at all radi
i. When allowed this freedom, however, the halo causes changes in assu
mptions relating to R much less than R-0 to affect profoundly the stru
cture of the best-fitting model at R much greater than R-0. For exampl
e, changing the disc slightly from an exponential surface-density prof
ile significantly changes the form of upsilon(c)(R) at R much greater
than R-0, where the disc makes a negligible contribution to upsilon(c)
. Moreover, minor changes in the constraints can cause the halo to dev
elop a deep hole at its centre that is not physically plausible. These
problems call into question the proposition that flat rotation curves
arise because galaxies have physically distinct haloes rather than ou
twards-increasing mass-to-light ratios. The mass distribution of the G
alaxy and the relative importance of its various components will remai
n very uncertain until more observational data can be used to constrai
n mass models. Data that constrain the Galactic force field at z great
er than or similar to R and at R > R-0 are especially important.