In this paper, we have used an interpolation method based upon that of
PRH92 to search for temporal lags and leads between the 2-4keV,
5-7keV and 8-15keV bands in a long RXTE observation of the
bright Seyfert 1 galaxy MCG-6-30-15. In essence, we use the PRH92 method
to compute an optimal reconstruction of the 2-4keV light curve in which
the datagaps are interpolated across. We then fold this reconstructed
light curve through trial transfer functions and compare with data from the
other bands in a 
sense.
Our search for lags and leads was tailored to find reverberation effects in the iron line which is thought to originate from the innermost regions of the black hole accretion disk. We find no evidence for any reverberation, and rule out reverberation delays in the range 0.5-50ksec. We can extend the conclusions of Lee et al. (1999b), and infer that the iron line possesses a constant flux on timescales on timescales as short as 500s. We also find that the hard band (8-15keV) is delayed by 50-100s relative to the 2-4keV band.
We attempt to put these various results together into a coherent picture
for this object. The constancy of the iron line flux leads one to consider
large black hole masses (in excess of 
). However, such a large
mass is found to be unacceptable from the standpoint of both X-ray
variability constraints, and constraints based on the mass of the galactic
bulge. Indeed, using the bulge/hole scaling factor of Magorrian et
al. (1998), we estimate that the hole has a mass of 
. Given that this is a more reasonable mass estimate,
some mechanism beyond the simple X-ray reflection model must be invoked to
explain the temporal variability of the iron line and Compton reflection
continuum. We suggest that flux correlated changes in the average
ionization state of the surface layers of the accretion disk may be such a
mechanism. While we support this suggestion with a toy model, the
plausibility if this suggestion can only be assessed once detailed
modeling has been performed.