MARK parameter miscounting glitch

[evan.adams]

I'm running a Burnham recapture and recovery analysis in Program MARK and I think I've noticed a bug. When I model recoveries using a sin link the model counts all the recovery parameters appropriately, but when I've used a logit link (so I can add individual covariates) it doesn't seem to count them at all. This only seemed to be a problem with parameters relating to recovery. Any parameters associated with survival, detection or site fidelity calculates fine. Has anybody run into this problem before?

If the problem is just appropriate parameter counting I can pull the models out of MARK and calculate AIC with the correct number of parameters. If that's not the only problem then I guess I just can't use a logit link?

Thanks in advance,

Evan