I had similar problems, but it was a delayed problem.  I got
crystallized precipitation, like stars, but not immediately after mounting.
It was 1 week or so after the staining that the crystals started to appear,
and getting worse as time passed by.  I found out the problem was because I
used xylene based mounting solution.  When I switched to aqueous based
mounting solution without dehydrating the slides, I got rid of the problem.
However, I ran into other problems:  dehydration of the slides eventually
with time.  I've tried to use nail-polisher to seal the edges of the
coverslip, but not working well.  So finally, I've just dropped alk phos,
and go back to hydrogen peroxidase.  I've been using Vector's DAB with
Nickel to make the substrate purplish black, similar to NBT/BCIP, and have
got great results so far.  I just have to learn to be careful in discarding
DAB.
	I hope this might help.
	Su


> Paul Klosen wrote:
> > Recently we have run into a nasty problem with our NBT/BCIP in situ
> > detections. The in situ part seems to work fine and the alk phos 
> > detections is sensitive.
> >
> > However, under the microscope, the precipitate is completely granular
> and
> > seems to be located above the tissue, which is absolutely unsuitable for
> > photomicrography. We have tried varying several parameters (pH, salt and
> > MgCl2 conc., ..) and different lots and sources of both NBT and BCIP, 
> > but nothing seems to bring us back to our previous results, where we had
> > a nice cell-filling precipitate, albeit a bit diffuse as expected with
> > NBT/BCIP.
> >
> > Has anyone encountered a similar problem, and how did you get rid of it?
> 
> 
>