I don't have access to my notes right now, but IIRC the half-dozen or so originals I measured for cone had about .02 widening from about 1/2" (or maybe 3/8") inside. The caveat is that my calipers only went about 1/2" back inside the bore, so the true bore size may have been smaller and the cone longer than I was able to measure. The taper was pretty consistent from gun to gun, though - I think the only one that didn't show coning was a post-1820 rifle, and I think that there was one with considerable deterioration at the muzzle that showed more, but I can't remember all that much.
I'd like to reiterate a point I made on the recent thread on short starters - period manufacturing techniques introduced a very slight choke in the bores, whereas modern techniques (AFAIK) do not. It is even possible that bores were intentionally choked, as we know fowling piece bores could be deliberately chocked and flared, if only to ensure that they weren't inadvertently funneled. Ergo, I don't think we can just assume that modern experience different load types in contemporary bores is representative of how original barrels may have behaved. I think it possible that tight loads such as used by modern target shooters may not have shown more accuracy over a looser load with a thicker patch in period bores - a thicker patch with more "give" might very well handle variations in bore-size better, particularly when using smaller charges with less chance of ball obturation. That would go a long way with reconciling the discrepancies between modern experience and what we can reconstruct of 18th and early 19th century practice.
I also wonder if changing manufacturing techniques in the 19th century might have produced more consistent bores, and if so then there might be connection between the widespread use of factory-made barrels and changing methods of loading, including the increased popularity of short starters and the successful use of conical bullets (toyed with for ages but seldom used for otherwise unknown reasons).
