I did look in his book - half wondered if I'd seen the comment there, but couldn't see it at a quick look - have to say I didn't know he charged for information - perhaps that's the reason for his absence here, though good to know funds go to worthy casuses.
Twenty odd years ago whilst in a north Wales cafe/antiques emporium, I was standing near an older guy who was looking at what appeared to be an early C19 Irish decanter, and as I passed I rememinded him to check for matching numbers - to which he replied caustically that those old workers just grabbed the nearest stopper that fitted and didn't bother with numbers. So I felt admonished and left quietly as he seemed to be an expert - but that's to over simplify the process of course because as we know their method, so I understand, was to chuck the bottle on a lathe and as it spun use grit/polish etc. to make the stopper a matching fit. I always thought it sounded very dangerous.
I don't specialize in C19 bottles - few here do, but perhaps they can comment - it's just that we all know C20 quality examples are mostly numbered whereas I'd not seen this feature on earlier bottles.
As to actual numbers - I've seen them up to forty something and many with lower value digits - though not sure quite what dictates these numbers, but suppose that if you've a busy day then such higher digits would occur naturally. Sometimes the face of the script can indicate late C19 - early C20, having rather thick and slightly copperplate outline.
I don't know whether Peter (oldglassman) has any bottles from 1800 to 1840, for example, that might show matching Nos.