This one has puzzled me for a while, and I still don't have a method that could be considered "standard" except for me. The bottom of the cup is quite "curvy" as it gets close to the throat. If one used a flat-ended tool wide enough to not go into the throat, then depending on the rate of drop toward the throat, you would end up with misleading measurements.
So for today's test I used a mechanical pencil. It has a tapered tip, which will extend into the throat a little. So two mp's of similar depth would measure differently this way if one has a larger throat. BUT a larger throat can make a mp FEEL larger or deeper, so I thought maybe it's a fair way to test them. If you use your own pencil, you'd probably get different measurements. So this may be useless for obtaining any absolute standards, but might still be interesting.
I measured 4 today, all with pretty similar cup diameters and that might be thought to have similar depth and throat. I measured the distance the pencil went "in" compared to the top of the rim's curve. Here's what I got:
Wick Ultra SM4U: 1.94"
Wick Heritage 4AL: 1.85"
Wick Classic 4AL: 1.85"
Wick SM4: 1.83"
And for comparison, here's a 5th sample of a slightly larger mp, just for fun:
Wick SM3.5: 1.885"
So there you go. I think I haven't accomplished much today in the mouthpiece measurement arena, but it's something I wanted to try for a while. It's interesting to me that the Ultra swallows more of the pencil than the SM3.5, but I'm not sure what it really means.
My next idea is to somehow block the throat of each in a manner consistent with the general shape just before the throat. Then fill the cup with fluid and somehow measure the volume/weight of the fluid. The problem that might vex me there is surface tension. If I were using an eyedropper to fill the cups, how would I know when to stop? I could squeeze a few more in because of surface tension (the surface would be a little convex).
Surely there is a better way! Can one of our engineer types speak up and tell me if there is an accepted way to do this? I don't think I've seen a specification anywhere to make me think someone is currently measuring in a way that can be replicated.