Ah… yeah. I guess “mostly” doesn’t count.
I’m leaning back that way myself.
Is there ANY further context about this thing? Otherwise, I’m just going to file it under strange grid ball segment maker and sulk. 
It reminds me of a gaming wheel I saw at a Carnival once. There were small fluffy or cartoony kid prizes, or larger adult prizes (electric razor and such.) The tickets were pre-printed with five-number set. If you had the five numbers in order, adult prize. Three of them, the kid prize. So, the four and five in the number set lowered the chance of winning a “kid” prize.
The “pointer” at the front makes me think this must be it. The top and sides seems to have been cleaned (the bottom obviously wasn’t.) I’m guessing it was brightly painted in the past.
Here are the closest examples I can find, not made for games with the same rules, though:
More different types, but none just like that one:
I’m not aware of any other context. Just that it was found in someone’s basement or attic or something.
None of the suggestions I’ve seen on either board really seem to fit. It’s puzzling. I probably shouldn’t waste time on it, but it’s too interesting for me to give up on just yet.
I suspect it’s for some extremely specialized purpose. I would say even a single case use, except that there was obviously some work and expense put into it.
Except with this wheel, all you’d have to do is bet the digits 1, 2, and 3 in any order and you’d always win a kids prize, because all possible permutations of them are represented.
To me it seems like a weirdly specific set of numbers for gaming.
I am thinking in the same vein. I still wonder if the unequally distributed 123 numbers are simply a mistake or a inequality in the odds that were part of the game. (or that most players wouldn’t notice! )
Your first sentence is almost certainly right - it’s got one job on this ship… I mean, one purpose in this world.
But don’t overestimate the amount of work in it. Around the end of the 19th century, any machinist worth his salt could have turned out one or ten of them using sand-casting, a wooden model and probably iron slugs for the numbers. A machinist today could do it even more quickly.
It’s fine work, but it’s not beyond the ability of a single craftsman to make. Which makes it even more unlikely we’ll ever get a definitive answer, unless an expert on early 20th century ________s happens to see it.
It doesn’t appear to have ever had a serial number or manufacturer’s stamp. I suppose that could point to it being a one of a kind single purpose thing.
It looks to me like it’s a part of a mechanical computer. The “spring” that you can see on the sideways view may have held one or two additional wheels. Those additional wheels would block most of that back wheel, but spinning them would reveal whatever answer they wanted to show. With 12 even partitions it may have had something to do as a calendar.
This is good. More wheels on front is an interesting possibility. An answer might even consist of just one digit from the outer ring and one from the inner ring.
So where would such numbers be appropriate? Numbers consisting perhaps of the digits 1 to 5 in one position and 1 to 3 in the other?
Assuming the answer would appear more or less above the pointer, you would have a value of 1 to 3 on top and 1 to 5 underneath.
ETA: Maybe the front wheel or wheels were interchangeable for different situations, which could explain why it’s missing.
I think the “question” which would be on the top wheel would go above the pointer - the answer would be made up of the number sequence revealed by the slots and could be anywhere on that back wheel. My gut feeling is that this would show sunrise and sunset times for a specific date, though I haven’t quite figured out how it works yet.
But times use more than the digits 1 through 5. Even if they’re rounded off, you’d need zeroes.
Well, sunrise and sunset may not work. But, the 1-3 on the top seems to fit dates since you can’t have more than 31 days in a month. And 1-5 fits time since the max minutes would be 59. We’re missing a lot of numbers though, but it would help a lot to see a second wheel.
I bet the bottom plate (showng here) shows the first number in the minutes - the second wheel the second number.
59 minutes would require a 9. I suppose there could be two more wheels, but there isn’t a lot of space between the one wheel that’s there and the pointer (look at the side view).
ETA: I didn’t see your second post when I posted this.
I agree - I wonder if the second wheel, designed to turn, would be a thin brass disk?
I’m just guessing of course - it’s a fun puzzle to think about!
Not a mechanical bell ringer: a number system for ringing changes. 5 bells, rung in order, (5,4,3,2,1), but with changes: (4,5,3,2,1),(4,3,5,2,1) (4,3,2,5,1) etc.
There are standard systems of changes. Except I don’t know what they are, and if this is an example, or how it would be used.
Searching on the five-digit sequences turns up books on bell ringing that include those sequences, so I assume that’s where the guess comes from.
ETA: missed the second page, sorry.
Maybe it’s a teaching device for bell ringers. But do the 3 digit sequences fit into that idea somehow?
Yes, those are also 3-bell sequences. This is apparently the classick, original worke:
http://quod.lib.umich.edu/e/eebo/A65805.0001.001/1:5?rgn=div1;view=fulltext