At risk of hijacking the thread, there seems to be a limited set of units that behave like this. As it happens, I do weigh 12 stone (not “stones”), and in my dialect we would say “there was two foot of snow on the ground” but “six inches” or “two metres” of snow.
Thinking about it further, my grandparents would have used this uncountable form for some other units: “It’s six mile from here to Dublin”; “I need eight ounce of sugar”.
I was taught the rule, and always assumed the “101” = “One Hundred and One” was a way of being cute - as in not just 100 - but one more than that.
In most cases where you see someone using “101” in titles and such - it is to be cute - and I would totally pronounce it as such (using the “and”), but if it was just a fact being cited in an article - such as “So far police have confirmed 101 deaths” I would not use the “and”.
Of course I am no grammar king, but I’ve always found the “don’t use and for anything without a decimal point or fraction” to make sense as it removes any ambiguities.
Ambiguities are improbable in real life, however. In varieties of English that don’t apply this rule (where “one hundred and one” is the natural pronunciation of 101), the phrase “one hundred and two thirds” is in theory ambiguous. But in practice, would such an utterance occur in a natural setting?
Equally, as mentioned above, the American usage could give rise to ambiguity when calling out a list of numbers: {100,51} v {151}.
Neither of these is a huge problem - we deal with ambiguity every day.
In any event, it’s worth pointing out that the rule is not universal, and that some people consider your way as odd as you do theirs.
Pianodave, you would benefit immensely from reading a recent book called The Language Hoax, by John McWhorter. You are interested in these matters; you have some mistaken ideas (ones you share with many other people); and you are admirably willing to learn. I’m serious – you would love this book! It is not a difficult read – it’s fun, really – and the examples he uses to prove his points are memorable.
I will, thank you! And then maybe I’ll be as smart as you guys 
Thing is, despite the rule some Americans are taught, I have never heard of anyone say 100.51 as “one hundred and fifty one” or interpret “one hundred and fifty one” as “one hundred and fifty one hundredths.”
I (American English speaker) would say:
100.51 = one hundred point five one
151 = a hundred (and) fifty-one: I sometimes say and, and sometimes not. Closing my eyes and listening to myself count fast, I’m more likely to use “and” with 101–112 and 120, 130, 140, 150, 160, 170, 180, and 190, while the wordier numbers lose the “and.”
151 apples = a hundred fifty-one apples
150⅕ = a hundred fifty and one fifth
I don’t think I’ve ever encountered a problem with the ambiguity.
I suspect I would be most envious of your piano playing skills, so it’s all good.
I doubt we’re any smarter than you, at most we’re more used to arguing with each other…
Tried to order that from Amazon UK and it went very badly.
Bought it for less on eBay though
I think not.
Eleven originally meant ‘one left (over)’ in Old English endleofan < Proto-Germanic aina-lif ‘one left’. In Gothic, it’s aini-lif. If a Proto-Goth removes ten skullcups of mead from the table, and one is left over, that makes eleven in all.
Twelve goes back through Old English twelf* to Proto-Germanic twa-lif—‘two left (over)’ which is also twalif in Gothic.
*Pronounced the same as Modern English twelve.
Both of these are consistent with a base-ten system and also suggest that “twelve” is just the final of two irregular constructions and any resemblance to a base-twelve system is coincidental. Thirteen and up became regular from there on out. (Thirteen’s only “irregularity” is its shortened first vowel, but that is explained by regular phonology. Its syntax is exactly the same as fourteen and up.)
I’d say these examples are pointing toward base-ten for Germanic languages. Dozen and gross are of French origin.