I (and probably others) are laughing at you now. 

My maths is excellent - degree level in fact. Your appears to be good enough to illustrate my point for me...
Rounding errors Dan, rounding errors. 

.003” is.003” - or if you can measure accurately enough 0.0762mm. Not .07mm as that is (as you so elegantly show) a rounding error that gives .0027” - and that is exactly my point !!
Stick to the values designed and quoted. 

As for using SI units in general science, they are cleaner and simpler. 
Nuclear physics is much easier in SI which is why I’m bilingual. 

No, the point is: you claim all too often that metric is not as accurate as imperial measurement, when in fact it is equally precise, if not more so. Why else would the military, science, medicine and engineering - including automotive - use metric measurement almost exclusively.
It shows you don't seem to know much about math and you can't take criticism.

Methinks you've somewhat missed the point and made youself look stupic there Dan...

As you would say "BOLLOCKS"!

You are comparing cabbages to pommes. Since there are 25+mm (intended rounding to keep it easy to understand) to the inch, there isn't a straight comparison when it comes to decimal places. To prove that, lets round of your metric number 0.0762 to just two decimal places = 0.07mm. Now divide that by the conversion factor 25.4 and you get 0.0027559 inches. even ignoring the last 3 digits and rounding up, you'd get 0.0027", not your 0.003".

Oh, and by the way 0.003 is a decimal fraction, just like 0.0762, so your favourite quip about fractions doesn't quite fit. if you want to really use "fractions", your 0.003 would be 3/1024" (0.002929689" to be exact). (1/1024 is 1/4 of 1/256 when you are working with fractions.)
 
But I do agree that plastigage would be the best way to verify bearing tolerance. It would work in metric or imperial too!

Metric is of Fench origin, and generally for people who can't do fractions...

Working in whole numbers removes an error trap.

As an example, setting an end float to 0.003" - 0.007" is ridiculously easy compared to 0.0762mm - 0.1778mm and involves no rounding errors or judgement calls.

French?  The multiplier .03937 for imperial to metric mm, gives an accuracy 6 digits to the right of the decimal point, and 25.4 for metric mm to imperial is exact.

Or do I have a lesson coming?

Plus 1 for plastigage as the final step, and a good way to remove it without scratching the journals is with the free little 6" trade show flexible plastic rulers.

Your machine shop should have the housing bore diameter specs in their bearing catalog. Look under Austin, BMC or even Mg Midget 1275.

I had it at one time, but not sure where it is at the moment

My machine shop carries two sets of measuring equipment, one Imperial and one French - tolerances and dimensions should ideally be measured in the manner that they were designed & calculated as conversions are often lacking.

Bench checking & measuring is no substitute for empirical mesuring with plastigauge - bolt it all together and make an educated decision based on what you see.  My racing spec crank was recently measured after a bearing failure and the empirical result was close but not identical to the measured one - luckily the result came back in my favour which saved me regrinding a hardened crossdrilled crank.

.002" with a high flow pump ought to be around your top limit.

Hi folks, 

What is the diameter and tolerance on the mainberaing bore in the 1275 non S engine block ? 

We have got a blok slightly line honed, and the crank ground 0.20" on the mains.
The crank now seems to have a radial clearance of 0.06 mm (close to top limit of 0.069 mm) will that be ok for a high reeving engine ?
The crank is ground ø50.31   - Would it not have been better to have ground the crank 0.02 mm above upper limit, so a nominal clearance would have been obtained ? 

Short about the engine
The engine is a short stroke (69 mm)  x 73.5 bore made as a full race engine with 7-port cyl. head and fuel injection. Power band to 9000 rpm. The rotating mass is made with the lightest steel parts available.