OK i talked with our "Gauge,instrument inspection and calibration Tech" and he said that frame is probably alluminum, Aluminum would acclimate rather quickly but other metals are slower. So once you wait and both the product and the Micrometer have acclimated they would stay very stable under gradual temp. changes.for example at a +/_.0001 accuracy,40 deg.over 8 hr you wouldnt see. Now with rapid temperature changes the aluminum would react violently and you would have to wait longer for your product to acclimate then re-zero. For slow gradual changes the variance is absorbed in the accuracy.
All micrometers of any quality (especially old Starretts like this one) have frames that are made from forged steel. If you picked that mic up you'd be able to tell right off that it isn't aluminum. The frame is heavy duty so it doesn't flex when measuring and also to minimize radical swings in thermal expansion.
Aluminum has a greater coefficient of thermal expansion than steel does and also greater thermal conductivity. With a micrometer that large the accuracy would be crap if it was made out of aluminum. It would grow just from the person holding it in their hand. I've never seen an aluminum micrometer.
Starrett is an American icon synonymous with the ultimate in quality like Snap-On. Although both companies now sell some equipment made in China (especially electronic stuff like digital calipers and mics and other tools) they both still sell ultra-quality made in USA products too.
Both companies have the best slogans in their respective industries.
Starrett's slogan is: "The World's Finest Toolmakers".
and of course Snap-On's is: "Nothing Else Even Comes Close".
Starrett mics are second to none just like Snap-On hand tools.
As far compensating for thermal expansion on the shop floor, that is done very carefully. Since each metal has it's own known thermal expansion factor, you can calculate the thermal expansion of both the gage and the part being measured and compensate for the reading on the mic. Theoretically when you bring them both into the climate-controlled metrology lab then they will measure as expected.
I am involved in the design of plastic mold tooling. The cores are all designed to have a precise centerline at operating temperature in order to properly mate with the cavities. So the thermal expansion of each material must be taken into consideration. The larger the mold is (like this big micrometer) the greater the thermal expansion and a lot more challenging it gets. A lot of our molds will not close at room temperature because one half of the mold is designed to be hotter than the other half. You must bring the hot half up to operating temperature before it will mate with the cold half. The centerline distance is intentionally designed smaller on the hot half so it can "grow" to size after heating up and will then mate perfectly with the cold half. Some of the tolerances are as tight as 50 millionths of an inch (or one half of a ten thousandth of an inch).
That big Starrett micrometer was a very expensive metrology instrument when it was new. Because of it's size, it isn't one that most toolmakers would own. That's something the company would buy most-likely for a specific part of the manufacturing process. It is still worth a lot of money if it isn't damaged. You can send it back to L.S. Starrett in Athol, Massachusetts and they will refurbish it and re-certify it for a fee and then it is as good as new.
How much were they asking for it?