The reason I raised this issue is as follows. When I moved to this address I built a new shed/workshop. I laid the floor as level as possible, by the simple method of good strong shuttering, and making the concrete mix fairly wet, then vibrating it, so that, in theory, it should all be level. (It worked for the pyramids, so its good enough for me!) The bench is an old one that is built like a battleship, with 6 x 4" square legs (of identical length) supporting a 3" thick top. I know that has always been square and true, and had no reason to assume it had changed. That, on a level (I hope) concrete floor should also be level, so at the start I was fairly sure that the top surface would be reasonably level. Bolted down the headstock end, leaving the tailstock end fixing bolts loose. Then turned a test piece, intending to shim the tailstock end as required. As it happened, it wasnt required so tightened down all the bolts, did another test piece, and all was OK. Since then I have used the lathe with no problems. As some here may know, I also play with vintage radio/electronics. A couple of weeks ago a pal wanted some items I had, and asked if I would be willing to swap them for "this". "This" was, indeed is, a beautiful wooden case, containing a precison inclinometer. It is graduated to 1/4 of one minute of arc, and one can "guestimate" about 1/4 of that. Now, geometry was always my week subject, but I reckon 1/4 of 1 minute = 1 thou per foot. (But I could easily be wrong there! I am sure someone will check and tell me if I am,) Any fraction of that is...fairly accurate. Checked the lathe and found that the tailstock end is lower than the headstock by about 15 thou. I have no intention of doing anything about it, but just thought it showed that it makes no difference to accuracy. Alan MODEL ENGINEERING DISCUSSION LIST. To UNSUBSCRIBE from this list, send a blank email to, modeleng-request@xxxxxxxxxxxxx with the word "unsubscribe" in the subject line.