Dear Brian, I am afraid I have to agree in that USNO-A2.0 is indeed not much better than USNO-A1.0 , but it is slightly better. I derived sequences for a number of regions in both systems based on GSPC fields and got reasonable agreement in the results (+ or - 0.2) for V. I use V= (mr+mb)/2, arounded to mr.. and correct then for the calibrated off-set. I stated that before. In a number of cases USNO1 gave too faint values for brightish stars. This seems to be improved in USNO2. I also found that field calibrations differ for the same region and the further away from the calibration field (GSPC) the more uncertain the transfer. Variables like AS 327 and V818 Sco lie on top of GSPC fields which makes the process more certain. PKS 2005 is a typical case with the nearest GSPC field about 1.5 degree away. Last year I estimated by eye about the same values for those comparison stars, based on H/T values on one side and limiting sky magnitude at any time on the other. With the magnitude scale proportional to visual sensation, the interpolation is kind of straightforward. I must admit with USNO in hand I would have selected at least one other (more bluish) reference star at 12.9..., the ones published are a bit more reddish, except for the brightish 11.9 star which is mostly outside the range unfortunately. Perhaps it will come in handy now with PKS 2005 evolving in the active optical state (I expect some time-delay from the XR brightening). Bulk calibration data have their limitations but they can be made useful tools when used with discretion (like when applying GSPC calibrations). For exploration of Southern fields we lack alternatives, I am afraid. Regards, Berto