GWGill,
You may well be right, so thanks for sticking with this as correcting an erroneous belief I may have is valuable to me.
I1Profiler, in single pass mode, does not appear to activate the uV LED at the start of a scan. If that is maintained during the scan then the data must be M0 and the tungsten lamp would have to supply M0 levels of uV. This would mean the tungsten lamp was not uV cut.
There are several possible ways M0, M1, and M2 data can be generated with only 2 scans, and M0 data with 1 scan.
I'll try to capture the USB data directly during an I1Pofiler scan. One of the things I noticed some time back was that all the spectral data above 550nm were identical on the saved, M0, M1, and M2 patch scans. This suggests that the uV LED is the only illumination present on the reverse scan on the same patch line. Levels above 550nm, which would mostly contain noise, were discarded leaving the reflectance levels above 550nm unchanged across M0, M1, and M2.
However, I can't explain the differences in the fluorescent paper spectra if the tungsten lamp isn't uV cut. Perhaps the USB capture will elucidate.