[daip] CALIB

[Eric Greisen]

CALIB is totally dependent on the NX table.  If that table separates
the data then CALIB will (or at least is expected to).  If that table
does not, then CALIB will be lost.  Note that, if the data have the
same source number and same freqid then nothing can tell them apart
except perhaps FILLM which does attempt to write the first NX table.
Any later NX table, e.g. by INDXR, UVCOP, etc, will also be at a
loss.

Observing multiple close frequencies without separating them by
something (e.g. qualifier) does lead to later difficulties.

Eric Greisen