Clutching at straws - I wonder if it because I define lowest level maps outside of the survey they are associated with?
Well, that is definitely different from me. Every survey contains its own maps of its scraps. survey foo -title "day 1 at the swamp" map fooMP -projection plan area1SP area2SP endmap centreline ...data... endcentreline endsurvey Each parent survey contains maps with any potential offsets: survey thecave input foo/foo.th ... #these say how long the arrows should be # XP = exploded plan, in case you were wondering map fooXP -projection plan fooMP@foo [4 -47 m] none endmap ... #these say where the arrows should point to - ie. the total offset #of the "parent" piece of cave it connects to map thecaveXP -projection plan fooXP [123 15m] none endmap endsurvey The grandparent survey contains the whole area, minor caves, surface, etc. and for testing purposes, I added another offset at this level, and everything worked: survey thevalley input thecave/thecave.th ... map valleyXP -projection plan thecaveXP@thecave [-20 123 m] none othercaveXP@othercave endmap endsurvey This is the basic structure that I try to maintain. I do not generally have sub-sub-sub-surveys since none of my projects so far are big enough to want to divide a cave into areas within areas. So I do not have any data to test deeper. (Inception might come into play.) Who knows why Therion likes my setup more? Tarquin