Jim360 and Philoctetes pretty well sum up my feelings about the puzzle, particularly with regard to effective checking. I didn't star until mid-day yesterday and was plugging away at it off and on until midnight. Getting the last eight clues t took me as long as the rest put together. The first 20 cold solves took me about 90 minutes, at which point I had enough of the message to guess the entry method. After that, unlike most puzzles that get easier as the grid fills, this one got progressively harder.
I don't know whether I've parsed the clue wrongly but 24a appears to take the form, wordplay from definition, whereas the Ximenean convention (if not rule) is definition from wordplay. I've seen this on the odd occasion in the Times daily cryptic, but the only time I've risked it myself in a barred puzzle it was rejected by the editors (not unexpectedly).
Another issue is a more general one. In a number of Listeners over the past year (including the current one) French words have not been flagged as such (I'm not talking about ordinary imports that are now part of our everyday English vocabulary). It seems illogical that Scottish words must be indicated but French words need not. Can anyone throw light on this apparent inconsistency.