Listener No. 4370: Hefbeet by Yorick
Posted by Dave Hennings on 20 November 2015
Just one puzzle to date from Yorick, as far as I can see, and that was almost exactly ten years ago… but in the Independent. That had Androcles and the Lion as its theme and bizarrely the preamble started in almost the same way as this week’s with “In 26 clues…”. For some reason, the number 26 rang a faint bell!
The first thing that I noticed was that the grid had a couple of double-unches. Oh dear! Alas, I know some well who would have found that a bit jarring. Luckily, I’m not one of them, so I got on with solving the clues. The first letter of the each of the 26 was to be misprinted, but I made a note of the correct letter in the corner… they might be needed later. The remaining 18 clues had an extra letter and their initial letters would spell out something useful.
The first half dozen acrosses looked a bit tricky, but 16ac was BLEAR from first letters. I decided to try the downs in the top left, and 2dn On the left, Stanley Matthews is a capital example (4), I was pretty sure was PORT with an excellent surface reading, and Matthews being an extra word, so it wasn’t a change-the-first-letter type of entry. Crossing the L of BLEAR meant that it wasn’t a simple jumble either, but probably some sort of code.
Perevering with the downs, [I]GNOBLE and [D]REAMT were soon entered, followed by [O]LEASTER. Following on from that, everything became a bit erratic, and entries were slotted into the grid in a somewhat random order. It was obviously essential that the body of misprinted entries gave the correct letters for initial misprints in the crossing entries. It also confirmed that the other entries would need to be coded somehow.
It was still at least two hours before I could see what the first letters of the extra words spelt out: Code using misprints. A short while later, and my list of misprints alongside their correct letters had ALL THE LETTERS OF THE ALPHABET, albeit with a few gaps. Thus each misprint, in clue order, encoded to its corresponding letter in that phrase: S at 1ac became A, Q at 5ac became L, J became L, M became T, etc. This code could then be used to enter the non-misprinted entries. Was this obvious to everyone straight away?
I still had about ten gaps in the code that needed resolving by the remaining clues, but eventually they were teased out. There were certainly some tricky ones along the way, such as:
|18ac||ZEROS||Ciphers that could convert I to C? (5)
How I becomes 100; I’d be interested to know whether 1 and 0 can be freely interchanged with I and O
|20ac||WOMEN||Henrietta Maria and Marie Antoinette, say, seeing Charles I and Louis XVI beheaded? (5)
|36ac||CAECA||Old Oz college [news] about biological blind ends (5)
CAE + CA
|19dn||CHIMERA||Big Ben? A regular monster! (7)|
20ac must be one of the longest clues to a 5-letter word that I’ve seen.
Despite being at ease with double unches, 1ac Cover girl has inner passion (7) proved quite tricky for me, easy though it turned out to be. Also 25ac Black medic loses no people of the same urgent kind (4), which was obviously SUCH, but took me ages to find the black medic connection. Reverse-engineering a couple of the other coded entries also took a bit of time.
The final code that I drew up was:
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
E E E P R A B O T L T S T F A H L E A L E H T T T L
All that was left was to decode Hefbeet to give PANGRAM below the grid.
A quick check of my grid before sending it to AL3 and I spotted that I had entered WOMEN with the N encoded to F, despite it being a misprint clue. A close shave, and a reminder of Rule 12! That aside, not too taxing an endgame this week, but an enjoyable puzzle nonetheless. Thanks for this one, Yorick.