If you like Latin squares and such things, take a look at Diamond Geezer’s post for today: a pair of orthogonal Latin squares with two disjoint common transversals, and some entries given (if you do the harder puzzle).
Top Posts
Recent comments
 dsp on The enhanced power graph is weakly perfect
 dsp on The enhanced power graph is weakly perfect
 What Lovelace Did: From Bombelli to Bernoulli to Babbage  on Polynomials taking integer values
 What Ada Did: From Bombelli to Bernoulli to Babbage  on Polynomials taking integer values
 Peter Cameron on The enhanced power graph is weakly perfect
Blogroll
 Astronomy Picture of the Day
 Azimuth
 British Combinatorial Committee
 Comfortably numbered
 Diamond Geezer
 Exploring East London
 From hill to sea
 Gödel's lost letter and P=NP
 Gil Kalai
 Jane's London
 Jon Awbrey
 Kourovka Notebook
 LMS blogs page
 Log24
 London Algebra Colloquium
 London Reconnections
 MathBlogging
 Micromath
 Neill Cameron
 neverendingbooks
 Noncommutative geometry
 numericana hall of fame
 Ratio bound
 Robert A. Wilson's blog
 Since it is not …
 Spitalfields life
 Sylvy's mathsy blog
 SymOmega
 Terry Tao
 The Aperiodical
 The De Morgan Journal
 The ICA
 The London column
 The Lumber Room
 The matroid union
 Theorem of the day
 Tim Gowers
 XKCD
Find me on the web

Join 659 other followers
Cameron Counts: RSS feeds
Meta
I assume that the “harder” puzzle is aimed at nonmathematicians, as presumably any mathematician will just apply the group S_3 x S_3 to the symbols that aren’t J and D to get an appropriate Latin square.
From feedback on the site, it seems most people failed to realise that the second puzzle was a trivial transformation of the first.
Indeed – see David Craven’s comment (he’s a mathematician!)
You may be interested in my post of today.
Pingback: Geezer Puzzle « Log24