Most crosswords in American publications have a few structural features that have become fairly standard since the days of Margaret Farrar's tenure as the New York Times crossword editor. They're usually symmetrical in that the arrangement of black squares will be the same if you rotate the grid 180 degrees, but sometimes that symmetry takes other forms, like when the black squares are mirrored about the center of the grid. Most puzzles tend to have the same number of rows and columns, though occasionally you'll find a puzzle with an extra row or column or two.

Another one of Farrar's conventions was that puzzles should have a limited number of black squares. While there's never any hard number on that, constructors have often used 16 percent of the grid as a benchmark. As one comparison, my Dec. 22 puzzle "Sign of the Times" had 89 black squares in a 21-by-21 space, which was 20.2 percent of the total grid -- a fairly heavy amount of space needed for black squares. It's also become standard for constructors to keep an eye on "cheater squares," which are black squares that lower the average word length but do not add to the total word count. Cheater squares are common in crosswords, though it's also become conventional wisdom among puzzle writers to avoid having too many cheater squares since they make the overall puzzle smaller. (Merl Reagle had an even more playful term for them: Black Adders.)

There's no law that says you have to abide by any of these conventions, though. They're just aesthetic qualities. So what did I do for this puzzle? I tossed them all aside. Today's crossword is an asymmetrical, 18-by-26 puzzle with a whopping 146 black squares, which is 31.2 percent of the entire grid. Seventy of those black squares are cheater squares, which I'm sure is a personal record for any puzzle I've written. There are only four answers that are 10 letters or longer (COMMON ROOMS at 15A, JOTTED DOWN at 44A, QANTAS AIRWAYS at 77A and STREET SMARTS at 81A). And finally, there's a meta component to this puzzle, where the note above the clues says the following: "One letter in the finished grid must be changed to complete this puzzle's theme. Which letter is it, and which letter should it be changed to?"

What could be the reason for such an unusual grid? The title is an important hint, but there are some other pieces of information that might help you understand the method to the madness:

There are some less common letters cropping up near the bottom of the grid, with a pair of X's and Z's.Those four long Across answers listed above are in alphabetical order.There are exactly 26 rows in the puzzle.

The key thing to notice is that the 26 rows proceed in alphabetical order such that the Across entries in the first row start with A, the second-row entries start with B, the third-row entries start with C, and so on ... but there's one exception.

1A: ["EastEnders" network] is BBC, but this breaks the pattern in the rest of the puzzle. To fix it, change that starting B in the 1 square to an A, which appropriately would give you the entry ABC. (You would also create another valid word with AERATE at 1D.)

I don't think there was any way I could have written this puzzle with standard symmetry, though I certainly did try that at first. My first draft (pictured at right) was symmetrical, but the only alphabetic theme answers were the first entries in their row. I threw in SERIES / OPENER and ROW as additional hints, but thematically it just felt too thin; the entire right side of the grid was basically pointless. I figured, why not continue the A-to-Z pattern throughout every row? That would give the puzzle a little more thematic heft and would give you additional avenues to spot the alphabetic pattern. And believe me, while I am hoping you enjoyed solving the puzzle, I can tell you it was astoundingly fun to write. Freeing myself from the standard restrictions on symmetry and the number of black squares meant I could build the grid in whatever shape I wanted, provided that I kept to the A-to-Z constraint. I still have my own aesthetic preference for symmetry in my puzzles, but it was refreshing to just let it go for a change.

As a heads-up about next weekend, I am going to warn you that the Feb. 23 puzzle might be pretty challenging. Good luck.

What did you think?