# Chiasmata Solution

## How It Works

The primary mechanic of this puzzle is arrangement of the pieces onto the chromosomes so that each dowel was filled exactly to length with pieces. There are lots of ways to do this, but there is only one way to arrange pieces when certain properties are maintained. The instructions and "Approved Chromosome Chart" hint at these properties:

• A single innermost piece and two or three outermost pieces on each dowel are colored.
• Except for the outermost pieces, no other colored piece appears consecutively.
• The silver pieces only appear in consecutive groups of ones and threes.
• The orientation marks point inward.

The instructions hint that the silver sections "are lengths of DNA that are translated into proteins," that the colored sections are "unexpressed areas between" silver sections, and that translation proceeds in-to-out. Combined with the prior rule that the silver sections come in discrete one and three lengths, we get Morse code. Translating the silver triples into Morse code dashes and the silver singles into Morse code dots yields a letter on each dowel. Combining the numbers on the dowels and color ordering across the chromosomes, you get a request for trivia: "your order in animal kingdom." Assuming you're human, you're part of the animal kingdom and your biological order is primates.

Optionally, the Approved Chromosome Chart could also be translated so that independent progress could be made on both arrangement and translation. It read "match same hues follow rules."

## Solution

Primates (primate also accepted)

## Design Notes

The uniqueness of the solution was proved by brute force computation. The complexity of the problem grows exponentially. Proving the nine-piece problem took seconds, while an eight-piece problem was fast and a ten-piece problem took about a minute. Combined with plenty of rules to allow for logical deduction, nine pieces seemed to be about right for humans as well.

Fabrication of this puzzle is generally believed by staff to be the second-biggest pain in the history of MSIG. Each puzzle included: 6 cut, drilled, sanded, painted, dotted, re-drilled, and sometimes re-re-drilled blocks; 24 cut, sanded, painted, and dotted dowels; 248 painted spools; 194 glue joints holding the spools together, about a third of those being re-glued, (glue joints, not pieces, mind you) about half of those being re-re-glued, and about half of those being re-re-re-glued before giving up. The excess gluing was due to breakage issues encountered throughout playtests, beta, and RC. For the final run, we actually tried breaking the pieces and re-gluing the joints that failed, the theory being that you keep removing the minimum-strength joints from the population. After a few cycles of this, we decided it was sturdy enough.

If that doesn't already sound like a lot of trouble, scale that up to 15 copies. 90 blocks, 360 dowels, 3720 spools, somewhere around 5000-6000 attempted glue joints, a week and a half of evenings, and multiple offers to right things by ritual suicide on my part. My bad.

## GC Notes

GC was pretty quiet. I heard some data confirmation and some gentle nudging, but it seemed like things went fairly smoothly. By contrast, the beta fast solve was one hour and the slow solve was close to three with lots of help.