The way I read the rule on p. 125 you quoted (or Damage Resolution procedure from p. 122, or the one you posted in the thread I referred to), it could mean one of three things:
1. Critical hits (crits) transfer only to the very next location along the Damage Transfer Diagram. If there is nothing there to crit (or the location does not exist anymore, which is possible in case of leg hits), the crits are lost.
2. Crits transfer to the closest location, which has critical slots, that can take a hit at the moment the crits are applied.
3. Crits transfer to the closest location, which at the beginning of a current phase had critical slots, that could take a hits. If all such slots took hits since then, or the location itself is destroyed - all remaining crits are lost.
From your answer I understand, that interpretation 1 is out, although that is something, that comes from your answer only - the rules read literally imply only that the crit transfers only from the damaged location to the next location - "one layer in" so to speak - and simply don't say what to do if there is nothing to crit there. I guess you mean interpretation 3, since it is the way it is decided if the crits are transferred from the location, that took them in the first place, so the entire mechanism of transferring critical hits is identical whether we consider transferring them from the location of the original hit, or further propagation from the side torso to central torso in case of a limb hit, but since I believe, that the rules you quoted could technically be interpreted in any of the three ways I mentioned.
Perhaps it one more thing, that should be clarified in the errata document? Not that I insist, since it is such a rare situation, but considering, that you work on BattleMech Manual at the moment, I think, that it is perhaps better to fix a few things to many, then risk releasing a book with one or two things, that will require an errata later, even though it could be caught before the release.
