In his continuing coverage of l’affaire Lott, Mark Kleiman asks:

if the Ayres and Donohue statistical results showing a mild “more guns, more crime” effect were correct, what mechanism might account for that effect?

That would be the trigger mechanism.