After long and careful consideration, I've decided not to do this test.
The problem is that in order for the test to be valid, I'd have to boot from the external drive for at least ten days and possibly more. That means doing without many of the things I've set up on the internal drive and on my personal account. I could set all those up on the external boot drive and a generic identity, but this would be a tedious and time consuming task and involve making significant changes on both local and remote machines. Things I'd have to set up or change include:
• Postfix SMTP server
• MacPorts
• Numerous shell scripts and the launchd plists that control them
• Public/Private keys on the local and remote machines
• Remote (reverse) SSH tunnels on remote machines
And many more, some of which I probably wouldn't discover until something important stopped working.
It's hard for me to justify taking the time and effort to do all this on a drive that might be in use for just a couple of weeks and for a test that might not even be conclusive.
Since the frequency of kernel panics has fallen off quite a bit, it makes more sense to just put up with them until I decide on buying a new machine on which setting everything up will actually be worth the effort.
@macguru9999 - Thanks for your help and suggestions.