The tree could have been hit with no visible signs.
Squirrels knocked out of tree (or left 'voluntarily') and died when they hit the ground.
If they were on the trunk of the tree, there could have been fatal current flow between front and rear paws.
If on the ground there could have been similar current. Since paws are close together they would have had to be near what was struck. (Radial voltage drop decreases rapidly away from the earthing point.)
Died of fright (well it happens to people in horror movies).
The lightining striking the ground anywhere near the tree (or on the tree itself) could have electrocuted the squirrels. The step-potential on the ground could have been on the order of several thousand volts, well within the range to kill a small animal.
Lightining is very strange. I've lived in Oregon for about 10 years and can count on two hands the number of times I've seen and heard a classic electrical thunderstorm.
The Pacific Northwest air is cold and while it rains a lot, you rarely will see the intense thunderstorms that are produced in the South, Midwest, and Eastern parts of the USA.
For that reason, grounding of substations and transmission lines does not get as much attention here as it does in other parts of the country. Many times I will see transmission lines run without a static ground wire at the top.
Except for two weeks ago, during the morning hours a bolt-from-the-blue struck near my house and blew up a tree. The debris damaged at least three nearby houses. The skies were generally clear and this was completely unexpected. There was no other apparent lighting or thunder in the area.
To me, it sounded like a 500 bomb had gone off, right outside my house. It made the news that night, and I was astounded to learn that it hit the ground about six blocks away. (I'm guessing it must have been travelling horizontally for a while, directly over my house). I just heard one big, sharp blast. There was no delay between flash and thunder, whatsoever.
It had to be a positive stroke, with exceptional power. I imagine that had it hit my satellite dish, the puny little lighting arrestor probably wouldn't have done much good.
Actually - most livestock, wildlife, and human deaths from lightning are due to the potential difference created when 10's of thousands of amperes are directly or indirectly (induced) between 2 (or 4) points on the ground. Rules of thumb vary, but for the sake of argument consider a squirrel's maximum span is 8 inches. At 2,000A and 5 ohms per foot - he'd experience the delight of just under 7,000V front legs to back. That's just about what he'd get phase to ground (7200V) on a standard U.S. medium voltage distribution circuit. Yep, I think that would do it.
Just for kicks - that same calculation for a cow (5 foot span) would yield plenty of "well done" steak at 50,000V
Park rangers and caddies know when the hair on your arm stands-up or you smell ozone - it's time to squat with your feet together.