Yes, if there's no ground strap on the starter, then it's grounded through its case directly to the engine block.
Sean's suggestion above would be a good way to eliminate the starter as the problem. If you power it directly from a separate battery via jumper cables and it works normally, then you'll know that the problem is elsewhere. Just be ready for sparks when you make the final connection! Wear gloves and eye protection, and make sure there's nothing combustible nearby.
I would suggest making the positive connections first, then attach the negative to the starter, then finally make the negative connection at the test battery. This keeps you away from the prop and eliminates the possibility of arcing/sparks causing weld damage to the starter terminals.
The starter behavior that you described in your responses to my first post is consistent with a high resistance connection somewhere in the starter circuit. The high current demand of the motor causes a large voltage drop across that high resistance, which reduces the power of the motor to such an extent that it can't rotate the engine.
I'm sorry that I don't have a better suggestion for troubleshooting, but you're going to have to get to the battery terminals and perhaps the battery contactor. If you can prepare a couple of long wires with crocodile clips at the ends, attach them to the battery terminals and route them up near the firewall, you'll be able to use them for the measurements (just don't let them short to each other or let the positive wire touch ground!). They can be thin wires, as virtually no current will flow through them during measurements; they're just extensions of the multimeter probe leads.
Best of luck!
