[Greg Reid]

Yeah, test #2 should have pegged it empty if you were making a good connection...although it's not really a great idea to tie the positive lead to ground with the switch on. You can damage the gauge. Best to use some type of resistor between the wire and the grounding point. 40~50 ohms should give you roughly 1/2 tank reading on the gauge.
I have done it many times without the resistor but I just 'brush' the bare wire to ground, very quickly, so as not to peg the meter. I just want to see it react and move in the right direction.

Really, you have to first verify whether the problem is in the gauge or the sender (or both!). If you had a good connection on that test #2, it may indicate that the gauge is the problem...but... keep in mind that you could possibly have a broken wire anywhere between the gauge pod and the tank on that positive side. Could be why you got no response when you grounded it. That's why I usually check the gauge 'at' the gauge. That is, behind the dash.
On test #1....You may have to pull the sender to really check it while working the arm up and down (I know yours is still in the tank). It should go from near zero to about 90 ohms. (If the '64 has the typical 90ohm rating?)

Really, unless you are making an error somewhere in your testing method, you may have more than one issue. You are getting mixed symptoms. Sending unit reads 'shorted] (.36ohms) but gauge is reading 'open' (full reading)
Choose one thing to address first. I would verify the gauge first. Why is it not responding to shorting the positive lead to ground? Fix that issue first. Make sure there is not a break in the wire somewhere between the gauge and the tank.