Copy/paste from another post I did
......I believe the purpose of the weight is to counteract the effects of driveshaft resonance due to it's Critical Speed. The CS is the RPM that the driveshaft spins at where it begins to vibrate. This can vary from driveshaft to driveshaft. Aluminum, because its thicker, does a better job of resisting this resonance. I have no idea at what driveshaft RPM this occurs at.
I dont have hard data on this, just some assumptions. There was a great article written by Jerry Wroblewski who was a Ford drivetrain engineer. He touched upon this and the weight and it's impact with the Driveshaft's Critical speed. Unfortunately I do not remember the context of the article and the link I have bookmarked no longer works.
But, for the 86-04 8.8 axles, Ford stuck 2.73, 3.08's, 3.27's and I think the Mach1 got 3.55's. I believe they all got the same weight. Changing your rear axle gearing would not change the (driveshaft) RPM that the critical speed of the shaft is reached. Changing the rear axle ratio would just change the relelative vehicle speed that this CS would occur at. It doesn't negate the purpose of the weight as it would still be doing it's function, but just at a different (vehicle) speed.
But like I said, i'm just speaking of the concept, as I have no concrete data on this, nor any way or desire to go out and get data on it.
In terms of failure mode, I would think what you would see is excessive vibration causing pinion and trans seals to leak, and pinions/trans bearings/bushings and U-joints to begin to fail over time. But it would be tough to prove causation, vs correlation. Many cars don't have their weight anymore, so is that leaky pinion seal because of the missing weight, or would it have happened regardless of if the weight was there.
EDIT: BTW Jerry Wroblewski was the Engineer who devised the "Jerry's Mod" for the SN95 automatic trans.