The bola was a hunting weapon which consisted of two or more weights connected by one or more cords. This device was thrown to entangle a target. In Florida, the bola is generally credited as belonging to the Late Paleo toolkit.
The examples I have from North Florida rivers are the classic egg-shape with a flattened end. I have alway understood that the flattened end was the point of attachment of the cord.
Since there is no groove in the classic bola stone, it must have been cradled in a leather wrap with the cord attached to the leather wrap. The leather would have provided a buffer when the bola stones cracked together.
While the leather would have its own wear problems, it would be more easily replaceable than finding, grinding, and pecking a replacement stone.
Two key questions remain unanswered:
1. Why would someone go to the trouble of pecking and grinding a pebble into the classic bola shape when any round stone would work?
2. What is the function of the "dimple" on the end of a bola stone?
It occurs to me that a teardrop shape may be optimal (or was believed to be optimal) to fly straight and true.
It may be that stream-rounded stones of the proper size were not common locally. (Try walking the bank or the riffles of the Santa Fe River, for example, looking for stream-rounded pebbles.)
In some instances, pecking and grinding may have had the primary purpose of reducing high points or sharp edges. A rounded, uniform surface would be less likely to produce a cut in the leather wrap in a stone-on-stone strike.
I also suggest that the dimple may have been a recess to protect a knot in the the leather wrap. In use, bola stones would inevitably crack together. It might take a good number of uses for a leather wrap to fail under such pounding; but, it might take just one sharp strike on an exposed knot to cause failure.
But, that's just my opinion.