When you look at slingers from slinging cultures, almost all use one or two rotations in a horizontal or angled plane followed by a sidearm or overarm throw.

This is a typical example: http://www.youtube.com/watch?v=pofjXqod774

Anywhere between 100 to 200 yards range is common with stone ammo, although further is possible.

Damn I'd LOVE to join!
But I can't afford it right now 
Greetings,
Mauro.

Sorry Pete, don't you have a drawing of the profile section of one of these hillforts?
I have one for the Italian gradinas that I'd like to post, just have to find it...
Greetings,
Mauro.

I like this topic. It's interesting. I thought I'd throw in my 2 cents. I live on the American plains, so there's not a lot of hills around here to throw from. I have however had numerous opportunities to throw off of creek bottom drop offs and once even off of a 120 foot tall grain elevator while I was on top of it doing an internet antenna install. While the height definitely allows for an increase in range, I found the angles for straight line accuracy quite awkward. Especially on the elevator where a guardrail simulated something resembling a parapet wall. In order to get the angle of release right with the rail in front of me my shots had to take a high lobbing throw and were more on a straight down drop by the time they hit the ground. I would be interested to know the ballistics on this as far as how much energy is lost when the stone is on almost a straight down fall. When throwing at closer range to the drop off, however, (less than 100 yards give or take) straight line shots become much easier with more of an overhead release. This effect, to my somewhat educated guess, would allow attackers to get close enough under cover of shields to negate the added range advantage of the defenders. With the angle of incoming projectiles being predictable and somewhat overhead like a continuous rain of stones. The ballistics of lead would also give whichever side is using it a very clear advantage. For defenders it allows a more straight line accurate throw from the parapets, in addition to longer range lobbing throws. making shield defense against such attacks tougher, unless the enemy is advancing in a phalanx type interlocking shield formation a-la Greek and Roman tactics. For an assaulting force using lead while defenders are using stones, this would even further negate the range advantage of defenders throwing from an elevated position.

I personally reckon one of the more effective ways slings could have been used en masse back in those times would be the blind lob C_A described. I remember hearing once that ammo up to 500g was used by some Balearic slingers. Now although this would have been the very top limit, when just throwing a lob underarm into the sky, its very do able, and with that weight falling a good height, even a shield will do nothing.
Assuming other people threw even 300g+ rocks in a similar way, with very little training I think most people could manage 50m+ with good height added to that to crush anything it falls on. It doesnt take much training to lob into a rough area and en masse it could be very effective.

Nemo

For what concerns Italian defenses, palisades have been hypotized for every parapet, but since not a single Gradina has been archaeologically excavated, we don't know it for sure.
I find the idea presumable, though.
By the way, I'm still looking for that section drawing, can't really find it...
Greetings,
Mauro.

Dear Pete, I think it's time for me to explain you some of the (few) evidences that characterize one of the most ancient military settlement of Italy (and the whole Europe): the Copper Age settlement of Conelle di Arcevia.
Imagine a plain at the top of a very low hill (around 15 meters high), surrounded by other higher hills on the South, West and Northern sides, and by other higher ones, but even more distant of the first ones, on the East side.
To tell the truth, Conelle di Arcevia was located in a valley, but the hills around are all well away from the settlement, out of archer's range, for example.
There're two small rivers in this valley, that connect at a certain point.
The plain of Conelle di Arcevia begin from the junction of these two rivers, and proceeds South for about half a kilometer.
After this distance, the ground lower slowly until it reach the river's level.
So you look from above, and what do you see? a wedge of ground between two rivers.
What did these Copper Age people do? They dug a V-shaped moat on the South side of their settlement, actually making a barrier that transformed the wedge into a triangle.
The moat is 120 meters long, 8 meters deep and 7 meters wide. It's made of two semicircles, with their convex part facing the outer side of the plain. This made archaeologists believe that such a design was thought to allow archers and slingers a crossed fire against incoming forces. The moat has just an entrance, which was in the middle of the two semicircles.
It's also plausible that, while digging the moat, Copper Age people had piled the extra soil on the inside part of the moat, so that they could reinforce it with an artificial scarp, which could also have been further fortified with a stockade.
This was perhaps the defense adopted for the South side of Conelle.
And what about the East and West sides of the triangle?
They were naturally delimited by the two small rivers, plus the hill's slopes, while not that high, are steep nonetheless. Blackberry bushes and brambles grows naturally along the rivers, forming an additional defense. Curiously, the ground along the triangle's borders does not show signs of any artificial intervention, such a scarp.
Perhaps the sides were fairly well protected by nature, or perhaps they added stockades to them.
I don't know if this description could help you, but I bet it won't disturb 
Next time I'll describe you our Iron Age hillforts 
Greetings,
Mauro.