If you have arrived at this location due to clicking on an URL posted in a forum thread, the subject article has been removed due to invalid assumptions I made based on incorrect data drawn from the experiments. I suggest you go back and read the rest of the comments in the forum thread. I will state that subsequent casting experiments confirmed the following:
* It's not clear why, but I found that stirring the alloy with the ladle while casting significantly reduces the resulting bullet weight variances. By stirring I mean lowering the ladle to the bottom of the pot a couple of times then filling the mould, all while maintaining a constant cadence & always filling the ladle to the same level.
*Lead/tin alloys are a homogeneous solution. Therefore, the tin will not separate from the lead. The tin will not stratify in the alloy and/or concentrate on top of the alloy. The lead/tin solution exposed to air on top of the alloy pool may oxidize at the same or slightly different rates but the lead/tin ratio of the alloy will not change a significant or measurable amount due to fluxing or removing the slag/dross.