
Finding this mold is not easy as it has been out of production for some years. I have used this design with good success. The best choice of those that you list would then be the Lyman #311334. As such the best choice of bullet design would be that of a short(er) bullet "body" and a longer forepart.i.e. Harrison of the NRA in some of his classic investigations into 30 cal bullets the lands of the rifling occupy 5/8 of the circumference of the bore. The type of rifling is 5 groove and as pointed out by Col. The bullet molds that you list are all excellent choices and all should perform well. I've got 3 of these and have been shooting cast in them for a very long time. You should have lots of fun working up and finding the load for that M1917 and then shooting the matches with it. Final selection would be based on low ES and smallest group size of the number of shots, including sighters, of a string in the match. The best accuracy will come between 16 fps. I'd test with a chronograph looking for a low ES and the SD about 25-35% of the ES. I'd use well fire formed cases match prepped that were NS'd. I would also use a 3/4 gr dacron fillet over the powder. I'd preseat the GC size them as close to "as cast" as I could.Lube would be Javelina or Lars 50/50 alox/beeswax or his BAC. During final selection of the load and for the matches I might weight the bullets for consistancy depending on the size of the 10/X rings of the match target. Initial load development would be with visually inspected bullets for no wrinkles or voids.

I'd use an alloy about the same as #2 and would air cool the bullets.
