MIMF

This is a replica bridge that I made and installed on this guitar a few years ago because the original bridge split just like this one, in the saddle slot. The saddle height is a little high (I suspect the neck was over-set at some time) and is probably contributing to the problem.
I'll be making another bridge, and this time I think I'll choose wood that if off quarter rather than the nicely vertical grain of this one, and I was thinking of making the bridge itself a little thicker so that the saddle doesn't protrude above the bridge so much and there is a little more wood to perhaps strengthen the bridge some.
Any other ideas? How can I prevent this from happening again?
The 12the fret is a little bit clear of the edge of the body so I could reset the neck for a lower string height at the bridge, and since I have to make a new bridge anyway I could move the saddle slot accordingly, but that's a lot of work...

Edit: Sorry, forgot the pic. Here it is.

Making the bridge taller sounds good to me.

Using off quarter wood might help, but making the bridge thicker and heavier might change the sound (better, worse?)
Although it is poor form, gluing the saddle into the slot might add some strength. The over set neck and tall saddle is adding a lot of stress, but if the player likes the sound, leaving the saddle tall and gluing it into place may be the best solution for adding strength without adding weight.

Slightly off-quarter is good, as would choosing a quartered blank with the grain running a bit "off" end to end, such that the saddle slot doesn't follow any one grain line. Sometimes, we just run across a blank or three that aren't as tough as they should be, so I wouldn't change too much about it at this point. You might have simply had a weaker example in that blank...

I'd avoid a taller bridge at all costs! I suspect you would, too.

I don't know how deep you usually cut the slot, but I would suggest going deeper than normal, to within .090" of the bridge bottom. The more saddle you have -in- the bridge, the less leverage it has against the bridge, despite having the same amount above the bridge. Basically, you're putting a larger percentage of saddle into the bridge, lessening it's mechanical advantage(leverage).

I've actually had good success in gluing this type of break, too. If it's a fresh break, and it will clamp tightly, the repair will be stronger than the wood was originally. HHG, of course. I just did my '51 000-18 this week, after the saddle folded forward and completely sheered the bridge in front of it. Literally, the bridge was in two pieces. It has held up nicely so far, and I have some similar repairs going-on a decade or more, at least. The only bit of woodworking I did with the '51 this time was to route the saddle slot slightly oversized and a bit deeper, fitted a piece of BRW perfectly, and glued and clamped that in place. Next morning I shaved the insert to the bridge, and re-routed the slot at .093" and to its original height, made a new(taller, since the action was a bit too low for me, anyhow) saddle, and in this case, glued it in place, as was done originally. I did this because the slot was slightly curved(the existing saddle was not glued-in), and I figured that the insert would add strength right where the bridge needed it most. Now I have a nice and straight slot, with a nice and flat bottom.

When making replica bridges, I normally measure the original and cut the slot to the same depth. I can see how deepening the slot moves the fulcrum point, and that does seem like a good idea. In fact, I think I'll cut them deeper in the two new bridges I'm making right now, though as they are there is only about .120" of thickness under the slots, .030" more depth could help them in the future.

It wouldn't hurt to angle the saddle back a few degrees, that would reduce the torque the saddle is exerting on the slot...

Doesn't fit very well with the design of my bridge slotting jig, but something to consider none the less.

Hijacking the thread -- I've avoided angling the saddle back since the intonation would change with every saddle height adjustment. Maybe it's not enough to worry about? Or the benefits outweigh the hassles?

It actually works in favor of the situation. First, the amount of angling back is very small, and still the higher you raise the action, the further back it angles and that is when you need more compensation anyway. Lower action needs less compensation and that works with a lower saddle. The benefits of the angled saddle are more than the problems it may cause. it really helps a guitar with an undersaddle pickup.

Yay to everything Mark said; I've been angling-back my saddles since the late 90's. About 6-8 degrees is ideal. As you raise the saddle, you move the intonation point further back, compensating for the raised action's need for slightly more compensation. In short, it becomes a self-compensating system.

I don't do(and refuse to do) undersaddle pickups, so can't help there.

Late to the party again. John, I think the original design is sufficient but you need a better blank to make the bridge from. Older and darker BRW has a tendency to split more easily than younger lighter colored wood, more or less. All the other ideas are great if you want to go to the lengths to make the changes, but you know well the original designs have been working for many years.