I like the sound of your bells (also "bell" as a term). In case your volunteers don't grow out well -- or simply because you like the shape:

Some similarly-shaped pastes are Jumbo Roma and Justine from Mariseeds, also Italian Red Pear from TGS (sometimes aka Italian Pear or Red Pear -- but not the cherry with that name).

These are what I call "gathered purse type" -- bell-shaped to some degree, but ribbed: Goldman's Italian American, Franchi's Red Pear (possibly aka Franchi Giant Pear?), UglyRipe, Thompson & Morgan's Cuore di Bue (very different from most of the varieties with similar names), Piriform (Johnny's Selected Seeds probably still sells it).

One year I grew a hybrid paste that was my dad's favorite ever.
I saved seeds and grew the F2. These looked nothing like the "parent" and had bad flavor. I have not yet tried to grow the F3 seeds.

I have also unintentionally crossed tomatoes in my garden; then unknowingly grew the F1s. These are fabulous. Not sure if the F2s will be nice, but willing to try.
Not sure what your bell volunteer is - maybe a cross or a hybrid showing it's parentage, or whatever, but it's always a fun experiment! Too bad the season is so long!

Thanks for the replies!

A question I have is this... Should I plant seeds from multiple fruit from the plant? Or just plant multiple seeds from the most desirable looking fruit?

I am also wondering this for my Marglobe tomatoes, one plant in particular had what I would call the most hardy and vigourus growth and also had the most and largest fruits. So should I save all my seed from just the one plant, and try to get many of those the next year? Or should I be saving seeds from a wider array of my marglobe plants for genetic diversity?

Or just plant multiple seeds from the most desirable looking fruit?

That depends what you mean by "the most desirable looking fruit." Certainly you want to pick a large, healthy fruit, because it's likely to have healthier seeds.

However, if you're aiming for a wide pear-shape and you're think that by choosing the widest fruit on the plant you'll be more likely to grow a plant with wide fruit than if you choose a narrower fruit from the same plant ... unfortunately, it doesn't work that way.

The genes that decide fruit shape come solely from the mother plant (from the seed the mother plant grew from). The genes in the pollen that fertilized the flower that became the desirable-looking fruit had no effect on the fruit's shape (or color, taste, size ... you get the idea). [The only exception to this rule that I'm aware of is that a mild pepper flower fertilized by pollen from a hot pepper will result in a mild fruit with hot seeds. The seeds are the only part of the fruit that contain genes from both parents.]

As to why different fruit on the same plant have different shapes or sizes, that's mostly the nutrients that particular fruit received; the weather and amount of water at particular times in the fruit's development; ditto fertilizer; etc. There can be negative impacts from insects like aphids or from disease, resulting in smaller, less healthy fruit.

Sometimes when the cells divide to form the parts of the flower that result in the tomato, they divide more times and form extra parts: this is why you'll sometimes see a Roma, grape, or cherry with three locules (seed-chambers) instead of two. The same process causes the occasional wide or vee-shaped beefsteaks that result from what are known as fused blossoms or mega blooms: these fruit have many more locules than usual for their variety (there is a genetic tendency in some varieties to produce fused blossoms more often).

Yesterday when I was cutting up Romas, I found a fruit with only one locule: presumably when that fruit was in its infancy, the cell(s) that were to form the second locule died. Since the problem happened so early, the resultant fruit looked normal, other than being small and curving toward the side where the seeds were attached.

===

For the Marglobe, save seeds from the superior plant. You don't really want genetic diversity, because that would give you an unstable variety. You don't want the genes from the not-so-good Marglobe -- unless that plant has a special trait the better Marglobe lacks. You want the best genes you can get: maybe your better plant has slightly different genes than a normal Marglobe (perhaps there was a slight mutation, or maybe it's a cross and not a true Marglobe). Or maybe it just got a healthier start than the other Marglobes, or a bit more sun and water and fertilizer.

Thanks for all the insight MTO. I'm excited now about the healthy marglobe, it was not in the most ideal part of the bed as far as sunlight goes, it would have actually been in the second least desirable spot (house side and second from the north end of the bed), so that it is doing better and producing bigger healthier looking fruit than the rest is a good sign I think.

So for the Roma, any fruit I get off that plant with the big fruits will give me a good chance at having similar offspring, the actual fruit it comes from doesn't matter. I am still harvesting a seed collection from my other "normal looking romas as well, just in case next year I get some weird frankenfruit off the larger roma type plants, I can then go back to square 1 the following year with reular romas.

Ok here are some pictures I took this morning. I had some breakers and it was raining so I ran out and grabbed them. First, one of the larger romas in between two of the normal sized ones.

And the next picture is two of the Marglobes on the top, which are from the big plant, notice they have more heavy fluting as well, please disregard the bug hole on the big guy, I almost cried when I saw it but i'm still going to use it and save the seeds anyway.