I have a small isolated archipelago. Is it feasible that the flora of the island through natural selection (with animals favoring bigger fruits) has been steered to evolve large, plump fruits of the same caliber as or much like what humans have produced through selective breeding?
Why do plants make fruit? Fruit is essentially a bribe. Plants have so much free energy that it makes sense for them to invest huge amounts of energy creating a fruit, even though the sweet parts of the fruit do not help the child plants grow at all. They do this because animals eat the fruit, but don't digest the seeds, causing the seeds to be spread in the animal's spoor. (This is why fruit seeds are often poisonous while the fruit itself is perfectly safe to eat: that's not an accident.)
So you need to come up with a reason why large fruit would be preferable. I'm having trouble doing that, but there might be a way.
The first possible reason that comes to mind is big herbivores: The plant might use (say) an elephant as its preferred seed transmission vector. A big herbivore would find it more convenient to eat big fruit, whereas the sheer unwieldiness of getting at it would dissuade smaller animals from freeloading. The big problem here is big herbivores need big territories, and these are small islands. I think this is out.
I can only think of one other possibility: The animal is encouraged to forage for these fruits and bring them back to its nest, to feed its young. Think of some explanation: the fruit is just the right softness to be perfect for young. It has some nutrient they need. Something. Big fruit will work better for this purpose, because the animal can get more weight back to the next per run.
Not really. For it to happen, there would need to be a massive surplus of fruit so that animals only selected the plumpest fruit. If there was a surplus then the animals would breed more until the food supply was used up.
It's an energy used verses return problem. The plant wants its seeds spread but it wants to use the least amount of energy to do it. A smaller fruit gets spread just as well as a large one but uses less energy to grow. The plant is better growing lots of small ones rather than large ones for the same energy.
There is a wide selection of tropical fruit that is plump and nourishing without having been bred by humans for that purpose. There is also yams and other root vegetables, that provide huge quantities of carbohydrates without being bred. The fruit evolved for seed dispersion in the feces of larger animals, and the roots try to stay hidden and are best accessible (caloric content) after cooking, so they were 'safe'-ish until humans invented fire - and then they were safe because humans planted them all over.
Plants investing into huge stores of chemical energy also have some way of steering the intended audience: Poisons specific to clades of animals (i.e. only birds can eat it), mechanical barriers (high up, spiky, tough shell), selective advertisement (markers only some can smell, reflecting wavelengths only some can see), digestability (have a cover of easily digestible material as prize, have an indigestable (but even more energy-rich) core) - most of which humans can outsmart by fire, sword and ladder.
For reference, think about wild bees: They sometimes have kilos of metabolic holy grail (honey=sugar) high up in cliffs, and the only non-humans able to reach it are bats, birds, apes with sticks and insects, all of which are easily repelled by the bees. Humans carry up a bit of smoking grass, and the whole thing is subverted. 'Plump fruit', whatever its actual evolutionary purpose, is just like that - prey to human mastery of fire\tools\planning.
No, at least not under your scenario. The opposite would in fact occur. For the plants, the selection pressure would be the animals. With the animals favoring the largest fruits, the plant species would be forced by natural selection towards small fruit considering the small fruit bearing plants would reproduce at a far greater rate.
If there was a selection pressure that found a way to utilize the fecal matter of the animals to enrich the seed that was passed, the pressure would push the species towards the larger fruit. This is infact how fruit seeds must, in some cases, be planted in nature. This, however, would increase the animal population which again would force the species towards the small end.
If the dung enriching secret ingredient was some alien superfood that happened to be scattered during a crash landing, then the plants could grow like crazy. Question is, is that natural selection?
A good read on the subject:
The Beak of the Finch: A Story of Evolution in Our Time
By Jonathan Weiner
As others have said, large fruit will only evolve if it's somehow advantageous for the fruit to be larger. Various answers have suggested that larger fruit might be more attractive to certain animals that the plant wants to be eaten by, but that requires additional pressures and very specific conditions for the animals to prefer those fruits. Let's look at something that makes the size itself the goal.
The effects of the fruit being larger will generally fall under the purview of the square-cube law. Larger fruits will have less surface area relative to their internal volume than smaller fruits. Perhaps the seed takes a long time to germinate, and it's beneficial for it to not dry out over that time. A large fruit stores more moisture while having proportionally less surface area to lose it through. In a similar vein, a larger fruit has greater thermal mass, so the seed at the centre would be insulated from changes in temperature - useful if it's quite sensitive to, say, being dried out or otherwise denatured by heat.
Mechanically, a larger fruit would also fall harder. Compare what happens when you drop a ping-pong ball and a stone of a similar size: the stone hits the ground much harder. Perhaps it's beneficial for the fruit to embed itself solidly in the ground when it falls off the tree, or maybe the tree grows in soft, marshy ground, where a large fruit will sink into the soil while a tiny, lightweight seed will sit on top, unable to break the surface tension. Maybe the fruit has a hard shell, and only the heaviest fruit will hit hard enough to crack the shell open and allow the seed to grow. Or, for a more spectacular option, maybe the tree disperses its seeds by using the splat of impact to spread them - and the nutrient-rich fruit pulp - over a wider area. For something slightly more complex, perhaps these fruits grow large, with a tough skin, so that it's not viable for small creatures to eat them, ensuring that they can grow to maturity unmolested... but once ripe, it relies on those same creatures to disperse the seeds, so they fall and splatter on the ground, and which point those small creatures rush to get at the now readily accessible fruit pulp, with all the seeds mixed into it.
Yes. If similar to coconut
When talking about big fruits, what comes to mind is something whith plenty of flesh and tiny seeds. As Dayton Williams explain, this is when the tree gives a big bribe to a big animal.
But in a small archipelago, big animals are unlikely ( not impossible. I once lived there and I can tell you it's full of seawater crocodile).
The strategy of coconut is different: they make a big seed that float and cannot be eaten. So if it fall in the water then is wash ashore, a tree can grow several km from the original tree.
There is one obvious advantage: the adult tree can take as much time as needed to grow a huge nut and a huge nut can make a 50cm tall tree quickly.
The natural selection is different here. It would rather be rabbit like animal or insects that destroy smaller shoot before they reach a certain size.
While all the other answers say no; I'm going to say yes, it just requires the correct conditions, but the no sayers are just unable to think of them.
"I can't think of one" isn't the same as no. Look at melons for example. They evolved big juicy fruit (compared to the apple or grape). The fact that we have fruit like melons means the answer MUST be a yes.
Remember that it doesn't need to be a significant advantage to this - as time will exaggerate it; in the same way that being able to detect if you're in light or dark will get exaggerated into sight in time.
The most obvious example that I can think of: The rind of the fruit must be so tough and thick to counter insects (again - think melons!) that the animal must exert so much effort to get into it, only the large ones are worth the animals time to break through it.