I'm not aware of any site that addresses this specifically, but I've seen references to it here and there when people talk about particular plants. Leaf shape is more or less genetically determined, though in some cases the mature and immature leaves have significant differences (Monstera, Syngonium).

As for what evolutionary purpose different leaf shapes have, well, some of this you know already. Leaves that are thick and roundish (jade plant, burro's tail) are trying to minimize surface area, to prevent loss of water. Leaves that are deeply lobed or perforated (Monstera, Philodendron selloum) are subject to occasional hurricanes or other forms of intense wind, and benefit from reducing wind resistance (especially since both of those examples also have really brittle stems which are easy to break off). Other plants in hurricane-prone areas have gone with a different strategy and developed flimsy leaves that curl into cones in high winds (Caladium), or developed leaves which are brittle in one direction and tough in others, so they can tear themselves into Philodendron selloum-like leaves immediately in the event of strong winds (banana, bird of paradise).

A lot of aroid leaves (Dieffenbachia, Philodendron, Anthurium, Monstera) have rounded leaves with a distinct "drip tip" at the end. I've seen this explained as an adaptation to keep water from building up on the leaves, though I'm not clear on why that would be bad. (You'd have to ask the plants.)

Obviously leaves with spines (Agave, Aloe, Pandanus, some bromeliads) are interested in deterring predators of some kind.

It's not entirely an answerable question, because there isn't always necessarily a reason. I mean, there's a *reason*, but it might have to do with being eaten by giant ground sloths that went extinct 10,000 years ago, or what color of flower was attactive to an insect that found some other way of making a living and no longer pollinates flowers, or a mountain range suddenly (in geological time) popping up in the middle of a plant's range and separating a small population from the main population. The only thing you can really be certain of is that every change that's happened to the genome of a particular species, *every* change, reproduced better or faster or more broadly than its relatives for some reason or another.

Kinda off-topic, but:

Currently, the best way to reproduce yourself, if you're a plant, is to be appealing to human beings in some way. There's a whole book about this, _The Botany of Desire: a Plant's-Eye View of the World_, by Michael Pollan, which is one of my favorite books ever, and which I enthusiastically recommend to any Garden Webbers who might be reading.

Here is a link that might be useful: The Botany of Desire (Amazon)

Ohhh, that's a greeeaaat book!

My version of the short answer to this question is quite simple. As with all living organisms, plants' success or failure is determined by the ability to adapt and evolve to particular eco-systems or environments. One learns that 'form does indeed follow function', when it comes to plants. Plants (and other life forms) NEVER succeed by accident.