In S.W. Frost’s excellent 1959 book, Insect Life and Insect Natural History, there is a drawing of a clover leaf with a radially symmetrical pattern that gives the impression of a bite having been taken out of each side of each of the three leaflets. It is said to have been caused by a weevil feeding on the leaf before it opened (the weevil is said to be Hypera punctata, which apparently is now Donus zoilus, the clover leaf weevil). Here is an example I found with the opposite pattern, caused by damage to the tips rather than the edges of the still folded-together leaflets:
Last fall, I found a beech leaf with a more elaborate pattern, apparently caused by a single hole in the bud, producing an effect like a paper snowflake:
I thought that was pretty impressive until the other day, when John Maxwell brought to my attention this photo of a banana leaf, taken by Keara Giannotti in the Monteverde Cloud Forest Reserve in Costa Rica:
Keara says this was the third leaf she found with a similar pattern. She found this web page that attributes similar holes in a banana leaf to the caterpillar of an owlet moth (Noctuidae), Spodoptera litura. I find it fascinating that the holes line up perfectly in perpendicular rows and columns, which (unlike the holes in my beech leaf) have nothing to do with the leaf venation. Curiously, in the upper left there is a set of additional, alternating holes on one side of the midrib, and at the very top left (again just to one side of the midrib) there is an extra row of holes that is slightly offset.
So, beyond the question of what insect is responsible, I wonder how many separate “chomps” this pattern represents? Maybe someone with access to unfolding banana leaves will experiment with poking some holes and report back to us?
BugTracks 
Re: the banana leaf … looks like a sapsucker went haywire on that one, mistaking it for a tree trunk!
On the banana leaf, I imagine the eggs being oviposited in a spaced-out row (see link for an example). Then as the eggs hatch, all near the same time, they burrow into the leaf. Since they hatch and burrow near the same time, the leaf is rolled up similarly for all of them. Imagine drilling a line of holes into a roll of adding-machine paper, then unrolling the paper, you would get a nice grid pattern.
http://bugguide.net/node/view/907628
Interesting thought!
In this photo, the right side would have been the center of the roll (pattern close together) and the left would have been on the outside of the roll (pattern farther apart).
Also, as you go from left-to-right in the photo (outside to inside) the diameter of each hole gets larger, suggesting that the larvae were getting larger as they neared the center of the roll.
Fascinating and beautiful – the process of the bud opening memorialized in the leaf.
No picts but I’ve seen this happen on large leafed plants (my mother has a small grove of ornamental banana trees in her backyard. A small but growing grove)- the insect damage occurs while the leaf is still rolled up and when it unrolls it produces an interesting pattern.
Pretty sure it’s a caterpillar that ate through the leaf while it was still rolled up (banana leaves roll up like scrolls with the midvein in the center). I don’t know if there were several or just one insect that was causing damage.
My Cannas leaves looks like that every year. I live in The Woodlands, TX