Prepona sp. (Nymphalidae)
Here is a smaller individual from the same plant (maybe a full sib). The first photo below shows the characteristic feeding damage.
It almost disappears once it gets amongst its "trash" on the leaf vein. In many ways masquerade is more similar to mimicry than crypsis. When people discuss crypsis as an anti-predator defence they usually mean that the predator does not detect the prey. It is perhaps more realistic to recognise that the predator does see the object it just doesn't think that that object is a meal. In classical Batesian mimicry theory, the number of mimics should be smaller than the number of models (model = the object being mimicked), otherwise predators don’t learn to avoid that type of prey. Biologists call this "frequency-dependence" because the efficacy of the defence depends on the relative frequency of the models to mimics. In theory this should be the case in masquerade too, it's just that the model is usually something incredibly common like a living/dead leaf or a twig.
On a night walk tonight I found another dead leaf mimic:
A colleague/friend recently blogged about The weird and wonderful world of katydids. I doubt that I could do a better job of enthusiastically describing these guys so I'll leave you to read his post. He writes very well and I highly recommend following his blog: Katatrepsis. He's right though, the detail of the false venation is mind-blowing.