Birds evolved during the Mesozoic, during the various "Ages" of the dinosaurs, as a subset of those dinosaurs. Many researchers believe that these early birds were different from their then very close dinosaur cousins because of their flight adaptations, and some have linked this idea to flight-based or tree- based foraging.
Today, most birds fly (counting by species) and their flight is linked to their primary dietary adaptation. Some birds actually feed on the wing, while others fly to food sites and once there do not locomote very much. Other birds forage on the ground habitually. The difference between these two kinds of birds should be evident in the morphology of their feet.
If you examine bird feet today and can successfully characterize these feet in a way that links reliably to mode of foraging, then you should be able to look at Mesozoic bird (and dinosaur) feet and say something about how they foraged. In this way, you can test the hypothesis that early birds were flight- and tree-foragers rather than foot-foragers. A paper in Current Biology does this.
The first thing the researchers had to do was to deconstruct commonly held conceptions of bird locomotion and positional behavior. In other words, they noticed that the way we characterize birds ... as "ground-dweller" vs "percher" vs "climber" ... requires "summarisation of potentially disparate aspects of behaviour and morphology ... with emphasis on the 'ability/inability' to perch. However, many 'perchers' visit the ground and some 'ground-dwellers' can perch."
The authors re-categorized the birds into groups that were both finer than those normally used, and based on a better set of observations of bird behavior, focusing not on how birds cling (or fail to cling) to various arboreal surfaces, but rather, how they use their feet to forage.
They then measured "claw angle" .. the angle between key physical landmarks on the third toe ... which they believe varies in relation to mode of foraging. These data separated out by category of foraging very nicely:
To someone who studies the relationship between morphology and behavioral adaptations, this graph is a real wet dream. One wonders if they made up the data. But they didn't.
The top part of this graph shows the claw angle (Y-axis) increasing from left to right across the foraging categories into which the authors placed the birds. Ground foraging birds are on the left, others towards the right. This is done for two phylogenetically distinct groups of birds and shows the same pattern for both of these large taxa.
This is very nice work. In and of itself this morphological pattern is an important contribution. But there is more.
The lower right part of the graph shows the claw angle measurements for Mesozoic genera. They appear to be mostly ground foragers. The horizontal line in this graph is the cutoff point that seems to work for living birds (lower than 100 degrees = ground forager, higher = varying degrees of tree foraging). One Mesozoic specimen is above this line, which does not mean much ... it could be an outlier.
In other words, it is reasonable to conclude that ground foraging was the predominant behavior among the Mesozoic birds.
In particular the hypothesis that Mesozoic bird evolution involved flight-and tree based foraging is not supported by this analysis. This forces evolutionary biologists to change their thinking about early bird evolution. Science marches on.
GLENN, CHRISTOPHER L. BENNETT, MICHAEL B.(2007): Foraging modes of Mesozoic birds and non-avian theropods. Current Biology, 17, R911-R912.