The Leafy Origins of Sepals

The sepals of a rose bud are green and photosynthetic
like fully developed leaves, and like the one on the left,
sometimes even appear to partially subdivided like full
leaves.

From an evolutionary point-of-view, it is generally accepted that the parts of the flower originated as modified leaves. Though there is controversy about the nature of the earliest carpels and stamens, the leaf-like nature of petals and sepals is abundantly evident. Sepals are generally the most leaf-like, no doubt because they are the most recently evolved of the flower parts, and may have originated separately in different lines of early angiosperms. 

In some archaic flowering plants, such as Magnolias, petals and sepals intergrade, with no clear distinction between the two, and they are called tepals.

In Magnolia, there is a series of similar outer, leaf-like
structures that are colored like petals, though the outer ones
serve to protect the bud during development, as specialized
sepals do. Photo from Wikiwand, License: Creative
Commons Attribution-Share Alike 3.0

Upon looking at a rosebud, the leaf-like nature of the sepal is evident. The occasional sepal that takes on even more of the subdivided shape of the full leaf, emphasizes the point.  Leaves, bracts, sepals, and other flower parts develop from outgrowths of the apical meristem, and so in their earliest stages look the same. As each develops for its specific function, specialized genes kick in to determine the final shape, color, and other physical features of the organ. The fact that some rose sepals look a little more like normal leaves than others shows that the genes for full leaf development that are normally suppressed, can sometimes be partially expressed.

In Clusia, the leaves are opposite with each pair at right angles to the
preceding pair. A series of small bracts and two pairs of colored sepals
continue the pattern. it's truly hard to see where bracts end and
sepals begin.

The development of similar, but modified, organs, from outgrowths of the apical meristem, is called serial homology. Homology in general refers to different body organs that were the same in ancestral species, but have become specialized for different functions in  more specialized species.  The classic examples in animals are front legs that have become specialized as wings in both birds and bats, and arms with grasping hands in primates.  Serial homology can be seen in animals with segmented bodies. Insects and crustaceans, for example, descended from many-legged, centipede-like ancestors, but now have specialized walking legs, reproductive organs, claws, mouth parts, and other specialized organs, all representing modified legs.

Lilies, like many monocots, have what appear to be six petals,
but three of them are actually petal-like sepals

Serial homology of leaf-like organs in plants suggests that at one time there were only leaves, as in early seed ferns, that did everything: photosynthesis as well as bearing pollen sacs and ovules, and all had the same shape.  As plants progressed, a division of labor came into being, with some leaves continuing the primary photosynthetic function, while others became specialized as bracts or floral organs.  In previous posts I have described even more bizarre leaf modifications, such as insect-catching traps.

Trilliums are monocots only distantly related to the true lilies,
and display three leaf-like sepals, most likely as did the
ancestral monocots.

The ancestral set of genes that orchestrated the
development of leaves was supplemented with new sets of genes that served to modify the embryonic leaves for specialized functions.  The new sets of genes both suppressed the full development of the original leaf size and shape and directed the development of specialized features. Serial homology along a single shoot, from leaves to bracts to flower parts, shows that these different sets of genes are turned on and off in an orderly way.