Natural selection hinges on the existence of variation in traits among individuals – but how much can different traits actually vary? Even more critically, how much do those traits vary during an animal’s development?  A recent IOB paper by Kelly et al. investigates how morphological integration changes among and between the fore- and hind limb skeletons of mammals, and helps us get a grip (boom, fore limb pun!) on how traits vary during development.

When it comes to animal skeletons, some elements display high morphological integration, meaning that the shapes of some bones are highly correlated with the shapes of other bones, and if one shape changes, the other is likely to change as well. This integration constrains the variability of bone shape. On the other hand, sometimes skeletal structures are weakly integrated, and different bones can change shape independently of each other. Less integrated shapes may be considered more “evolvable”, and could open new avenues of morphological specialization.

Kelly and colleagues set out to track potential changes in integration patterns through several developmental stages in three different mammals: mice, and two different species of marsupials. Unlike us placental mammals who insist on completing our embryonic development in the womb, marsupial young are born at a very early developmental stage (the technical term is “half-melted gummy bear”). They then have to crawl up their mother’s body where they find a teat and continue developing until they look like a regular non-terrifying baby mammal. The fact that young marsupials must perform such a heroic feat of athleticism just after birth suggests that the morphology of marsupial fore limbs is constrained, since marsupials need comparatively beefy forelimbs for the post-birth crawl.

So are marsupial limbs more or less integrated than placental limbs? And does the pattern of limb integration in developing animals match that seen in adults? To address these questions, Kelly and colleagues used landmarks to quantify the shape of bones in the fore- and hind limbs of mouse and marsupial embryos and neonates. They then analyzed the correlations between elements of the same limb (humerus, radius, metacarpals; or femur, tibia, metatarsals), and correlations between hind limbs and forelimbs. They also compared these correlations in limb shape across developmental stages. The researchers found that although the overall pattern of integration among limb bones is similar across the developmental stages they examined and similar to adult integration patterns, the relative integration within each limb versus between fore- and hind limbs varies dynamically at different stages in different species. This study highlights the complexity of morphological integration in development, and is a good reminder that levels of variation in adults and embryos may be very different.

The big takeaway? These results support the idea that selection for marsupial newborns with forelimbs strong enough to crawl into the pouch has reduced the morphological integration between the fore- and hind limb, allowing the marsupial forelimb to become more specialized before birth. So the next time you see an arm-wrestling contest between a newborn mouse and a newborn possum, thank the authors of this study for cutting-edge scientific insight into where you should place your bets!

Abby Vander Linden is an evolutionary biologist, functional morphologist, and PhD student at UMass Amherst. You can find her inside a microCT scanner or at www.abbyvanderlinden.com.