Morphological integration of the turtle shell: the effects of plastral hinging

2010 REU Intern Keegan Melstrom


Freshman Geology major at University of Michigan

REU Mentor: Dr. Kenneth Angielczyk (Curator, Geology)

Symposium Presentation Title: Morphological Integration of the Turtle Shell

Symposium Presentation Abstract: The turtle shell is one of the most distinct evolutionary novelties among tetrapods. Though the shell is a recognizable feature of all turtles, there is significant variation in the function and morphology of the shells of various species. One of these differences is the presence of a hinge on the plastron or ventral shell. Using landmark-based geometric morphometrics our research sought to explore the link between plastral hinging and the organization of modules within the shell. We digitized over 1,800 turtles, from the families of Emydidae, Geoemydidae, and Kinosternidae, from photographs. A principal components analysis was then run which demonstrated that a major component of variation in plastron shape was correlated with presence or absence of a hinge. This is an expected result because the evolution of a hinge typically accompanies the replacement of the bony bridge between the carapace and plastron by a ligamentous connection.  Using canonical variates analysis confirmed this result by showing that plastra of kinetic and akinetic turtles could be classified with a high degree of accuracy. We then used the method of Klingenberg (2009) as implemented in the program MorphoJ to test a priori hypotheses for the presence of different modules. The number of modules varied from two to four, and we tested their locations throughout the plastron. The majority of turtles possessing a kinetic hinge had two distinct modules, one in the anterior portion of the shell, the second in the posterior. This line of separation is the location of the hinge. Given that most turtles with a kinetic hinge have this particular modular placement, despite belonging to several distantly related clades, one may conclude that plastron shape is independent of phylogeny. This analysis and others demonstrate that plastral kinesis has a strong effect on plastron shape, and this effect seems to extend to patterns of modularity, but to a lesser degree. This may indicate that functional selection, in the case for plastron shape, is less strongly affected then the modules in the plastron.

Original Project Description: The morphologies of many organisms display patterns of integration, where developmental, functional, or other interactions between parts cause groups of characters to undergo evolutionary changes as a single unit. The turtle shell is one of the most distinctive evolutionary novelties among vertebrates, and there are a priori reasons to think that different turtles may display different patterns of integration in their shell morphology. In particular, some turtles possess a hinge on their bottom shell, or plastron, that allows the plastron to be closed up against the carapace (upper shell). We would expect the shapes of the plastron and carapace to be integrated in hinged turtles so that a tight fit can be maintained when the shell is closed, but hingeless less turtles may show little or no integration between their plastron and carapace because they do not have this constraint.

Research methods and techniques: In this project we will first collect shell shape data from hingeless and hinged turtle species using the collections of The Field Museum. Then, in collaboration with Dr. Peter Roopnarine (California Academy of Sciences), we will use a new method for detecting patterns of integration in the turtle data set. The intern will be trained in the collection and analysis of geometric morphometric data, analysis of morphological integration, and the use of phylogeny as a framework for analyzing comparative data.