A new parsimony-based method for reconstructing sequences of developmental events and estimating rates of intraspecific variation, with applications to trilobites

Author

• Alexander Benjamin BradleyFollow

Semester

Summer

Date of Graduation

2026

Document Type

Dissertation

Degree Type

PhD

College

Eberly College of Arts and Sciences

Department

Geology and Geography

Committee Chair

James Lamsdell

Committee Member

Kathleen Benison

Committee Member

Dorothy Vespe

Committee Member

Joseph Lebold

Committee Member

Thomas Hegna

Abstract

Ontogenetic Sequence Analysis (OSA) is a method for reconstructing multiple sequences of abrupt phenotypic transformations (developmental events) for sampled populations of any kind of organism. It makes use of cladistic parsimony software to produce a network, or graph, whose vertices represent unique phenotypes at different stages of maturity (“semaphoronts”) and whose edges describe the developmental events occurring between semaphoronts. It is a powerful tool for constraining the range of developmental variation possible for a given species and generating sample statistics capable of identifying both the most common and the outlier ontogenetic sequences. However, OSA has seen little use since its conception in the late 1990s because it is not fully automated, lacks dedicated software, and leverages the principle of parsimony in an unintuitive way.

Here, a new, graph theoretical (rather than cladistic) approach, Revised Ontogenetic Sequence Analysis (ROSA), is proposed and tested. ROSA is like OSA in that it takes a data matrix of conspecific individuals by ontogenetic characters and outputs an ontogenetic sequence graph describing reconstructed sequences of semaphoronts. It differs from OSA in that parsimony is applied directly to sequence reconstruction, additional unobserved semaphoronts are predicted in a separate step, the involved matrix operations are automated with custom R code, and the output graph gives a more conservative estimate of sequence polymorphism.

ROSA is tested on two trilobite species, the Cambrian “ptychoparioid” Elrathia kingii, and the Ordovician trinucleid Cryptolithus tessellatus. Specimens of the former species were photographed at the American Museum of Natural History in New York. New silicified material of the latter species was collected from the lower Martinsburg Formation in the Shenandoah Valley of Virginia. Field and petrographic observations indicate that the strata containing C. tessellatus were deposited in repeated turbidity currents that accumulated in a deep marine environment off the southeastern coast of Laurentia. These trilobites were liberated from limestone blocks and imaged using Scanning Electron Microscopy.

In addition to estimating rates of variation and describing alternative ontogenies for both trilobite species, the effects of sloppy character selection on the results of ROSA are tested and discussed. E. kingii is also used to develop a protocol for analyzing populations characterized by multiple target phenotypes (adult phenotypic polymorphism). Results are consistent with the hypothesis that trilobites became increasingly developmentally canalized in the early Paleozoic, with E. kingii exhibiting elevated rates of sequence polymorphism and C. tessellatus exhibiting minimal polymorphism.

Recommended Citation

Bradley, Alexander Benjamin, "A new parsimony-based method for reconstructing sequences of developmental events and estimating rates of intraspecific variation, with applications to trilobites" (2026). Graduate Theses, Dissertations, and Problem Reports (ETD). 13402.
https://researchrepository.wvu.edu/etd/13402