Purdue University Graduate School

File(s) under embargo





until file(s) become available

Phylogenomics and species distribution models to infer evolutionary relationships, delimit species, and better understand lichen-host interactions in tiger moths

Version 2 2024-01-17, 18:21
Version 1 2023-12-16, 20:02
posted on 2024-01-17, 18:21 authored by Makani L FisherMakani L Fisher

The lichen-feeding tiger moth tribe Lithosiini (Erebidae: Arctiinae) represent the largest radiation of invertebrate lichenivory. Caterpillars feed on lichen and as they feed, also sequester lichen polyphenolics, a behavior unique to these insects. The role of these compounds is believed to defend lithosiines against predators as larvae have been found to be protected against predators such as ants and moths to predators such as birds and bats. Experimental testing with controlled diets is necessary to fully make this connection, however little is known about host specifics for lithosiines. Furthermore, although lithosiines are monophyletic, the lack of a fully resolved phylogeny hampers investigation into many of the shallower level relationships, e.g. those among genera and species, within the group.

I addressed these knowledge gaps using the subtribe Cisthenina. Members of this group have been used to investigate predator-prey interactions and been included in morphological and molecular studies. Thus, while the group still needs attention, there is an ample amount of legacy loci data available for its members. I used these data to investigate the evolutionary relationships at the genus level, but to increase resolution in my analyses I additionally sampled taxa throughout the group with a recently developed anchored hybrid enrichment (AHE) probe set. I combined it with the legacy loci to both increase taxon sampling and resolution. I confirmed that trees made strictly from the legacy loci were unsuccessful and resulted in poorly supported relationships that made little sense. The addition of the AHE data greatly helped resolve relationships, however, there remained areas that were poorly supported and they appear to be genera with only a few loci. Thus, there is still room for improvement, but this offers a way for moving forward in lithosiine research, particularly to involve others who may have limited funding, equipment, and/or personnel and may only be able to afford legacy loci in diverse collaborations.

As the AHE probe set worked well with genus-level relationships I further attempted to use it in species delimitation of the notorious Hypoprepia fucosa-miniata species complex. Members of this group are varying shades of yellows, oranges and reds and have a convoluted taxonomic history. I gathered and organized over 4,000 specimens and using the AHE probe set found support for five distinct species. Interestingly, I used other morphological characters such as genitalia, but found no differences between species and a large amount of intraspecific variation. This suggests other courtship behaviors may be present and external morphology, i.e., color patterns, remain the best way to identify species. As part of this I am describing a new species and raising one from subspecies and as species are now readily distinguishable, they can be used for further investigations into lithosiines.

I used a member of this complex, H. fucosa to then evaluate the use of species distribution models (SDMs) to better understand their niche and how it relates to plausible lichen hosts. I evaluated 17 lichen species from two lichen genera, Physcia (13 species) and Myelochroa (4 species). These genera were selected based on previous feeding assays and the metabolites found in them have also been found in H. fucosa further suggesting caterpillars may feed on them. SDMs typically only use environmental factors to define and predict species niches. I compared the niches described by traditional SDMs to assess how similar they were, but I also investigated the use of lichens as biotic factors in the models. I assessed the influence each lichen had on the moth’s distribution found the niche of every lichen to be significantly different than that of the moth and their inclusion in SDMs of H. fucosa to improve model performance. This suggests H. fucosa caterpillars to be polyphagous, but to have some connection with these lichens. Further investigation with live specimens is needed, but these results support this as an effective way to describe lithosiine niches to better understand lichen feeding.


Degree Type

  • Doctor of Philosophy


  • Entomology

Campus location

  • West Lafayette

Advisor/Supervisor/Committee Chair

Stephen L. Cameron

Additional Committee Member 2

Jennifer M. Zaspel

Additional Committee Member 3

Aaron D. Smith

Additional Committee Member 4

James C. Lendemer