Skip to main content

Genomics confirms surprising ecological divergence and isolation in an endangered butterfly


Phylogeographic patterns in phytophagous organisms are often contextualized in light of geographic isolation and ecological (host, habitat) specialization. However, assessing the relative impact of these phenomena is not straightforward, even in areas where phylogeography is well-studied, such as the California Floristic Province. Here, we use genome-wide markers to elucidate population genomic and phylgeographic patterns for a group of monophytophagous butterflies in southern California. This group is of high conservation interest because it includes the El Segundo blue, Euphilotes battoides allyni, one of the first insects listed under the U.S. Endangered Species Act, and a newly discovered population putatively assigned to E. b. allyni. Despite using the same unique host and coastal habitat, our results indicate that the newly discovered populations are not E. b. allyni and are more closely related to geographically proximate populations of the E. battoides group using a different habitat/host combination. Aside from E. b. allyni and the newly discovered populations, the rest of the group shows only fine-scale structure and apparently maintains genetic connectivity throughout southern California, across a vast range of habitats and climates, and on multiple hosts. Thus, habitat and host specialization did not elicit genetic isolation in neighboring populations suggesting that: (1) other phenomena are needed to explain the remarkable and idiosyncratic divergence of these highly restricted, proximate, taxa, and (2) fine-scale genomic markers suggest broader implications for understanding the mechanisms of speciation and reinvestigation of phylogeographic patterns in regions like the California Floristic Province.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Data availability

Raw sequence data for ddRAD dataset is available on NCBI, BioProject PRJNA555218 (SRA accessions: SRR9707610-SRR9707926), and COI data is available on GenBank (accessions: MN224680-MN224984). Data files are available as a Dryad Data repository (repository: Data citation: Dupuis et al. (2019).


Download references


This project would not have been possible without the enthusiasm, assistance, and advice of Nancy Ferguson (United States Fish and Wildlife Service), and support by the USFWS and VAFB. Eric Porter and William Miller (USFWS), and Travis Longcore (Urban Wildlands Group) provided advice and support necessary for our success. We thank the following individuals for assistance in collecting specimens: Alice Abela (ManTech), Sam Fisher, Robert Fisher, Marie Fisher, Carolyn Lin (Los Angeles World Airports), Adrienne Mohan (Palos Verdes Peninsula Land Conservancy), Katrina Olthof (Wildlands Conservation Science), Nancy Price (Los Angeles World Airports), and Patrick Tyrrell (Friends of Balona Wetlands). We also thank M. Wells, Torrey Pines State Reserve; D. Lydy U.S. Navy, Pt. Loma; T. Walker of the Tri-canyon Parks (San Diego), Environmental Protection Services, U.S. Marine Corps Camp Pendelton; and especially S. Weber, Chief of Natural Resources at Cabrillo National Monument. D.R. thanks J. and P. Brown and family for support and invaluable advice. J. Vanderweir and E. Hein, U.S. Fish and Wildlife Service provided essential technical advice and assistance for collection of San Diego County samples. ddRAD library preparation and wet lab procedures were performed at the United States Department of Agriculture—Agricultural Research Service (USDA-ARS) Daniel K. Inouye US PBARC Genomics facility by Angela Kauwe and Kimberley Morris, and HiSeq sequencing was conducted at the Vincent J. Coates Genomics Sequencing Laboratory at University of California at Berkeley, supported by National Institutes of Health S10 Instrumentation Grants S10RR029668 and S10RR027303. Phylogenetic data file with flanking sequence produced using, and figures produced using Inkscape v0.91 (The Inkscape Team 2017) and R (R Core Team 2018). USDA is an equal opportunity employer. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the USDA.

Author information

Authors and Affiliations



KHO and DR conceptualized study with input from JRD and SMG. KHO and DR collected specimens and JRD conducted data processing and analyses with input from SMG. JRD and DR wrote the manuscript with input from SMG and KHO.

Corresponding author

Correspondence to Julian R. Dupuis.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

All ethical standards have been followed.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Communicated by Nigel E. Stork.

Electronic supplementary material

Below is the link to the electronic supplementary material.


Figure S1. Full results of STRUCTURE analyses for all individuals (a), and the allyni (b), VAFB (c), and battoides (d) clusters, including plots of LnPr(K|X), ΔK, and the Peuchmaille statistics, and barplots for all values of K. Individuals labeled at an optimal value of K, if appropriate. Alternative sort order (by general area within the VAFB (northern, southern, off-base) and by longitude) provided for VAFB cluster below (c). (PDF 2316 kb)


Figure S2. Results of DAPC for all individuals (a), the allyni and VAFB clusters (b), and the battoides cluster (c). Inset graphs show plots of number of clusters versus BIC and the relative contribution of each discriminant axis to the analysis. In (b), find.clusters predicted K = 1, so DAPC plots not shown; however in (c), K = 2 is forced in the final iterative hierarchical analysis of the battoides cluster to show partial separation of San Diego populations from San Bernardino/Riverside populations. (PDF 324 kb)


Figure S3. ML consensus tree from phylogenetic dataset (260 individuals, 54,305 SNPs), with SH-aLRT/ufBS branch support displayed (strong support = >0.8/>0.95). (PDF 16 kb)


Figure S4. ML consensus tree from supplementary phylogenetic dataset 1 (260 individuals, 1,882 SNPs matching the population genetic dataset), with SH-aLRT/ufBS branch support displayed (strong support = >0.8/>0.95). (PDF 16 kb)


Figure S5. ML consensus tree from supplementary phylogenetic dataset 2 (260 individuals, 189,837 characters, matching the population genetic dataset but with flanking sequence around each SNP), with SH-aLRT/ufBS branch support displayed (strong support = >0.8/>0.95). (PDF 16 kb)


Figure S6. ML consensus tree from COI dataset (305 individuals, 650 bp), with SH-aLRT/ufBS branch support displayed (strong support = >0.8/>0.95). (PDF 14 kb)

Table S1. Detailed specimen information. (XLSX 51 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Dupuis, J.R., Geib, S.M., Osborne, K.H. et al. Genomics confirms surprising ecological divergence and isolation in an endangered butterfly. Biodivers Conserv 29, 1897–1921 (2020).

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Conservation genetics
  • Lycaenidae
  • Euphilotes battoides allyni
  • Host specialization
  • Phylogeography
  • California floristic province