Morphometrics of the Dolphin Genus Lagenorhynchus: Deciphering A

Home , Cephalorhynchus, Lagenorhynchus

Morphometrics of the dolphin genus Lagenorhynchus: deciphering a contested phylogeny Allison Galezo1,2 and Nicole Vollmer1,3 1 Department of Vertebrate Zoology, Smithsonian Institution National Museum of Natural History 2 Department of Biology, Georgetown University 3 NOAA National Systematics Laboratory

Background Results & Analysis Discussion

• Our morphological data support the hypothesis that Figure 1. Phenogram from cluster analysis of dolphin skull Figure 2. Species symbols key with sample sizes.

measurements. Calculated using Euclidian distances and the genus Lagenorhynchus is not monophyletic, evident Height l La. acutus (24) p C. commersonii (5) a b c Ward’s method. from the separation in the phenogram of La. albirostris Height p C. eutropia (2) Recent phylogenetic studies1-7 have indicated that the Distance l La. albirostris (10) and La. acutus from the other Lagenorhynchus species, 0 2 4 6 8 p genus Lagenorhynchus, currently containing the species 0 2 4 6 8 l La. australis (7) C. heavisidii (1) u Li. borealis (11) and the mix of genera in the lowermost clade (Figure 1). L. obliquidensa, L. acutusb , L. albirostrisc, L. obscurusd , L. l La. obliquidens (28) p C. hectori (2) n Unknown (1) • Our results show that La. obscurus and La. obliquidens e f cruciger , and L. australis , is not monophyletic. These C. commersonii l La. obscurus (15) are very similar morphologically, which supports the C. commersonii species were originally grouped together because of C.C. cocommemmersoniirsonii C. commeC. hersoniictori hypothesis that they are closely related: they have C. commeC. hersoniictori similarities in external morphology and coloration, but C. commeC. hersoniictori Figure 3. PCA. Principal Component 1 (PC1) explained 59.47% C. commeC. hersoniictori noticeable overlap in both PCAs (Figures 3 & 4) and DFAs C. commeLa. arsoniicutus of the variance in the data, and was largely a measure of skull genetic studies have proposed that this grouping may C. commeLa. arsoniicutus (not pictured) with and without a size correction, La.La. aacucutustus not be valid. La.La. aacucutustus width. PC2 (19.26%) was a measure of tooth count. Note the La.La. aacucutustus representing a wide range of skull measurements. They La.La. aacucutustus overlap between La. obliquidens l, La. obscurus l, and La. La.La. aacucutustus La.La. aacucutustus australis l. All 5 Lagenorhynchus species lllll cluster also cluster together in the phenogram, with the These studies have suggested that La.La. aacucutustus La.La. aacucutustus together, and are particularly isolated from Li. borealis u and exception of 8 La. obliquidens individuals originating 2-7 La.La. aacucutustus • L. australis and L. cruciger are sister taxa. La.La. aacucutustus La.La. aacucutustus the two Cephalorhynchus clusters: pp & pp. La. acutus l from a Southern Californian region believed to house a • L. obscurus and L. obliquidens are sister taxa.1-7 La.La. aacucutustus La.La. aacucutustus and La. albirostris l lie on the margins of the Lagenorhynchus 9 La.La. aacucutustus distinct La. obliquidens morphotype. • L. australis, L. cruciger, L. obscurus, and L. obliquidens La.La. aacucutustus La.La. aacucutustus cluster. PCA belong in the subfamily Lissodelphininae, along with La.La. aacucutustus • The grouping of La. australis, La. obliquidens, La. La.La. aacucutustus 5 3-7 La.La. aacucutustus two other genera, Lissodelphis and Cephalorhynchus. La.La. aacucutustus obscurus, C. eutropia, and C. heavisidii in the phenogram La.La. aacucutustus • La.La. aacucutustus sister to Li. borealis supports the proposed makeup of L. acutus and L. albirostris do not belong in the same La. obliquidensLa. acutus La. obliquidensLa. acutus 4-8 genus as the other Lagenorhynchus species, or even in La.La. obliquidensobliquidens Cephalorhynchus cothemmersonii Lissodelphininae subfamily , but the exclusion of C. La.La. obliquidensobliquidens Cephalorhynchus eutropia 3-7 La. obliquidens the Lissodelphininae subfamily. La. obliquidens Cephalorhynchus heavisidii La.La. obliquidensobliquidens 0 commersonii and C. hectori does not. This may be due to La.La. obliquidensobliquidens Cephalorhynchus hectori dist(data.log, method method dist(data.log,="euclidian") La.La. obliquidensobliquidens Lagenorhynchus acutus La.La. obliquidens albirostris the exceptionally small size of C. commersonii and C. dist(data.log, method method =dist(data.log,"euclidian") La.La. obliquidens albirostris Lagenorhynchus albirostris We sought to assess whether the skull morphology of La.La. albialbiroroststririss Lagenorhynchus australis Cluster Dendrogram hectori. The addition of morphological data from La. La.La. albialbiroroststririss La. albirostris Lagenorhynchus obliquidens the Lagenorhynchus species, as well as the morphology La. albirostris Cluster Dendrogram

hclust "ward.D2")(*, La. albirostris Lagenorhynchus obscurus La. albirostris Principal 2 Component cruciger is vital to clarifying the relationships within this La.La. albialbiroroststririss Lissodelphis borealis of species in Lissodelphis and Cephalorhynchus, hclust "ward.D2") (*, La.La. albialbiroroststririss La.La. albialbiroroststririss -5 Unknown subfamily because of the proposed close relationship La.La. albialbiroroststririss reflected the results of these genetic studies. La.Li. albi bororestalisris 2-7 La.Li. albi bororestalisris between La. cruciger and La. australis . d e f Li.Li. boborerealisalis Li.Li. boborerealisalis • The close proximity/overlap of species clusters in the Li.Li. boborerealisalis Li.Li. boborerealisalis Li.Li. boborerealisalis -5 0 5 10 15 PCAs may be because they have been subject to similar Li.Li. boborerealisalis Principal Component 1 Li.Li. boborerealisalis Li.Li. boborerealisalis evolutionary pressures, or due to the recent and rapid Li.Li. boborerealisalis Li.Un boknreownalis La.Li. ob boscurealisrus radiation of delphinids not giving these species ample La.Un obknscuownrus Figure 4. PCA with size correction. All variables were regressed La.C. ob eutscurorupias time to accumulate great morphological differences. La.C. ob eutscurorupias against condylobasal length (total skull length) to correct for Methods La.C. obeutscuropiarus C.C. heaeutrovisipiadii size. PC1 (47.35%) was dominated by supraorbital thickness, La.La. ob auscustraruliss C.La. hea auvisistradiilis La.La. auauststraralislis PC2 (20.48%) by lacrimal length (Fig. 5). Note the continued We collected 38 skull measurements from 5 of the 6 La.La. auauststraralislis La.La. auauststraralislis overlap between La. obliquidens l and La. obscurus l, and the Acknowledgments Lagenorhynchus species (excluding L. cruciger due to a lack La.La. auauststraralislis La.La. auauststraralislis shift of La. australis l away from l and l and towards C. La.La. obliquidens australis of samples), as well as Lissodelphis borealis and all 4 La.La. obauscustraruliss La.La. obliquidens obscurus hectori p and C. commersonii p. The Lagenorhynchus and We would like to thank Allen Collins for introducing Ali to the world of Cephalorhynchus species, totaling 106 individuals. Only La.La. obobscuscururuss La.La. obobscuscururuss Cephalorhynchus species have formed one large super-cluster, cnidarians, as well as John Ososky, Ana Costa, Jim Mead, and Charley La.La. obliquidens obscurus mature individuals were measured. Measurements were La.La. obliquidens obscurus to the exclusion of Li. borealis u. La. acutus l and La. Potter for sharing their insight on marine mammals. We thank Liz Cottrell La.La. obliquidensobliquidens taken using the Microscribe 3D-LX. We used the Perrin La.La. obliquidensobliquidens PCA with Size Correction and Gene Hunt, the Natural History Research Experiences (NHRE) co- La.La. obliquidensobliquidens albirostris l lie on the margins of this super-cluster. 8 La.La. obliquidens obscurus 1975 skull measuring scheme, with the addition of 3 novel La.La. obliquidens obscurus directors, as well as NHRE administrator Virginia Power, for their support La.La. obobscuscururuss measurements. La.La. obliquidens obscurus and guidance. A special thanks to Gene Hunt for his assistance with the La.La. obobscuscururuss La.La. obliquidensobliquidens PCA. This project was made possible by the National Science Foundation. Cluster analysis (CA), principal component analysis (PCA), La.La. obliquidens obscurus La.La. obliquidensobliquidens 2.5 discriminant function analysis (DFA), and MANOVA La.La. obliquidensobliquidens Cephalorhynch1)us co mme Cipriano,rsonii F. (1997). Antitropical distributions and speciation in dolphins of the genus Lagenorhynchus: a preliminary analysis. La.La. obliquidensobliquidens CephalorhynchMolecularus eutropia Genetics of Marine Mammals Volume Special Publication, 3. La.La. obliquidensobliquidens statistical tests were run using R Version 3.2.1. Data were La. obliquidens Cephalorhynch2)us hea visi Miyazaki,dii N., & Shikano, C. (1997). Preliminary study on comparative skull morphology and vertebral formula among the six species La. obliquidens of the genus Lagenorhynchus (Cetacea: Delphinidae). Mammalia, 61(4), 573-587. La.La. obliquidensobliquidens Cephalorhynchus hectori 3) LeDuc, R. G., Perrin, W. F., & Dizon, A. E. (1999). Phylogenetic relationships among the delphinid cetaceans based on full cytochrome log transformed for analyses. When appropriate based on La.La. obliquidensobliquidens Lagenorhynchus acutus b sequences. Marine Mammal Science, 15(3), 619-648. La.La. obliquidens obscurus 0.0 sample size, we tested for sexual dimorphism. We handled La.La. obliquidens obscurus Lagenorhynchu4)s albi ro Pichler,stris F. B., Robineau, D., Goodall, R. N. P., Meyer, M. A., Olivarria, C., & Baker, C. S. (2001). Origin and radiation of Southern La.La. obobscuscururuss LagenorhynchuHemispheres australis coastal dolphins (genus Cephalorhynchus). Molecular Ecology, 10(9), 2215-2223. La. obscurus La. obliquidens Lagenorhynchu5)s obliquidens Harlin-Cognato, A. D., & Honeycutt, R. L. (2006). Multi-locus phylogeny of dolphins in the subfamily Lissodelphininae: character missing data in two ways: 1) deleting all variables and La.La. obliquidens obscurus Lagenorhynchusynergys obscuru improvess phylogenetic resolution. BMC evolutionary biology, 6(1), 87. La.La. obliquidensobliquidens Principal 2 Component La. obliquidens 6) May-Collado, L., & Agnarsson, I. (2006). Cytochrome b and Bayesian inference of whale phylogeny. Molecular phylogenetics and individuals with missing data, or 2) filling in missing data La. obliquidens Lissodelphis borealis La.La. obliquidens obscurus evolution, 38(2), 344-354. La. obliquidens -2.5 Unknown 7) Banguera-Hinestroza, E., Hayano, A., Crespo, E., & Hoelzel, A. R. (2014). Delphinid systematics and biogeography using averages from the same species. Overall clustering La. obscurus with a focus on the current genus Lagenorhynchus: Multiple pathways for antitropical and trans-oceanic radiation. Molecular phylogenetics and evolution, 80, 217-230. patterns were the same regardless of how missing data SOT 8) Perrin, W. F. (1975). Variation of spotted and spinner porpoise (genus Stenella) in the eastern Pacific and Figure 5. Lagenorhynchus skull. Hawaii. Scripps Institution of Oceanography. were treated, so method 2 was used during statistical 9) Walker, W. A., Leatherwood, S., Goodrich, K. R., Perrin, W. F., & Stroud, R. K. (1986). Geographical variation and Supraorbital thickness (SOT) -5.0 biology of the Pacific white-sided dolphin, Lagenorhynchus obliquidens, in the north-eastern Pacific. Research on analyses. -5 0 5 10 dolphins, 441, 465. and lacrimal length (LL). LL Principal Component 1 Dolphin thumbnail images © Jörg Mazur. Background image by Steve Hillebrand. Skull image by A. Galezo.