2009 NORTHEASTERN NATURALIST 16(4):621–628 Morphological Analysis of a Hybrid Minnow Swarm Robert L. Hopkins II1,*, David J. Eisenhour2, Lisa J. Hopman1, and David K. Peyton2 Abstract - Lythrurus fasciolaris (Scarlet Shiner) and Lythrurus umbratilis (Redfin Shiner) are closely related minnow species usually exhibiting a parapatric distribution in the Ohio River basin; however, areas of sympatry do exist and hybridization has been documented. In Eagle Creek (Kentucky River drainage), a stream capture event is likely responsible for creating the contact zone, providing a unique opportunity to study hybridization. In this paper, we conduct a morphological analysis of nuptial males from six Lythrurus populations in Eagle Creek. Our results show longitudinal variation in morphology. Individuals from downstream areas exhibit a novel phenotype with a robust body and low meristic counts. In contrast, upstream areas contain individuals resembling Redfin Shiner. Overall, 62% of individuals examined in Eagle Creek have the novel phenotype and 27% have the Redfin Shiner phenotype. Local selection pressures and apparent geographical isolation may be fostering the development of a distinct phenotype via introgressive hybridization. Introduction Hybridization is common among freshwater fishes (Dowling and De- Marais 1993, Hubbs 1955, Scribner et al. 2001). However, the evolutionary significance of hybridization in fishes remains in dispute. Lythrurus fasciolaris Cope (Scarlet Shiner) and Lythrurus umbratilis Girard (Redfin Shiner) are closely related, common minnow species with a nearly parapatric distribution within the Ohio River basin (Burr and Warren 1986, Etnier and Starnes 1993, Trautman 1981). Typically, these species are ecologically segregated, with Scarlet Shiner occupying higher gradient streams with coarser substrates. Exceptions to parapatry include areas along the periphery of the species’ ranges (Burr and Warren 1986, Trautman 1981). Within these sympatric populations, hybridization is evidently due to congruent nesting biologies; both species are nest associates with Lepomis species (Etnier and Starnes 1993, Trautman 1981). Hopkins and Eisenhour (2008) examined putative hybrid populations of these species within the Ohio River basin and found that several populations showed evidence of gene flow based on morphological analyses. They also recovered evidence of an eastward range expansion of Redfin Shiner with introgressive elimination of Scarlet Shiner. In most cases, the contact zone between these two species is restricted to the abutment of distributional ranges. However, the population from Eagle Creek in the Kentucky River drainage—which is documented as containing 1Department of Zoology, Southern Illinois University, Carbondale, IL 62901. 2Department of Biological and Environmental Sciences, Morehead State University, Morehead, KY 40351. *Corresponding author - rhopkins@siu.edu. 622 Northeastern Naturalist Vol. 16, No. 4 only Scarlet Shiner (Burr and Warren 1986)—is an exception. Geomorphological evidence suggests a stream capture event is responsible for the contact zone in Eagle Creek, which is thought to have eroded eastward and captured the north flowing tributaries of the Licking River or perhaps an Ohio River tributary. As a result, Redfin Shiner from the Licking River drainage was introduced into Eagle Creek. No other taxa have the geographical distributions required to test the hypothesis of stream capture in Eagle Creek, and Hopkins and Eisenhour (2008) provide the first biogeographical evidence of such an event. According to Hopkins and Eisenhour (2008), individuals from Eagle Creek exhibit a novel phenotype not detectable in either parental species or other hybrid population. Morphological examination of specimens from two localities in Eagle Creek showed discordant combinations of shape, meristics, coloration, and tuberculation. In this paper, we evaluate the status of Lythrurus populations in Eagle Creek. We discuss the spatial extent of the hybrid swarm within Eagle Creek, characterize patterns of phenotypic variation along the stream gradient, and consider the potential evolutionary implications of the hybrid swarm. Methods Specimens of putative hybrids were collected from six localities in Eagle Creek in June 2008 (Fig. 1). Five meristic characters were used in the analyses: anal-fin rays, scales above the lateral line, scales below the lateral line, scales around the caudal peduncle, and circumferential scales. Meristic counts followed Etnier and Starnes (1993). Meristic data were analyzed using principal components analysis (PCA). Morphometric data were acquired from 21 linear measurements between homologous landmarks (Bookstein et al. 1985) following, in part, the methods of Hopkins and Eisenhour (2008). Measurements were made to the nearest 0.1 mm using digital calipers. Multivariate analyses of morphometric data were accomplished using sheared PCA (Bookstein et al. 1985, Humprhies et al. 1981) to decrease allometric effects of size on shape variation. To optimize resolution of the analyses, a combined plot was used in which the x-axis is PC1 of the meristic data and the y-axis is sheared PC2 for the shape analysis (Hopkins and Eisenhour 2008). Data for parental species, which were collected from “pure” reference populations within the Ohio River drainage, were extracted from a pre-existing dataset used by Hopkins and Eisenhour (2008). In total, 270 specimens were examined: 100 Redfin Shiner from 18 populations, 106 Scarlet Shiner from 11 populations, and 70 putative hybrids from six localities in Eagle Creek. Only nuptial males were used in the analyses to reduce the effects of sexual and seasonal variation. Results Comparable to the results of Hopkins and Eisenhour (2008), a combined plot of PCA results was able to separate over 98% of reference specimens 2009 R.L. Hopkins II, D.J. Eisenhour, L.J. Hopman, and D.K. Peyton 623 (Fig. 2, top), providing a high level of resolution for our assessment of the Eagle Creek population. Typically, Scarlet Shiner displayed lower meristic counts and a more slender, gracile body. In contrast, Redfin Shiner had a more robust body with higher meristic counts. Each of the five counts evaluated provided some level of discrimination between reference individuals. The novel phenotype in Eagle Creek is clearly evident when all individuals are pooled (Fig. 2, bottom); these individuals showed discordant combinations of meristic counts and body shape. Generally speaking, most individuals showed meristic counts characteristic of Scarlet Shiner, while Figure 1. Map of sample sites in the Ohio River basin. Inset shows localities for reference specimens. Eagle Creek sites are individually labeled for reference. 624 Northeastern Naturalist Vol. 16, No. 4 2009 R.L. Hopkins II, D.J. Eisenhour, L.J. Hopman, and D.K. Peyton 625 body shape was most similar to Redfin Shiner. In total, 7.2% of Eagle Creek specimens fell within the Scarlet Shiner polygon, 27.2% fell within the Redfin Shiner polygon, 2.9% fell within the zone of overlap, and 62.7% fell entirely outside of either reference polygon. When each Eagle Creek locality was plotted separately, a general trend of phenotyptic variation emerged (Fig. 3). The most downstream site, EC 1, showed the highest proportion of individuals having the novel phenotype. At EC 1, seven of eight specimens scored outside of either reference polygon. The prevalence of Redfin Shiner-like phenotypes generally increased with position upstream. At EC 6, the most upstream site, five individuals had Redfin Shiner-like phenotypes and seven had the novel phenotype. Sites positioned between EC 1 and EC 6 showed a modest gradation of these two patterns. Only EC 4 and EC 5 contained individuals scoring within the Scarlet Shiner polygon. Site EC 4 showed the most phenotypic variation, with four Scarlet Shiner-like individuals, two intermediate individuals, three Redfin Shiner-like individuals, and four novel phenotypes. Site EC 5 had a single individual located on the boundary of the Scarlet Shiner polygon. Coloration and tuberculation patterns varied greatly within and between sites. However, upstream sites did contain more individuals with Redfin Shiner-like tuberculation and coloration. For example, 42% of specimens at EC 6 had Redfin Shiner-like morphology compared to only 13% at EC 1. Discussion Analysis of morphological variation of Lythrurus in Eagle Creek suggests the formation and persistence of a novel phenotype. However, the relative abundance of the novel phenotype shows longitudinal variation. Many individuals from headwater areas more closely resemble Redfin Shiner. Historically, these headwater streams probably drained into the adjacent Licking River drainage or an Ohio River tributary and contained resident populations of Redfin Shiner. Thus, the presence of this gradation is not surprising. Despite apparent hybridization with and immigration of Scarlet Shiner— which was the probable original inhabitant of Eagle Creek—the Redfin Shiner phenotype has persisted. However, individuals with the Redfin Shiner phenotype did not display the full range of variation detected in reference populations and consistently scored on the edge of the polygon. The most upstream site, EC 6, showed the widest variation of individuals scoring within the Redfin Shiner polygon. Moreover, only a few individuals examined resembled Scarlet Shiner and even fewer individuals had intermediate morphologies. In each population analyzed, the number of individuals with the novel phenotype invariably was higher than any other phenotype. With such a low proportion of individuals showing typical morphologies for reference Figure 2 (opposite page). Combined plot of PCA scores for meristic and shape analyses. Top, results for 206 reference specimens. Bottom, results for all 70 specimens from Eagle Creek. Polygons of reference specimens are bounded by dashed lines. 626 Northeastern Naturalist Vol. 16, No. 4 species, it is possible that local ecological conditions may be fostering selection of hybrid (Fx or Bx) offspring. Sampling efforts downstream of EC 1 failed to produce any Lythrurus specimens. Near the confluence with the Kentucky River, Eagle Creek exhibits a marked shift in hydrology with an abundance of long, slack pools. The apparent absence of Lythrurus in this downstream area combined with Figure 3. Combined PCA plots for specimens from Eagle Creek separated by locality. Locality numbers correspond to map of sample sites and are sequenced from mostdownstream (A) to most-upstream (F) site. 2009 R.L. Hopkins II, D.J. Eisenhour, L.J. Hopman, and D.K. Peyton 627 the hydrological modification of the Kentucky River proper by navigation dams probably precludes a high level of gene flow with populations in adjacent systems. Given these conditions, we speculate that the Eagle Creek population will continue a shift toward the novel phenotype and the hybrid zone will be constrained to Eagle Creek. Similar tendencies of hybrid populations have been documented for other taxa including butterflies (Mavarez et al. 2006), birds (Pearson and Manuwal 2000), turtles (Lutterschmidt et al. 2007), and fishes (Dowling and DeMarais 1993, Salzburger et al. 2002). In many of these cases, hybridization is believed to have been the mechanism for speciation. We are not suggesting that the Lythrurus in Eagle Creek warrant specific recognition, but we do hypothesize that stream capture and subsequent hybridization has created a markedly different evolutionary trajectory compared to adjacent populations. Presently, the population in Eagle Creek is almost morphologically distinct and nearly diagnosable from other Lythrurus populations. Outside of Eagle Creek, contemporary and historical hybridization has perhaps resulted in the morphological convergence of Scarlet Shiner and Redfin Shiner in the upper Ohio River basin (Eisenhour and Eisenhour 2007) or morphological introgression toward Redfin Shiner (Hopkins and Eisenhour 2008). However, none of these other hybrid populations exhibit a novel phenotype. It is possible the stream capture event and environmental conditions in Eagle Creek have provided a unique system for the population evolution of Lythrurus. We are currently investigating the Eagle Creek population using mitochondrial and nuclear DNA markers to describe the genetic structure and quantify gene flow within the population. Acknowledgments We thank M.A. Hopkins for assistance in the field and B.M. Burr for a helpful review of the manuscript. Specimens were removed from the field in accordance with permits from the Kentucky Department of Fish and Wildlife Resources, Morehead State University IACUC, and Southern Illinois University IACUC. Literature Cited Bookstein, F.L., N. Chernoff, R.L. Elder, J.M. Humphries, G.R. Smith, and R.E. Strauss. 1985. Morphometrics in evolutionary biology. Academy of Natural Sciences, Philadelphia Special Publication 15:1–277. Burr, B.M., and M.L. Warren Jr. 1986. A Distributional Atlas of Kentucky Fishes. Kentucky State Nature Preserves Commission, Frankfort, KY. Science and Technical Series 4. 398 pp. Dowling, T.E., and B.D. DeMarais. 1993. Evolutionary significance of introgressive hybridization in fishes. Nature 362:444–446. Eisenhour, D.J., and L.V. Eisenhour. 2007. Clinal variation in Ohio River basing populations of the Redfin Shiner (Lythrurus umbratilis). Northeastern Naturalist 14:519–530. Etnier, D.A., and W.C. Starnes. 1993. The Fishes of Tennessee. University of Tennessee Press, Knoxville, TN. 681 pp. 628 Northeastern Naturalist Vol. 16, No. 4 Hopkins II, R.L., and D.J. Eisenhour. 2008. Hybridization of Lythrurus fasciolaris and Lythrurus umbratilis (Cypriniformes: Cyprinidae) in the Ohio River basin. Copeia 2008:162–171. Hubbs, C.L. 1955. Hybridization between fish species in nature. Systematic Zoology 4:1–20. Humprhies, J.M., F.L. Bookstein, B. Chernoff, G.R. Smith, R.L. Elder, and S.L. Poss. 1981. Multivariate discrimination by shape in relation to size. Systematic Zoology 30:291–308. Lutterschmidt, W.I., S.A. Escobar, and E.D. Wilson. 2007. Multivariate analyses of shell morphology of putative hybrid Box Turtles. Northeastern Naturalist 6:571–576. Mavarez, J., C.A. Salazar, E. Bermingham, C. Salcedo, C.D. Jiggins, and M. Linares. 2006. Speciation by hybridization in Heliconius butterflies. Nature 441:868–871. Pearson, S.F., and D.A. Manuwal. 2000. Influence of niche overlap and territoriality on hybridization between Hermit Warblers and Townsend’s Warblers. The Auk 117:175–183. Salzburger, W., S. Baric, and C. Sturmbauer. 2002. Speciation via introgressive hybridization in east African cichlids? Molecular Ecology 11:619–625. Scribner, K.T., K.S. Page, and M. Barton. 2001. Hybridization in freshwater fishes: A review of case studies and cytonuclear methods of biological inference. Reviews in Fish Biology and Fisheries 10:293–323. Trautman, M.B. 1981. The Fishes of Ohio, 2nd Edition. Ohio State University Press, Columbus, OH. 782 pp.