2012 SOUTHEASTERN NATURALIST 11(4):733–746 Host Identification and Glochidia Morphology of Freshwater Mussels from the Altamaha River Basin Jennifer A. Johnson1, Jason M. Wisniewski2, Andrea K. Fritts1, and Robert B. Bringolf 1,* Abstract - Recovery of imperiled freshwater mussels requires knowledge of suitable host fishes and other early life-history traits. We provide quantitative host information for 6 mussel species from the Altamaha River Basin, GA, 3 of which previously had no host information. Glochidia of Alasmidonta arcula (Altamaha Arcmussel) metamorphosed on 2 species of suckers (Moxostoma spp.); Elliptio hopetonensis (Altamaha Slabshell) on Lepomis macrochirus (Bluegill), Pimephales promelas (Fathead Minnow), and Micropterus salmoides (Largemouth Bass); E. shepardiana (Altamaha Lance) on 2 species of Bullheads (Ameiurus spp.) and L. macrochirus; Lampsilis dolabraeformis (Altamaha Pocketbook) on Bluegill and Largemouth Bass; and L. splendida (Rayed Pink Fatmucket) and Villosa delumbis (Eastern Creekshell) on Largemouth Bass. We also provide descriptions of glochidia morphology for the above mussel species and E. spinosa (Altamaha Spinymussel). Glochidia were correctly identified to species in 88.7% of cases by discriminant function analysis of 3 shell dimensions. Glochidia morphology may be useful for identification of glochidia attached to wild fish, thereby providing additional host information. Introduction Freshwater mussels (family Unionidae) provide critical ecosystem services and often dominate the benthic biomass in minimally impacted streams (Strayer et al. 2004). Live mussels and empty shells provide habitat for other invertebrates and fish, and as filter feeders, they influence nutrient cycling by linking the water column and the substrate (Spooner and Vaughn 2006). Nearly 300 species of unionids occur in North America, but unfortunately they are highly imperiled, with approximately 70% of North American species being considered of conservation concern (Neves et al. 1997, Williams et al. 1993). Factors thought to contribute to the decline of mussels include sedimentation, pollution, urbanization, and habitat fragmentation (Williams et al. 1993). Because nearly all mussel larvae (glochidia) are obligate parasites on fish, declines in host fish populations may also contribute to mussel declines. The Altamaha River in Georgia (drainage area = 36,976 km2) is inhabited by approximately 18 mussel species, of which 7 are endemic (Johnson 1970). At least 3 of the endemic species, Alasmidonta arcula Lea (Altamaha Arcmussel), Pyganodon gibbosa Say (Inflated Floater), and Elliptio spinosa Lea (Altamaha Spinymussel), are declining (Dinkins et al. 2004, Keferl 1981, O’Brien 2002a, Skelton et al. 2002, Wisniewski et al. 2005), and E. spinosa was listed as 1Warnell School of Forestry and Natural Resources, University of Georgia, Athens, GA 30602. 2Nongame Conservation Section, Wildlife Resources Division, Georgia Department of Natural Resources, Social Circle, GA 30025. *Corresponding author - bringolf@ uga.edu. 734 Southeastern Naturalist Vol. 11, No. 4 endangered in 2011 under the US Endangered Species Act. Fish hosts of mussel species in the Altamaha River are poorly known, and of the 3 declining endemic species, limited host information is available only for A. arcula. Identification of host fishes allows managers to determine if appropriate host species are present in the river, and knowledge of suitable hosts is also necessary for captive propagation of mussels, which can produce juveniles for population augmentation or toxicity testing. The Georgia State Wildlife Action Plan (2005; http://www.georgiawildlife. com/conservation/wildlife-action-plan) identified knowledge of early life histories of mussels endemic to the Altamaha River Basin as a high priority need for recovery of these species. Description of glochidia morphology also is necessary to identify patterns of glochidia occurrence on wild fishes, to inform phylogenetic relationships among species, and to provide additional characters for identification. In this study, we identify host fish and describe glochidia morphology of 7 mussels from the Altamaha River Basin: A. arcula, E. spinosa, E. hopetonensis Lea (Altamaha Slabshell), E. shepardiana Lea (Altamaha Lance), Lampsilis dolabraeformis Lea (Altamha Pocketbook), L. splendida Lea (Rayed Pink Fatmucket), and Villosa delumbis Conrad (Eastern Creekshell). Methods Mussel collection Gravid female mussels were collected from the mainstem Altamaha River in 2008–2009 (Table 1) by visual and tactile searches with SCUBA and snorkel. All species except A. arcula and E. spinosa were collected downstream of Oglethorpe Bluff Landing, ≈12 km north of Jesup, GA (Fig. 1); A. arcula and E. spinosa were collected upstream of Upper Wayne Landing, ≈17.1 km SSW of Glennville, GA. Mussels were gently pried open to detect marsupial swelling, which indicate females are gravid and brooding glochidia. On each sample date, at least 10 individuals of each species were examined, and if no gravid mussels were found, inspections ceased. Gravid females of each species were transported in coolers to the University of Georgia. To minimize premature release of glochidia, mussels were maintained in dechlorinated tap water at 10–12 ºC in a Living Stream (Frigid Units, Inc., Toledo, OH). Mussels were fed weekly with a mixture of concentrated microalgae (Reed Mariculture Instant Algae® Shellfish Diet and Nanno Table 1. Collection dates and water temperature at time of collection of gravid female mussels from the Altamaha River. Species Number collected Date collected Water temperature (oC) Alasmidonta arcula 5 12 Nov 2009 16 Elliptio hopetonensis 5 18 June 2009 29 Elliptio shepardiana 5 18 June 2009 29 Elliptio spinosaA 1 18 May 2009 21 Lampsilis dolabraeformis 5 18 June 2009 29 Lampsilis splendida 2 1 Oct 2008 26 Villosa delumbis 2 1 Oct 2008 26 AGlochidia immature at time of collection; mature glochidia released on 10 June 2009. 2012 J.A. Johnson, J.M. Wisniewski, A.K. Fritts, and R.B. Bringolf 735 3600 Nannochloropsis). Glochidia were extracted from gravid females within 2 months for morphological characterization and host trials. Host trials Fish for host trials were obtained from state, federal, and private fish hatcheries in the southeastern US, or by electrofishing or seining in ponds and streams in the Altamaha basin where mussels do not occur. Small (presumably young) individuals of each fish species were used when possible to maximize likelihood of successful metamorphosis by glochidia (Strayer 2008). Based on length at capture, all Micropterus salmoides (Largemouth Bass) and Lepomis macrochirus (Bluegill) used were assumed to be young of year. Ideally, fish used in host trials have not had any previous exposure to mussels because fish develop acquired immunity to glochidia following exposure (Dodd et al. 2005, 2006; Meyers et al. 1980; Zale and Neves 1982). Fish species used in each trial depended on availability and were not the same for each mussel species (Tables 2, 3). Glochidia were obtained from 1 to 4 females of each species by gently flushing the marsupial gills with water from a 5-ml syringe. Elliptio spinosa, E. hopetonensis, and E. shepardiana released mature glochidia within one month of transport to the lab, so glochidia of these species were collected without the use of a syringe. Subsamples of 50–100 glochidia per female were tested for viability by exposure to a saturated sodium chloride solution; viable glochidia quickly closed their valves upon exposure to the solution. Only female mussels with glochidia viability of >90% were used for host trials. Potential host fishes Figure 1. Altamaha River Basin, GA and approximate location (oval) of freshwater mussel collections from the mainstem Altamaha River. 736 Southeastern Naturalist Vol. 11, No. 4 were separated by species and exposed to glochidia for one hour in aerated 19-L buckets. Each bucket contained 1–22 individuals of each fish species (Tables 2, 3), and a target glochidia concentration of 4000/L. For E. spinosa, a limited amount of viable glochidia were available (from a single female), so glochidia were pipetted directly on the fish gills to increase the probabi lity of attachment. Glochidia-infested fish were removed from the glochidia suspension, gently rinsed to remove any unattached glochidia, and placed in holding (monitoring) tanks in a recirculating aquaculture system (AHAB; Aquatic Habitats Inc., Apopka, FL). The AHAB unit is an array of self-cleaning tanks through which water flows and then re-circulates back into a main sump for treatment with activated carbon and UV sterilization. Whenever possible, we placed one fish per tank, but unfortunately this was not feasible for all species because of limited space; no tanks contained more than 3 fish (always the same species). Water temperatures ranged from 21–25 ºC during all host trials. Filter cups (5-cm PVC pipe with 153-μm-mesh screen on one end) were fitted to the outlet of each tank to capture dead glochidia and metamorphosed juveniles as they were sloughed from the Table 2. Summary of Alasmidonta arcula host trial. Metamorphosis success is reported as mean ± 95% confidence interval. Juv = juveniles produced, Period = period in days of juvenile release, # = number of fish used. Metamorphosis Fish species success (%) Juv Period # Moxostoma robustum (Cope) (Robust Redhorse) 4.6 ± 0.63 61 8–13 5 Moxostoma rupiscartes Jordan and Jenkins (Striped Jumprock) 0.12 ± 0.11 4 10–11 7 Acipenser brevirostrum Lesueur (Shortnose Sturgeon) 0 0 - 2 Acipenser fulvescens Rafinesque (Lake Sturgeon) 0 0 - 5 Acipenser oxyrinchus Mitchill (Atlantic Sturgeon) 0 0 - 2 Ameiurus brunneus (Jordan) (Snail Bullhead) 0 0 - 2 Ameiurus natalis (Leseur) (Yellow Bullhead) 0 0 - 1 Cyprinus carpio L. (Common Carp) 0 0 - 2 Etheostoma inscriptum (Jordan and Brayton) (Turquoise Darter) 0 0 - 3 Hypentelium nigricans (Lesueur) (Northern Hogsucker) 0 0 - 6 Ictalurus punctatus (Rafinesque) (Channel Catfish) 0 0 - 5 Lepomis auritus (L.) (Redbreast Sunfish) 0 0 - 1 Lepomis cyanellus Rafinesque (Green Sunfish) 0 0 - 5 Lepomis macrochirus Rafinesque (Bluegill Sunfish) 0 0 - 7 Lepisosteus osseus (L.) (Long Nose Gar) 0 0 - 3 Minytrema melanops (Rafinesque) (Spotted Sucker) 0 0 - 3 Micropterus salmoides (Lacépède) (Largemouth Bass) 0 0 - 3 Morone chrysops (Rafinesque) (White Bass) 0 0 - 5 Moxostoma collapsum (Cope) (Notch-Lip Sucker) 0 0 - 1 Nocomis leptocephalus (Girard) (Bluehead Chub) 0 0 - 4 Notemigonus crysoleucas (Mitchill) (Golden Shiner) 0 0 - 3 Notropis hudsonius (Clinton) (Spottail Shiner) 0 0 - 1 Notropis lutipinnis (D.S. Jordan & Brayton) (Yellowfin Shiner) 0 0 - 5 Notropis rubescens Bailey (Rosyface Chub) 0 0 - 4 Noturus leptacanthus D.S. Jordan (Speckled Madtom) 0 0 - 1 Pimephales promelas Rafinesque (Fathead Minnow) 0 0 - 4 Plyodictis olivaris (Rafinesque) (Flathead Catfish) 0 0 - 2 Semotilus atromaculatus (Mitchill) (Creek Chub) 0 0 - 4 2012 J.A. Johnson, J.M. Wisniewski, A.K. Fritts, and R.B. Bringolf 737 fish. Beginning one day after exposure (day 1), filter cups were checked daily for 7 days, and dead glochidia and metamorphosed juveniles were counted and photographed. After day 7, the cups were checked every other day. Filtered material Table 3. Summary of host trials by species of mussel from the Altamaha River, GA. Metamorphosis success is reported as mean ± 95% confidence interval. Period = period in days of juvenile release, # = number of fish used. Mussel species/ Metamorphosis Juveniles fish species Common name success (%) produced Period # Elliptio hopetonensis Lepomis macrochirus Bluegill 3.7 ± 1.0 49 7–8 8 Pimephales promelas Fathead Minnow 3.1 ± 0.7 55 7 4 Micropterus salmoides Largemouth Bass 0.8 ± 0.4 16 7 4 Acipenser fulvenscens Lake Sturgeon 0 0 - 3 Cyprinus carpio Common Carp 0 0 - 4 Ictalurus punctatus Channel Catfish 0 0 - 3 Elliptio shepardiana Ameiurus nebulosus Brown Bullhead 45.2 ± 35.8 378 11–17 2 Ameiurus natalis Yellow Bullhead 18.9 17 11–14 1 Lepomis macrochirus Bluegill 2.2 ± 1.4 4 11–12 6 Moxostoma robustum Robust Redhorse 0.1 ± 0.06 1 17 3 Acipenser fulvenscens Lake Sturgeon 0 0 - 2 Hypentelium nigricans Northern Hogsucker 0 0 - 2 Nocomis leptocephalus Bluehead Chub 0 0 - 4 Lepomis microlophus Redear Sunfish 0 0 - 5 Pimephales promelas Fathead Minnow 0 0 - 5 Notropis hudsonius Spottail Shiner 0 0 - 1 Notemigonus crysoleucas Golden Shiner 0 0 - 1 Notropis lutipinnis Yellowfin Shiner 0 0 - 15 Elliptio spinosa Acipenser fulvenscens Lake Sturgeon 0 0 - 4 Cyprinus carpio Common Carp 0 0 - 4 Ictalurus punctatus Channel Catfish 0 0 - 4 Lepomis auritus Redbreast Sunfish 0 0 - 4 Lepomis macrochirus Bluegill 0 0 - 3 Micropterus salmoides Largemouth Bass 0 0 - 4 Minytrema melanops Spotted Sucker 0 0 - 4 Morone chrysops White Bass 0 0 - 1 Morone saxatilis Striped Bass 0 0 - 4 Pimephales promelas Fathead Minnow 0 0 - 2 Lampsilis dolabraeformis Micropterus salmoides Largemouth Bass 74.8 ± 3.6 1209 19–29 4 Lepomis macrochirus Bluegill 1.5 ± 0.1 23 10–19 22 Acipenser fulvenscens Lake Sturgeon 0 0 - 3 Ictalurus punctatus Channel Catfish 0 0 - 2 Notemigonus crysoleucas Golden Shiner 0 0 - 5 Pimephales promelas Fathead Minnow 0 0 - 4 Lampsilis splendida Micropterus salmoides Largemouth Bass 43.0 ± 5.5 2352 13–24 5 Villosa delumbis Micropterus salmoides Largemouth Bass 73.1 ± 1.5 4673 12–24 18 738 Southeastern Naturalist Vol. 11, No. 4 was gently rinsed into a Bogorov tray and examined under a stereomicroscope. Juveniles were identified by the presence of tissues such as gills, foot, and heart. To determine if juveniles were alive, we observed them for foot movement, heartbeat, or valve closure. We checked filter cups every 2 days until no glochidia or juveniles were observed for 5 consecutive days. When fish mortality occurred, deceased fish were examined for glochidia. No high infestations were observed on dead fish, and data from these fish was not included in the final analysis of metamorphosis success. We quantified juvenile metamorphosis (%M) for individual fish as ([number of juveniles/ (juveniles + glochidia)] x 100). When more than one individual of a fish species was used per tank, we determined the total %M for the tank. We then calculated the mean %M across replicates (tanks) for each species. Initial glochidia attachment rates were determined by summing the total number of sloughed glochidia and juveniles that were recovered from each tank. Glochidia morphology A sub-sample (n = 100–150) of glochidia from each individual mussel was fixed in formalin for at least 24 hours and then stored in 95% ethanol. Morphological measurements were made for 25 glochidia/female for each species (Table 4). Glochidia were photographed at magnifications of 16–50X with a stereomicroscope (Leica MZ 7.5, Leica Microsystems, Wetzlar, Germany) equipped with a digital camera (Leica DCF 290, Leica Microsystems, Wetzlar, Germany). Glochidia shape classifications were based on previous descriptions (Hoggarth 1999, Hornbach et al. 2010). Glochidia length (parallel to hinge), height (perpendicular to hinge), and hinge length (Hoggarth 1999, Kennedey and Haag 2005) were measured with image analysis software (Leica LAS, Leica Microsystems, Wetzlar, Germany). Differences in mean length, height, and hinge-length measurements were compared among species with 3 separate analysis of variance (ANOVAs) followed by Tukey’s test to identify differences between species for each measurement (α = 0.05). We also examined the utility of shell measurements to identify glochidia by species with a discriminant function analysis (DFA) as previously described by Kennedy and Haag (2005). Briefly, all measurements were transformed (log10[x + 1]) to achieve normality, and we derived quadratic discriminant functions for each species because variance-covariance matrices Table 4. Glochidia (n = 25 per female) measurements (mean ± 95% confidence interval) for freshwater mussels of the Altamaha River, GA. Within a column (shell dimension), different superscripted capital letters indicate significant differences among species (Tukey’s test, α = 0.05); species with the same letter within a column were not significan tly different. Species # of females Height (μm) Length (μm) Hinge (μm) Alasmidonta arcula 4 A309 ± 14.3 A274 ± 5.4 A212 ± 4.1 Elliptio hopetonensis 1 B199 ± 4.7 B186 ± 3.8 B134 ± 2.4 Elliptio shepardiana 4 C259 ± 6.5 C217 ± 4.0 C153 ± 3.1 Elliptio spinosa 1 B208 ± 3.8 D197 ± 3.0 B133 ± 3.7 Lampsilis dolabraeformis 4 D239 ± 5.2 E207 ± 3.3 D100 ± 3.4 Lampsilis splendida 4 C268 ± 3.0 C222 ± 4.0 D97 ± 4.2 2012 J.A. Johnson, J.M. Wisniewski, A.K. Fritts, and R.B. Bringolf 739 were unequal (c2 = 92.7, df = 30, P < 0.0001; Morrison 1976). We used cross-validation scores to determine identification success for each individual glochidium and reported results as the % of total number of measured glochidia identified correctly. All statistical analyses were performed with SAS (version 8.2, SAS Institute, Cary, NC). Glochidia ultrastructure (i.e., presence of microstylets, interior/exterior valve sculpture) was described for L. dolabraeformis, E. shepardiana, E. spinosa, and A. arcula from scanning electron microscope images (SEM, Zeiss 1450EP, Zeiss SMT, Peabody, MA). We were unable to describe ultrastructure for glochidia of L. splendida and V. delumbis due to the low quality of preserved specimens. Glochidia samples were mounted on SEM stubs with carbon adhesive tabs (EMS, Hatfield, PA), and a SPI Module Sputter Coater (SPI Supplies, Inc. West Chester, PA) was used to coat samples with 20 μm of gold. Specimens were then examined under the SEM run at 20 Kev with a probe size of 450 uA. Results Host trials Juvenile A. arcula (61 individuals total, %M = 4.6) were produced by all 4 individuals of Moxostoma robustum Cope (Robust Redhorse) and 4 A. arcula juveniles (%M = 0.8) were produced from 1 of the 7 Moxostoma rupiscartes (Striped Jumprock). No juvenile A. arcula were produced from 26 other fish species tested (Table 2). Juvenile E. hopetonensis were produced from Bluegill (%M = 3.7), Pimephales promelas (Fathead Minnow; %M = 3.1), and Largemouth Bass (%M = 0.8), but 3 other species were non-hosts (Table 3). Juvenile E. shepardiana were produced by Ameiurus nebulosus Lesueur (Brown Bullhead; %M = 45.2), A. natalis (Yellow Bullhead; %M = 18.9) and Bluegill (%M = 2.2%). A single E. shepardiana juvenile was produced from a Robust Redhorse (%M = <0.1%), and no juveniles were produced from 8 other fish species (Table 3). Juvenile L. dolabraeformis were produced from Largemouth Bass (%M= 74.8%) and Bluegill (%M = 1.5), but no juveniles were produced from 4 additional fish species (Table 3). Largemouth Bass also produced juvenile L. splendida (% M = 43) and V. delumbis (%M = 73.1); no other fish species were tested with L. splendida and V. delumbis (Table 3). None of the 10 fish species tested produced juvenile E. spinosa (Table 3). Eight fish species sloughed 100% of the attached E. spinosa glochidia within 3 days after initial glochidia exposure (data not shown), but 4 E. spinosa glochidia remained attached on Acipenser fulvescens (Lake Sturgeon) and 5 on Lepomis auritus (Redbreast Sunfish) until 5 days after attachment. Glochidia morphology and shell ultrastructure Glochidia height (F5,149 = 161.3, P < 0.0001), length (F5,149 = 323.5, P < 0.0001), and hinge length (F5,149 = 530.4, P < 0.0001) all varied significantly among the 6 species (Table 4). The 95% confidence interval for a given shell dimension overlapped with 0–2 other taxa (Table 4). The DFA correctly classified 133 (88.7%) of the 150 glochidia in the data set, and correct classification percentages by 740 Southeastern Naturalist Vol. 11, No. 4 species ranged from 60% to 100% (Table 5). Correct classification was 100% for 3 of the 6 taxa. Glochidia were misclassified as 0–2 other taxa, and the most common misclassification was E. hopetonensis as E. spinosa (40%). Conversely, E. spinosa were misclassified as E. hopetonensis in 5 of 25 cases (20%). All other misclassifications were ≤4%. Glochidia shell structure varied by species. Alasmidonta arcula glochidia were pyriform and contained a styliform hook ventrally located on each pitted valve (Fig. 2). Lampsilis dolabraeformis had a pitted subelliptical valve with a ventral edge (flange) covered with micropoints (Fig. 3). Lampsilis splendida glochidia were also subelliptical, but we were unable to determine ultrastructure Table 5. Identification success (%) for glochidia of 6 mussel taxa from the Altamaha River Basin. Values were determined with cross-validated scores of quadratic discriminant functions for 25 glochidia of each species. Numbers in parenthesis are % of glochidia misclassified for the given species. Species Correct (%) Misclassified as Alasmidonta arcula 100 - Elliptio hopetonensis 60 E. spinosa (40) Elliptio shepardiana 100 - Elliptio spinosa 76 E. hopetonensis (20), L. dolabraeformis (4) Lampsilis dolabraeformis 100 - Lampsilis splendida 96 L. dolabraeformis (4) Figure 2. Scanning electron micrographs of Alasmidonta arcula glochidia A) exterior, 200X, B) interior 200X, C) side view, 160X, and D) detail of styliform hook, 1100X. Scale bars = 100μm. 2012 J.A. Johnson, J.M. Wisniewski, A.K. Fritts, and R.B. Bringolf 741 with SEM. Glochidia of E. hopetonensis had a depressed subelliptical shape. Elliptio shepardiana glochidia had a pitted valve structure and a depressed subelliptical shape with a ventral flange extended beyond the gill margin covered with numerous microstylets (Fig. 4A). Similar to the other members of the genus Elliptio, E. spinosa was also pitted with a depressed subelliptical shape and the ventral flange was covered with microstylets, but not extended (Fig. 4B). We also observed several subcylindrical white conglutinate packets (with immature glochidia) released by E. spinosa in the lab; conglutinate length was 18 mm and width was 5 mm. Discussion We produced quantitative host information for 6 mussel species from the Altamaha River Basin, GA, 3 of which previously had no host information. Our finding that robust Redhorse may serve as a host for A. arcula is noteworthy because we are aware of only one other report of a mussel-host fish relationship involving 2 imperiled species (Fritts et al. 2012). Robust Redhorse were historically abundant in the Altamaha Basin but are now listed as state endangered. They were thought to be extinct for 122 years until their “rediscovery” in 1991 in the Altamaha Basin and subsequently in other Atlantic Slope drainages north to Virginia (Grabowski and Jennings 2009). The relatively low metamorphosis rate (4.6%) of A. arcula on Robust Redhorse may not be indicative of the importance of this fish species as Figure. 4. Scanning electron micrographs of the exterior valves of A) Elliptio shepardiana, 200X, and B) Elliptio spinosa, 200X. Scale bars = 100 μm. Figure. 3. Scanning electron micrographs of Lampsilis dolabraeformis glochidia A) interior flange, 1180X, B) interior, 200X, and C) valve exterior, 200X. The wavy margin (dissociation of pellicle from shell) is an artifact of preservation. 742 Southeastern Naturalist Vol. 11, No. 4 a host because host trials were performed at warmer temperatures than likely experienced by glochidia and fish in the river during the time of natural encystment. Gravid A. arcula were collected in November when water temperature was 16 oC; thus, under natural conditions glochidia would develop on host fish when water temperatures were likely at or below this temperature. Unfortunately, we lacked the ability to chill the water in our trial, and temperatures were 22–24 oC. The warm temperatures may have created sub-optimal conditions for glochidia, resulting in lower %M. The only other species that produced any A. arcula juveniles was Striped Jumprock, suggesting that A. arcula may be a host specialist, using members of the sucker family (Catastomidae). Other suckers did not produce juveniles in our trials, but these tests should be repeated at cooler water temperatures that more closely mimic natural conditions before any definitive conclusions are made. Further, we have tested only a fraction of the 93 extant fish species in the Altamaha River system, and other fishes may be important hosts. Other investigators have previously reported that A. arcula metamorphose on Gambusia holbrooki Girard (Eastern Mosquitofish; O’Brien 2002b). Similar to other mussel species, identification of host(s) for A. arcula is imperative for successful conservation; therefore, efforts to identify additional hosts should continue. O’Brien (2002b) reported the only other published host information for Altamaha mussel species. In that study, juvenile E. hopetonensis were produced from Eastern Mosquitofish and Bluegill, but not from Largemouth Bass. In the present study, juvenile E. hopetonensis were produced by Bluegill, Largemouth Bass, and Fathead Minnows. O’Brien (2002b) reported that L. dolabraeformis juveniles were produced from Eastern Mosquitofish and Largemouth Bass, but not from Bluegill. In the present study, juvenile L. dolabraeformis were produced on Largemouth Bass and Bluegill to a lesser extent. One of the objectives for this study was to identify a host fish for the federally endangered E. spinosa. Unfortunately, suitable hosts for E. spinosa remain unknown at this time. The major limiting factor in our trials was the inability to collect gravid females. In 2009 and 2010, record high flows in the Altamaha limited search efforts in April–June, the period when gravid E. spinosa had been collected by other investigators (P. Johnson, Alabama Department of Conservation and Natural Resources Alabama Aquatic Biodiversity Center, Marion, AL, pers. comm.). Future host trials with E. spinosa will depend upon a source of gravid females. One option is to collect E. spinosa throughout the year and relocate them to a centralized location where the chances of recapture are greater. Another option is to attach external sonic tags or passive integrated transponder (PIT) tags to the mussels to enhance the chances of recovery during periods of gravidity. Additionally, E. spinosa may be collected throughout the year and returned to a culture facility to determine if they will undergo fertilization and brooding in captivity. A number of factors can influence metamorphosis success in host trials, including those in the present study. Mussels may have higher metamorphosis rates on fish with which they co-occur than fish of the same species from other basins (Strayer 2008). Metamorphosis success also appears to be higher on smaller and younger fish than older and larger fish of the same species even if the larger fish 2012 J.A. Johnson, J.M. Wisniewski, A.K. Fritts, and R.B. Bringolf 743 have never had prior glochidia exposure (Dodd et al. 2006). Further, metamorphosis rates may be lower in wild fish collected from streams where mussels occur because fish may develop acquired immunity to glochidia (Dodd et al. 2005). In the present study, not all fish species we tested occur in the Altamaha basin (e.g., Lake Sturgeon), or were native to the basin (e.g., Fathead Minnow), but we attempted to maximize the number of species tested for potential propagation of imperiled species. For studies with the exclusive goal of identification of natural hosts, we recommend collecting small (young) fish that co-occur in the basin where mussels are collected. Host studies frequently report only numbers of juveniles produced per fish species, not the metamorphosis rate on an individual fish or for a fish species. Numbers of juveniles produced provides a qualitative assessment of hosts, but evaluation of %M is essential for quantitative assessment and defensible classification (e.g., “primary”, “secondary”, etc.) of suitability of host species. We recommend that future host studies report %M for each fish species, and for individual fish when possible, to provide a quantitative assessment of the relative potential contribution for metamorphosis on a particular species of fish. Reporting of infestation rates provides additional valuable information that, along with %M, would provide a holistic view of relative host suitability . We provided some of the first detailed descriptions for glochidia of mussels from the Altamaha River basin. Shell measurements were similar among some species we examined, but no species were similar in all 3 dimensions. A qualitative assessment of shell measures among species, comparison of overlapping 95% confidence intervals, was generally useful for species identification; overlap of height and length among species was minimal. The DFA was successful for distinguishing the correct species in >88% of the cases. The most common error (10 of 17 total) was E. hopetonensis misidentified as E. spinosa. Elliptio shepardiana, the only other member of the genus Elliptio in the present study, was larger in all 3 shell dimensions than the other 2 members of the genus and was readily distinguished by DFA from all other taxa (100% correct identification). Lampsilis dolabraeformis and L. splendida were correctly identified in 98% (49 of 50) of cases and were distinguished by the shortest hinge lengths of all taxa examined. With DFA, glochidia from all 6 of the taxa tested here could be placed in groups of ≤2 taxa. Thus, consistent with previous findings (Kennedy and Haag 2005), glochidia morphology may be useful for identifying glochidia, such as those attached to wild fish. However, some glochidia may remain indistinguishable by analysis of morphometric data, and use of molecular techniques such as DNA barcoding may allow positive identification (Boyer et al. 2011). Although molecular technology is rapidly evolving, analysis of morphometrics is currently more readily accessible and economical than genetic approaches. Further efforts should be made to describe glochidia morphology of the other species in the Altamaha Basin, as morphology may also be a potential method to examine problematic taxonomic relationships of mussels within the Altamaha Basin and other basins in the South Atlantic Slope. We also recommend that future studies seek to more completely describe the variability in glochidia shell measures from individual gravid females of a given species. 744 Southeastern Naturalist Vol. 11, No. 4 Dimensions and features of A. arcula glochidia were consistent with other members of the genus Alasmidonta (Ortmann 1911). Species in this genus exhibit a styliform hooked shell and tend be larger in size than hookless glochidia (Barnhart et al. 2008, Williams et al. 2008). Hooked glochidia often attach to fins and body surfaces of hosts whereas hookless glochidia generally attach to gills (Barnhart et al. 2008). Measurements and morphology of E. hopetonensis and E. shepardiana were similar to those previously reported for these species by O’Brien et al. (2003), while morphology of viable E. spinosa glochidia had not been reported prior to this study. The size and shape of E. spinosa are generally similar to those previously reported for other members of the genus Elliptio (O’Brien et al. 2003, Williams et al. 2008); however, the ventral flanges of E. spinonsa and E. shepardiana were distinctly different in length, which may be helpful when trying to distinguish between these species. Many physical characteristics of L. dolabraeformis and L. splendida glochidia were similar to other species in the genus Lampsilis. For example, L. dolabraeformis and L. splendida were comparable to Hamiota subangulata Lea (Shinyrayed Pocketbook) glochidia, which had an average height of 261 ± 7 μm (O’Brien and Brim-Box 1999). However, L. straminea Conrad (Rough Fatmucket), L. ornata Conrad (Southern Pocketbook), and L. teres Rafinesque (Yellow Sandshell) all were markedly smaller than both L. dolabraeformis and L. splendida (Kennedy and Haag 2005). When feasible, future analyses of glochidia morphology should be conducted using a scanning electron microscope, which can provide precise measurements and detailed images of shell ultrastructure that are not feasible with traditional light microscopy techniques. Additional studies on the reproductive biology and early life history of declining species in the Altamaha River are warranted. Efforts to protect imperiled species will be greatly enhanced by knowledge of spawning and brooding periods, optimal brooding temperatures, host fish, and descriptions of glochidia morphology. Knowledge of mussel early life history may provide insight into causes of specific mussel population declines (e.g., loss of fish host or preferred habitat) and may be used for development of propagation programs for population augmentation or reintroduction for restoration and preservation of freshwater mussel biodiversity. Acknowledgments Funding for this project was provided by the Georgia Department of Natural Resources, Wildlife Resources Division Nongame Conservation Section. Additional funds were provided by the Altamaha River Cooperative and the United States Fish and Wildlife Service. We are indebted to many people who provided assistance in the laboratory and field including Dr. Chris Barnhart, Kaitlin Brotman, Mieko Camp, Derek Colbert, Julie Creamer, Peter Hazelton, Jimmy Rickard, Colin Shea, Amos Tuck, and Deb Weiler. Bob Ratajczak provided valuable assistance with statistical analyses in addition to laboratory support. Dr. John Shields in the University of Georgia Center for Advanced Ultrastructural Research provided invaluable assistance with scanning electron microscopy. We are also grateful to Brian Simmons and the staff at Owen and Williams Fish Farm for donating Largemouth Bass for host trials. We thank two anonymous reviewers for their constructive comments to improve the manuscript. 2012 J.A. Johnson, J.M. Wisniewski, A.K. Fritts, and R.B. Bringolf 745 Literature Cited Barnhart, M., W. Haag, and W. Roston. 2008. Adaptations to host infection and larval parasitism in Unionoida. Journal of the North American Benthological Society 27(2):370–394. Boyer, S.L., A.A. Howe, N.W. Juergens, and M.C. Hove. 2011. A DNA-barcoding approach to identifying juvenile freshwater mussels (Bivalvia: Unionidae) recovered from naturally infested fishes. Freshwater Science 30(1):182–194 . Dinkins, G.R., J.R. Dinkins, J.E. Daniel. 2004. Survey for native mussels with a focus on Altamaha Spinymussel (Elliptio spinosa) and Altamaha Arc-mussel (Alasmidonta arcula) in approximately 15 km of lower Ocmulgee River, Coffee/Telfair/Jeff Davis counties, Georgia. The Nature Conservancy, Darien, GA. Dodd, B., M.C. Barnhart, C.L. Rogers-Lowery, T. Fobian, and R. Dimock. 2005. Crossresistance of Largemouth Bass to glochidia of unionid mussels. Journal of Parasitology 91:1064–1072. Dodd, B., M. Barnhart, C. Rogers-Lowery, T. Fobian, and R. Dimock. 2006. Persistence of host response against glochidia larvae in Micropterus salmoides. Fish and Shellfish Immunology 21(5):473–484. Fritts, A.K., M.W. Fritts II, D.L. Peterson, D.A. Fox, and R.B. Bringolf. 2012. Critical linkage of imperiled species: Gulf Sturgeon as host for Purple Bankclimber mussels. Freshwater Science 31(4):122–1232. Grabowski, T.B., and C.A. Jennings. 2009. Post-release movements and habitat use of Robust Redhorse transplanted to the Ocmulgee River, Georgia. Aquatic Conservation- Marine and Freshwater Ecosystems 19(2):170–177. Hoggarth, M.A. 1999. Descriptions of some of the glochidia of the unionidae (Mollusca : Bivalvia). Malacologia 41(1):1–118. Hornbach, D., V.J. Kurth, and M.C. Hove. 2010. Variation in freshwater mussel shell sculpture and shape along a river gradient. American Midland Naturalist 164(1):22–36. Johnson, R.I. 1970. The systematics and zoogeography of the Unionidae (Mollusca: Bivalvia) of the southern Atlantic Slope Region. Bulletin of the Museum of Comparative Zoology 140:263–449. Keferl, E.P. 1981. A survey of the naiads of the Ohoopee River, Georgia. The Bulletin of the American Malacological Union 11–15. Kennedy, T.B., and W.R. Haag. 2005. Using morphometrics to identify glochidia from a diverse freshwater mussel community. Journal of the North American Benthological Society 24(4):880–889. Meyers, T.R., R.E. Millemann, and C.A. Fustish. 1980. Glochidiosis of salmonio fishes. IV. Humoral and tissue responses of Coho and Chinook Salmon to experimental infection with Margaritifera margaritifera (L.) (Pelecypoda: Margaritanidae). The Journal of Parasitology 66(2):274–281. Morrison, D.F. 1976. Multivariate Statistical Methods. McGraw Hill, New York, NY. Neves, R., A. Bogan, J. Williams, S. Ahlstedt, and P. Hartfield. 1997. Status of aquatic mollusks in the southeastern United States: A downward spiral of diversity. Southeast Aquatic Research Institute, Aquatic Fauna in Peril: The Southeastern Perspective, Special Publication 1:43–85. O’Brien, C. 2002a. A survey of the Altamaha Spinymussel (Elliptio spinosa) and Altamaha Arcmussel (Alasmidonta arcula) in the Altamaha and Ocmulgee Rivers, Georgia. Final Report to the Georgia Department of Natural Resources, Social Circle, GA. 746 Southeastern Naturalist Vol. 11, No. 4 O’Brien, C. 2002b. Host identification for three freshwater mussel species endemic to the Altamaha River, Georgia. Ellipsaria 4(1):17. O’Brien, C., and J. Brim-Box. 1999. Reproductive biology and juvenile recruitment of the Shinyrayed Pocketbook, Lampsilis subangulata (Bivalvia: Unionidae) in the Gulf Coastal Plain. American Midland Naturalist 142(1):129–140. O’Brien, C., J. Williams, and M. Hoggarth. 2003. Morphological variation in glochidia shells of six species of Elliptio from Gulf of Mexico and Atlantic Coast drainages in the southeastern United States. Proceedings of the Biological Society of Washington 116(3):719–731. Ortmann, A.E. 1911. A monograph of the naiades of Pennsylvania, Parts I and II. Memoirs of the Carnegie Museum 4(6):279–347, 4 plates. Skelton, C.E., S. Cammack, and E. VanDeGenachte. 2002. Survey for Elliptio spinosa (Altamaha Spinymussel) in the lower Ocmulgee River. Georgia National Heritage Program:16. Spooner, D.E., and C.C. Vaughn. 2006. Context-dependent effects of freshwater mussels on stream benthic communities. Freshwater Biology 51:1016–1024. Strayer, D.L., J.A. Downing, W.R. Haag, T.L. King, J.B. Layzer, T.J. Newton, and S.J. Nichols. 2004. Changing perspectives on pearly mussels, North America’s most imperiled animals. BioScience 54:429–439. Strayer, D.L. 2008. Freshwater Mussel Ecology: A Multifactor Approach to Distribution and Abundance. University of California Press, Los Angeles, CA. Williams, J., M. Warren, Jr,, K. Cummings, J. Harris, and R. Neves. 1993. Conservation status of freshwater mussels of the United States and Canada. F isheries 18(9):6–22. Williams, J.D., A.E. Bogan, and J.T. Garner. 2008. Freshwater Mussels of Alabama and the Mobile Basin in Georgia, Mississippi, and Tennessee. University of Alabama Press, Tuscaloosa, AL. Wisniewski, J.M., G. Krakow, B. Albanese. 2005. Current status of endemic mussels in the lower Ocmulgee and Altamaha rivers. In K.J .Hatcher (Ed.). Proceedings of the 2005 Georgia Water Resources Conference. Institute of Ecology, Athens, GA. Available online at http://www.gwri.gatech.edu/uploads/proceedings/2005/WisniewskiJGWRCpaper. pdf. Zale, A., and R. Neves. 1982. Fish hosts of four species of lampsiline mussels (Mollusca: Unionidae) in Big Moccasin Creek, Virginia. Canadian Journal of Zoology 60(11):2535–2542.