2013 SOUTHEASTERN NATURALIST 12(2):251–266 Observations on the Identification of Larval and Juvenile Scaphirhynchus spp. in the Lower Mississippi River Paul Hartfield1,*, Nathan M. Kuntz2, and Harold L. Schramm, Jr.3 Abstract - Scaphirhynchus albus (Pallid Sturgeon) and S. platorynchus (Shovelnose Sturgeon) are sympatric and not uncommon in the lower Mississippi River from the confluence of the Ohio River to the Gulf of Mexico, and in its distributary, the Atchafalaya River. Reports of sturgeon larvae have been rare in the Mississippi River but have been increasing with more effective collection methods. A suite of characters identified in hatchery-reared larval Pallid Sturgeon and Shovelnose Sturgeon from the Yellowstone and upper Missouri rivers has been used to distinguish larval Scaphirhynchus spp. In the Mississippi River; however, a large proportion of wild Scaphirhynchus spp. larvae are intermediate in these characters and have been identified by some as hybridized Pallid Sturgeon and Shovelnose Sturgeon. We applied three diagnostic characters developed from Missouri River sturgeon larvae to hatchery-reared progeny of Atchafalaya River Pallid Sturgeon and found them inadequate to identify most of the known Pallid sturgeon larvae. Additionally, fewer than 10% of a large sample of wild Scaphirhynchus spp. larvae from the lower Mississippi River conformed to either Pallid Sturgeon or Shovelnose Sturgeon at two or more of the characters. We also found a small mouth width relative to head width and a concave forward barbel position may be useful for the identification of 30% or more Scaphirhynchus spp. larvae and postlarval young-of-year as Shovelnose Sturgeon. Established adult character indices and diagnostic measurement proportionalities also failed to correctly identify any hatchery-reared Pallid Sturgeon juveniles recaptured 6–7 years following their release. Introduction Historical records of larval and small (less than ca. 300 mm fork length [FL]) Scaphirhynchus spp. from the Mississippi River are rare, as are records of juvenile (<700 mm) Scaphirhynchus albus (Forbes and Richardson) (Pallid Sturgeon [PLS]). This rarity was initially interpreted as recruitment failure, particularly in PLS (USFWS 1990). Methods have been refined during the past decade to collect larval and young-of-year (YOY) Scaphirhynchus spp. (e.g., Herzog et al. 2005), and the numbers of larval Scaphirhynchus spp. collected in the Mississippi River are increasing (Hrabik et al. 2007, Phelps et al. 2010). Identification of Scaphirhynchus spp. <300 mm FL has proven difficult (Bailey and Cross 1954, Kuhajda and Mayden 2001), and it is further complicated by a protracted larval development period and three larval stages. Protolarva is the 1US Fish and Wildlife Service, 6578 Dogwood View Parkway, Jackson, MS 39213. 2Department of Wildlife, Fisheries, and Aquaculture, Mississippi State University, Mississippi State, MS 39762. 3US Geological Survey, Mississippi Cooperative Fish and Wildlife Research Unit, Mail Stop 9691, Mississippi State, MS 39762. *Corresponding author - paul_hartfield@fws.gov. 252 Southeastern Naturalist Vol. 12, No. 2 early life stage from hatching to development of first median fin rays at 21–26 mm total length (TL), mesolarva is the morphological stage from the appearance of first median fin rays through the appearance of last caudal fin rays at 50–60 mm TL in Scaphirhynchus platorynchus (Rafinesque) (Shovelnose Sturgeon [SNS]) and greater than 81 mm TL in PLS, and metalarva is the stage from the appearance of the last caudal fin rays through complete disappearance of the pre-anal finfold at about 200 mm TL (Snyder 2002). Snyder (2002) conducted a rigorous morphometric and meristic study of hatchery-reared (HR) PLS and SNS larvae that were progeny of brood stock collected from the Yellowstone River, MT, and upper Missouri River, ND, and identified several characters diagnostic for differentiating PLS and SNS larvae at different stages. These included: (1) pigmentation of tissues over the heart and posterior lip lobe of protolarvae and mesolarvae (SNS with pigmentation, PLS without); (2) inner barbel to outer barbel length ratios of protolarvae through metalarvae (SNS < 1.4, PLS > 1.6); (3) the ventral shape of the lateral rostral plate of mesolarvae and metalarvae (SNS semi-circular, PLS deltoid); and (4) caudal fin ray counts of metalarvae (SNS < 65, PLS > 66). Larval descriptions from Snyder (2002) have been used to identify sturgeon larvae in the Mississippi River (e.g., Hrabik et al. 2007); however, a number of Scaphirhynchus spp. larvae had characteristics of both species and were identified as hybrid Scaphirhynchus spp. Using characters 1, 2, and 3 proposed by Snyder (2002), above, as well as secondary characters (dorsal and anal fin ray counts), Hrabik et al. (2007) identified 17 of 39 Scaphirhynchus spp. larvae (17.5–84 mm TL) collected from the middle and lower Mississippi River as hybrids. These findings appear to support reports of high proportions of hybrid Scaphirhynchus spp. relative to total catch in this river reach (e.g., 86.4% of Scaphirhynchus spp. reported as hybrids by Keenlyne et al. 1994). Identification of larger Scaphirhynchus spp. specimens (i.e., postlarval YOY, juveniles, and adult) is based upon continuous anatomical ratios rather than discrete meristic measurements. Diagnostic proportionalities for species identification were established from anatomical ratios of only a few juvenile and mature sturgeon specimens, most of which were collected in the Missouri River (Bailey and Cross 1954), and character indices were developed from these proportionalities (e.g., Wills et al. 2002). However, these anatomical ratios appear to change with fish size in the lower Mississippi and Atchafalaya rivers, possibly resulting in reports of disproportionate numbers of hybrids (USFWS 2007). Concern over misidentification of Scaphirhynchus spp. in the Mississippi River as hybrids is supported by Murphy et al. (2007), who compared the identification of 41 lower Mississippi River SNS and PLS using principal component analysis (PCA) with a character index relying on anatomical ratios (mCI) and a character index using both anatomical ratios and a meristic character (CI). The PCA resulted in the identification of only two morphological intermediates, i.e., putative hybrids; whereas, the CI and mCI identified 37% and 73%, respectively, of the specimens as hybrids (Murphy et al. 2007). Knowledge of whether Scaphirhynchus spp. 2013 P. Hartfield, N.M. Kuntz, and H.L. Schramm, Jr. 253 larvae can be distinguished in the lower 1931 km of their range, and what proportion can be reliably identified, is essential to measuring the recruitment rates and demographics of both PLS and SNS, as well as the extent of possible hybridization in the Mississippi River. Larval sturgeon characters, as well as the character indices used to identify Scaphirhynchus spp. >300 mm FL, were developed primarily from upper Missouri River specimens. Our initial objective was to evaluate the utility of the Snyder (2002) characters 1, 2, and 3 (above) for identifying HR PLS larvae and wild Scaphirhynchus spp. larvae in the extreme southern portion of their range. We also evaluated larval mouth width relative to head width and barbel placement, two characters commonly used to identify large (>300 mm FL) Scaphirhynchus spp. (Bailey and Cross 1954, Forbes and Richardson 1905). Following the recapture of age 6–7 PLS HR progeny in 2010 through 2012, we tested the effectiveness of anatomical ratios and character indices for identification of these larger juvenile fish. Methods Designation of life stages The larval stage length ranges of Scaphirhynchus spp. used in our analysis were modified from Snyder (2002) to accommodate our comparison of known HR PLS to wild caught specimens potentially containing individuals of both species. The smallest specimen we examined was 31 mm, larger than the protolarval stage of either species (27 mm) defined by Snyder (2002). We considered specimens up to 80 mm TL as mesolarvae, and those between 80 to 200 mm TL as metalarvae. For purposes of discussion and clarity, we also define post-larval lower Mississippi River PLS. The age-0 postlarval life stage (200 to 400 mm) begins upon the loss of the pre-anal finfold and the development of the last caudal fin rays (Snyder 2002); the upper length limit for the first year of growth is derived from hatchery records (Dean 1998, 2004) and a growth curve developed by Killgore et al. (2007). The juvenile life stage (>400 to 750 mm FL) begins at age 1 and includes individuals of both sexes showing early gonadal development; the maximum length for this stage is based on the growth curve (Killgore et al. 2007) and minimum age of male sexual maturity (7 years), as well as the lower sizes at which Mississippi and Atchafalaya PLS begin to conform to diagnostic proportionalities and character indices (P. Hartfield, N.M. Kuntz, and H.L. Schramm, Jr., pers. observ.). The mature adult life stage (>750 mm FL) is based upon the minimum age of female sexual maturity (10 years) and successful spawning of PLS >800 mm FL at Natchitoches National Fish Hatchery (Dean 1998, 2004). Age-0 hatchery-reared Pallid Sturgeon A series of HR larval and post larval PLS (33 mm TL–260 mm FL) were obtained from the University of Alabama Ichthyology Collection (UAIC), and 100 254 Southeastern Naturalist Vol. 12, No. 2 specimens were randomly selected for measurement and analysis (UAIC Lot #s 15246, 15247, 15249, 15254–15265, 15648, 15649, 15652). These specimens were primarily progeny of brood stock collected in the Atchafalaya River at its divergence from the Mississippi River in 2004 (n = 95), along with a smaller number (n = 5) from brood stock collected in 1998. All specimens for the analysis were spawned and reared at Natchitoches National Fish Hatchery during the years of brood stock collection. Brood stock for the 2004 HR PLS included three females and four males visually identified as PLS, which were confirmed by character index (Kuhajda and Mayden 2001) and genetic analysis (Dean 2004). The water temperature was maintained at 19 °C throughout incubation, and eggs began hatching on 11 May 2004. Approximately 14,000 fingerlings were produced in 2004 (Dean 2004). Larvae were randomly subsampled at 10 to 71 days post hatch, fixed, and preserved in 10% buffered formalin, and deposited in the University of Alabama Ichthyology Collection. Surviving HR age-0 fish were stocked into the Atchafalaya (n = 965) and lower Mississippi rivers (n = 2600) in October 2004. The 1998 HR PLS were progeny from a single pair of visually identified (i.e., morphological) PLS. Water temperature was held at 17.2 °C throughout incubation, and eggs hatched 23–25 March 1998. On 26 March 1998, 8030 fry were transferred to rearing troughs with flow-through water from a local stream. Rearing water temperatures were not reported. On 14–15 October 1998, 35 surviving age-0 fish were PIT tagged and released into the Atchafalaya River in Concordia Parish, LA (Dean 1998). Wild-caught Age-0 Scaphirhynchus spp. As no hatchery has produced SNS from the lower portion of the range, HR SNS were not available for this analysis. Wild Scaphirhynchus spp. larvae and age-0 postlarvae (<300 mm TL) were collected during 2008–2010 from various locations in the lower Mississippi River between Rosedale (rkm 935) and Vicksburg (rkm 698), MS, and deposited at the Mississippi Museum of Natural Science (MMNS Registration # 9868). A 3.05-m otter trawl with 3.18-mm bar mesh (Innovative Net Systems, Lafayette, LA) was used during May and June, and a 3.66-m otter trawl with 31.75-mm bar mesh and a 3.18-mm bar mesh cod liner (Memphis Net and Twine Company, Memphis, TN) was used during other months. Specimens were fixed in 10% buffered formalin, then rinsed and transferred to 70% denatured ethanol after approximately 2 weeks of fixation. We randomly selected 94 meso-, meta-, and post-larval YOY wild sturgeon (31 mm TL–301 mm FL) for examination and measurement. An additional 6 specimens cataloged in the MMNS Fish Collection (MMNS Cat # 43530–43533, 43535, 53694) were also included in the analysis. Age-0 specimen analysis All specimens were measured (mm) for TL (larvae) or FL (postlarvae) using a 30-cm ruler and assigned to a developmental stage as defined above. We recorded 2013 P. Hartfield, N.M. Kuntz, and H.L. Schramm, Jr. 255 (1) the presence or absence of pigmentation of the integument ventral to the heart and on the posterior lip lobes, (2) the lengths of inner and outer barbels, (3) the head width, (4) the mouth width, (5) the shape of the lateral rostral plate, and (6) the relative position of the inner and outer barbels. All measurements other than TL and FL were measured to the nearest 0.01 mm using digital calipers (Traceable Control Co., Friendswood, TX) and a 1.25x illuminated magnifier. We did not attempt to use fin ray counts (defined as secondary characters by Hrabik et al. 2007) because of discrepancies in counts between the original description (Forbes and Richardson 1905) and the re-description (Bailey and Cross 1954), the variability of counts due to developmental water temperature (Snyder 2002), and the difficulty of accurate fin ray counts even in larger specimens (Bailey and Cross 1954). Snyder (2002) considered an outer barbel-inner barbel ratio (OB/IB) <1.4 diagnostic for SNS, and >1.6 diagnostic for PLS for mesolarvae and larger specimens; however, ratios of 1.5 from meso- and metalarvae could be indicative of either species. For the purposes of our study, we selected a measurement ratio of <1.45 as characteristic of SNS and a ratio >1.55 as characteristic of PLS. The most intact pair of left or right barbels was selected for meas urement. The head width/mouth width ratio (HW/MW) was recognized by Forbes and Richardson (1905) as a diagnostic character for distinguishing adult PLS (1.4–1.6) and SNS (1.6–1.9), but, to our knowledge, this ratio has not been evaluated for larval or postlarval Scaphirhynchus spp. Mouth width is typically measured as the distance between the outside of the lips; however, lips may be distorted by fixation and preservation. To avoid this problem, we defined mouth width as the inside width of the cartilaginous border at the posterior of the buccal cavity; head width was the widest measurement of head. In our analysis, we compared mouth width relative to head width (MW/HW), the reciprocal of the ratio used by Forbes and Richardson, to reflect a higher value as mouth width increases relative to head width. Using this ratio, the Forbes and Richardson (1905) diagnostic HW/MW ratios for adult specimens would correspond to MW/ HW ratios ≥0.63 for PLS, and <0.63 for SNS. Snyder (2002) identified the shape of the lateral rostral plate in ventral aspect as diagnostic for meso- and metalarvae of PLS and SNS. He described the more angular lateral rostral plate of PLS larvae as “deltoid” and widest near the barbel; whereas, in SNS, the semicircular plates are widest near the middle. To compare quantitatively the shape of the lateral rostral plate, we measured total length of the plate and distance from the anterior margin to its widest point. Shape of the rostral plate was quantified as length at the widest point divided by its total length. Measurement ratios <0.30 were considered deltoid, and ratios >0.40 were considered semicircular. Bailey and Cross (1954) noted that the origin of the four barbels typically lie in a straight line for SNS, occasionally with the bases of the outer pair originating slightly forward of the inner pair (i.e., straight or concave forward); whereas, in PLS the outer barbels originate posterior to the inner barbels (i.e., convex 256 Southeastern Naturalist Vol. 12, No. 2 forward). Snyder (2002) described barbel position only for prot olarvae and only relative to other structures. We determined barbel position by placing a straightedge along the anterior base of the barbels. Barbel position was recorded as convex forward, straight, or concave forward. A chi-square analysis was used to test whether morphological characteristics changed with larval stage. Exact tests were used to accommodate the low cell size for some comparisons, and HR age-0 postlarvae were excluded from analysis because of the low sample size. Juvenile and adult Pallid Sturgeon During 2010 through 2011, 8 HR PLS (500–674 mm FL) were captured on trotlines within the combined outflow channel of the Old River Control Complex (Atchafalaya River) or just downstream (Red River Miles 7–10). Coded wire tags (CWT) extracted from two of the individuals confirmed that they were from the same family lots as our larval specimens, released as age-0 postlarvae into the lower Mississippi and Atchafalaya rivers in 2004. Over the past 2 years, we have also captured, measured, and released 7 HR PLS (625–741 mm FL) with CWT and/or elastomer marks in their rostrums from the Mississippi River near Baton Rouge, LA (Mississippi River Mile [MRM] 234. 246), Vicksburg, MS (MRM 431, 454, 459), and Rosedale, MS (MRM 580). We applied diagnostic proportionalities (Bailey and Cross 1954) and two character indices currently in use for identifying PLS and SNS to these specimens (Table 1). The mCI (Wills et al. 2002) was developed to distinguish PLS, SNS, and presumptive hybrids in the middle Mississippi and lower Missouri rivers; the BKI (Kuhajda and Mayden 2001) was developed to conservatively select brood stock for pr opagation in the lower Mississippi and Atchafalaya rivers and identifies morphological intermediates as “intermediate”, rather than as “hybrid”. Measurements used in the proportionalities and character indices include: mouth width (MW), the distance from the center of the mouth to the front edge of the right inner barbel insertion Table 1. Diagnostic proportionalities from Bailey and Cross (1954) and character index values (mCI from Wills et al. 2002, BKI from Kuhajda and Mayden 2001) used to identify Scaphirhynchus spp. throughout their range. Measurements used in the proportionalities and indices are defined in the methods. Pallid Sturgeon Intermediate or hybrid Shovelnose Sturgeon Measurement proportionalities MW:MIB >1.63 1.42–1.63 <1.42 IRL:MIB >2.29 2.19–2.29 <2.19 HL:MIB >5.54 5.04–5.54 <5.04 OBL:IBL >1.72 1.48–1.72 <1.48 IRL:IBL >2.63 2.50–2.63 <2.50 HL:IBL >6.35 5.76–6.35 <5.76 Character indices mCI <-0.70 -0.70–0.41 >0.41 BKI >4.70 3.60–4.70 <3.60 2013 P. Hartfield, N.M. Kuntz, and H.L. Schramm, Jr. 257 (MIB), the distance from the tip of the snout to the front edge of the right outer barbel insertion (IRL), head length as measured from the tip of the snout to the rear margin of the right opercle (HL), outer barbel length (OBL), and inner barbel length (IBL) (USFWS 1990). Results Characteristics of Age-0 hatchery-reared Pallid Sturgeon Preserved HR PLS ranged from 33 mm TL–260 mm FL and included 36 mesolarvae (<80 mm TL), 62 metalarvae (80–200 mm TL), and 2 age-0 postlarvae (206 and 260 mm FL). Only 24 HR PLS specimens conformed to Snyder’s (2002) three characters as PLS. Barbel length ratios, shape of lateral rostral plate, or both either conformed to Shovelnose Sturgeon or were intermediate in value in the other 76 specimens. No pigmentation was evident in the middle of the posterior lip lobes of any of the HR PLS larvae. A single mesolarva (49 mm TL) exhibited a few small, scattered specks of pigment over the heart. Few to moderate numbers of small specks of pigmentation were observed on the outer mar gin of the lips near the posterior corners of the mouth of 39 of the specimens (Table 2). These included 18 mesolarvae and 21 metalarvae (up to 155 mm TL). Pigmentation was not observed on the lips of any HR metalarvae >155 mm TL or the 2 age-0 postlarvae. The general lack of pigmentation in HR PLS larvae conformed to Snyder’s (2002) description; however, almost 40% of specimens exhibited light to moderate pigmentation on the outer margin of the lip lobes near the posterior corners of the mouth, a character Snyder reported as an occasional occurrence in either species. Hatchery-reared PLS OB/IB ratios ranged from 1.16–1.96. Barbel ratios for 36 of the HR specimens fell within the range for SNS (<1.45), 29 had intermediate values (1.45–1.54), and 35 had ratios >1.55 that are considered diagnostic of PLS (Table 2). Therefore, 65 of the known HR PLS did not conform to the expected barbel ratio value. Relative shape of the lateral rostral plate in HR PLS varied from 0.16 (deltoid, PLS condition) to 0.49 (semicircular, SNS condition). Most (n = 72) were deltoid (<0.30), but 1 mesolarva and 4 metalarvae exhibited a strongly semicircular shape (>0.40) (Table 2). Eight specimens were marginal to our defining ratio (0.30, 0.31), and 5 of these conformed to PLS in barbel ratios. Only 39 of the 100 HR PLS larvae and postlarvae had the convex forward barbel position indicative of PLS; all others were straight, and none were concave forward (Table 2). Convex-forward specimens ranged from 33 mm TL–260 mm FL, included fish in all stages of development, and 12 conformed to PLS in all measurements. Specimens with straight barbel position ranged from 42 mm TL–206 mm FL and included 61 fish in all stages of development. Twenty of these conformed to PLS in all other measurements. The MW/HW ratios for HR PLS ranged from 0.49 to 0.65, and only a single HR PLS had a MW/HW ratio < 0.5 (0.49). However, only 8 mesolarvae and 2 258 Southeastern Naturalist Vol. 12, No. 2 Table 2. Ventral head measurements, barbel position, and pigmentation of hatchery-reared (HR) Pallid Sturgeon and wild Scaphirhynchus spp. larvae and postlarvae from the lower Mississippi River. Pigmentation Barbel ratio Barbel position Rostral plate Mouth width:head width None Heart Lips <1.45 1.45–1.55 >1.55 Concave Straight Convex <0.3 0.3–0.4 >0.4 <0.49 0.49–0.51 >0.51 # HR 61 1 39 36 29 35 0 61 39 72 23 5 0 4 95 -Meso 18 1 18 10 12 14 0 21 15 28 7 1 0 1 34 -Meta 41 0 21 26 17 19 0 39 23 42 16 4 0 3 59 -Post 2 0 0 0 0 2 0 1 1 2 0 0 0 0 2 Chi-square 4.14 1.972 0.200 1.316 0.222 df 2 2 1 2 1 Pexact 0.089 0.373 0.670 0.562 1.000 # Wild 84 10 0 62 22 14 25 62 13 42 50 6 29 33 37 -Meso 16 10 0 10 12 10 1 22 9 18 12 2 3 8 21 -Meta 38 0 0 26 9 3 14 23 1 15 20 2 15 14 9 -Post 30 0 0 26 1 1 9 17 3 9 18 2 11 11 7 Chi-square 14.930 26.702 18.060 4.299 17.507 df 2 4 4 4 4 Pexact <0.001 <0.001 <0.001 0.375 <0.001 2013 P. Hartfield, N.M. Kuntz, and H.L. Schramm, Jr. 259 metalarvae had a ratio > 0.63 considered diagnostic of large PLS by Forbes and Richardson (1905). Developmental stage did not appear to be a factor in identification of HR PLS as there were no differences between meso- and metalarval stage in any characters measured (P > 0.089; Table 2). Thus, although SNS and intermediate characters were found in some HR PLS, the frequency of occurrence remained the same between larval stages. Characteristics of wild-caught Age-0 Scaphirhynchus spp. Wild-caught Scaphirhynchus spp. larvae and postlarvae ranged from 31 mm TL to 301 mm FL and included 32 mesolarvae, 39 metalarvae, and 29 postlarvae. Excluding pigmentation, 7 specimens (1 mesolarvae, 3 metalarvae, 3 postlarvae) conformed to SNS in barbel ratio and lateral rostral plate, while 9 conformed to PLS (7 mesolarvae, 1 metalarva, 1 postlarva). Ventral pigmentation over the heart was evident in only 10 of the wild Scaphirhynchus spp. specimens, and only in mesolarvae 38–70 mm TL. All of these were intermediate in barbel ratio and/or lateral rostral plate. Pigmentation in most of these specimens consisted of a few small specks, but 1 specimen had a distinct patch of pigmentation over the heart conforming to Snyder’s (2002) description of SNS. No pigmentation was observed on the lips of any of the wild specimens. Barbel ratios ranged from 1.2–1.7, with 14 greater than 1.55 and 22 with intermediate values (Table 2). High, low, and intermediate barbel ratios were found for all developmental stages. Shape of the lateral rostral plate ranged from 0.19 to 0.45, similar to the HR specimens (Table 2). Almost half of the wild specimens exhibited a deltoid lateral rostral plate, and only 6 exhibited a strongly semicircular sha pe (>0.40). Thirteen specimens had convex forward barbel placement, 62 were straight, and 25 were concave forward (Table 2). Most of the specimens with concave forward barbels were metalarvae and postlarvae; only 1 mesolarva had concave forward barbel placement. Three of the 7 specimens identified as SNS, and 1 of the 9 identified as PLS by barbel ratio and lateral rostral plate exhibited concave forward barbel placement, while 1 specimen identified as SNS and 4 identified as PLS exhibited convex forward barbel placement. The MW/HW ratios ranged from 0.43 to 0.64, and 29 individuals had MW/ HW ratios of <0.50 (Table 2). Of those 29, almost half had concave barbels and none exhibited convex barbels. Specimens of all developmental stages were represented in the 70 wild sturgeon with MW/HW ratio > 0.49. Only 2 mesolarvae had MW/HW > 0.63. For wild Scaphirhynchus spp., barbel ratio, barbel position, MW/HW ratio, and pigmentation all differed between larval stages (P < 0.001; Table 2). All 4 characters indicated a reduced frequency in PLS characters and an increase in SNS characters as fish advanced in development stage. 260 Southeastern Naturalist Vol. 12, No. 2 Characteristics of juvenile and adult Pallid Sturgeon All 8 HR PLS (500–674 mm FL) captured in the Atchafalaya River 7 years after release were visually identified by field crews as PLS and confirmed to be progeny from the 2004 Atchafalaya hatchery release by the presence of coded Table 4. Diagnostic proportionalities from Bailey and Cross (1954) and character indices applied to 7 Pallid Sturgeon from the Atchafalaya River used as brood stock for propagation. Identification of all measurements and their specific diagnostic value can be found in Table 1. Pallid Sturgeon characters are in bold, Shovelnose Sturgeon characters are in italics, all others are considered intermediate. Capture Fork date length Sex IRL:MIB HL:MIB OBL:IBL IRL:IBL HL:IBL mCI BKI 3/19/04 855 M 2.88 6.25 2.22 3.19 6.94 -1.22 5.10 4/29/04 880 M 2.75 6.13 2.21 2.82 6.28 -1.10 4.96 4/29/04 890 M 2.62 5.95 1.89 2.44 5.56 -0.72 4.51 4/29/04 903 M 2.53 6.13 3.06 3.26 7.90 -1.37 5.59 4/29/04 935 F 2.59 6.22 2.29 2.59 6.22 -0.82 4.88 4/29/04 970 F 2.88 6.80 2.25 2.88 6.80 -1.02 5.13 4/29/04 1015 F 2.50 5.83 2.76 3.11 7.24 -1.18 5.26 Table 3. Diagnostic proportionalities from Bailey and Cross (1954) and character indices applied to 8 recaptured hatchery-reared Pallid Sturgeon from the Atchafalaya River. Diagnostic values for these metrics are presented in Table 1. Pallid Sturgeon characters are in bold, shovelnose stur geon characters are in italics, all others are considered intermedia te. Capture Fork date length MW:MIB IRL:MIB HL:MIB OBL:IBL IRL:IBL HL:IBL mCI BKI 12/16/2010 500 1.38 2.23 5.14 1.68 2.16 4.98 -0.22 3.91 12/16/2010 513 1.35 2.16 5.18 1.61 2.18 5.24 0.01 3.77 3/14/2011 564 1.81 2.46 5.46 1.80 2.12 4.70 -0.65 4.26 3/14/2011 591 1.71 2.23 5.13 1.89 2.40 5.52 -0.37 4.12 3/14/2011 591 1.72 2.41 5.86 1.94 2.52 6.12 -0.38 4.35 3/14/2011 619 1.96 2.43 5.43 2.04 2.83 6.32 -0.65 4.47 3/13/2011 654 1.35 1.79 4.33 1.93 2.57 6.20 0.02 3.72 3/14/2011 674 1.53 2.15 5.01 2.24 2.91 6.79 -0.51 4.39 Table 5. Diagnostic proportionalities from Bailey and Cross (1954) and character indices applied to 7 recaptured hatchery-reared Pallid Sturgeon from the Mississippi River. Diagnostic values for these metrics are presented in Table 1. Pallid Sturgeon characters are in bold, Shovelnose Sturgeon characters are in italics, all others are considered intermedia te. Capture Fork date length MW:MIB IRL:MIB HL:MIB OBL:IBL IRL:IBL HL:IBL mCI BKI 3/12/2012 625 1.65 2.32 5.77 1.74 2.12 5.26 -0.09 4.06 12/16/2011 630 1.60 2.34 5.24 1.40 1.96 4.40 -0.20 3.74 11/17/2011 645 1.70 1.50 6.12 1.86 1.43 5.84 1.64 3.36 3/12/2012 680 1.57 2.31 5.43 1.79 2.08 4.87 -0.34 4.11 12/14/2011 688 1.73 2.10 5.08 1.88 2.24 5.41 -0.16 3.98 2/16/2012 691 1.81 2.37 5.63 2.00 2.50 5.94 -0.47 4.37 6/8/2010 791 1.90 2.48 6.13 2.71 2.83 7.00 -1.01 5.19 2013 P. Hartfield, N.M. Kuntz, and H.L. Schramm, Jr. 261 wire tags. However, head measurements applied to the proportionalities defined by Bailey and Cross (1954) failed to identify any of the recaptured fish as PLS, and 6 of the 8 specimens exhibited 1 or more proportions within the SNS range (Table 3). Additionally, none of the measurement ratios of the 2 smallest HR fish were within the ranges identified for PLS. There was some indication of the effect of allometric growth in that proportionalities associated with inner barbel length (IBL) increased with size. Both character indices (mCI, BKI) failed to correctly identify any of the recaptured fish as PLS (T able 3). Established measurement proportionalities and character indices do not appear diagnostic for lower Mississippi and Atchafalaya river age-0 and <700 mm FL PLS. Alternatively, it is possible that the HR PLS could be hybrids. However, the brood stock reportedly conformed both morphologically and genetically to PLS identifications. To test the morphological identification of the brood stock, we obtained and compared the morphometric measurements of the PLS brood stock (J. Dean, US Fish and Wildlife Service, Natchitoches National Fish Hatchery, Natchitoches, LA, unpubl. data) to 5 of the 6 diagnostic proportionalities of Bailey and Cross (1954) (mouth width measurements were unavailable) and the two character indices (Table 4). Three of the 4 males and 2 of the 3 females were identified as morphological PLS by all proportionalities. A single male scored as SNS on IRL:IBL and HL:IBL ratios, and 1 female scored intermediate at the same ratios. All brood stock were identified as morphological PLS by the mCI. The BKI identified 1 male as an intermediate and all others as PLS. It appears that the intermediate proportionalities in 2 specimens and the BKI identification of 1 of these as an intermediate are because of a disproportionately longer IBL in both specimens. We also applied the 6 diagnostic proportionalities and the character indices to 6 juvenile and 1 small adult HR fish captured in the Mississippi River (Table 5). Only the adult HR fish scored as PLS in all proportionalities and indices. One juvenile (645 mm FL) was identified as SNS by both indices. While at least some of these 7 fish may be from the 2004 lower Mississippi release, we cannot be certain because >45,000 HR YOY were released in the riverine reach of the Missouri River between 2003 and 2006, many marked with CWT and elastomer marks in the rostrum. Genetic samples of these specimens have been provided to the US Fish and Wildlife Service for brood stock parental screening but have not yet been analyzed. Discussion Our analysis indicates that morphological characters considered diagnostic for Missouri River larval Scaphirhynchus spp. failed to correctly identify all but a small proportion of HR PLS larvae reared from the Atchafalaya River brood stock. Atchafalaya River HR PLS meso- and metalarvae deviated from Snyder’s (2002) characters by 1) a higher incidence of pigmentation on the outer margins of the posterior lip lobes, 2) a high degree of overlap and variation in barbel 262 Southeastern Naturalist Vol. 12, No. 2 length ratios, and 3) the occurrence of semicircular lateral rostral plates in some HR PLS specimens. Therefore, characteristics found useful for identifying larval PLS in the upper Missouri River do not appear reliable for identifying meso- and metalarval PLS from the lower Mississippi River Basin. Application of the Missouri River characters to wild larval and postlarval Scaphirhynchus spp. from the lower Mississippi River indicates limited value of these characters for identification of either PLS or SNS in this portion of the range. Based on the low ratio of PLS:SNS adult captures in this reach of the Mississippi River (1:16; Killgore et al. 2007), we would expect that the majority of wild larval Scaphirhynchus spp. would consist of SNS and conform to that larval description, but this was not the case. Snyder (2002) reported that ventral pigmentation on the lower lip lobes was diagnostic for SNS mesolarvae, and pigmentation over the heart was characteristic for SNS meso- and metalarvae. All three characters (pigmentation, semi-circular lateral rostral plate, and low OB/IB ratios) were not observed in any wild specimen. Excluding pigmentation, less than 50% of the wild larvae conformed to SNS in either barbel length ratio or shape of lateral rostral plate, and only 7 conformed to SNS in both characters. Therefore, the suite of characteristics diagnostic for SNS larvae in the Missouri River appear to be ineffective in the lower Mississippi River. We also observed 2 characters in the wild larvae that were not present in the HR PLS meso- and metalarvae: MW/HW ratio < 0.49, and concave barbel position. Because these characters were not present in any of the HR PLS, they may have some utility for identification of SNS larvae in the Mississippi. However, while a low MW/HW ratio may be unique to some SNS specimens, we do not believe that a wide mouth (>0.49) is diagnostic of PLS because, as noted above, it is unlikely that the large percentage (70%) of wild Scaphirhynchus spp. specimens with MW/HW ratios > 0.49 in our study are all PLS. Similarly, barbel position is only partially informative. While no HR PLS exhibited a concave barbel position, the majority (>60%) of both HR PLS and wild larvae exhibited a straight barbel alignment. During our analysis, we noted differences in barbel papillation of metalarvae and age-0 postlarvae between the HR PLS and the wild Scaphirhynchus spp. samples. Papillae were not evident on the barbels of any HR PLS of any stage, or on any wild mesolarvae. However, prominent papillae or fine serrations indicative of papillae development were evident on both the anterior and posterior margins of all four barbels on the majority of wild metalarvae above 100 mm TL. Only 2 of the larger (>100 mm TL) wild metalarvae exhibited smooth barbels similar to the HR specimens of similar size. Both of these specimens (181 mm, 206 mm TL) exhibited other PLS characters, including convex barbel positions, deltoid lateral rostral plates (0.23), and wide mouths (>0.5 M/H). Barbel ratio in the smaller specimen was also within PLS range (1.68), while barbel ratio in the larger specimen was within SNS range (1.42). We suggest that the degree and location of barbel papillae on larval and postlarval Scaphirhynchus spp. warrants further investigation. 2013 P. Hartfield, N.M. Kuntz, and H.L. Schramm, Jr. 263 There are several possible explanations for the deviation of Atchafalaya River HR PLS from the 3 characters described by Snyder (2002) and our ability to identify only a few of the wild larvae that we examined with these characters. 1) Hatchery effects: Snyder’s characters (barbel length, pigmentation, lateral rostral plate) may be affected by the hatchery environment; however, these characters have been successfully applied to wild larvae collected in the Missouri River. While hatchery effects might explain differences between HR PLS larvae reared in south Louisiana vs. South Dakota, they do not explain the inability to positively identify any lower Mississippi River SNS larvae with all 3 characters. 2) Adaptation to regional environmental differences: The habitat and environment for brood stock that produced the progeny used in Snyder’s analysis (Missouri River, North Dakota) differs from that in the Atchafalaya River, LA, relative to hydrograph, discharge, temperature regimes, and other factors. Differences have been noted in size and longevity between adults from the upper Missouri River and the lower Mississippi River populations (USFWS 2007). Additionally, genetic studies have shown genetic differences between PLS populations increase with spatial separation (e.g., Campton et al. 2000, Schrey and Heist 2007, Tranah et al. 2001). Therefore, local adaptation to regional environmental differences cannot be discounted as an influence on larval characters. In other words, while the adults from both the middle Missouri River and the Atchafalaya River all appear to be PLS, local adaptations in the adults may be manifest in morphological differences in their progeny. 3) Introgression: The high proportion of intermediate morphological characters in Mississippi River Scaphirhynchus spp. has been attributed to hybridization (e.g., Carlson et al. 1985, Keenlyne et al. 1994, Hrabick et al. 2007, Schrey and Heist 2007). Introgression might also explain the deviation of the age 6–7 recaptured HR PLS from the Bailey and Cross (1954) morphometric proportionalities and character indices (Kuhajda and Mayden 2001, Wills et al. 2002). This hypothesis is supported to some degree by Allendorf et al. (2001) and Schrey et al. (2011), who concluded the amount of genetic introgression observed in Scaphirhynchus spp. would require multiple generations. If such is the case, however, both species have continued to maintain their adult morphologies. Such prolonged hybridization between sympatric species is not unexpected (e.g., Wallace 1912), has been demonstrated in other species (e.g., Saether et al. 2007), is not uncommon in sturgeon species, and has had a primary role in sturgeon speciation (Birstein et al. 1997, Robles et al. 2005, Vasil’ev 1999). The occurrence, extent, and significance of hybridization in Scaphirhynchus spp. populations have yet to be fully resolved (USFWS 2007). Our analysis of the 8 recaptured Atchafalaya River HR PLS and the 7 Mississippi River HR PLS indicates that lower Mississippi River juvenile PLS are likely to be identified as morphological intermediates, or “hybrids”, by measurement proportionalities and character indices currently in use. These results support the findings of Murphy et al. (2007) that morph ometric indices should not be exclusively used to classify juvenile or adult sturgeon specimens as hybrids. 264 Southeastern Naturalist Vol. 12, No. 2 The paucity of larval and juvenile PLS records in the Mississippi River portion of the range has been previously attributed to recruitment failure (e.g., USFWS 1990). We contend, however, that this rarity reflects insufficient historical collection efforts, a poor understanding of phenotypic and genotypic variation in lower Mississippi River Scaphirhynchus spp., and the inability to accurately identify larval, postlarval, and juvenile PLS with identification methods developed in other portions of its range. Large Scaphirhynchus spp. (>700 mm FL) classified as PLS have been collected from numerous locations throughout the Mississippi River (e.g., Killgore et al. 2007, and our unpublished collections). Our results demonstrate that larval and juvenile fish reared from morphological and genetic PLS collected from the Atchafalaya River are identified as “hybrids” or SNS by larval identification methods and character indices. Therefore, it follows that wild larval and juvenile PLS cohorts are also recruiting, but have likely been identified and discounted as “hybrids” or even as SNS. While multi-generation genetic introgression may play a role in the morphological differences between lower Mississippi and upper and middle Missouri river PLS populations, further refinement of both genetic and morphological methods of identification of PLS and SNS are needed to understand its significance. Acknowledgments Larval collections of Scaphirhynchus spp. were conducted under US Fish and Wildlife Subpermit TE198846-1, Louisiana Department of Wildlife and Fisheries Freshwater Scientific Collecting Permit 1235, and Mississippi Department of Wildlife, Fisheries, and Parks Administrative Scientific Collection Permit 1206102. Funding and support for our Scaphirhynchus collection efforts were provided by the US Fish and Wildlife Service, US Geological Survey, and Arkansas Game and Fish Commission. Additional support was provided by Mississippi State University and Catfish Point Hunt Club, Benoit, MS. We express our appreciation to Jan Dean, Natchitoches National Fish Hatchery, and Bobby Reed, Louisiana Department of Wildlife and Fisheries, for their information on PLS broodstock from the Atchafalaya River, and to Bernie Kuhajda and the University of Alabama Ichthyology Collection for the loan of the HR PLS series. Wild specimens collected for this study have been deposited at the Mississippi Museum of Natural Science. The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the US Fish and Wildlife Service. Literature Cited Allendorf, F.W., R.F. Leary, P. Spruell, and J.K. Wenburg. 2001. The problems with hybrids: Setting conservation guidelines. Trends in Ecology and Evolution 16:613–622. Bailey, R.M., and F.B. Cross. 1954. River sturgeons of the American genus Scaphirhynchus: Characters, distribution, and synonymy. Michigan Academy of Science Arts and Letters 39:169–208. Birstein, V.J., R. Hamner, and R. DeSalle. 1997. Phylogeny of Acipenseriformes: Cytogenic and molecular approaches. Environmental Biology of Fishes 48:127–155. Campton, D.E., A.L. Bass, F.A. Chapman, and B.W. Bowen. 2000. Genetic distinction of Pallid, Shovelnose, and Alabama sturgeon: Emerging species and the US Endangered Species Act. Conservation Genetics 1:17–32. 2013 P. Hartfield, N.M. Kuntz, and H.L. Schramm, Jr. 265 Carlson, D.M., W.L. Pflieger, L. Trial, and P.S. Haverland. 1985. Distribution, biology, and hybridization of Scaphirhynchus albus and S. platorynchus in the Missouri and Mississippi rivers. Environmental Biology of Fishes 14:51–59. Dean, J. 1998. Pallid Sturgeon at Natchitoches National Fish Hatchery—FY 1998. Annual report, Natchitoches National Fish Hatchery, Natchitoches, LA. 21 pp. Dean, J. 2004. Pallid Sturgeon at Natchitoches National Fish Hatchery—FY 2004. Annual report, Natchitoches National Fish Hatchery, Natchitoches, LA. 15 pp. Forbes, S.A., and R.E. Richardson. 1905. On a Shovelnose Sturgeon from the Mississippi River. Bulletin of Illinois State Lab of Natural History 7:37–44. Hrabik. R.A., D.P. Herzog, D.E. Ostendorf, and M.D. Petersen. 2007. Larvae provide first evidence of successful reproduction by Pallid Sturgeon, Scaphirhynchus albus, in the Mississippi River. Journal of Applied Ichthyology 23:436–443. Herzog, D., V.A. Barko, J.S. Scheibe, R.A. Hrabik, and D.E. Ostendorf. 2005. Efficacy of a benthic trawl for sampling small-bodied fishes in large river systems. North American Journal of Fisheries Management 25:594–603. Keenlyne, K.D., L.K. Graham, and B.C. Reed. 1994. Hybridization between the Pallid and Shovelnose Sturgeon. Proceedings of the South Dakota Academy of Sciences 73:59–66. Killgore, K.J., J.J. Hoover, S.G. George, B.R. Lewis, C.E. Murphy, and W.E. Lancaster. 2007. Distribution, relative abundance, and movements of Pallid Sturgeon in the freeflowing Mississippi River. Journal of Applied Ichthyology 23:476–483. Kuhajda, B.R., and R.L. Mayden. 2001. Morphological comparisons of hatchery-reared specimens of Scaphirhynchus albus, S. platorynchus, and S. albus x S. platorynchus hybrids, final report. US Fish and Wildlife Service, Missouri River Fish and Wildlife Conservation Office, Bismark, ND. 118 pp. Murphy, C. E., J.J. Hoover, S.G. George, and K.J. Killgore. 2007. Morphometric variation among river sturgeons (Scaphirhynchus spp.) of the middle and lower Mississippi River. Journal of Applied Ichthyology 23:313–323. Phelps, Q.E., S.J. Tripp, W.D. Hintz, J.E. Garvey, D.P. Herzog, D.E. Ostendorf, J.W. Ridings, J.W. Crites, and R.A. Hrabik. 2010. Water temperature and river stage influence mortality and abundance of naturally occurring Mississippi River Scaphirhynchus sturgeon. North American Journal of Fisheries Management 30:767–775. Robles, F., R. De la Herrán, A. Ludwig, C.R. Rejón, M.R. Rejón, and M.A. Garrido- Ramos. 2005. Genomic organization and evolution of the 5S ribosomal DNA in the ancient fish sturgeon. Genome 48:18–28. Sæther, SA, G. Sætre, T. Borge, C. Wiley, N. Svedin, G. Andersson, T. Veen, J. Haavie, M.R. Servedio, S. Bureš, M. Král, M.B. Hjernquist, L. Gustafsson, J. Träff, and A. Qvarnström. 2007. Sex chromosome-linked species recognition and evolution of reproductive isolation in flycatchers. Science 318:95–97. Schrey, A.W., and E.J. Heist. 2007. Stock structure of Pallid Sturgeon analyzed with microsatellite loci. Journal of Applied Ichthyology 23:297–303. Schrey, A.W., R. Boley, and E.J. Heist. 2011. Hybridization between Pallid Sturgeon Scaphirhynchus albus and Shovelnose Sturgeon, Scaphirhynchus platorynchus. Journal of Fish Biology 79:1828–1850. Snyder, D.E., 2002. Pallid and Shovelnose Sturgeon larvae–morphological description and identification. Journal of Applied Ichthyology 18:240–265. Tranah, G., H.L. Kincaid, C.C. Krueger, D.E. Campton, and B. May. 2001. Reproductive isolation in sympatric populations of Pallid and Shovelnose Sturgeon. North American Journal of Fisheries Management 21:367–373. 266 Southeastern Naturalist Vol. 12, No. 2 US Fish and Wildlife Service (USFWS). 1990. Determination of endangered status for the Pallid Sturgeon; final rule. Federal Register 55:36641–36647. USFWS. 2007. Pallid Sturgeon (Scaphirhynchus albus) 5-year review summary and evaluation. Billings, MT. Vasil’ev, V.P. 1999. Polyploidization by reticular speciation in Acipenseriformes evolution: A working hypothesis. Journal of Applied Ichthyology 15:29–31. Wallace, A.R. 1912. Darwinsim: An Exposition of the Theory of Natural Selection with Some of Its Applications. McMillan and Co., Limited. London, UK, 3rd Edition. Pp. 174–179. Wills, P.S., J.R. Sheehan, R. Heidinger, and B.L. Sloss. 2002. Differentiation of Pallid Sturgeon and Shovelnose Sturgeon using an index based on meristics and morphometrics. Pp. 249–258, In W. Van Winkle, P. Anders, D.H. Secor, and D. Dixon (Eds.). Biology, Management, and Protection of North American Sturgeon, American Fisheries Society Symposium 28, Bethesda, MD.