2007 SOUTHEASTERN NATURALIST 6(2):321–332 Factors Influencing Paddlefish Spawning in the Tombigbee Watershed Daniel M. O’Keefe1,*, Johanna C. O’Keefe1, and Donald C. Jackson1 Abstract - The early life-history requirements of Polyodon spathula (paddlefish) are not well understood, in part due to the difficulty of sampling early life stages. Passive sampling with benthic, mat-style devices effectively collected paddlefish eggs in the Tombigbee watershed (Mobile River basin) during spring 2005, facilitating identification and characterization of egg-incubation microhabitats. Eggs were collected over gravel, sand-impacted gravel, and bedrock substrates at corrected depths ranging from 1.2 to 7.7 m. Sampling occurred continuously (489 sampler-days) in the lotic bendway of the Tennessee-Tombigbee Waterway from late February through April, when water temperatures ranged from 11.5 to 20.8 °C. Of 106 paddlefish eggs collected from this unique macrohabitat, 95% were taken on either April 6 or April 16. Nine paddlefish eggs were collected in a tributary (Noxubee River) on April 13 after four sampler-days of effort. Water temperatures associated with peak spawning activity ranged from 16.9 to 19.4 °C, slightly higher than temperatures recorded for Mississippi River basin populations. A substantial (> 2.74-m) rise in water level triggered spawning activity, similar to that observed in other systems. Benthic mats proved useful for delineating paddlefish egg-incubation habitat in areas not subject to shifting substrate, and could be used in the future to address hypotheses regarding micro- and macrohabitat suitability. Introduction Many details regarding early life-history and spawning requirements of Polyodon spathula Walbaum (paddlefish) have historically eluded researchers. Although efforts to collect early life stages of paddlefish date back to the early 1900s (Stockard 1907), spawning was not documented until 1960 (Purkett 1961). Since that time, no single method has been used consistently to sample eggs. Although paddlefish are present throughout large river systems of the Mississippi and several other Gulf of Mexico drainages, collection of eggs and subsequent description of spawning or egg-incubation habitat have occurred only in the Osage River, MO (Purkett 1961), Old Hickory Reservoir, TN (Pasch et al. 1980, Wallus 1986), and the lower Yellowstone River, MT and ND (Firehammer et al. 2006). Published studies have reported collection of eggs from dry gravel bars after a sudden drop in river level (Purkett 1961) and by sampling with epibenthic sleds (Pasch et al. 1980, Wallus 1986), ichthyoplankton drift nets (Pasch et al. 1980, Wallus 1986), and dredges (Purkett 1961). Collection of larvae (Alexander and McDonough 1983, Lein and DeVries 1998) or youngof- year (Houser and Bross 1959, Jennings and Wilson 1993, Reed et al. 1Department of Wildlife and Fisheries, Mississippi State University, Mississippi State, MS 39762. *Corresponding author - danokeefe@hotmail.com. 322 Southeastern Naturalist Vol. 6, No. 2 1992, Ruelle and Hudson 1977) is commonly used to provide evidence of successful spawning. These methods do not provide information regarding the conditions necessary for successful egg deposition and development, and only allow coarse-scale delineation of spawning locations (i.e., areas upstream of capture). Passive sampling with benthic, mat-style devices has been used to collect the adhesive eggs of sturgeons (family: Acipenseridae), facilitating description of egg incubation habitat and examination of spatial and temporal trends in spawning activity (Marchant and Shutters 1996, McCabe and Beckman 1990). Suspended, tube-style sampling devices have recently been used to passively sample paddlefish eggs (Firehammer et al. 2006). Although suspended devices have the advantage of sampling effectively when shifting sand substrate renders benthic devices ineffective (Firehammer et al. 2006), their suspension in the water column does not mimic the position of natural substrates. The use of suspended sampling devices therefore results in capture of eggs between the location of spawning and the location of egg incubation; spawning and egg-incubation locations are not always identical due to the drifting of eggs before their adhesion to substrates (Purkett 1961). Determination of egg-incubation locations and monitoring of these locations over time could provide detailed site-specific information regarding environmental cues necessary to induce spawning. This is especially important in regulated rivers, where paddlefish spawning success may hinge on human decisions regarding dam operation and flow regime (Alexander and McDonough 1983). Several reviews of paddlefish biology and management indicated a need for basic knowledge regarding availability of spawning areas and factors that influence reproductive success (Birstein et al. 1997, Carlson and Bonislawsky 1981, Graham 1997, Jennings and Zigler 2000, Russell 1986). Toward this end, the purpose of this study was to: evaluate the effectiveness of passive sampling with benthic, mat-style devices; determine the location and characteristics of microhabitats used for paddlefish egg incubation; and draw inferences regarding environmental factors that influence spawning in the Tennessee-Tombigbee Waterway. Study Area The Tennessee-Tombigbee Waterway is a comprehensively modified ecosystem consisting of tailraces, dredged navigation channel segments, bendways (portions of the former Tombigbee River), and impoundments (Jackson 1995). Gill netting for paddlefish in 4 tailraces, 15 bendways, 3 impoundments, and 10 tributaries revealed large concentrations of paddlefish in 2 bendways (O’Keefe 2006). Sampling for paddlefish eggs was limited to one of these: the bendway between Howell Heflin Dam and Howell Heflin Lock near Gainesville, AL (Fig. 1). This area is not subject to channel dredging, snagging, or commercial navigation. Other sections of the former Tombigbee River in Alabama are part of the navigation channel, impounded, or subject to flow diversion from the creation of cutoff canals, 2007 D.M. O’Keefe, J.C. O’Keefe, and D.C. Jackson 323 which carry the majority of the waterway’s discharge and commercial barge traffic. Most bendways are impacted by siltation or formation of sand plugs at their upstream ends, which further reduce flow (Pennington et al. 1981). The bendway between Howell Heflin Dam and Howell Heflin Lock is not subject to excessive sedimentation due to spatial decoupling of the dam and its navigation lock. This “lotic bendway” is the only portion of the original river channel that retains the majority of the waterway’s discharge but is not subjected to commercial navigation and channel maintenance activities. Due to its relatively undisturbed condition, apparent availability of spawning habitat, and high adult paddlefish density, the lotic bendway was chosen as a macrohabitat in which an egg-sampling technique could be evaluated, microhabitats used for egg incubation could be located and characterized, and environmental cues related to spawning behavior could be studied. Potential paddlefish spawning locations were identified prior to the current study. Purkett (1961) described paddlefish spawning habitat consisting of shallow gravel bars that were not submerged during normal flow conditions. Such an exposed gravel bar was located in the lotic bendway at the mouth of the Noxubee River during concurrent population-dynamics and radio-telemetry studies (Fig. 1). This is a unique habitat in the mainstem of Figure 1. Location of egg samplers used to collect paddlefish eggs at shallow (< 3 m) and deep (􀂕 3 m) locations in the lotic bendway of the Tennessee-Tombigbee Waterway, AL, below Howell Heflin Dam during spring 2005. One egg sampler was deployed in the Noxubee River, 7.6 km upstream from its confluence with the Tennessee-Tombigbee Waterway. 324 Southeastern Naturalist Vol. 6, No. 2 the Tennessee-Tombigbee Waterway; all other exposed gravel bars have apparently been eliminated through dredging, sedimentation, and impoundment. In the Noxubee River, a fourth-order tributary that has not been heavily impacted by channelization or damming, an exposed gravel bar was located 7.6-km upstream from its confluence. Most egg sampling occurred in the lotic bendway of the Tennessee-Tombigbee Waterway, but limited sampling for eggs also occurred at the shallow gravel bar in the Noxubee River. Within the lotic bendway, egg sampling was conducted in shallow gravel microhabitats where spawning activity was expected to be heaviest, and in microhabitats that appeared less suitable for egg incubation (i.e., deeper locations and areas with bedrock substrate). Median discharge above Howell Heflin Dam in Gainesville, AL, was 141 m3 sec-1 over the period March 17, 1978 through March 16, 2004 (station number 02447025; USGS 2005). Median discharge of the Noxubee River at Geiger, AL, was 10 m3 sec-1 over the period August 1, 1944 through July 31, 2004 (station number 02447025; USGS 2005). Discharge data were not available for the lotic bendway of the Tennessee-Tombigbee Waterway below Howell Heflin Dam, where median gage height was 23.10 m (range 22.36 to 34.47 m) over the period of available data from August 19, 1999 through August 18, 2004 (station number 02447026; USGS 2005). Due to high variability of gage height, depths in the lotic bendway were highly variable. Subsequently, depths reported at sampling locations were corrected for high and variable water levels by subtracting the difference between actual gage height and median gage height from the actual depth measured in the field. Surface-water temperature in the lotic bendway ranged from 6 to 30 °C, Secchi depth ranged from 9 to 76 cm, and specific conductance at 25 °C ranged from 70 to 232 􀂗S/cm (D.M. O'Keefe, unpubl. data). Methods Benthic egg-sampler design was based on mat-style sampling devices used to collect eggs of Acipenser spp. (Marchant and Shutters 1996, McCabe and Beckman 1990). Egg samplers were constructed using artificial substrates of latex-coated hog hair filter material 51 cm wide by 2.54 cm deep. The material was wrapped smooth-side down around a 51-cm square angle-iron frame such that both sides of the frame were covered with the filter material. The filter material was attached to the frame using nuts and bolts. Washers were used to prevent material from slipping off bolts because holes in the material wore wider with use. Frames were fastened to anchors with ropes and swivels to prevent twist in the current. Anchors consisted of scrap iron pieces from 9 to 23 kg. Float lines of 15 to 30-m length and 1-cm diameter were attached to anchors with a swivel. Most float lines were attached to bullet-shaped floats 35 cm long and 13.5 2007 D.M. O’Keefe, J.C. O’Keefe, and D.C. Jackson 325 cm in diameter, but others were tied to riparian vegetation including overhanging limbs and exposed roots. The egg samplers were deployed on shallow gravel bars (< 3 m deep at median flow) where spawning was expected to occur, at deeper locations where substrate consisted of gravel or bedrock, and at shallow (< 3 m deep at median flow) tailrace locations with bedrock substrate (Fig. 1). Sampling began on February 28, 2005, and concluded on April 25, 2005. Two to eleven egg samplers (mean of seven) were effectively sampling (i.e., deployed and retrieved within 14 days) in the lotic bendway at any given time during this period. One egg sampler was deployed in the Noxubee River from April 9 to April 13, 2005. Individual egg samplers were deployed at 4- to 8-day intervals when possible, but high water prevented retrieval of egg samplers between April 6, 2005 and April 16, 2005. Upon retrieval, egg samplers were examined for the presence of large (approximately 3–4 mm diameter), grey, adhesive eggs. Size and coloration were sufficient to determine that eggs of such description were from acipenseriform species (Firehammer et al. 2006). Eggs of this description were removed by cutting the strands of hair to which they were attached. Periphyton, detritus, and fine substrates accumulated on artificial substrates, which were rinsed thoroughly before redeployment. Eggs were preserved in 80% ethanol or transported to aerated tanks for hatching. Hatched larvae were preserved in 80% ethanol upon death. On each date that eggs were collected, a portion of eggs were preserved as vouchers, and others were hatched to verify that eggs were paddlefish, and not Scaphirhynchus suttkusi Williams and Clemmer (Alabama sturgeon). The Alabama sturgeon is the only other acipenseriform species recorded in the Tombigbee drainage (Mettee et al. 1996, Ross 2001), and is so rare that capture of its eggs was unlikely. Paddlefish egg catch per sampler-day was calculated as the sum of eggs collected over the total number of sampler-days. Water temperature was measured at 0.5 m below the surface when samplers were retrieved using a Model 85 Yellow Springs Instrument (YSI, Inc., Yellow Springs, OH). Water velocity was measured at 0.5 m below the surface using a Flo-Mate (Marsh-McBirney, Inc., Frederick, MD) at successful sample sites (i.e., those at which at least one paddlefish egg was captured) on April 6 and April 13, 2005. Results Paddlefish eggs were collected from egg samplers in the lotic bendway on three dates (March 30, April 6, April 16) and in the Noxubee River on the one date it was sampled (April 13) during 2005 (Fig. 2). Of 106 paddlefish eggs collected from the lotic bendway, 95% were taken on either April 6 or April 16. Paddlefish egg catch per sampler-day was 0.23 for the entire study period. Catch per sampler-day was 0.61 for March 30 326 Southeastern Naturalist Vol. 6, No. 2 through April 16 and zero for the remainder of the study. Correct identification of eggs was verified by hatching a subsample (29%) of the eggs collected. Thirty-three paddlefish larvae were hatched from eggs collected on March 30, April 6, April 13, and April 16. Twelve larvae lived at least two weeks, developed rostrums, and successfully switched from endogenous to exogenous food sources. Paddlefish eggs were collected under a wide variety of depth, substrate, and velocity conditions. Successful artificial substrates were set at depths of 1.2 to 7.7 m at median gage height, representing the full range of depths sampled. Eggs were collected from samplers placed over gravel at shallow (< 3 m) and deep (􀂕 3 m) locations, and over bedrock at deep locations. Egg samplers set in the tailrace of Heflin Dam over bedrock in shallow water did not collect eggs. High flow deposited coarse sand over some submerged gravel bars during the study. One egg collected from the Noxubee River hatched successfully despite adhesion to sand grains, which covered it completely. On April 6, water velocity ranged from 0.39 to 1.06 m/s at successful sites in the lotic bendway. On April 13, water velocity at the successful site in the Noxubee River was 0.99 m/s and temperature was 19.4 °C. Water temperature in the lotic bendway was 18.0 °C on April 6, 16.9 °C on April 9, and 19.4 °C on April 13 and 16. High water level (which peaked at 8 m above median gage height) between April 6 and April 16 submerged floats and terrestrial tie-off points for the egg samplers, making it impossible to examine them during this Figure 2. Gage height and water temperature in the Tennessee-Tombigbee Waterway, AL, below Howell Heflin Dam during spring 2005, with number of successful egg samplers (numerator) and number of functioning egg samplers (denominator) shown above the abscissa. Collection of one or more paddlefish eggs on an egg sampler was considered a success. 2007 D.M. O’Keefe, J.C. O’Keefe, and D.C. Jackson 327 period. Five of the egg samplers were retrieved on April 16, two of which contained no eggs and were deeply buried in gravel and sand. These were not considered as functioning effectively between April 6 and April 16 and subsequently were not included when determining the percentage of successful sampling mats. Two others were partially covered with sand or gravel and contained 1 or 2 eggs. These were considered as functioning effectively, as were other partially-covered egg samplers examined on other dates. The only entirely unimpacted egg sampler retrieved on April 16 contained 19 paddlefish eggs. All substrates not affixed to terrestrial vegetation shifted position during high water, washing into locations 1.4 to 6.6 m deeper than original locations. Discussion Passive sampling with benthic, mat-style devices is an effective method for locating and describing egg-incubation habitat of paddlefish. This method allows researchers to sample continuously for extended periods of time and is effective in locations with abundant debris, deep water, and high velocity. Although mats affixed to floats may experience some downstream drift during high water, attachment to riparian vegetation eliminates this problem. Benthic mats are very effective relative to other methods used for collecting paddlefish eggs. Two of seven methods used to sample paddlefish eggs and larvae over a 9-year period were successful at capturing eggs in the Cumberland River (Wallus 1986). Over that period of time, 41 eggs were collected using an epibenthic sled and an obliquely towed 0.5-m square ichthyoplankton net (Wallus 1986). During a single season in the present study, 493 mat-days resulted in the capture of 115 eggs in the lotic bendway and the Noxubee River. Firehammer et al. (2006) reported that mat-style egg samplers similar to those used in our study and previous studies on Acipenser spp. (Marchant and Shutters 1996, McCabe and Beckman 1990) did not effectively sample paddlefish eggs in the lower Yellowstone River in Montana and North Dakota, where shifting sand substrate buried egg samplers. Subsequently, tube-style egg samplers designed to collect eggs drifting in the water column were used effectively in the lower Yellowstone River (Firehammer et al. 2006). Mat-style egg samplers successfully collected paddlefish eggs over relatively stable gravel and bedrock substrates in the Tennessee-Tombigbee Waterway, but tube-style egg samplers might have reduced problems associated with sand deposition on samplers following peaks in discharge. The tube-style sampling devices used by Firehammer et al. (2006) captured 130 acipenseriform eggs (89 genetically confirmed as paddlefish) over two years with 926 collector days of effort in the lower Yellowstone River. Our higher catch-per-effort using mat-style devices suggests that mats may be more effective than tubes where substrate permits their use, but catch rates are not 328 Southeastern Naturalist Vol. 6, No. 2 directly comparable between studies and are dependant on the density of spawning paddlefish in sampled habitats. Paddlefish eggs were collected in microhabitats similar to the exposed gravel bars described by Purkett (1961), and also were collected in large numbers from deep, high velocity areas with gravel or bedrock substrate. Purkett (1961) noted that eggs collected from deep areas downstream from gravel bars were covered with debris or adhered to waterlogged wood, implying that they came to rest in a depositional zone. In the lotic bendway, deep bedrock runs were not sites of sediment deposition, but it is unlikely that eggs would adhere to the clay-rich marl substrate. Artificial substrates set over deep bedrock were relatively free of leaf litter, fine substrate, and periphyton growth that accumulated at shallow sites, suggesting that eggs would experience suitable incubation conditions if they could adhere to a suitable surface. The artificial substrates provided such a surface, as might large woody debris (present at the sample site in the form of sunken, waterlogged timber, even in areas of swift current). Tailrace spawning in other systems has been inferred by collection of larval paddlefish found downstream from dams (Hoxmeier and DeVries 1997, Pasch et al. 1980, Wallus 1986). We collected eggs 2.8 km to 4.5 km downstream from Howell Heflin Dam, but not from artificial substrates set within 400 m of the dam. This suggested that eggs did not incubate in the tailrace of Howell Heflin Dam. Most of the eggs were collected in the Tombigbee watershed at water temperatures between 16.9 and 19.4 °C when water level was at least 2.74 m above gage height at median flow. This corresponds to the 2.74-m rise shown to trigger paddlefish spawning in the Osage River, MO (Purkett 1961). Although paddlefish in the lotic bendway required the same increase in discharge needed to induce spawning in other systems, spawning occurred in warmer water than has been observed for Mississippi River basin populations such as those of the Osage River, MO (15 to 16 °C; Purkett 1961), and Cumberland and Tennessee rivers, TN (12 to 15 °C; Wallus 1986). A study of paddlefish in the Mobile River basin found that spawning occurred in tributaries of the Alabama River, AL, when water temperature ranged from 12 to 17 °C (Lein and DeVries 1998), which is consistent with results from Mississippi basin populations. However, Hoxmeier and DeVries (1997) found that paddlefish adults did not vacate a suspected spawning area below Claiborne Dam on the Alabama River, AL, until water temperature exceeded 24 °C, and suggested that paddlefish of the Mobile River basin may, in some instances, spawn at warmer temperatures than recorded in the Mississippi River basin. Our results in the Tombigbee watershed, which is part of the Mobile River basin, add some credence to this hypothesis. Genetic differences among paddlefish in the Mobile and Mississippi river basins (Carlson 1982, Epifanio et al. 1996) may explain slight differences in spawning cues. However, construction of the Tennessee-Tombigbee Waterway created a 2007 D.M. O’Keefe, J.C. O’Keefe, and D.C. Jackson 329 freshwater connection between these two historically isolated stocks, and interbreeding has therefore been possible since 1985. Current velocity at paddlefish spawning locations has not been reported by other authors who have collected eggs (Firehammer et al. 2006, Pasch et al. 1980, Purkett 1961, Wallus 1986). Spawning and egg-incubation habitatsuitability curves developed by Crance (1987) indicated a suitability index of 0.4 or higher for mean water-column velocities of 0.4 to 1.6 m/s. In the lotic bendway, subsurface water velocities at egg-incubation locations were at the lower to middle portion of this range. Sub-surface velocity is greater than mean velocity in deep rivers (Allan 1995), suggesting that mean velocities at some egg-incubation locations in the Tombigbee watershed were slightly lower than expected by Crance (1987). However, velocity measurements were taken during our study only on selected dates, and do not represent the full range of current conditions that incubating eggs experienced. Velocity measurements were taken in the lotic bendway on April 6, when gage height was at its lowest for the period of March 31 to April 17. This fact suggests that recorded sub-surface velocities represented the minimum values encountered by the majority of eggs incubating at sampled locations in the lotic bendway. Egg collection facilitated description of date, discharge, substrate, depth, water temperature, and current velocity conditions associated with spawning and subsequent egg incubation. In highly regulated rivers such as the Tennessee- Tombigbee Waterway, determining environmental spawning cues is especially important because timing and duration of high discharge (which are controlled, to some extent, by human activity) must correspond with the appropriate date and water temperature for successful paddlefish spawning (Graham 1997). Collection of paddlefish eggs in the lotic bendway and an unregulated tributary (Noxubee River) suggests that macrohabitats not subjected to channelization, navigation traffic, or impoundment are used as egg incubation habitat. The suitability of these relatively undisturbed environments as compared with more degraded environments was not directly addressed by this study. Although shallow (< 3 m deep) gravel microhabitat is rare in altered mainstem Tennessee-Tombigbee Waterway macrohabitats, deeper gravel and bedrock areas are common in certain areas of the navigation channel (D.M. O'Keefe, unpubl. data). Based on egg-sampler success at deep gravel and bedrock locations in the lotic bendway, these areas could provide suitable egg-incubation habitat, assuming that adult paddlefish would elect to spawn in the absence of proximal shallow gravel habitat. Radio-telemetry data from paddlefish tagged in the lotic bendway indicate that paddlefish show strong preference for the lotic bendway macrohabitat, although some individuals moved into the Noxubee River or navigation channel when conditions were ideal for spawning (O’Keefe 2006). Regardless of the degree to which paddlefish preferenially spawn in areas relatively free from disturbance, these areas should be protected 330 Southeastern Naturalist Vol. 6, No. 2 from excessive human disturbance because of their demonstrated role in the life cycle of paddlefish. Graham (1997) suggested that all potential and known paddlefish spawning areas should be protected from detrimental human disturbance because of widespread degradation of historically suitable spawning habitats. The limited distribution of adult paddlefish in the Tennessee-Tombigbee Waterway and their restricted movements (O’Keefe 2006) further suggest that the lotic bendway and Noxubee River are critical to their continued survival in the Tombigbee watershed and worthy of protection. Documentation of paddlefish spawning and egg-incubation locations is uncommon, but necessary for protection of critical habitats. We found that passive sampling with benthic mats effectively collected paddlefish eggs in areas not subjected to shifting substrates. The use of benthic mats allowed us to document and characterize paddlefish egg-incubation habitats, and examine patterns of paddlefish spawning activity in relation to abiotic factors. Beyond the scope of our study, our methods could be used to investigate the relative suitability of multiple macrohabitats (e.g., channelized vs. unchannelized stream reaches) in addition to facilitating more detailed description of egg-incubation microhabitat characteristics. Acknowledgments Field work for this project was completed with the assistance of M. Kashiwagi and A. Pollack. Phillip Bettoli and two anonymous reviewers improved this paper with their helpful comments. Funding was provided through the Mississippi Department of Wildlife, Fisheries, and Parks under Federal Aid Project T-1. This paper is approved as manuscript number WF-230 of the Forest and Wildlife Research Center, Mississippi State University. Literature Cited Alexander, C.M., and T.A. McDonough. 1983. Effect of water condition during spawning on paddlefish year-class strength in Old Hickory Reservoir, Tennessee. TVA/ONR/WRF-83-4(a). Tennessee Valley Authority, Knoxville, TN. 29 pp. Allan, J.D. 1995. Stream Ecology: Structure and Function of Running Waters. Chapman and Hall, London, UK. Birstein, V.B., W.E. Bemis, and J.R. Waldman. 1997. The threatened status of acipenseriform species: A summary. Environmental Biology of Fishes 48:427–435. Carlson, D.M. 1982. Low genetic variability in paddlefish populations. Copeia 1982(3):721–725. Carlson, D.M., and P.S. Bonislawsky. 1981. The paddlefish (Polyodon spathula) fisheries of the midwestern United States. Fisheries 6(2):17–27. Crance, J.H. 1987. Habitat suitability index curves for paddlefish, developed by the Delphi technique. North American Journal of Fisheries Management 7:123–130. Epifanio, J.M., J.B. Koppelman, M.A. Nedbal, and D.P. Philipp. 1996. Geographic variation of paddlefish allozymes and mitochondrial DNA. Transactions of the American Fisheries Society 125:546–561. 2007 D.M. O’Keefe, J.C. O’Keefe, and D.C. Jackson 331 Firehammer, J.A., D.L. Scarnecchia, and S.R. Fain. 2006. Modification of a passive gear to sample paddlefish eggs in sandbed spawning reaches of the lower Yellowstone River. North American Journal of Fisheries Management 26:63–72. Graham, L.K. 1997. Contemporary status of the North American paddlefish, Polyodon spathula. Environmental Biology of Fishes 48:279–289. Houser, A., and M.G. Bross. 1959. Observations on growth and reproduction of the paddlefish. Transactions of the American Fisheries Society 88:50–52. Hoxmeier, R.J.H., and D.R. DeVries. 1997. Habitat use, diet, and population structure of adult and juvenile paddlefish in the Lower Alabama River. Transactions of the American Fisheries Society 126:288–301. Jackson, D.C. 1995. Distribution and stock structure of blue catfish and channel catfish in macrohabitats along riverine sections of the Tennessee-Tombigbee Waterway. North American Journal of Fisheries Management 15:845–853. Jennings, C.A., and D.M. Wilson. 1993. Spawning activity of paddlefish (Polyodon spathula) in the lower Black River, Wisconsin. Journal of Freshwater Ecology 8(3):261–262. Jennings, C.A., and S.J. Zigler. 2000. Ecology and biology of paddlefish in North America: Historical perspectives, management approaches, and research priorities. Reviews in Fish Biology and Fisheries 10:167–181. Lein, G.M., and D.R. DeVries. 1998. Paddlefish in the Alabama River drainage: Population characteristics and the adult spawning migration. Transactions of the American Fisheries Society 127:441–454. Marchant, S.R., and M.K. Shutters. 1996. Artificial substrates collect Gulf sturgeon eggs. North American Journal of Fisheries Management 16:445–447. McCabe, G.T., Jr., and L.G. Beckman. 1990. Use of an artificial substrate to collect white sturgeon eggs. California Fish and Game 76:248–250. Mettee, M.F., P.E. O’Neil, and J.M. Pierson. 1996. Fishes of Alabama and the Mobile Basin. Oxmoor House, Birmingham, AL. 820 pp. O’Keefe, D.M. 2006. Conservation and management of paddlefish in Mississippi with emphasis on the Tennessee-Tombigbee Waterway. Ph.D. Dissertation. Mississippi State University. Mississippi State, MS. 161 pp. Pasch, R.W., P.A. Hackney, and J.A. Holbrook II. 1980. Ecology of paddlefish in Old Hickory Reservoir, Tennessee, with emphasis on first-year life history. Transactions of the American Fisheries Society 109:157–167. Pennington, C.H., J.A. Baker, F.G. Howell, and C.L. Bond. 1981. A study of cutoff bendways on the Tombigbee River. Technical Report E-81-14. United States Army Corps of Engineers, Vicksburg, MS. 78 pp. Purkett, C.A. 1961. Reproduction and early development of the paddlefish. Transactions of the American Fisheries Society 90:125–129. Reed, B.C., W.E. Kelso, and D.A. Rutheford. 1992. Growth, fecundity, and mortality of paddlefish in Louisisana. Transactions of the American Fisheries Society 121:378–384. Ross, S.T. 2001. Inland Fishes of Mississippi. University Press of Mississippi, Oxford, MS. 624 pp. Ruelle, R., and P.L. Hudson. 1977. Paddlefish: Growth and food of young-of-theyear and a suggested technique for measuring length. Transactions of the American Fisheries Society 106:609–613. 332 Southeastern Naturalist Vol. 6, No. 2 Russell, T.R. 1986. Biology and life history of the paddlefish: A review. Pp. 2–20, In J.G. Dillard, L.K. Graham, and T.R. Russell, (Eds.). The Paddlefish: Status, Management, and Propagation. North Central Division American Fisheries Society, Bethesda, MD. Special Publication 7. 159 pp. Stockard, C.R. 1907. Observations on the natural history of Polyodon spathula. American Naturalist 41:753–766. United States Geological Survey (USGS). 2005. Water data. Available online at http://waterdata.usgs.gov/nwis/rt. May 6, 2005. Wallus, R. 1986. Paddlefish reproduction in the Cumberland and Tennessee River systems. Transactions of the American Fisheries Society 115:424–428.