Percina crypta, Freeman, Freeman & Burkhead, 2008

Freeman, Mary C., Freeman, Byron J., Burkhead, Noel M. & Straight, Carrie A., 2008, A new species of Percina (Perciformes: Percidae) from the Apalachicola River drainage, southeastern United States, Zootaxa 1963 (1), pp. 25-42 : 28-40

Percina crypta, Freeman, Freeman & Burkhead , new species

Halloween Darter

( Fig. 2A and 2B )

Holotype. GMNH 21606 , male, 69.7 mm SL, Georgia, White County, Chattahoochee River at GA Hwy 17/75 at Nacoochee , Georgia (Nora Mill), 11.1 air km NNE Cleveland, 17 May 1994, B. J. Freeman, J. Devivo, J. W. Garrett, M. J. Zieg, R. E. Jenkins, J. S. Boyce, L. M. Hartle, M. Flood.

Paratopotypes. GMNH 21610 (18; 48–73 mm), Collected with the holotype . GMNH 21578 (11; 52–72 mm), 5 April 2000 ; TU 200399 (4; 55–66 mm) same data as GMNH 21578; USNM 393568 (4, 58– 64 mm) same data as GMNH 21578.

Paratypes. Georgia: Lumpkin County: GMNH 21607 (1; 58 mm) Chestatee River at and alongside GA Hwy 60, 3.7 km S of Dahlonega, 30 October 1996 ; GMNH 21602 (106; 29–76 mm) Chestatee River at County Route 41, 8.8 km E of Dahlonega, 13 November 1996 ; UAIC 15066.01 (5, 41– 74 mm) same data as GMNH 21602 ; AUM 47715 (5, 46– 58 mm) same data as GMNH 21602 ; UT 91.7859 (5, 40– 69 mm) same data as GMNH 21602 ; GMNH 21603 (21; 26–85 mm) Chestatee River upstream of County Route 190, 10.3 km NE of Dahlonega, 13 November 1996 ; UF 172181 (4, 39– 72 mm) same data as GMNH 21603; CU 94240 (4, 39– 66 mm) same data as GMNH 21603 ; GMNH 21604 (3; 47–64 mm) Chestatee River at Turners Corner, junction of US Hwys 19 and 129, 14.8 km NE of Dahlonega , 16 November 1996 ; GMNH 21605 (11; 49–65 mm) Chestatee River at County Route 134, 16 November 1996 ;. White / Habersham counties: GMNH 21580 (11; 30–68 mm) Chattahoochee River at GA Hwy 115, 14 September 1996 ; GMNH 21596 (7; 33–42 mm) Chattahoochee River between GA Hwys 115 and 384, 13 October 1996 ; GMNH 21608 (56, 37– 82 mm) Chattahoochee River 1.6 km downstream GA Hwy 115, 24 October 1963 . White County: GMNH 21588 (6; 52–65 mm) Sautee Creek at GA Hwy 255, 14 September 1996 .

Additional material examined (nontypes).

Upper Flint River system, Georgia: Meriwether/Pike counties: AUM 6635 (12) Flint River , 6.4 km ESE of Gay, hardtop road, 18 March 1971 . AUM 24790 (1) Flint River between Flat Shoals and GA Hwy 18 ( R.M. 280), W of Molena, 21 June 1984 . Meriwether/Upson counties: AUM 24799 (319) Flint River 17.1 km WNW of Thomaston at Pleasant Valley ( R.M. 270), 13 June 1984 . Upson County : AUM 6959 (11) Potato Creek at State Rt. 36, 11 September 1971 ; TU 27520 (25) same locality, 23 April 1962 ; GMNH 21554 (92) Potato Creek between GA Hwys 74 and 36, 4.3 km W of Thomaston, 25 June 1994 ; GMNH 21553 (11) same locality, 27 May 1994 . Upson/Talbot counties: GMNH 22061 (2) Flint River at Sprewell Bluff State Park, 15 km W of Thomaston, 1 July 2002 ; AUM 6903 (7) Flint River at Adam's Island, 3.4 km W of mouth of Potato Creek, 13.7 km SSW of Thomaston, 17 June 1971 ; AUM 24802 (3) Flint River 12.9 km SSW of Thomaston, 1.6 km below Pobiddy Bridge, 20 June 1984 ; AUM 6882 (7) Flint River at Pasley Shoal, 14.5 km W of Thomaston, 16 June 1971 ; AUM 17272 (4) Flint River ca. 5 km below GA Hwy 36, 9.2 km E of Pleasant Hill, 25 June 1978 ; AUM 24789 (269) Flint River 11.7 km WSW of Thomaston, above GA Hwy 36 ( R.M. 260), 12 June 1984 ; AUM 6966 (4) Flint River , 10 km SW of Thomaston, GA Hwy 36, 14 September 1971 ; GMNH 21555 (4) Flint River at Big Lazer WMA, 28 April 1995 . Talbot County : GMNH 21556 (14) Lazer Creek at GA Hwy 36, 17 September 2001 . Taylor/Upson counties: AUM 24815 (1) Flint River 18.5 km SSE of Thomaston, 2.4 and 10.5 km above US Hwy 80 ( R.M. 240), 26 June 1984 ; Taylor/Crawford counties: AUM 24829 (1) Flint River 21.2 km NNW of Reynolds, 3.2 km above GA Hwy 128 ( RM 230 ), 19 June 1984 .

Lower Flint River system, Georgia: Baker County: GMNH 13418 (24) Ichawaynochaway Creek immediately downstream of dam at upstream boundary of J. E. Jones Ecological Research Center Property, 18 October 1990 ; GMNH 21568 (4) same locality, 7 January 2000 ; GMNH 21569 (2) & GMNH 21570 (2) same locality, 16 June 2000 ; GMNH 13430 (2) Ichawaynochaway Creek 4 km N GA Hwy 91, 16 October 1990 ; GMNH 13470 (17) Ichawaynochaway Creek downstream from County Route 25, 0.3 km NE of GA Hwy 91 crossing, 18 October 1990 ; GMNH 13483 (7) Ichawaynochaway Creek , first shoal immediately upstream from GA Hwy 200 crossing, 19 October 1990 ; GMNH 14174 (3) same locality, 6 December 1991 ; GMNH 14225 (34) Chickasawhatchee Creek at GA Hwy 37, 5 December 1991 ; GMNH 14250 (17) same locality, 7 May 1992 ; GMNH 14275 (15) same locality, 18 May 1992 ; GMNH 21567 (8) same locality, 26 April 1995 ; GMNH 21572 (1) same locality, 3 April 2000 ; GMNH 14262 (2) Ichawaynochaway Creek between dam at upstream boundary of J. E. Jones Ecological Research Center Property and GA Hwy 200, 8 April 1992 . Lee County: AUM 12348 (4) Muckalee Creek, 6.1 km SE of Leesburg, Beverly Acres , 1 October 1973 . Lee / Worth counties: GMNH 21576 (12) Flint River about 1 km upstream from GA Hwy 32, near mouth of Philema Branch, 24 May 2004 .

Chattahoochee River system, Georgia: Lumpkin County: GMNH 22055 (2) Chestatee River at County Route 134, 16 November 1996 .

Chattahoochee River system, Alabama: Russell County : AUM 698 (1) Uchee Creek at AL Hwy 165, 1 September 1967 ; GMNH 21557 (4) same locality, 22 April 1989 ; INHS 64577 (2) same locality, 18 March 1989 ; GMNH 21559 (1) same locality, 30 July 1991 ; GMNH 21561 (5) same locality, 31 October 1997 ; GMNH 21565 (11) same locality, 22 November 1996 ; AUM 1516 (1) Uchee Creek 16.1 km SW of Phoenix City, vicinity of US Hwy 431, 4 November 1968 ; GMNH 21564 (3) same locality, 22 November 1996 ; GMNH 21563 (1) Little Uchee Creek at US Hwy 431, 23 November 1996 .

Diagnosis. Percina crypta differs from all other described species of Percina in possessing the following combination of characters: branchiostegal membranes slightly connected; preopercular margin non-serrate; premaxillary frenum well-developed; rectangular dorsal saddles usually seven, closely spaced; snout subconical, subocular bar prominent; and first-dorsal fin with yellow-orange to orange submarginal band in nuptial males and females. Nuptial males lack discrete tubercles, but exhibit tubercular ridges on the anal-fin rays and ventrally on pelvic rays. Lateral blotches rectangular, forming discrete blocks or bars, sometimes conjoined with dorsal saddles and tapering ventrally. Caudal-fin base with three vertically aligned spots or dashes that enclose two pale areas, middle mark may conjoin with small blotch at caudal base.

Percina crypta is most readily distinguished from sympatric P. nigrofasciata in having narrowly separated dark dorsal saddles, pale inter-saddle width usually less than or rarely equal to saddle width versus pale intersaddle width usually greater than saddle width; width of last two dorsal saddles always greater than pale intersaddle space versus width of last two dorsal saddles always less than inter-saddle space ( Fig. 3 ); usually a single modified scale between the pelvic bases versus two or more scales; and pectoral-fin rays strongly banded versus fins clear or pectoral rays lightly tessellated. Lateral pigmentation highly variable in P. nigrofasciata ; lateral blotches may conjoin with dorsal saddles in large adults and frequently taper ventrally onto lower side; subocular bar variably present; and nuptial males may develop a pale submarginal yellow wash suffused with pale iridescent green in the first-dorsal fin.

Description. Percina crypta is moderately large and robust with a terete body (males are larger than females, maximum size approximately 101 mm versus 85 mm SL); snout subconical, mouth subterminal; dorsal fins large, spinous-dorsal fin distinctly separate from soft-dorsal fin (more so in males than in females); caudal fin emarginate, lobes rounded; anal fin in nuptial males larger than second dorsal fin. Frequency distributions of scale and fin-ray counts of P. crypta are given in Tables 1–6; proportional measurements are summarized in Table 7. Lateral line complete with 50–68 scales, usually 55–65 (mean=59.3, S.D. =2.80, n=188); pored scales on caudal fin 0 (116 specimens), 1 (68) or 2 (7); and transverse-scale rows 14–23, averaging approximately one scale row less in females (mean=18.7, S.D. =1.42; n=104) than males (mean=20.3, S.D. =1.44, n=86). Scale rows above lateral line 5–8 in females and 6–9 in males. Scale rows below lateral line 8–14, usually 11 in females and 12 in males. Caudal-peduncle scale rows 16–25, usually 19–24, approximately one scale row less in females (mean=20.7, S.D. =1.49, n=103) than males (mean=22.0, S.D. =1.38, n=87). Dorsal spines 11–15, usually 11–13; dorsal rays 10–12; anal-fin spines 2; anal rays 7–10, usually 8–9; pectoral-fin rays 12–15, usually 12–14. Branchiostegal rays 6– 6 in 94% of examined specimens (ES). Modified midventral scales on belly 8–17 (mean=12.1, S.D. =1.69, n=87) in males; 15% of females (n=104) have 1–10 modified scales on belly. Modified scales between pelvic fin bases number 0 (1 specimen), 1 (155; 62% of males and 95% of females), 2 (35), or 3 (2). Infraorbital pores 7–9, modally 8 (49 of 55 ES); preoperculomandubular pores 9–11, modally 10 (44 of 53 ES). Vertebrae 41–42 (n=6).

Squamation variable on nape, opercle and cheek, scales usually fully or partially embedded, less frequently fully scaled or naked; breast naked in most females (64 of 65 ES), naked or with partially embedded scales in males; anterior abdomen naked in females, usually naked in males (42 of 50 ES).

Coloration in alcohol. Head countershaded and body disruptively colored with varied dark markings over pale background. Head dorsum dark in adults, snout and interorbital area dusky or darkish in juveniles; dark to black pre- and postorbital bars, former not encircling snout and latter extending across preopercle and opercle (nearly contiguous with dark midlateral stripe) forming boundary between dark upper half and pale lower half of head; skin encircling eyes dark in adults and dusky in juveniles; suborbital bar black (adults) to darkish (juveniles, small adults); upper jaw pale to dusky at angle and dark distally, tip of upper jaw and snout dark in adults, pale in juveniles; lower jaw pale at angle, tip and chin dusky; branchiostegal rays and gular area pale to dusky, sometimes with patchy dark stippling; cheek sometimes pale, usually dusky or with patchy stippling in adults, pale in juveniles; opercle darkish, ventral portion pale to dusky, margin usually dark (pale in juveniles). Body ground color pale; dorsum with seven dark rectangular saddles, first saddle entirely anterior to spinous dorsal and last saddle overlying caudal peduncle. Inter-saddle spaces usually marked with scattered dark crescents (exposed scale edges pigmented) to uniform crescent maculation in adults, dusky or flecked in juveniles. Widths of dorsal saddles nearly always greater than widths of inter-saddle spaces on nape and mid-dorsum; always greater than inter-saddle space widths on peduncle ( Fig. 3 ). Dorsolateral body with crisscross pattern formed by darkly pigmented, alternating anterior and posterior oblique scale rows; crisscross marks narrow or reduced to oblique rows of crescents or flecks in juveniles. Side conspicuously marked with 7 to 13 dark to black bars or angular blotches; bars or blotches superimpose narrow dark midlateral stripe, extending from cleithral area to caudal base; midlateral stripe less pronounced in juveniles (blotches more separated) or obscure in dark specimens; lower edges of bars or blotches dusky or darkish, sometimes tapering ventrally in females. Ventrolateral area often with dark scrawl marks, small blotches, or large flecks between bars in females, versus dusky and usually unmarked in males. Venter flecked or with patchy stippling in females versus dusky, immaculate in males; both sexes with dark midventral line (dark peritoneum visible through naked integument). Dark saddles and lateral bars may be contiguous or virtually obscured in very dark specimens.

Dorsal fin typically dusky in males and with medial dusky band in females. Second-dorsal fin dusky with light, distal band. Anal fin dusky (in males) or with a distal light colored band (females). Pelvic fins dusky in males and heavily speckled in females. Pectoral-fin rays pigmented along their length to form one or two light bands in the medial portion of the fin. Caudal fin dusky with an irregular medial band and two light-colored spots that separate three vertically aligned dark spots at base of fin.

Coloration in life. A nuptial male and female P. crypta are depicted in Figure 2 . In males, head dark bluegray to black with gold to green iridescent specks concentrated on cheek and opercle, specks less dense on posterior maxilla, below snout, and above black postorbital bar; eye dark, ventral edge orbit sometimes lightly stippled with gold-green iridescence flecks, preorbital bar black, lower jaw and chin black with bluish cast.

Lateral bars black to olive-black, suffused with iridescent green sheen; background straw to pale orange; entire side stippled with minute gold, orange, green, and lime iridescent specks. Dorsum black (saddles obscure) or black saddles evident, inter-saddle spaces amber to orange, lacking iridescent specks; breast grayblack, abdomen amber, sometimes with ventral pink tinge. In females, head dorsum dark olive-black; snout medium olive-black, sometimes with median yellow to amber mark; cheek and opercle variable, medium olive-black with yellow crescent marks (over scales), or with pale yellow mottling and stippling; iridescent gold to green specks concentrated on opercle, over cleithrum, and just above opercle spine. Eye black with yellow to amber edges, along ventral margin, divided by midventral black mark; suborbital bar obscure or black, preorbital bar obscure or black, postorbital bar black. Upper and lower jaws olive-black to pale yellow, dappled with olive-black specks and marks. Lateral bars olive-black, conjoined by small blotches and dark marks; bars taper ventrally and contrasted with pale yellow-green abdomen to pink tinged venter; lower side and abdomen speckled with olive-black marks on scales and blotches or marks on lower peduncle; lateral bars on upper side variegated by irregular pale yellow marks. Entire side with minute to tiny iridescent gold, orange, green, and lime specks, most evident when contrasted against lateral bars. Dorsal saddles dark oliveblack or black, inter-saddle spaces pale yellow to amber with black flecks or crescent marks.

In males, first-dorsal fin with distinct olive-yellow to orange marginal band that becomes submarginal posteriorly, distally bounded by diffuse olive-black edge; basal interspinous membranes black, except for small clear window behind spine base of the third to penultimate spines. Second-dorsal fin with diffuse oliveblack margin (rays yellow), wide yellow-green wash in distal third of fin (rays amber at first branch), and remaining base of fin and rays dusky black. Caudal-fin membranes dusky black with olive yellow rays, caudal base with two light yellow-olive areas. Anal fin dusky black, darkest in basal third of fin; rays with prominent apparently keratinized blue-gray ridges. Pelvic fins dusky black to olive at fin tip, ventral rays thickened and blue-gray along leading edge; pectoral-fin membranes clear, rays dusky black; rays with thickened blue-gray edges along leading and distal margins of lower fin. In females, median fins are distinctly banded. First-dorsal fin has yellow-orange marginal band that becomes submarginal posteriorly, bounded by olive-black distal edge; basal pigmentation consists of clear, dusky black, and clear irregular bands. Second-dorsal fin has irregular dusky olive and olive-black marks and clear areas on membranes and rays in distal third of fin, an uneven clear band in mid-fin, and variegated clear and dusky black pigmentation in basal third of fin. Caudal fin with alternating olive-black and clear banding on rays, membranes clear to slightly dusky along ray edges; distal ray tips black edged, with submarginal clear and olive-black bands, medial clear band, and basal third of fin rays olive-black; caudal base with two light yellow areas separated by upper, medial, and lower black marks. Anal fin with dusky green-olive pigment in basal third of membranes, rays with alternating olive-black and clear banding; clear submarginal band present in longest rays and medially across fin. Pelvic fins with yellowolive banding on rays, leading ventral ray thickened; pelvic fin banded yellow and olive-black on rays, membrane yellow, except for clear ventral edge.

Geographic Variation and Comparison to P. nigrofasciata . Meristic variation among populations of P. crypta from the lower Flint River, upper Flint River, upper Chattahoochee River and Uchee Creek systems entail differences in central tendency of scale and fin-ray counts, although count distributions broadly overlap among populations for all variables ( Tables 1–6). Uchee Creek specimens average about one to two fewer lateral-line, transverse and caudal-peduncle-scale rows than upper Flint River specimens. Specimens from the upper Chattahoochee modally have 13 dorsal spines compared to 12 in the other three populations ( Table 5). Count distributions also overlap between P. crypta and P. nigrofasciata from the lower Flint River system, although P. nigrofasciata average two to four fewer lateral-line scales ( Table 1), and modally have nine anal rays compared to eight in P. crypta ( Table 6). Lateral-line scale counts reported by Crawford (1956) for P. nigrofasciata from Apalachicola (including the Chattahoochee River system) and Flint populations also tend lower (averages = 55.3 and 54.4, respectively) than counts reported here for P. crypta . Crawford (1956) also reports a modal count of nine anal rays in P. nigrofasciata . Other counts reported by Crawford (scales above and below the lateral line, scale rows around the caudal peduncle, dorsal-fin spines and rays) for P. nigrofasciata broadly overlap with counts for P. crypta .

Morphometric analysis shows some differentiation among P.crypta specimens from the upper Flint River, lower Flint River and upper Chattahoochee River systems, based on a sheared PCA of the 22 variables listed in Table 7 ( Fig. 4 , Table 8). Males from all three populations overlap on both sheared principal components ( Fig. 4A ), although specimens from the upper Flint River system tend toward shorter pectoral and pelvic fins (PC III; Tables 7 and 8). Additionally, males from the lower Flint River system tend toward greater orbit length and greater gill-mandible length than in males from the upper Chattahoochee River system (PC II; Fig. 4A ). Females from the two Flint River populations separate primarily on the basis of body depth, head depth, and length of longest dorsal spine and anal spine (PC II; Fig. 4B ). Females from both Flint River system populations overlap on both principal components with females from the upper Chattahoochee River system ( Fig. 4B ). For all individual variables, measurement ranges broadly overlap among populations ( Table 7). We have not discerned any characters that are diagnostic of P. crypta from the separated portions of its range.

Percina crypta from all three examined populations separate from P. nigrofasciata from the lower Flint River system in the multivariate space defined by the sheared PCA ( Fig. 4 ). The measured P. nigrofasciata have greater inter-orbital width (average, range of proportional measurement: males, 47, 42–50; females 46, 44–48) than P. crypta (38–41; Table 7). Percina nigrofasciata also tends toward lesser caudal peduncle depth (males 86, 82–89; females 81, 77–86), and greater gill-mandible distance (males 142, 132–159; females 145, 122–166). Male P. nigrofasciata additionally separate from male P.crypta on the basis of greater orbit length (66, 61–70; PCII, Fig. 4A ). Female P. nigrofasciata additionally separate from female P. crypta in having lesser body depth (172, 161–184) and greater snout length (69, 64–76; PCIII, Fig. 4B ).

Comparisons. Percina crypta possesses characters that would variously align it with species in the subgenera Ericosma , Hadropterus , and Alvordius as diagnosed by Page (1974) and as described by Boschung and Mayden (2004). Percina crypta resembles P. palmaris in pigmentation, but differs from the subgenus Ericosma in lacking tubercles, having the lateral bars usually not conjoined across the dorsum, and in the firstdorsal fin of breeding males with a dark basal band, not entirely orange. Percina crypta differs from members of the subgenus Hadropterus in having narrowly connected branchiostegal membranes (versus moderately connected), nuptial coloration, pored scales on the caudal fin normally 0 to 1 (usually 0, versus 0 to 6), and in lacking distinct serrae on the preopercular margin. Percina crypta differs from the species of the subgenus Alvordius in usually possessing a narrow connection between the branchiostegal membranes (frequently separate in Alvordius ), and in having an orange submarginal band in the first dorsal fin (bright colors usually lacking in Alvordius ). Recent analysis using gene sequence data ( Near 2002) provides evidence that the subgenera Ericosma , Hadropterus , and Alvordius are not monophyletic. The same analysis ( Near 2002) has supported monophyly in five of the six remaining Percina subgenera. Percina crypta lacks the apomorphic characters defining four of these subgenera; an elongated anal fin in males of the subgenus Imostoma ; a caudal keel and lack of modified scales along the belly midline in male Odontopholis ; an elongated snout in Swainia (which also has a moderate connection between the branchiostegal membranes); a bulbous, conical snout in Percina s.s. ( Page 1974; Near 2002). Percina crypta differs from species in the subgenus Cottogaster as diagnosed by Suttkus et al. (1994) in large size, possession of a broad premaxillary frenum, and in having a connection between the branchiostegal membranes. Percina crypta differs from the single species ( P. aurantiaca ) assigned to the subgenus Hypohomus in a number of characters, including possessing a row of modified scales on the midline of the belly.

Distribution. Percina crypta is endemic to the Apalachicola River drainage, where the species occurs in the Flint River system, Georgia, and the Chattahoochee River system, Alabama and Georgia ( Fig. 1 ). Within the Flint River system, the species is known from the Flint River mainstem above and below the Fall Line (the boundary between Piedmont and Coastal Plain physiographic provinces), and in at least four tributary stream systems (Lazer Creek and Potato Creek, in the Piedmont province, and Muckalee Creek and the Ichawaynochaway Creek system in the Coastal Plain province). Within the Chattahoochee River system, P. crypta is known from two broadly separated areas, the upper portion of the system in Georgia, including the mainstem Chattahoochee River, Chestatee River, and Sautee Creek (all upstream from Lake Lanier and in the Blue Ridge province), and from the Uchee Creek system, an Alabama tributary that enters the Chattahoochee River in the Fall Line Hills district of the upper Coastal Plain ( Couch et al. 1996).

We hypothesize that P. crypta historically occurred more widely in the Chattahoochee River mainstem, prior to the construction of 13 dams on the river beginning in the mid-nineteenth century ( Couch et al. 1996). The construction of Buford Dam, which began operating in 1959, likely eliminated habitat for P. crypta in the Chattahoochee and Chestatee mainstems in what is now Lake Lanier, and modified the thermal and hydrologic regime in the Chattahoochee River downstream from the dam ( Collier et al. 1996; Couch et al. 1996). Pollutants originating in the Atlanta metropolitan area have also altered water quality in the Chattahoochee downstream from Buford Dam ( Couch et al. 1996). Shoal habitat (rocky areas that are relatively shallow with swift velocities at base-flow discharges) exists in this reach of the river; however, habitat quality for many species is impaired by altered water quality and hydrologic conditions.

Percina crypta may also have historically occurred more widely in the Coastal Plain portion of the system, prior to the removal of rock shoals from the Chattahoochee and Flint rivers. Navigation surveys conducted in 1871 and 1872 ( U.S. Congress 1874) list numerous rock shoals as impediments to navigation by steamboats in the Coastal Plain portions of both rivers. A subsequent report ( U.S. Congress 1910) documents efforts during the late 1800’s and early 1900’s to remove rock shoals, along with snags, logs and boulders, to improve navigability for steamboats in the lower portions of the Chattahoochee and Flint rivers. Channel modifications involving removing rock and “cutting through rock reefs” may have removed habitat for P. crypta . Currently, populations of P. crypta are known from shoals in the Flint River mainstem and larger tributary streams below the Fall Line ( Fig. 1 ), and the species likely also historically occurred in similar habitat in the lower Chattahoochee River prior to channel improvements for navigation. We have relatively fewer collection records from the lower Flint and Chattahoochee river mainstems ( Fig.1 ), and it is possible that future efforts will locate additional extant populations of P. crypta if appropriate habitats persist in these river reaches. We have no records of P. crypta from the Apalachicola River, formed by the confluence of the Chattahoochee and Flint rivers. We hypothesize that the species has never occurred far downstream into the Apalachicola River, where shoal habitat is restricted to the upstream-most portion of the channel.

Habitat and ecology. Percina crypta inhabits relatively swiftly flowing areas over bedrock or a mixture of coarse (boulder to gravel) bed sediments ( Hill 1996; Marcinek 2003). The species is frequently associated with the aquatic macrophyte Podostemum ceratophyllum ( Marcinek 2003) . All of our observations of P. crypta have been in shoal habitats, which contrasts markedly with the broader range of stream habitats (pools, runs and riffles) occupied by the co-occurring congener P. nigrofasciata .

Hill (1996) studied life history aspects of P.crypta in the upper Flint River system. Percina crypta in this population consumed aquatic insect larvae, including Diptera, Ephemeroptera and Trichoptera. Spawning occurred during April and May, based on observations of gonad condition, when mean monthly water temperatures were 18 to 20 o C. Individuals reached sexual maturity at age one or two; maximum observed age was estimated at three years based on otolith examination.

Percina crypta occurs commonly and abundantly in shoals of the upper Flint River, above the Fall Line, where the species often outnumbers P. nigrofasciata in catch data ( Marcinek 2003; GMNH records). This does not appear to apply to populations in the lower Flint system or in the Chattahoochee system, where known localities are fewer and P. crypta are generally outnumbered by P. nigrofasciata in collections. Frequently associated species include Cyprinella callitaenia in the upper Flint system, C. venusta , Luxilus zonistius in the upper Chattahoochee system, Ameiurus brunneus , Noturus leptacanthus , Micropterus cataractae , Lepomis auritus , Etheostoma swaini in the lower Flint and Uchee Creek systems, and P. nigrofasciata .

Conservation status. Percina crypta is considered vulnerable by Warren et al. (2000) and Jelks et al. (2008). Within Alabama, P. crypta is considered a species of highest conservation concern ( Johnston & Kuhajda 2002), and the species is listed as threatened under the State of Georgia Endangered Wildlife Act. Percina crypta is considered imperiled primarily because of its fragmented distribution within a range limited to two river systems. As discussed above, P.crypta may have been extirpated from a major portion of the species’ historic range in the Chattahoochee River mainstem as a result of dam construction and operation, and channel modification to support navigation. The status of the P. crypta population in the Uchee Creek system is uncertain; this population is isolated from all others by mainstem dams on the Chattahoochee River upstream and downstream of the mouth of Uchee Creek. Distribution of P. crypta in the Flint River system is also fragmented by two mainstem dams located below the Fall Line. Upstream from these dams, shoals in the Piedmont portion of the upper Flint River mainstem appear to support the largest populations of P. crypta .

Present threats to the persistence of P. crypta primarily consist of effects of urban growth on stream hydrology and water quality, particularly in the north Georgia mountains and in the vicinity of the Atlanta Metropolitan area. The Flint River headwaters originate in Atlanta, and population growth in the upper Flint River system is expected to place increasing demands on the river system for water supply and waste assimilation. Similarly, population growth in the Blue Ridge province of North Georgia will affect water availability and quality in the Chattahoochee River headwaters.

Etymology. From the Greek root crypt, meaning hidden or concealed, in reference to the close similarity in appearance between P. crypta and the co-occurring congener, P. nigrofasciata . The common name refers to the striking black and orange coloration of nuptial individuals, colors associated with the Western celebration of Halloween (a Celtic festival adopted by the Roman Catholic Church as the eve of All Saints Day).

Figures

FIGURE 1. Distribution of Percina crypta (closed circles), Chattahoochee and Flint River systems, Apalachicola River drainage (inset), USA. Type locality indicated with a star. Circles indicate examined collections of Percina. Percina nigrofasciata is known from all plotted localities; P. crypta has been found only at the localities marked with closed circles.

FIGURE 2. Nuptial adults of Percina crypta and Percina nigrofasciata. A: P. crypta male, 68 mm SL; B: P. crypta female, 65 mm SL; C: P. nigrofasciata male, 77 mm SL; D: P. nigrofasciata female, 66 mm SL. Chattahoochee River at GA Hwy 17/75 at Nacoochee, White County, Georgia, 5 April 2000; except P. crypta female: Chickasawhatchee Creek (Lower Flint River system) at GA Hwy 37, Baker County, Georgia, 3 April 2000. Photos by NMB.

FIGURE 3. Dorsal saddle pigmentation differences between Percina crypta (left) and Percina nigrofasciata (right); note diagnostic difference in widths of pale inter-saddle spaces between P. crypta (usually less than saddle width) versus P. nigrofasciata (usually greater than saddle width). Illustration by Bricelyn H. Strauch based on photographs of dorsally oriented live specimens from the type locality, and preserved specimens from GMNH 14225 (Lower Flint River system).

FIGURE 4. Sheared principal components analysis of morphometric variation among Percina crypta from the upper Chattahoochee River system (open circles), the upper Flint River system (open squares), and the lower Flint River system (open diamonds), and Percina nigrofasciata from the lower Flint River system (filled squares). Males (n=46, A) and females (n=49, B) were analyzed and plotted separately.