2010 SOUTHEASTERN NATURALIST 9(1):119–128 Life-history Aspects of Fundulus stellifer (Southern Studfish) (Actinopterygii: Fundulidae) in Northern Georgia Kerstin L. Edberg1 and Steven L. Powers2,* Abstract - Fundulus stellifer (Southern Studfish) was studied using 12 monthly collections from Moore’s Creek near Waleska, GA. Female specimens ranged from 24.72 mm to 94.29 mm SL and up to 15.90 g total weight. Male specimens ranged from 27.81 mm to 96.07 mm SL and up to18.60 g total weight. Increases in size were greatest in spring and summer. Maximum age appears to be just over 2 years for both sexes, as the oldest specimens collected were estimated at 28 months of age. Spawning likely occurred between late April and early June. Sexual maturity was reached in the second full spawning season for both sexes, with females having a maximum ova diameter of 2.75 mm. The primary diet of F. stellifer consisted of Chironomidae, Branchiopoda, and unidentified insect parts, with other food items seasonally. The amount of food consumed is greatest in the months preceding and during spawning and lowest during fall and early winter. Introduction Fundulus stellifer Jordan (Southern Studfish) is endemic to the Mobile Basin (Boschung and Mayden 2004), a member of the subgenus Xenisma, and most closely related to Fundulus catenatus Storer (Northern Studfish) and Fundulus bifax Cashner et al. (Stippled Studfish) (Ghedotti et al. 2004). Fundulus stellifer is relatively abundant within its range, which includes small to medium streams in the Coosa and Alabama River systems of northeast Georgia and central to south Alabama. It is easily recognized by the superior placement of the mouth, random placement of brown to dark red spots on the side of the body, and black pigmentation along the distal margin of the caudal fin, a characteristic seen only in sexually mature males (K.L. Edberg, pers. observ.; Thomerson 1969). There is little information regarding the biology of F. stellifer except for a diet that may consist of snails and other mollusks, as suggested by the structure of the pharyngeal teeth of adults (Thomerson 1969). The spawning season is reported to occur between April and May, as indicated by the capture of nuptial males during these months (Boschung and Mayden 2004). The primary objective of this study was to investigate aspects of the life history of F. stellifer. Study Area Fundulus stellifer were collected from Moore Creek, just upstream of its confl uence with Shoal Creek (34.3240°N, 84.5636°W), outside of Waleska in Cherokee County, GA (Fig. 1). Moore Creek is characterized as a 1Department of Biology, Western Kentucky University, Bowling Green, KY 42101. 2Biology Department, Roanoke College, Salem, VA 24153. *Corresponding author - powers@roanoke.edu. 120 Southeastern Naturalist Vol. 9, No. 1 second-order tributary to the Etowah River, with wetted widths ranging from 3.1 m to 6.4 m and depths <1.0 m at base-fl ow water levels. The substrate of Moore Creek is primarily gravel to cobble with sporadic bedrock in riffl es, gravel to sand in runs, and silt and sand in pools. Fundulus stellifer were collected primarily from backwaters with heavy vegetation cover. Upstream of the study area, the Moore Creek watershed is mostly forested with moderate agricultural use and sparse residential development. Water temperatures during the study period ranged from 5 °C in November 2006 to 25 °C in June and August 2006. Species richness of fishes within the study area was 34 species from nine families collected between 2004 and 2007. A complete list of species collected from Moore Creek near its Shoal Creek confl uence can be found in O’Kelley and Powers (2007). Methods Fundulus stellifer were collected using a Smith-Root model 24 backpack electrofisher and a 3.3-m x 1.3-m seine with 9.5 mm mesh during twelve mid-monthly surveys from February 2006 to January 2007. A total of 240 specimens was collected, preserved in 10% formalin in the field, rinsed with water after two weeks, and transferred into 70% ethyl alcohol (EtOH) for long-term storage in the University of Alabama Ichthyological Collection (UAIC 15028–15039). Standard length (SL) of each specimen was measured using digital calipers and recorded to the nearest 0.01 mm. Specimens were blotted dry, and their total weight (TW), eviscerated weight (EW) and gonad weight (GW) were measured using a digital analytical scale and recorded to the nearest 0.001 g. Sexual size dimorphism was not detected using a Figure 1. Map of Fundulus stellifer study area in Moore Creek (34.3240°N, 84.5636°W), near Waleska in Cherokee County, GA. 2010 K.L. Edberg and S.L. Powers 121 two-sample t-test of SL (t = 1.3, P = 0.18); therefore, statistical analyses were performed without regard to sex, except for those involving GSI, in which sexes were treated separately. All statistical analyses were carried out using Data Desk 6.0 (Data Description, Inc., Ithaca, NY) with alpha values for all tests set at 0.05. Standard deviation is abbreviated as SD. To illustrate age and growth patterns, SL was plotted against month. Gaps of >10 mm SL in specimens from a given month were considered indicative of age groups (e.g., for October all specimens were 29.35–44.79, 64.56–75.58, and 90.7–93.52 mm SL, with each cluster lacking gaps <10 mm) (Fig. 2). Due to the high GSI values seen in April and May, spawning is assumed to occur in May for estimating the age of individuals. Specimens <12 months of age were counted as age 0+, specimens between 12 and 23 months were counted as age 1+, specimens >24 months were counted as age 2+. Regressions by least sum of squares were performed for SL and EW to describe length-weight relationships of F. stellifer. The anterior one-third of the gastrointestinal track was opened and the contents were removed, weighed to the nearest 0.001 g on a digital analytical scale, and placed in 70% EtOH for storage. Empty guts were recorded with a weight of zero. One-way analysis of variance (ANOVA) was performed on the weight of gut contents among different months to test for differences in amount of food consumed throughout the year. Taxa richness of gut contents was the total number of different food items in each specimen. An ANOVA was performed on the taxa richness of gut contents per month to test for dietary changes throughout the year. To test for infl uence of body size on the amount of food consumed, gut content weight was regressed with SL by least sum of squares. Taxa richness of gut contents was also regressed with SL to test for an infl uence of size on food items. Food items were enumerated and identified to the lowest possible taxonomic category following Thorp and Covich (1991) and Merritt and Cummins (1996). Due to mastication by the pharyngeal teeth, food items were often unidentifiable past the level of family, order, or class. Presence/absence of sand, detritus, and unidentifiable insect parts were recorded. Niche breadth (NB) was calculated following Levins (1969) to compare diet of F. stellifer to that of Hypentelium etowanum Jordan (Alabama Hog Sucker) and Notropis xaenocephalus (Jordan) (Coosa Shiner), two syntopic species for which diet has been documented within the study area. Values approaching 1.0 are considered generalists, while smaller values indicate a more specialist diet. Reproductive condition was investigated by calculating gonadosomatic index (GSI = GW/EW). A one-way ANOVA was performed to test mean differences in GSI among months for both males and females. In gravid females, greatly enlarged (≈2 mm in diameter), fully yolked, mature oocytes were counted, and five representatives were measured using digital calipers to provide an approximation of oocyte size and number (Heins and Baker 1988). Smaller oocytes (<0.5 mm in diameter) were also counted and measured to provide a size comparison to mature oocytes. Oocyte counts include all oocytes (small and large) present in the ovary of the female. Oocyte identification procedures can be found in Heins and Baker (1988). Regressions 122 Southeastern Naturalist Vol. 9, No. 1 were performed to test for an infl uence of SL on the number of mature and immature oocytes present in a mature female. Results The smallest specimen collected during this study was taken in July, measuring 16.3 mm in length and weighing 0.076 g TW. The gender of this specimen could not be determined. The largest male and female specimens were 96.07 mm and 18.6 g and 94.29 mm and 15.9 g, respectively. The earliest young-of-year capture occurred in June, with specimens ranging from 17.66 to 19.08 mm (mean = 18.38, SD = 1.0). For all collections used in this study, female to male ratio was 0.78:1, but no sexual size dimorphism was detected in regards to SL (P = 0.1813) and EW (P = 0.0742). Standard length increased with age in months (R2 = 0.92, P < 0.001), and preliminary examination of the data suggested a curvilinear relationship existed between SL and EW. We therefore log-transformed EW before regressing it with SL. This analysis produced a significant linear relationship (R2 = 0.95, P < 0.001). Increases in SL were greatest among months in the spring (Fig. 2). Of the 240 specimens collected, 52% were age 0+, 39% were age 1+, and 9% were age 2+ age. The maximum ages of males and females were 28 and 26 months, respectively. The diet of F. stellifer consisted mostly of Chironomidae larvae (68% of total diet by number) and Branchiopoda (12% of total diet) (Table 1). Levins’ NB value for F. stellifer was 0.0996. Data for Hypentelium etowanum and Notropis xaenocephalus from recently published life-history studies (Jolly and Powers 2008; O’Kelley and Powers 2007) produced NB values of 0.0791 and 0.275, respectively. Unidentified insect parts were in 22% of the individuals examined, and sand particles were in 27%, peaking in June and Figure 2. Standard length (SL) in mm by month of collection (1 = January, 2 = February, etc.) for Fundulus stellifer collected from Moore Creek, Cherokee County, GA between February 2006 and January 2007. 2010 K.L. Edberg and S.L. Powers 123 September (present in 11 individuals each month). Feeding was not uniform throughout the year (F = 5.8, P < 0.001), with February and May having the greatest weight of gut contents. Much of the variation in the weight of gut contents can by explained by SL (R2 = 0.37, P < 0.001). There was also a significant relationship between SL and taxa richness of food items (P = 0.027), but the relationship had low explanatory power (R2 = 0.02). Taxa richness of food items did not differ statistically (F = 1.79, P = 0.06) throughout the year. Collembola, Hirudinae, and Physa sp. made up moderate to large components of the diet in winter, spring, and summer respectively, but were largely absent from other seasons. The greatest proportion of empty guts occurred in November, with a total of 41% of individuals lacking food items that month (Table 1). Female GSI peaked in May (mean GSI = 0.0974, maximum GSI = 0.211; Fig. 3). Gravid females were collected in March, April, and May, with oocytes ranging in size from 0.34 mm to 2.75 mm. The number of oocytes present in a female was significantly related to SL (R2 = 0.58, P = 0.010), with a mean of 308 mature oocytes found in gravid females (SD = 139.5). The youngest age at sexual maturation for females was 22 months. The highest mean GSI value for males (0.0025) was in February, and maximum male GSI (0.0045) was in April (Fig. 4). Elevated GSI values were found in males collected from February through July. Discussion Our study suggests that specimens <1 year of age are generally <40 mm SL, those 1+ are generally 45–80 mm SL, and those 2+ are 85–95 mm SL. Maximum age appears to be just over 2 years. Growth appears to increase in Figure 3. Gonadosomatic Index (GSI) by month of collection (1 = January, 2 = February, etc.) for female Fundulus stellifer collected from Moore Creek, Cherokee County, GA between February 2006 and January 2007. 124 Southeastern Naturalist Vol. 9, No. 1 Table 1. Stomach contents of Fundulus stellifer from Moore Creek, Cherokee County, GA by month. Numbers for each food item indicate total number of individuals found. Detritus, sand, and unidentified insect parts are noted by occurrence due to difficulty quantifying them (e.g., the occurrence of sand in three specimens is denoted as “3”). Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total Proportion Sand 3 8 2 4 11 8 3 11 3 2 6 61 Detritus 6 8 6 7 16 13 8 8 12 3 3 5 95 Unidentified parts 6 2 2 4 8 7 3 5 4 3 1 4 49 Nematoda 8 8 0.003 Trematoda 1 1 0.000 Mollusca Ancylidae 1 1 0.000 Physa 2 5 4 12 16 2 41 0.015 Annelida Oligochaeta 1 1 0.000 Hirudinae 18 16 1 35 0.013 Arachnida Araneae 1 1 2 0.001 Acari 2 1 3 0.001 Crustacea Branchiopoda 5 21 45 68 14 10 19 16 73 21 34 2 328 0.122 2010 K.L. Edberg and S.L. Powers 125 Table 1, continued. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total Proportion Parainsecta Collembola 20 10 3 1 50 84 0.031 Insecta Hemiptera 1 1 0.000 Megaloptera 5 5 0.002 Ephemeroptera 3 4 1 3 18 2 4 1 11 3 50 0.019 Plecoptera 1 1 1 3 8 14 0.005 Coleoptera 1 2 1 4 0.001 Colepotera Larvae 1 2 3 0.001 Hymenoptera 5 2 7 0.003 Trichoptera 1 5 2 3 1 1 13 0.005 Diptera 1 1 0.000 Chironomidae larvae 6 168 62 545 221 356 126 85 109 29 12 103 1822 0.679 Unidentified pupae 1 4 5 15 9 1 5 1 2 43 0.016 Unidentified adults 3 1 2 2 1 2 1 12 0.004 # of specimens 21 20 16 20 19 21 14 19 16 21 17 17 221 Empty stomachs 6 1 1 1 0 0 1 0 1 8 7 3 29 % Empty 28.57 5.00 6.25 5.00 0.00 0.00 7.14 0.00 6.25 38.10 41.18 17.65 13.12 Total items 57 235 134 663 301 428 180 150 216 78 61 181 2684 Mean items/stomach 5.2 16.8 16.8 66.3 25.1 35.7 20.0 10.7 21.6 8.7 6.1 20.1 107.4 126 Southeastern Naturalist Vol. 9, No. 1 spring and level off in fall. This pattern coincides with the feeding patterns of F. stellifer addressed below. The low proportion of age 2+ individuals suggests that few individuals survive to reproductive maturity. This high mortality prior to sexual maturity has been observed in other syntopic species such as Hypentelium etowanum (O’Kelley and Powers 2007) and Notropis xaenocephalus (Jolly and Powers 2008), and is common among stream fishes (Boschung and Mayden 2004). The low proportion of specimens <25 mm SL collected could be explained by these small individuals passing through the 9.5-mm mesh of the seine. Sexual dimorphism appears to be restricted to male specimens developing a black band on the distal caudal margin as they approach sexual maturity. Increased feeding during spring and summer months appears to coincide with increased energetic requirements associated with gamete production, growth, and spawning. The high numeric proportion of food items as Chironomidae and Branchiopoda (67.9% and 12.2%, respectively) may indicate selective feeding, as these data contrast with the diets of the syntopic Hypentelium etowanum and Notropis xaenocephalus in which Chironomidae made up 88% and 19.5%, and Branchiopoda 3.59% and 0% of their food items, respectively (Jolly and Powers 2008, O’Kelley and Powers 2007). Contrasting NB values (especially between F. stellifer and N. xaenocephalus) provide further evidence of this selective feeding. The high number of Chironomidae in the diet of F. stellifer may also be explained by densities often >50,000 Chironimidae/m2 in aquatic environments (Coffman and Ferrington 1996), making them the most abundant food resource available and the bulk of the diet of most stream fishes (Alford and Beckett 2007, Matthews et al. 1982). The significant relationship between weight of gut contents and SL suggests Figure 4. Gonadosomatic Index (GSI) by month of collection (1 = January, 2 = February, etc.) for male Fundulus stellifer collected from Moore Creek, Cherokee County, GA between February 2006 and January 2007. 2010 K.L. Edberg and S.L. Powers 127 that the amount of food consumed increases with size. The significant relationship between taxa richness of food items and SL also suggests that diet may become more varied in larger specimens, but the low explanatory power value (R2 = 0.02) suggests these changes are slight. Taxa richness of food items does not appear to change throughout the year, but certain taxa appear to be seasonally important. Weight of gut contents differs throughout the year, with elevated levels from late winter through summer. This variation coincides with increased energetic requirements during periods of spawning and increased growth. The near restriction of Collembola, Hirudinae, and Physa sp. as food items to specific seasons may be evidence of selective feeding, but may also be explained by several relatively simple phenomena making them widely available as prey items during specific times of year. As Collembola are mostly soil dwelling or semi-aquatic (Merritt and Cummins 1996), the rising waters of winter may fl ood previously dry riparian areas, making them easy prey during winter months. Hirudinae are largely annual with eggs hatching in the spring (Thorp and Covich 1991), which may provide an abundance of small leaches as prey items at that time of year. Physa sp. may be largely absent from Moore’s Creek except for summer months because pulmonate snails are known to burrow into the substrate during cold periods (Thorp and Covich 1991). Therefore, these data do not provide clear evidence of F. stellifer selecting these specific items during particular seasons, but rather may identify F. stellifer as opportunistic in feeding on whatever food items are present in their habitat. The GSI of both males and females was highest in spring, but males displayed high GSI values earlier and later in the year compared to females. Elevated female GSI in April and May, followed by a precipitous drop in female GSI in June indicates spawning occurs in April and May, with water temperatures near 16 °C, and is finished by mid-June. One specimen collected in July appeared to be spent, as there were no oocytes and only a membrane remaining. Mature oocytes were not observed in females younger than 22 months of age. Subsequently, spawning does not appear to occur until the second spring of life. We found that maximum female lifespan was approximately 26 months. Thus, it appears that F. stellifer will go through no more than one spawning cycle during their lifespan. No spawning activity was observed, but during the spring months, specimens were collected from backwater pools with heavy submerged and emergent macrophyte vegetation, suggesting spawning habitat is likely similar to that inhabited during the rest of the year. Young-of-year were also collected primarily from these same areas of heavy vegetation or leaf cover. Although F. stellifer and F. catenatus are closely related, they appear to have several differences in life-history traits. Fundulus catenatus has a longer lifespan (5+ years) and larger maximum size (nearly 115 mm SL) (Fisher 1981), compared to F. stellifer (just over 2 years and maximum SL of less than 95 mm). Fundulus catenatus also has a longer spawning period which lasts from April to July (Fisher 1981), whereas the F. stellifer spawning period occurs from April to early June. Fundulus stellifer appears to be more 128 Southeastern Naturalist Vol. 9, No. 1 fecund with a mean of 308 mature oocytes in gravid females and a maximum ova diameter of 2.75 mm, while F. catenatus have 28–245 mature oocytes and larger ova up to 3.5 mm (Fisher 1981). Diptera larvae also appear to make up the bulk of the diet of both species. Seasonal shifts in food items appear common between the two species, but differences in pharyngeal teeth morphology (Thomerson 1969) is likely adaptive, as the more molariform teeth of F. stellifer likely crush mollusks such as Physa. Acknowledgments We thank S.D. Barton, D.S. Holder, and C.K. Ray for assistance with field and lab work. Fishes were collected under Georgia Scientific Collecting Permit number 16494 issued to S.L. Powers. We thank Reinhardt College for the use of equipment essential to this research. This study was conducted in part as an undergraduate independent research project at Reinhardt College by K.L. Edberg. Literature Cited Alford, J.B., and D.C. Beckett. 2007. Selective predation by four darter (Percidae) species on larval chironomids (Diptera) from a Mississippi stream. Environmental Biology of Fishes 78:353–364. Boschung, H.T., and R.L. Mayden. 2004. Fishes of Alabama. Smithsonian Institution. Washington, DC. 736 pp. Coffman, W.P., and L.C. Ferrington. 1996. Chironomidae. Pp. 635–754, In R.W. Merritt and K.W. Cummins. (Eds). An Introduction to the Aquatic Insects of North America. Third Edition. Kendall/Hunt Publishing Co., Dubuque, IA. 862 pp. Fisher, J.W. 1981. Ecology of Fundulus catenatus in three interconnected stream orders. American Midland Naturalist 106(2):372–378. Ghedotti, M.J., A.M. Simons, and M.P. Davis. 2004. Morphology and phylogeny of the studfish clade, subgenus Xenisma (Teleostei: Cyprinodontiformes). Copeia 2004(1):53–61. Heins, D.C., and J.A. Baker. 1988. Egg sizes in fishes: Do mature oocytes accurately demonstrate size statistics of ripe ova? Copeia 1988(1):238–40. Jolly, D.M., and S.L. Powers. 2008. Life-history aspects of Notropis xaenocephalus (Coosa Shiner) (Actinopterygii: Cyrpinidae) in northern Georgia. Southeastern Naturalist 7(3):449–458. Levins, R. 1969. Some demographic consequences of environmental heterogeneity for biological control. Bulletin of the Entomological Society of America 15:237–240. Matthews, W.J., J.R. Bek, and E. Surat. 1982. Comparative ecology of the darters Etheostoma podestemone, E. fl abellare, and Percina roanoka in Upper Roanoke River Drainage, Virginia. Copeia 1982(4):805–814. Merritt, R.W., and K.W. Cummins. 1996. An Introduction to the Aquatic Insects of North America. Third Edition. Kendall/Hunt Publishing Co., Dubuque, IA. 862 pp. O’Kelley, C.T., and S.L. Powers. 2007. Life-history aspects of Hypentelium etowanum (Alabama Hog Sucker) (Actinopterygii: Catostomidae) in North Georgia. Southeastern Naturalist 6(3):479–490. Thomerson, J.E. 1969. Variation and relationship of the studfishes Fundulus catenatus and Fundulus stellifer (Cyprinodontidae, Pisces). Tulane Studies in Zoology and Botany (16)1:1–19. Thorp, J.H., and A.P. Covich. 1991. Ecology and Classification of North American Freshwater Invertebrates. Academic Press, Inc., San Diego, CA. 911 pp.