Life history, food habits and production of Progomphus obscurus Rambur (Odonata: Gomphidae) in Harmon Creek of east Texas.
Progomphus obscurus Rambur is a burrowing dragonfly species, which ranges over much of the southern United States. In general, dragonfly naiads of the genus Progomphus are burrowers found primarily in shifting sandbars of both lotic and lentic habitats (Needham & Westfall 1955). Analysis of microhabitat preference of P. obscurus naiads indicates a preference for areas of sand with particle sizes ranging from 0.625-1.0 mm (Huggins & DuBois 1982). Their front and middle legs are used for burrowing into the sand, and they burrow completely beneath the surface, although never deeper than 2 cm (Huggins & DuBois 1982).
In general, odonates are of tropical evolutionary origin and have evolved different life cycle patterns and cold resistant stages as they adapted to temperate regions (Norling 1984). These adaptations produced two general dragonfly life history patterns described by Corbet (1954) as spring species and summer species. These two patterns characterize many dragonfly species, but a continuum of life history patterns between the extreme spring and summer species has been found, especially in southern temperate latitudes (Paulson & Jenner 1971). Life history patterns can also vary within species at different latitudes. For example, Kormondy & Gower (1965) found Epitheca cynosura to be semivoltine (bivoltine) in Pennsylvania, while Benke & Benke (1975) found the same species to be a univoltine spring species in South Carolina. Life cycles can also vary among individuals within the same stream. Ferreras-Romero (1997) found Boyeria irene to be mainly semivoltine, but some naiads required three years to complete development in the Sierra Morena Mountains of Spain.
While life history patterns have been worked out for some odonate species, the life cycle, food habits, and production of P. obscurus is still relatively unknown. Therefore, the objectives of this study were to elucidate the general life history pattern of P. obscurus, gather information about food habits of the naiads, and to estimate P. obscurus production in Harmon Creek.
MATERIALS AND METHODS
Naiads of Progomphus obscurus were collected from Harmon Creek in Walker County, Texas. The research site was a section of stream with a sandy bottom passing through eastern Texas piney woods. Monthly collections of 20 Surber samples were made from November 1995 through May 1997. Adults were collected during the same period using an aerial insect net. Both adults and naiads were preserved in 5% formalin.
The head capsule width (HCW) of naiads was measured at the widest point to the nearest 0.01 mm using a dissection microscope with an ocular micrometer. Specimens were then dried at 55[degrees]C for 24 hours and weighed. Head capsule width and body weight data were then In-transformed and a linear regression analysis of body weight on head capsule width was performed.
Because P. obscurus does not have a tightly synchronous development, production was estimated using the size-frequency method originally described by Hynes (1961) and modified by Hamilton (1969) and Benke (1979). Standing stock biomass was measured, and production/biomass (P/B) ratio was calculated.
Food habits were determined by stomach content analysis. Stomachs of both naiads and adults were dissected and their contents removed. Food organisms contained within each stomach were then identified to the lowest taxonomic level possible, by using either a dissecting microscope, or if higher magnification was necessary by mounting the prey item on a slide for identification using a compound microscope.
The number of instars of P. obscurus was determined by a combination of two methods. For the first method, which was used to determine the number of early instars, eggs were collected from copulating females that were captured in the field. The collected eggs were incubated in the laboratory at 22[degrees]C until they hatched. After hatching, each naiad was placed in a separate petri dish, where the exuviae were collected after each molt. Larvae were raised until they reached the 6th instar. The size-frequency method was the second method used to determine the number of instars of P. obscurus. This method was not very effective for separating early instars (1st through 5th), but was very effective for identifying later instars (6th through 11th). A [chi square] test was used to determine if there were significant differences in sex ratios of cast exuviae and adults collected at streamside.
Progomphus obscurus primarily exhibits a univoltine life cycle; first instar naiads were collected from May through early September at Harmon Creek. May, June, and July were the months of greatest recruitment. Penultimate naiads were first collected during late February. Emerging adults were first observed on 13 April 1996 and on 20 April 1997, which was also the first days that exuviae of penultimate naiads were collected. Exuviae were last collected on 2 October 1996. Adults were observed in the area from mid April until late October, with oviposition occurring from early May through late August.
The sex ratio of exuviae indicated that significantly ([chi square]=7.166, 0.001 < p < 0.01) more females (57%) than males (43%) emerged (207 female and 156 male, n=363). However, the sex ratio of adults captured at stream side showed that there were significantly ([chi square] = 4.546, 0.01 < p < 0.05) more males (61%) than females (39%) (59 male and 38 female, n = 97) present.
Just prior to emergence, naiads crawled onto sandy beach areas, dug their claws into the sand to anchor themselves, and began to emerge. The distance traveled from water's edge to emergence spot ranged from 5 to 48 cm, with a mean distance of 19 cm. Head capsule width of cast exuviae ranged from 4.49 mm to 5.02 mm, with a mean width of 4.66 mm (n = 358).
Incubation time for eggs collected in the field and hatched in the laboratory ranged from 8 to 12 days. Mean head capsule widths for laboratory raised 1-5 instar naiads are shown in Table 1. The combination of data from laboratory-raised naiads and from size frequency distributions done on field-collected naiads indicates that P. obscurus produced 11 instars in Harmon Creek (Table 1 and Fig. 1).
The primary food items consumed by P. obscurus naiads at Harmon Creek were chironomid larvae, followed by mayfly naiads of the families Caenidae and Baetidae. Small numbers of rotifers, ceratopogonids, and other odonates, were occasionally ingested. Chironomids were the primary food items for all P. obscurus size classes, but larger larvae had larger numbers of mayflies in their diets. During the months of May through September, greater numbers of mayflies were consumed than during other months. The number of individuals with food in their stomachs was variable throughout the year. The fewest individuals with food in their stomachs (<30%) were found during the months of November through February. During March, September and October 37-42% of individuals had food in their guts; Whereas, between the months of April and August over 60% of P. obscurus individuals had food in their guts.
Linear regression analysis showed a significant relationship between In HCW and In dry weight (W) (F = 278.63, P = 0.0001, [R.sup.2] = 0.952). The linear relationship between HCW and dry weight can be explained by the equation: In W = -7.613 + 2.791 (ln HCW). The standard errors for the Y intercept and the slope were 0.200 and 0.167, respectively.
[FIGURE 1 OMITTED]
The annual production estimate for P. obscurus was 6.842 g/[m.sup.2]/yr, the standing stock biomass was 1.682 g/[m.sup.2], and the cohort production/biomass ratio (P/B ratio) was 4.067 (Table 2).
Differences in emergence times of P. obscurus are probably related to the time that the eggs were oviposited. For example, if a naiad hatched from the egg in June, it would emerge as an adult the following June; if the naiad hatched in August it would emerge as an adult the following August. However, final instar naiads begin to appear in late February, but do not emerge until April; these naiads then spend part of the month of February, March and then part of April in the final instar. This indicates that naiads may mature to the final instar in less than a complete year, but do not emerge immediately. Possibly, low water temperatures corresponding with low air temperatures delay the emergence process until the likelihood of freezing temperatures is very low.
The life history pattern of P. obscurus is quite different from that of the gomphid, Onychogomphus uncatus, in France (Schutte et al. 1998). In this French study it was found that O. uncatus had a three year life cycle which is much longer compared to the one year cycle for P. obscurus; This may be partially explained by O. uncatus usually producing 13 instars compared to the 11 of P. obscurus. The closely related gomphid in Florida, Progomphus bellei, produced mature larvae during March and emergence began in mid-April (Knopf & Tennessen 1980). This development time is similar to what was observed in this study. Much different from P. obscurus was the gomphid Lanthus vernalis, which showed mixed voltinism and lacked a clear growth pattern of its naiads, but was at least semivoltine in a cold unproductive stream in South Carolina (Folsom & Manuel 1983).
Voltinism, emergence periods, and adult flight periods are quite variable for odonates in general. Spring species, as described by Corbet (1954), emerge synchronously during the spring over a short period of time. The synchronous emergence occurs because these species spend the winter before emergence in the final instar (some in a diapause). Summer species spend the winter before emergence in earlier larval instars and then emerge asynchronously throughout the summer. Corbet (1964) noted that the degree of synchrony of emergence is inversely proportional to the number of overwintering instars in the winter preceding emergence. As observed by Benke & Benke (1975), the synchronous or asynchronous emergence patterns reflect a similar type of development (i.e. spring species develop synchronously and summer species develop asynchronously). Progomphus obscurus fits best into the category of summer species proposed by Corbet (1954). The emergence period of P. obscurus extends from mid April until mid September, with a correspondingly long flight period, but gomphids in general can fit into either life history pattern. In a stream in Spain, Onychogomphus uncatus had a brief early spring emergence period consistent with the spring pattern (Ferreras-Romero & Corbet 1995). Suhling (1995) studied two different populations of O. uncatus in the southern part of France and found that one showed the emergence pattern of a summer species and the other showed the spring species pattern. This indicates that a single species can have different life history patterns in different habitat types within the same geographic region. The flight period of P. obscurus at Harmon Creek from mid April until late October is longer than that observed in more northern areas of the U.S., such as North Carolina where the flight period extended from April through September (Paulson & Jenner 1971). Even farther north in the U.S. in Indiana, the flight period is reduced further to May through mid October (Bick 1941).
Incubation time for P. obscurus eggs collected from adults and reared in the laboratory ranged from 8-12 days, which is longer than 5-7 day incubation time for the libellulid Tramea lacerata (Bick 1951). Bick (1941) determined that the incubation period of another libellulid species, Erythemis simplicicollis, varied from 10 to 16 days.
Laboratory studies of odonates have shown that they will feed on a variety of animals including Paramecium sp. and other protozoans, Culex eggs, mosquito larvae, mayfly naiads, and amphipods (Bick 1951). In the current study, food habits of P. obscurus varied with size class, but prey were predominantly chironomids for all size classes. Although early instars were found to feed on rotifers to some extent, the majority of their prey were still chironomids. Overall, when compared to libellulid and corduliid species investigated by Benke (1978) food habits of P. obscurus were similar to those of Ladona deplanata, Epitheca sp. and Celithemis fasciata which fed mainly on immatures of Chironomidae and Ephemeroptera (mostly Caenis sp.) and to a much lesser extent on cladocerans and ostracods. The species studied by Benke (1978) were from a lentic habitat.
Production has been calculated for only a few species of odonates (Benke 1976; Dudgeon 1989). In the current study, annual production of 6.842 g/[m.sup.2]/yr seems high for a single species of predaceous invertebrate, and is roughly comparable to the combined production of three odonate species in the pond studied by Benke (1976). Using the removal-summation method, Benke (1976) found that production of L. deplanata (Libellulidae) was 2.20 g/[m.sup.2]/yr, Epitheca sp. (Corduliidae) was 1.94 g/[m.sup.2]/yr, and C. fasciata (Libellulidae) was 2.27 g/[m.sup.2]/yr. Dudgeon (1989) estimated production of two species of gomphid naiads in a Hong Kong forest stream, and found that production of Heliogompus scorpio was 0.182 g/[m.sup.2]/yr and production of Onychogomphus sinicus was 0.236 g/[m.sup.2]/yr. Total odonate production in the stream was estimated to be 1.019 g/[m.sup.2]/yr. The production of P. obscurus in this study was much higher than that found for either of the gomphids H. scorpio or O. sinicus, or the total production of odonates in the forest stream study in Hong Kong (Dudgeon 1989). One reason for the high single species production of P. obscurus may be that, based on extensive sampling of Harmon Creek, P. obscurus is the only abundant invertebrate predator that inhabits the sandy substrate. Also, general trends for comparison are difficult to make because few estimates of odonate production have been published.
The sex ratio of P. obscurus exuviae in this study was female biased, which is typical of the emergence sex ratio of most gomphids previously studied (Miller 1964; Lutz & McMahan 1973). However, Suhling (1995) found some discrepancies in the emergence sex ratio of O. uncatus when sampling two different streams. Based on male biased sex ratios, it appears as though female P. obscurus mate relatively quickly, oviposit, and then move to more upland habitats to feed, and that males remain near the stream to look for more mates. This observation was also made by Needham & Westfall (1955). This behavior would explain the discrepancy between sex ratios observed in captured adults versus exuviae.
Table 1. Mean head capsule width and growth ratio of both laboratory raised* naiads and naiads collected from Harmon Creek. Instar Mean HCW (mm) Growth Ratio 1* 0.34 2* 0.52 1.53 3* 0.64 1.23 4* 0.83 1.30 5* 1.06 1.28 6 1.39 1.31 7 1.77 1.27 8 2.28 1.29 9 2.82 1.24 10 3.58 1.27 11 4.61 1.29 Avg. Growth Ratio 1.30 Table 2. Production estimate calculations for Progomphus obscurus in Harmon Creek, Texas. Size Class N change HCW (mm) (#/[m.sup.2]) (a) W (g) (b) B (g/[m.sup.2]) (c) in N (d) 0.01-0.50 47.1 0.000038 0.0018 0.51-1.00 161.3 0.000221 0.0356 -114.2 1.01-1.50 95.6 0.000921 0.0880 65.7 1.51-2.00 52.4 0.00236 0.1237 43.2 2.01-2.50 37.9 0.00475 0.1800 14.5 2.51-3.00 26.0 0.00831 0.2161 11.9 3.01-3.50 20.3 0.01326 0.2692 5.7 3.51-4.00 15.2 0.01976 0.3596 5.1 4.01-4.50 8.7 0.02802 0.2438 6.5 4.51-5.00 4.3 0.03824 0.1644 4.4 Total 1.6822 Size Class weight at weight X10 HCW (mm) loss (g) (e) loss (f) (g/[m.sup.2]) (g) 0.01-0.50 0.51-1.00 0.00013 -0.0148 -0.148 1.01-1.50 0.00057 0.0374 0.374 1.51-2.00 0.00164 0.0708 0.708 2.01-2.50 0.00356 0.0516 0.516 2.51-3.00 0.00653 0.0777 0.777 3.01-3.50 0.01078 0.0614 0.614 3.51-4.00 0.01651 0.0842 0.842 4.01-4.50 0.02389 0.1553 1.553 4.51-5.00 0.03313 0.1458 1.458 Total 6.842 (a) Number present per square meter of each size class. (b) Mean weight (milligrams) of individuals of each size class. (c) Total mean annual biomass for each size class. (d) Change in number of individuals present between size class. (e) Mean weight of individuals of each instar when lost from the population. (f) Total weight (grams) lost with each size class. (g) Weight loss multiplied by the number of size classes equals mean annual production for each size class.
Benke, A. C. 1976. Dragonfly production and prey turnover. Ecology, 57:915-927.
Benke, A. C. 1978. Interaction among coexisting predators--a field experiment with dragonfly larvae. J. Anim. Ecol., 47:335-350
Benke, A. C. 1979. A modification of the Hynes method for estimating secondary production with particular significance for multivoltine populations. Limnol. Oceanogr., 24:168-171.
Benke, A. C. & S. S. Benke. 1975. Comparative dynamics and life histories of coexisting dragonfly populations. Ecology, 56:302-317.
Bick, G. H. 1941. Life history of the dragonfly Erythemis simplicicollis (Say). Ann. Entomol. Soc. Am., 34:215-230.
Bick, G. H. 1951. The early larval stages of Tramea lacerata Hagan (Odonata: Libellulidae). Entomol. News, 52:293-303.
Corbet, P. S. 1954. Seasonal regulation in British dragonflies. Nature, 174:655, erratum 777.
Corbet, P. S. 1964. Temporal patterns of emergence in aquatic insects. Can. Entomol., 96:264-279.
Dudgeon, D. 1989. Gomphid (Odonata: Anisoptera) life cycles and production in a Hong Kong Forest Stream. Arch. Hydrobiol., 114:531-536.
Ferraras-Romero M. 1997. The life history of Boyeria irene (Fonscolombe, 1838) (Odonata: Aeshnidae) in the Sierra Morena Mountains southern Spain. Hydrobiologia, 345:109-116.
Ferraras-Romero M. & P. S. Corbet. 1995. Seasonal Patterns of emergence of Odonata of a permanent stream in southwestern Europe. Aquat. Insects, 2:123-127.
Folsom, T. C. & K. L. Manuel. 1983. The life cycle of the dragonfly Lanthus vernalis Carle from a mountain stream in South Carolina, United States (Anisoptera: Gomphidae). Odonatologica, 12:279-284.
Hamilton, A. L. 1969. On estimating annual production. Limnol. Oceanogr., 14:771-782.
Huggins, D. G. & M. B. DuBois. 1982. Factors affecting microdistribution of two species of burrowing dragonfly larvae, with notes on their biology (Anisoptera: Gomphidae). Odonatologica, 1:1-14.
Hynes, H. B. N. 1961. The invertebrate fauna of a Welsh mountain stream. Arch. Hydrobiol., 57:344-388.
Knopf, K. W., & K. J. Tennessen. 1980. A new species of Progomphus Selys, 1854 from North America (Anisoptera: Gomphidae). Odonatologica, 9:247-252.
Kormondy, E. J. & J. L. Gower. 1965. Life history variations in an association of Odonata. Ecology, 46:882-886.
Lutz, P. E. & E. A. McMahan. 1973. Five year patterns of emergence in Tetragoneura cynosura and Gomphus exilis (Odonata). Ann. Entomol. Soc. Am., 66:1343-1348.
Miller, P. L. 1964. Notes on Ictinogomphus ferox Rambur (Odonata: Gomphidae). Entomologist, 97:52-66.
Needham, J. G. & M. J. Westfall, Jr. 1955. A Manual of the Dragonflies of North America. University of California Press, Berkeley, 615 pp.
Norling, U. 1984. Life history patterns in the northern expansion of dragonflies. Adv. Odonatol., 2:127-156.
Paulson, D. R. & C. E. Jenner. 1971. Population structure in overwintering larval Odonata in North Carolina in relation to adult flight season. Ecology, 52:96-103.
Schutte, C., P. Schridde & F. Suhling. 1998. Life history patterns of Onychogomphus uncatus (Charpentier) (Anisoptera: Gomphidae). Odonatologica, 27:71-86.
Suhling, F. 1995. Temporal patterns of emergence of the riverine dragonfly Onychogomphus uncatus (Odonata: Gomphidae). Hydrobiologia, 302:113-118.
Edward C. Phillips
Department of Biology, Gannon University
Erie, Pennsylvania 16541
ECP at: email@example.com
|Printer friendly Cite/link Email Feedback|
|Author:||Phillips, Edward C.|
|Publication:||The Texas Journal of Science|
|Date:||Feb 1, 2001|
|Previous Article:||Fishes of the Keechi Creek Wildlife Management Area of east Texas.|
|Next Article:||Spider assemblages along the Rio Grande at Laredo, Texas.|