Year-round Diet of the Marsh Rice Rat, Oryzomys palustris, in Virginia Tidal Marshes

Abstract

The marsh rice rat, Oryzomys palustris, is the dominant semi-aquatic rodent living in tidal marshes of the Virginia coastal plain. Described as highly carnivorous, this species is known to consume a range of animal foods , including crustaceans, mollusks, fish, and arthropods, as well as some plant foods . Analysis of stomach contents from rice rats collected from Eastern Shore tidal marshes throughout an annual cycle revealed that all 103 stomachs contained di cots, 82 percent had monocots, 61 percent had crabs and insects, and 38 percent had snails. Thirty-eight percent of stomachs contained foods in all five categories, no stomach was empty or contained fish , and 84 percent of stomachs had amounts of hair, probably ingested during self-grooming. In sum, Virginia rice rats are carnivorous but consume greater amounts of plant foods compared to populations that have been studied in Georgia and Louisiana. INTRODUCTION The marsh rice rat, Oryzomys palustris, is a semi-aquatic rodent with its highest abundances in wet fields and marshes, mostly of the southeastern US (Wolfe 1982). Distributed along the eastern seaboard southward from coastal Pennsylvania to the tip of Florida and westward to Corpus Christi, Texas, its range extends northward along the Mississippi River basin into southern Missouri and Illinois. In Virginia, it is common in tidal marshes of the coast and Chesapeake Bay and is present in some grassland habitats as far west as the fall line (ca.Interstate Highway 95; Linzey 1998). The marsh rice rat readily takes to water to forage and escape from predators, and can be caught in floating live traps (personal observation, RKR). Its swimming ability has been studied by Esher et al. (1978) in Mississippi and Carter and Merritt (1981) in Virginia, and inter-island movements of marked rice rats have been documented for the Virginia barrier islands (Forys and Dueser 1993). Medium in size among rodents (up to 80 g), Oryzomys is considered to be highly carnivorous, second to North America's most carnivorous rodent, the grasshopper mouse, Onychomys, a desert grassland mouse of the western states. The meat-eating proclivities of marsh rice rats were observed by Schantz (1943), who reported them eating the bodies of trapped muskrats, a behavior also observed by RKR (unpublished) on trapped small mammals on Fisherman Island, Virginia. The natural history of the marsh rice rat is summarized in Wolfe (1982). 1 Corresponding author: Robert K. Rose, Department ofBiological Sciences, Old Dominion University, Norfolk, Virginia, Phone: 757-683-4202, Email: brose@odu.edu 116 VIRGINIA JOURNAL OF SCIENCE The objectives of our year-long study were to learn the kinds and proportions of foods eaten by marsh rice rats taken from tidal marshes of the Eastern Shore of Virginia and their seasonality of food selection. Oryzomys palustris is codominant in these tidal marshes with the meadow vole, Microtus pennsylvanicus (March 1995 Bloch and Rose 2005), with the latter being almost exclusively herbivorou~ (Zimmerman 1965, among others). Where the diet of marsh rice rats has been examined in Louisiana (Negus et al., 1961), Florida (Pournelle, 1950) and Georgia (Sharp, 1967), Oryzomys consumes both plant and animal materials, in differing proportions. In our study, we learned that Virginia rice rats ate higher proportions of plant material than at other geographic locations, plus varying amounts of crabs, snails, and arthropods (mostly insects). MATERIALS AND METHODS The $tudy1\rea This study was conducted over a one-year period, from May 1994-April 1995. The research goal was to collect samples of up to 15 animals each month from two seaside sites in Northampton County, Virginia, one located ca. 300 m south of Oyster and the other 500 m east of Townsend. In October, an opportunity was presented to examine animals from the marshes of nearby barrier islands (Myrtle, Ship Shoal, Smith, and Mockhorn), so the sample size for that month was much larger than the others. For unknown reasons, rice rats were scarce during the summer months of June-August, and only two animals were collected during that time (Table 1 ), despite an increased trapping effort then compared to other months. March (1995), in earl ier studies of the population dynamics of rice rats in similar tidal marshes on the Eastern Shore, had also found density to be low or near zero in June and July, and Negus et al. (1961) caught 13 rice rats in July in 2145 trap-nights and 3 in June of another year in 504 trap-nights, both < 1 rice rat per 200 trap-nights, indicating behavioral or other changes lowering their trappability in summer. Both study sites were in tidal marshes backed by areas of dense common reed, Phragmites australis. The flora of the marshes included Spartina alterniflora (salt grass), S. patens (salt meadow hay), Panicum sp. (panic grasses), Juncus roemeranius (black needle-rush), Salicornia sp. (glasswort), Baccharis halimifolia (saltbush), and Typha latifolia (cattail). Both marshes are flooded twice daily in the area of the Juncus, whereas the S. patens areas are flooded only during monthly high or wind tides. The border between Baccharis shrubs and Juncus often supported a more substantial wrack line than that between S. patens and Juncus. This wrack line provided additional structure to a marsh with relatively little structure, except for the Baccharis shrubs. Trapping cand m.aboratory[Procedures Transects of Fitch live traps (Rose 1994) were placed 2-3 m apart along the borders, i.e., at the normal extent of the daily high tide. Baited with wild birdseed and tended early each morning, these traps yielded mainly marsh rice rats and meadow voles, with lesser numbers white-footed mice (Peromyscus leucopus) and house mice (Mus musculus), and even fewer least shrews (Cryptotis parva) and short-tailed shrews (Blarina brevicauda). Only marsh rice rats were collected for this study. Marsh rice rats were returned to the laboratory, euthanized by chloroform anesthesia, and frozen. FOODS OF RICE RATS IN TIDAL MARSHES 117 TABLE I. For each sex, sample sizes of the age categories of rice rats, following the criteria of Negus et al. (1961 ). Age category I = juvenile, 2 = subadult, 3 = near adult, and 4 = adult. The months have been grouped into seasons, with June-September SUmmer, October and November AUtumn, December-February Winter, and March-May SPring. MONTH TOTAL # # FEMALES # MALES AGEi AGE2 AGE3 AGE4 May SP 6 I 5 2 0 0 4 June SU I I 0 I 0 0 0 July SU 0 0 0 0 0 0 0 August SU I I 0 I 0 0 0 September SU 7 1 6 2 2 I 2 October AU 36 10 26 8 8 3 17 November AU 3 2 I 3 0 0 0 December WI 9 4 5 5 I 3 0 January WI 15 6 9 12 0 2 February WI 12 7 5 7 2 2 March SP 9 2 7 I 0 4 4 April SP 4 0 4 0 I 0 3 Totals 103 34 69 41 14 15 33 In order to compare what was eaten with what foods were available, samples of all potential food sources in the tidal marshes were collected from the same marshes as the rice rats, returned to the lab, processed, and made into reference slides. After samples of plant and animal materials had been pulverized in a Waring® blender to a consistency comparable to that of stomach contents ofrice rats, the material then was washed in water, air-dried, and placed on microscope slides with Kleermount, a mounting medium, and covered with standard coverslips. Reference slides were made of three dicots (Baccharis, Salicornia, and Typha), four monocots (Juncus, Panicum sp., Spartina alternifolia, and S. patens), and four animals: fiddler crab, Uca minax; periwinkle, a univalve snail, Littorina irrorata; mummichog, a small brackish-water fish, Fundulus heteroclitus; and several arthropods, including grasshoppers, crickets, flies, and spiders, hereafter called 'insects'. After the rice rats were thawed, standard measurements were taken (total length, lengths of tail, foot and ear, body mass) and the reproductive information was recorded for a related research project (Dreelin 1997). The contents of each stomach were removed, washed in water, air-dried, and then placed in separate l 0-ml beakers, covered, labeled, and placed in the freezer to avoid contamination. For analysis, two samples from each stomach were placed on two slides with Kleermount, covered with standard coverslips, and compared to reference slides (method modified from Fleharty and Olson 1969). The contents were analyzed using a percent volume method, in which the amount of each type of food was visually estimated using a standard 10 X IO ocular grid (Whitaker and French 1984). Food items were identified as belonging to one of six categories: dicotylenous plant, monocotyledonous plant, crab, snail, fish, and arthropod. In each of the l O randomly selected 10 X l O ocular fields, the volume of each food type was estimated and recorded. The volumes from both slides of each stomach were then summed, and an average was calculated to determine the percent volume of each food type for each stomach. 118 VIRGINIA JOURNAL OF SCIENCE TABLE 2. Mean percent volumes and standard errors (in parentheses) based on examination of JO microscopic fields in each of two slides per marsh rice rat, using the technique of Whitaker and French ( 1984). Values are given for each month of study and for each food category. MONTH DICOT MONOCOT CRAB SNAIL INSECT May 77.2 (0.11) 8.1 (0.07) 7.1 (0.08) 3.7 (0.07) 4.0 (0.07) June 95.8 (----) 4.2 (----) 0.0(----) 0.0 (----) 0.0 (----) July --(----) --(----) --(----) --(----) --(----) August 73.4 (----) 10.9 (----) 4.2 (----) 0.0 (----) 11.5 (----) September 79 .5 (0.05) 9.2 (0.02) 6.1 (0.06) 2.9 (0.04) 2.1 (0.03) October 66.1 (0.04) 12.0 (0.04) 6.8 (0.03) 3.9 (0.03) 11 .3 (0.04) November 76.8 (0.08) 4.9 (0.05) I 1.6 (0.1 1) 3.6 (0.07) 3.2 (0.05 ) December 85.4 (0.08) 6.8 (0 .06) 4.6 (0.06) 1.7 (0.02) I .5 (0.03) January 75.8 (0.05) I 0.5 (0.04) 6.6 (0.04) 6.0 (0.04) I. I (0.01 ) February 75 .7 (0.06) 16.9 (0.05) 3.7 (0.03) 2.3 (0.03) 1.4 (0.02) March 62 .3 (0.07) 30.1 (0.05) 0.2