Oryzomys palustris ( Harlan, 1837 )

Rose, Robert K., 2023, Oryzomys palustris (Rodentia: Cricetidae), Mammalian Species 55 (1031), pp. 1-20 : 2-8

Oryzomys palustris ( Harlan, 1837)

Common Marsh Rice Rat

Mus palustris Harlan, 1837:386 . Type locality “freshwater swamps... near ‘Fast Land,’ in the vicinity of Salem,” Salem County, New Jersey.

Hesperomys palustris : Wagner, 1843:543. Name combination.

Arvicola oryzivora Bachman, 1854:214 . Type locality “rice plantation in St. John’s parish, South Carolina.”

Hesperomys (Oryzomys) palustris : Baird, 1857:459. Name combination.

O [ryzomys]. palustris : Coues, 1890:4164. First use of current name combination.

Oryzomys argentatus Spitzer and Lazell, 1978:787 . Type locality “the edges of a small fresh-water marsh by a roadside... in Cudjoe Bay , Monroe County, Florida.”

CONTEXT AND CONTENT. Context as for genus. Five subspecies are currently recognized ( Patton 2017; Burgin et al. 2020):

O. p. coloratus Bangs, 1898:189. Type locality “ Cape Sable , Monroe County, Florida; ” argentatus Spitzer and Lazell, 1978 is a synonym.

O. p. natator F. M. Chapman, 1893:44. Type locality “at Gainesville , [Alachua County], Florida.”

O. p. palustris ( Harlan, 1837:385) . See above; oryzivora ( Bachman, 1854) is a synonym.

O. p. planirostris Hamilton, 1955:83 . Type locality “a garbage dump and adjoining wet land” ... “one mile west of third bridge that spans Matachla Pass, [ Little ] Pine Island , Lee County, Florida.”

O. p. sanibeli Hamilton, 1955:85. Type locality “freshwater marsh, four miles west of lighthouse on Sanibel Island , Lee County, Florida.”

NOMENCLATURAL NOTES. Oryzomys is the combined form of a Latin word (derived from Ancient Greek) for rice, oryza, and the Greek word for mouse, mys, therefore, literally “rice mouse.” Other common names are marsh Oryzomys , marsh rice rat, rice meadow mouse, swamp rice rat, and in Spanish, rata arrocera de marisma ( Patton 2017).

DIAGNOSIS

The separation of the Texas marsh rice rat ( Oryzomys texensis ) from O. palustris was determined by differences at the molecular genetics level ( Hanson et al. 2010) when the five subspecies of O. palustris were grouped into one clade and populations of the former subspecies texensis formed another clade. Although the skulls of O. palustris and O. texensis could be separated by such morphological features as breadth at the condyles and alveolar length of the upper and lower molar toothrows ( Humphrey and Setzer 1989), few variables consistently separated these taxa. Perhaps presaging their present status as separate species, Merriam (1901:276), who had emphasized the need to use only adult males in analyses of morphology of Oryzomys , said in a footnote: “I am unable to distinguish Allen’s subspecies texensis , either externally or by the skulls, from O. palustris from Raleigh, N. C. and Dismal Swamp, Va. The braincase may average a trifle narrower, but the difference is very slight.” Both O. palustris and O. texensis have slender tails about equal to the head–body length.

Across much of its distribution in the southeastern United States, O. palustris is present in the same small mammal communities as the hispid cotton rat ( Sigmodon hispidus ), a large brown sigmodontine rodent. The dorsum of the hispid cotton rat is brown compared with the grayish or buffy gray pelage of O. palustris . Its tail is relatively shorter and thicker than the long slender tail of O. palustris , plus it has blackish feet compared with the white feet of O. palustris . Maximum body mass of O. palustris is about 90 g compared to twice that for hispid cotton rats. Members of the genus Rattus also are large and long-tailed but have relatively shorter, thicker, and hairless tails compared to the hairy and slim tail of O. palustris . Other brownish rodents in the community might include Peromyscus spp. (body mass <35 g) and the meadow vole ( Microtus pennsylvanicus ) or prairie vole ( M. ochrogaster ), both of which are 40–50 g, with tails about twice the length of hind feet compared to tails four times longer than hind feet in O. palustris . Oryzomys palustris is the only grayish long-tailed member in its small mammal community, apart from young of some species in juvenile pelage. In Virginia tidal marshes north of the Chesapeake Bay (where the hispid cotton rat is absent), the meadow vole is codominant with O. palustris ( Bloch and Rose 2005; Rose and March 2013).

GENERAL CHARACTERS

Oryzomys palustris is brownish gray with a tail nearly as long as the head–body length ( Fig. 1 ). Adult body mass is 50–80 g ( Rose and Dreelin 2011). Adults from eastern Virginia have statistically significant size dimorphism, with males 13% heavier and 9.7 mm longer in total length ( Rose and Dreelin 2011). The dorsal pelage is grizzled brownish gray with flecks of black, and the sides are buffy ( Merriam 1901). The ears have long, coarse hairs on both surfaces. Hairs of the venter are long and white, with slate-colored underfur showing through on close examination ( Merriam 1901). The nearly white venter and dark dorsum gives this semiaquatic rodent protective coloration while swimming, whether seen from below or above. The tapered tail, about 4 mm in diameter at the base and about 1 mm at the tip, is dusky above and whitish below, with no clear line between (RKR, pers. obs.).

The white toes, four on the front foot and five on the hind foot, are long and supple, enabling an individual to climb into tall vegetation during times of rising water or to catch mobile prey. The sole of the hind foot is hairless with six tubercles; the posterior one is narrow and “greatly elongated” ( Baird 1857:483). The three middle toes on the hind foot are long and of nearly equal length. Although semiaquatic, O. palustris lacks both the webbing on the toes and flattening of the tail seen in other native semiaquatic rodents, such as muskrat ( Ondatra zibethicus ) and beaver ( Castor canadensis ). The skull of O. palustris is mediumsized and somewhat flattened ( Fig. 2 ), with moderately developed supraorbital beads and an elongated rostrum ( Merriam 1901).

Among the subspecies, O. p. palustris has the shortest mean total length, 230.2 mm, the two subspecies from Florida islands ( planirostris and sanibeli) are of intermediate length (247.5 and 257.5 mm), and the two subspecies from mainland Florida (coloratus and natator) are both longer than 280 mm ( Table 1). Select cranial measurements of the five subspecies are presented in Table 1. Mean measurements of the four Floridian subspecies are from Hamilton (1955), mean total lengths of O. p. palustris are from 25 specimens collected in January–February from Northampton County, Virginia ( Dreelin 1997), and mean skull measurements from Goldman (1918). A May–August sample (24 males) of O. p. palustris from Dreelin (1997) was 4 mm shorter in mean total length than the January–February Virginia sample (latter given in Table 1), and the mean of the nine males from Raleigh, North Carolina, of Merriam (1901) was just 1 mm longer than Goldman’s sample from different locations.

Subspecies also differ in color of the back and sides but the middorsal stripe is a darker color than the rest of the body in all five subspecies. Oryzomys p. palustris is the grayest subspecies ( Merriam 1901), with brown and black hairs on the middorsum and buffy sides. Oryzomys p. planirostris is brownish gray with buffy sides, and the back of O. p. sanibeli is amber brown ( Hamilton 1955). Upperparts of O. p. natator are very dark brown mixed with gray, becoming buffy gray on the sides ( Chapman 1893), whereas upperparts of O. p. coloratus, which occupies the tropical part of Florida, are ochraceous-tawny and brighter, richer, and more rufescent than on the other subspecies ( Goldman 1918). All subspecies have white or cream toes and venter; the belly underfur is a slaty gray ( Merriam 1901).

DISTRIBUTION

The type specimen was from a wetland on upper Delaware Bay near Salem, New Jersey ( Harlan 1837), near the northernmost point of distribution. Later efforts to find Oryzomys palustris near Cape May, Fort Norris, and Salem, New Jersey, were futile, but 62 years later, H. W. Warrington collected a series of specimens from a site with old muskrat houses between the latter two cities ( Stone 1898). Skulls in owl pellets and a carcass represent records of its continued presence in the Delaware Bay region, including an excellent account of adults disturbed from a globe-shaped nest built on an abandoned bird nest in Tinicum marsh near Philadelphia ( Ulmer 1944, 1951). In the 1970s, O. palustris was found at multiple sites in Delaware and New Jersey, all in the Delaware Bay area (Arndt et al. 1978).

In the central states, the northern limit of distribution probably is the Mississippi River valley in Tennessee because recent work has revealed that the Oryzomys in southern Illinois is O. texensis ( Williams and Ibrahim 2023) . Farther up the Ohio River, Oryzomys is absent, despite extensive field studies conducted in the 1970s in six Indiana counties bordering the river; no O. palustris was trapped in more than 24,000 trap-nights using snap and pitfall traps in various habitats ( Rose and McKean 1980). In more intensive studies at 15 sites in southern Spencer County, Indiana, Oryzomys was not among the 1,887 small mammals trapped over 4 years of study ( Rose 1981). Nor were Oryzomys skulls among the remains of 1,526 small mammals taken by barn owls ( Tyto alba ) in 1974–1976 at a site in southern Spencer County ( Rose 1981: table 4). In brief, multiple studies and methods failed to detect Oryzomys in southwestern Indiana during the 1970s.

The western geographic boundary of O. palustris is poorly defined and a matter of speculation ( Fig. 3 ) because the contact zones with O. texensis need study.At present, the western boundary of O. palustris probably occurs in Alabama, Tennessee, and Kentucky.

Populations of O. palustris from coastal plains or islands are the most studied. Dueser et al. (1979) found O. palustris on 9 of 11 barrier islands off the Virginia coast. Two of the five subspecies are restricted to islands in southeastern Florida: Oryzomys palustris planirostris to Little Pine and Pine islands and O. p. sanibeli to Sanibel Island. Both O. p. coloratus and O. p. natator are from mainland Florida, and the nominal subspecies occupies the rest of the distribution ( Fig. 3 ). Its presence in piedmont regions is common in river valleys; in Virginia, piedmont populations are known from Caroline ( Bellows et al. 2001a), Chesterfield ( Jackson et al. 1976), and Cumberland ( Pagels et al. 1992) counties.

FOSSIL RECORD

Fossil sites with Oryzomys palustris are most numerous in Florida; all are from late Pleistocene or early postglacial sites. Located in all regions of the state, several sites are at limestone quarries, such as Reddick ( Gut and Ray 1963) and Williston ( Holman 1959) in Levy County, central Florida, which yielded fossils of O. palustris from the late Pleistocene. Oryzomys palustris was among the mammalian fossils identified in the Haile XIVA Fauna, a limestone quarry in Alachua County, Florida ( Martin 1974; Robertson 1976). Other fossil sites with O. palustris are sinkholes filled with fluvial material and bones of uncertain age ( Webb 1974), such as the Devil’s Den Fauna in Levy County ( Martin and Webb 1974). This sinkhole trap site included humans among the 46 species of mammals from 7,000 to 8,000 years ago—a cooler time characterized by xeric park-savanna habitat. Oryzomys palustris also was found in association with fossils from more than 45 species of small and large mammals excavated in an old stream bed near Vero Beach, Indian River County, Florida ( McFadden et al. 2012). That site was of late Pleistocene age (10,000 –12,000 years ago), and also yielded contemporaneous remains of humans, ground sloths, mammoths, and mastodons.

Elsewhere, fossil remains of Oryzomys , probably referable to O. palustris , were among the mammalian fossils from Anderson Pit Cave in Monroe County, southcentral Indiana; the large extinct armadillo, Dasypus bellus , was also present, indicating warmer temperatures during the late Pleistocene (Rancholebrean) than at present ( Richards 1979; Holman and Richards 1981). Fossils of O. palustris from Georgia include those from the Isle of Hope site (Rancholabrean) in Chatham County ( Hulbert and Pratt 1998), Clark Quarry in Glynn County ( Rhinehart 2019), and the Ladds fauna in Bartow County ( Ray 1967). In the coastal plain of South Carolina, the Ardis local fauna near Harleysville in Dorchester County contained fossil evidence of O. palustris ( Bentley et al. 1994) .

Fossils of O. palustris from states beyond its current distribution include those from Nebraska, Iowa, Indiana, Ohio, and West Virginia ( Vickery et al. 2016), indicating its broader distribution during warm interglacial periods. Many sites are 200–400 km beyond the current distribution and usually are located in proximity with major river systems, places where villages of indigenous people often were located (see map with 60 such fossil sites in Vickery et al. 2016). In brief, despite controversy about the reasons for its advances and regressions ( Semken 2016; Vickery et al. 2016), the fossil record of Oryzomys cf. palustris is excellent.

FORM AND FUNCTION

Form.— Body form of Oryzomys palustris is similar to that of other long-tailed rodents of eastern North America, except the tail is slender, nearly equal to the head–body length, and pencil-like near the tip. The fur is slate-colored, tipped with brown and black, giving it a dark grayish brown tint ( Bachman 1854); venter has long and dense whitish fur, with gray underfur; subspecies in mainland Florida tend to be larger than the other subspecies. The smallish ears have hairs on both surfaces. The 30 or so black and white hairs of the vibrissae are short, with the longest barely reaching the ears. Tail is dusky above and lighter below. Soles of the feet are naked, the first digit of the forefoot is reduced to a nail, and the three middle toes of the hind foot are of nearly equal length ( Bachman 1854).

The zygomatic arches of the skull of O. palustris are relatively delicate ( Fig. 2 ) with their outer sides nearly parallel ( Merriam 1901), and the squamosal bones have small but definite notches medial to the zygomatic process. The wedge-shaped nasal bones are long and tapering, ending on a plane with the premaxillae ( Park et al. 1974). Frontal bones taper into a narrow point posteriorly, in the median line between the parietal bones. The small, narrow interparietal bone forms a relatively straight line along its anterior border. The palatal slits are long and carrot-shaped. Auditory bullae are prominent but of average size ( Park et al. 1974). The coronoid process of the mandible is vertically oriented and pointed, so the mandibular notch is deep and wide; the angular process of the mandible is relatively short and round and does not extend posteriorly beyond the condylar process ( Park et al. 1974). The maxillary and mandibular toothrows are almost parallel, but the latter is angled toward the median sagittal plane ( Park et al. 1974).

Dental formula is i 1/1 c 0/0 p 0/0 m 3/3, total 16, as in other cricetid rodents. Incisors are ever-growing, consisting of segments of spirals in both maxilla and mandible that emerge as tooth wear requires ( Gupta and Shaw 1956a). The incisal edges of the maxillary incisor teeth are sharply beveled by attrition, due to the differential wear of the pigmented enameled face of the incisor and the softer dentin behind ( Gupta and Shaw 1956a). The incisal bevel is concave in the mandibular incisors, and longer than in the maxillary incisors, ending sharply in a notch.

The diastema separating the incisors from the molars is longer in the maxilla than in the mandible ( Gupta and Shaw 1956a). Molars have multiple roots and are subject to wear, enabling individuals to be assigned relative ages based on tooth wear, and are low-crowned with cusps separated by shallow sulci and ridges. The first maxillary molar is the largest, with six cusps and three roots (see photos in Gupta and Shaw 1956a), the second molar has four cusps and three roots, and the third maxillary molar is the smallest, with three cusps and two roots. In the mandible, the first molar is triangular, with five cusps and three roots, the second molar is squarish, with four cusps and three roots, and the third molar has three cusps and two roots ( Gupta and Shaw 1956a). Park et al. (1974) provided greater details and drawings of each tooth, cusp, and its root size and structure. As the enamel of each molar is worn by attrition, the crown and sulci disappear, leaving an enamel perimeter surrounding a shallow depression of dentin and some enamel ridges.

As O. palustris matures, sloughing of gum from tooth naturally occurs, starting at 8 weeks of age. Gupta and Shaw (1956b) were among the first to recognize its potential value as a lab animal for the study of gum disease (e.g., periodontitis). Soon it was learned that laboratory animals fed a high sugar–casein diet had an earlier onset and more rapid progression of periodontitis, making it a useful model for the study of this disease. Currently, the dental school of the University of Florida, Gainesville, uses lab-reared O. palustris to study gum diseases ( Aguirre et al. 2015).

The axial skeleton of O. palustris is typical of other long-tailed rodents except that the humerus and femur are relatively shorter, and the hind feet are relatively longer than seen in Sigmodontine species such as Sigmodon hispidus ( Stein 1988) . These morphological features are interpreted as adaptive modifications for semiaquatic life: the shorter proximal leg bones reduce drag on the recovery stroke (while swimming) and the longer feet promote more efficient power strokes ( Stein 1988). The musculature associated with these morphological features also is modified compared with strictly terrestrial species ( Stein 1988).

The simple gastrointestinal tract is similar to that of other Oryzomyine rodents: pharynx, esophagus, stomach, small intestine, cecum, large intestine. The notable abdominal feature is the absence of a gall bladder ( Weksler 2006).

The anatomy of the male reproductive system includes a complex phallus, paired testes, and the following pairs of accessory glands ( Arata 1964): preputial, bulbourethral, ampullary, vesicular, and multiple prostate glands (one pair dorsal, two pairs ventral, one pair anterior), all under hormonal control of products of the testes. Arata (1964:35, figure 4b) shows the positions of these reproductive parts.

Compared with many species of small mammals, the fur on the venter of O. palustris is long and dense, especially in the inguinal region (RKR, pers. obs.), sometimes making it difficult to distinguish sexes in the nonbreeding season. Because fur density is important in semiaquatic mammals (see Behavior section), studies comparing and quantifying the fur density of O. palustris and common terrestrial small mammals would be useful.

Function.— Basic information on metabolic, heart, and breathing rates is lacking for Oryzomys palustris . The challenge of maintaining a favorable balance of Na + and Cl− ions while living in an environment with prolonged exposure to seasalt aerosols was examined by Kuan et al. (2019). Their study suggested that the airway of O. palustris might possess unique ion-transport adaptations that facilitate their survival in highsaline environments.

Although captive O. palustris were able to assimilate both plant and animal food efficiently (ingested energy minus fecal energy over fecal energy times 100 = 88–95%), they ingested more calories when animal food was offered ( Sharp 1967). A daily intake level of 0.15 Kcal/Kcal of body tissue was necessary to maintain body mass in the laboratory; ingestion rates above this level resulted in weight gain. Sharp (1967) concluded that during the summer months, the trophic niche of O. palustris in a salt marsh ecosystem was that of a carnivore.

Long photoperiods stimulate growth of the reproductive organs in juvenile O. palustris and maintain gonadal function in adults ( Edmonds and Stetson 1993 a, 1994). The hormone melatonin, secreted from the pineal gland, is produced during the nighttime hours, but its release is quickly inhibited at the time of “lights on” in the laboratory ( Edmonds et al. 1995), indicating that melatonin levels are regulated by photoperiod. Melatonin, when introduced via implants, injections, or infusions, inhibits testicular and somatic growth in juvenile males ( Edmonds and Stetson 1994).

In a series of lab experiments to examine the relative roles of food intake and photoperiod on attainment of reproductive maturity, no significant effects of food intake on growth in either testes or seminal vesicles were seen until a 20% food reduction at 16L:8D photoperiod and a 40% food reduction at 14L:10D compared to ad-lib controls ( Edmonds et al. 2003). In juvenile females, absolute uterine mass was affected only in the 40% food reduction at 14L:10D, whereas absolute ovarian masses were significantly smaller on both long photoperiods only at 40% food reduction. Thus, in theory, the positive effects of long photoperiod on reproductive development can be altered by periods of low caloric intake, such as in habitats with few resources.

Hematology and serum chemistry values for O. palustris (20 males and 19 females) in a University of Florida lab colony are presented in Aguirre et al. (2015). Mean hematocrit of Tennessee individuals at 285 m elevation ( Sealander 1964) was 40.5% compared to 52.3% from the lab colony in Gainesville at 54 m ( Aguirre et al. 2015). Oryzomys palustris had among the lowest hematocrit (40.5) and hemoglobin levels of any of the 34 small mammal species tested by Sealander (1964). A mixedsex sample of 103 captive O. palustris at Oak Ridge National Laboratory, Tennessee, had the following blood parameters (mean ± SE): red blood count, 8.64 ± 0.07 × 106 /μl; hemoglobin, 15.5 ± 0.15 g /100 ml; hematocrit, 46.1 ± 0.49%; total white blood count, 4.99 ± 0.42 × 103 /μl; neutrophils, 0.36 ± 0.35 × 103 / μl; lymphocytes 4.33 ± 0.38 × 103 /μl; albumin, 57.09 ± 0.74%; alpha-globulin, 15.01 ± 0.29%; beta-globulin, 25.36 ± 0.62%; and gamma-globulin, 2.59 ± 0.20% ( Kitchings et al. 1970).

ONTOGENY AND REPRODUCTION

Ontogeny.— Details of early growth and development are drawn mostly from Hamilton (1946), who raised six litters of Oryzomys palustris collected from Chincoteague Island, Virginia. Mean mass of the blind and naked 42-mm-long neonates was 3.7 g (range 2.35–4.0 g). At 2 days of age, mean body mass was 5.0 g, dorsum was becoming darker, and each young was active and vocalizing. By 4 days, the young could crawl, the back was well-haired, with shoulder hairs dark gray washed with fulvous, but the belly was still naked; average body mass then was 6.3 g, head–body length was 52 mm, and the tail length was 30 mm ( Hamilton 1946). At 6 days, the head–body length was 59 mm, and the body was well-furred, with grizzled brownish gray above and a white belly. By 8 days, all young, now with juvenile pelage, were so active they fled the nest if disturbed. At 10 days, young ate solid food, had an average body mass of 10.2 g, a head–body length of 69 mm, and tail length of 47 mm. These young were weaned at 13 days, after which they gained 1.0– 1.5 g /day until about 3 weeks of age. After that, growth slows, juveniles become subadults (> 30 g) at about 2 months and are fully grown (50–80 g) at 4 months ( Hamilton 1946). In the laboratory, slow growth in head–body length flattened at 12 months in females and 16 months in males, when males averaged 2.4 cm longer than females ( Park and Nowosielski-Slepowron 1974). In nature, few O. palustris live as long as 12 months in eastern Virginia ( Bloch and Rose 2005).

In a study of rates of growth of pre-weanlings in another laboratory (n = 120), neonates (¯x = 3.25 g ± 0.14 SE) grew at the rate of 1.105 ± 0.32 g /day for the first 5 days, then higher growth rates (1.606 ± 0.51 g /day) through day 13, and finally somewhat slower growth (1.223 ± 0.40 g /day) until day 20, when they averaged about 22 g ( Park and Nowosielski-Slepowron 1975). Incisors were observed by days 3–4, and by day 10, the first mandibular molar (m1) had emerged. The next molars to emerge were M1 and m2 at days 13–14, followed quickly by the remaining molars; nibbling of solid food began immediately as seen by wear on incisors ( Park and Nowosielski-Slepowron 1975). No body size dimorphism was detectable until day 20, but thereafter males diverged, increasing in body mass more rapidly than in body length ( Park and Nowosielski-Slepowron 1974).

The size dimorphism of combined subadults and adults (≥ 30 g) was significantly different (t = 4.78, P <0.05) in O. palustris collected for necropsy in Virginia ( Dreelin 1997), with the mean body mass of 81 males (56.7 ± 1.31 g, range 32–91) being greater than the corrected (somatic) body mass of 43 females (47.8 ± 1.61 g, 30–78). Mean body length also differed (t = 4.94, P <0.05), with males (222.8 mm) longer than females (213.1 mm). Paradiso (1960:517) examined specimens in the US National Museum, and some females were equal in size to males but “small size was invariably associated with females.” He speculated that if immature females were impregnated, they never achieved their full-size potential.

Reproduction.— Using 21 Oryzomys palustris caught near Knoxville, Tennessee, and their descendants reared in the lab, Conaway (1954) determined the estrous cycle by doing daily vaginal lavages and staining the cells to learn when the cell types changed, indicating different stages in the ovarian cycle. The cycle in O. palustris was 7.62 days ± 0.19 SE (range 6–9), similar to that of the hispid cotton rat and many other cricetid rodents. Based on daily vaginal smears starting at day 30, the first spontaneous estrus was detected at day 57 ( Conaway 1954). Examining male reproductive competency, Conaway (1954) found one male with sperm in the testes at day 42 and another had sperm in the epididymis (indicating fertility) at day 49; he concluded that both sexes reach maturity at 50–60 days. Mean size of 10 litters raised in the Conaway lab was 5.0 ± 0.2 young (range 4–6). Gestation is about 25 days in O. palustris ( Hamilton 1946) .

Details of litter size and reproductive indices of both sexes for O. palustris in Virginia were derived from 120 individuals collected monthly from coastal tidal marshes ( Rose and Dreelin 2011) and 44 others collected in January–February 1982 on Fisherman’s Island (both in Northampton County). No individuals were caught in January–February 1996, a time of extremely low density on study grids ( Bloch and Rose 2005). Females with embryos were collected in April–October in Virginia; there was almost no evidence of breeding in either sex during the remaining 5 months ( Rose and Dreelin 2011). In Delaware, necropsies of wild-caught O. palustris indicated a breeding season from March through September; no pregnant or lactating females were detected from October through February ( Edmonds and Stetson 1993b). The breeding season probably is longer in southern populations; for example, Brimley (1923) recorded breeding in March–November in central North Carolina.

Mean litter size in Virginia (n = 16) averaged 4.63 and did not vary among months, with parity of females, or between subadult and adult females ( Rose and Dreelin 2011). The smallest pregnant female weighed 34 g (in September), and because embryonic bumps do not appear in the uterus until about day 11 or 12 of pregnancy, this female probably weighed about 25 g (≤30 days old) when inseminated. The smallest pregnant female in Mississippi weighed 34 g ( Wolfe 1985). Litters from seven females caught in May–July in Maryland averaged 5.6 young ( Harris 1953).

The age at maturity in males probably is determined by a combination of season (photoperiod, Edmonds and Stetson 1993a) and physical development (body mass); minimum age is probably lower for spring-born males than for overwintered ones. Near Gainesville, Florida, most males were fertile at 30–40 g when they had molted to adult pelage ( Birkenholtz 1963). In Virginia, the smallest fertile male, collected in August, was 43 g ( Dreelin 1997).

Fertilityinmalesisdeterminedbythepresenceofconvolutions (rather than loops) in the cauda epididymides ( Jameson 1950) in which mature sperm are stored in testes. Oryzomys palustris males in Virginia were fertile 2–4 weeks before and after the breeding season of females ( Rose and Dreelin 2011), a common pattern in small mammal species (e.g., Rose and Gaines 1978; Bergstrom and Rose 2004). Average mass of paired testes (mg/ 10 g of body mass) was greater in spring and summer (50.9–113.7) than in November (14.1) and December (7.3; Rose and Dreelin 2011). The values for the Fisherman’s Island males from January 1982 were 11.73 ± 8.2 (SE) but increased to 51.17 ± 4.6 in February when growth of the testes foretold the coming breeding season. Seminal vesicles showed similar patterns: greater in spring and summer (36.7–139.6) than in November (13.9) and December (2.5). Thus, in eastern Virginia, O. palustris is a seasonal breeder with almost no adults in breeding condition during the coldest 5 months ( Rose and Dreelin 2011).

Mammals sometimes adjust to local conditions that vary among years: pregnant and lactating females of O. palustris were observed in March in salt marshes near Lewes in southern Delaware ( Edmonds and Stetson 1993b) where the breeding season extended through September. Testicular masses more than doubled from January to February. Mean litter size of 10 necropsied females in Delaware was 6.7 ± 0.6 (range 4–10), suggesting females from northern populations have larger litters than females from southern populations, a pattern seen in some other rodents with broad distributions in North America (e.g., New World Microtus — Keller 1985).

Figures

Fig. 1.—An adult male Oryzomys palustris from southern Georgia. Used with permission of Bugwood Images; photograph by Sturgis McKeever.

Fig. 2.—Dorsal, ventral, and lateral views of skull and lateral view of mandible of an adult male Oryzomys palustris palustris from near New Berlin, Duval County, Florida. Greatest length of skull is 32.17 mm. Photo by Adam Ferguson of the Field Museum of Natural History, used with permission.

Fig. 3.—Distribution of Oryzomys palustris in eastern North America. Subspecies are those listed by Patton (2017): 1, O. p. coloratus and 2, O. p. natator are from regions of mainland Florida, and 3, O. p. palustris, from the rest of the mainland distribution; 4, O. p. planirostris and 5, O. p. sanibeli are restricted to small coastal islands in southern Florida. Dots represent locations from which the samples were taken for the lab analysis that elevated the former subspecies O. p. texensis to species level (Hanson et al. 2010). The solid line represents the best estimate of the line separating O. palustris from O. texensis.