Robert K. Rose

The Fecal Viral Flora of Wild Rodents

The frequent interactions of rodents with humans make them a common source of zoonotic infections. To obtain an initial unbiased measure of the viral diversity in the enteric tract of wild rodents we sequenced partially purified, randomly amplified viral RNA and DNA in the feces of 105 wild rodents (mouse, vole, and rat) collected in California and Virginia. We identified in decreasing frequency sequences related to the mammalian viruses families Circoviridae, Picobirnaviridae, Picornaviridae, Astroviridae, Parvoviridae, Papillomaviridae, Adenoviridae, and Coronaviridae. Seventeen small circular DNA genomes containing one or two replicase genes distantly related to the Circoviridae representing several potentially new viral families were characterized. In the Picornaviridae family two new candidate genera as well as a close genetic relative of the human pathogen Aichi virus were characterized. Fragments of the first mouse sapelovirus and picobirnaviruses were identified and the first murine astrovirus genome was characterized. A mouse papillomavirus genome and fragments of a novel adenovirus and adenovirus-associated virus were also sequenced. The next largest fraction of the rodent fecal virome was related to insect viruses of the Densoviridae, Iridoviridae, Polydnaviridae, Dicistroviriade, Bromoviridae, and Virgaviridae families followed by plant virus-related sequences in the Nanoviridae, Geminiviridae, Phycodnaviridae, Secoviridae, Partitiviridae, Tymoviridae, Alphaflexiviridae, and Tombusviridae families reflecting the largely insect and plant rodent diet. Phylogenetic analyses of full and partial viral genomes therefore revealed many previously unreported viral species, genera, and families. The close genetic similarities noted between some rodent and human viruses might reflect past zoonoses. This study increases our understanding of the viral diversity in wild rodents and highlights the large number of still uncharacterized viruses in mammals.

An Analysis of External Features as Predictors of Reproductive Status in Small Mammals

External reproductive features are frequently used as indicators of the reproductive status of small mammals, but the extent to which these external features correctly indicate reproductive status is poorly known. We evaluated the validity of the external reproductive features of nipple size, vaginal condition, condition of the pubic symphysis, and position of the testes in predictions of reproductive status based on necropsy of individuals of nine populations of seven species of small mammals. Results showed that testis position is a relatively accurate predictor of reproductive status in males (87–94%), but that the external features of females are less accurate and consistent (58–85%), with nipple size being best. Predictive equations produced by logistic multiple regression analysis, using combinations of external reproductive features and body measurements, produced no significant improvement in accuracy of predictions in males, but significant improvements were obtained in analyses of data from females of three species of small mammals from Manitoba, Canada. We suggest that the use of body weight with external reproductive features in predictive equations will improve the accuracy of predictions of the reproductive status of small, seasonally breeding species.

Comparative Life Histories of Georgia and Virginia Cotton Rats

Abstract Adult hispid cotton rats (Sigmodon hispidus) were collected from the field monthly for >2 years from populations near the northern edge of their range in Virginia and contemporaneously from south-central Georgia. Body measurements and weights were taken at capture, and after dissection embryos, corpora lutea, and placental scars were counted and measured; testes and seminal vesicles were dissected out, measured, and weighed. This allowed comparison of several life-history parameters between the populations and tests of several life-history hypotheses. The breeding season was up to 2 months longer in Georgia than in Virginia, where there was typically a 3-month or longer winter inactive period. Some reproductive activity was observed among Georgia females in all 12 calendar months, whereas pregnancies were never observed in Virginia during November–February. Average litter sizes were significantly higher in Virginia (5.91 ± 1.41, up to 13) than in Georgia (5.16 ± 1.79, up to 9); this difference may partly result from a higher incidence of embryo resorption and prenatal mortality in the Georgia population, primarily in the cooler 6 months of the year. Virginia rats averaged significantly smaller for both sexes, but this was likely the result of a younger age distribution. Among reproductive males and females, no body-size differences were found between populations except that pregnant females from Virginia averaged significantly longer. Fifty percent and 75% of the random sample of adult females and males, respectively, were reproductively active in Georgia, whereas only 35% and 40% were reproductively active in Virginia. Spermatogenically active males in Virginia had significantly greater relative gonadal mass than their Georgia counterparts. Overwinter survival of parous females was lower in Virginia. Virginia populations, in a more seasonal environment, displayed a more r-selected life history, with greater reproductive allocation, faster growth (except over winter), higher mortality, and less iteroparity.

Seasonal Changes in Brown Fat and Pelage in Southern Short-Tailed Shrews

We examined cellular changes in interscapular brown adipose tissue and pelage characteristics in Blarina carolinensis collected throughout the year in eastern Virginia. Cellular volume occupied by mitochondria and maximum mitochondrial size were significantly greater in the brown adipose tissue of winter shrews than in summer shrews. Lipid droplets occupied greater volume and were larger in shrews in summer than winter shrews. There were no seasonal differences in hair density; Type I and Type II guard hairs were significantly longer in winter than summer by a factor of 1.3. Woolly hairs were 1.2 times longer in winter than summer, a non-significant difference.

Crop damage and yield loss caused by two species of rodents in irrigated fields in northern Nigeria

We investigated the damage and yield loss caused by two rodent species, Arvicanthis niloticus and Mastomys (Praomys) natalensis, in cereal crops in fields near Kano, northern Nigeria from August, 1990 to April, 1992. Using the random cluster technique, we selected 100 clusters of 30 hills in each rice and wheat field, and assessed damage fortnightly, using the Cut Tiller Count method. In premature crops, damage increased from tillering stages to the dough stages. Differences in the percentages of damage (yield loss) in mature rice (4.8% in 1990 and 12.6% in 1991) and in wheat (30.0% in 1991 and 21.7% in 1992) were statistically significant between years. Severe yield losses in wheat, compared to low and moderate levels in rice (grown for the most part during the rains), were probably a result of greater consumption of the wheat, grown entirely by irrigation during the dry season when other rodent food sources were less abundant. Results of ANOVA showed significant effects of developmental stage of the crop, year of cultivation, and their interactions on the magnitude of crop damage.

Population Ecology of the Golden Mouse

An understanding of the population dynamics of a species requires knowledge of the major life-history parameters of a population, including age at maturity, distribution of age classes, and lifetime reproductive contribution of the sexes, sex ratio, length of the breeding season, mean litter size, rates of growth and survival, and life span. Because few long-term studies have been conducted with Ochrotomys nuttalli as the focal species of investigation, only fragmentary information is available for many population parameters. As importantly, densities of golden mice often are low, making them difficult to evaluate statistically. Little has been published on age at maturity for golden mice, lifetime reproductive success, or the distribution of age classes in nature. Nevertheless, even early studies provide some useful information focusing on the natural history of this species (e.g., Linzey 1968, McCarley 1958). In this chapter, I summarize studies in which information on one or more parameter(s) is presented, standardize the results as much as possible, and attempt to uncover patterns for populations in one region (e.g., Kentucky, Illinois, Tennessee) to compare with populations from another region (e.g., Florida, Georgia, Texas).

GROWTH AND SURVIVAL IN A NORTHERN POPULATION OF HISPID COTTON RATS

Abstract Using data from a 28-month capture–mark–recapture study that included 3 winters, we compared rates of body growth and survival for a population of hispid cotton rats (Sigmodon hispidus) in eastern Virginia with another marginal population in eastern Kansas, and where possible, with southern populations in coastal Texas and central Florida. Patterns of seasonal growth were similar in Virginia and Kansas, being low, often near 0, in winter but moderate in other seasons, unlike the uniform seasonal growth rates in Texas. Survival rates were similar between the sexes in both Virginia and Kansas but the overall monthly survival rate in Kansas (0.75) was much higher than the means for Virginia (0.69 for females and 0.62 for males). In sum, despite mild and mostly snow-free winters in eastern Virginia, the patterns of body mass and rates of growth and survival were more similar to those of Kansas populations than to those of cotton rat populations from Texas or Florida.

Geographic variation and taxonomy of the southeastern shrew (Sorex longirostris)

Three subspecies of the southeastern shrew ( Sorex longirostris ) currently are recognized: S. l. eionis from Citrus Co., Florida; S. l. fisheri from the Dismal Swamp in eastern Virginia and North Carolina; and S. l. longirostris from the remainder of the southeastern United States. Based on geographic variation of seven skull, dental, and external measurements, differences among the three subspecies were confirmed. S. l. eionis was distributed throughout peninsular Florida, but S. l. fisheri had a restricted distribution. Hence, land development might represent a threat to the survival of this population. A few specimens hint at a larger range for this subspecies, but more data are needed to explore this possibility.

Coexistence of Two Congeneric Praying Mantids: A 7-Year Field Study of Reproductive Success and Failure

Abstract Two species of Asian praying mantids, Tenodera angustipennis (Saussure) and Tenodera aridifolia sinensis (Saussure), which have become common to old fields in the northeastern United States, share a common resource base that raises the question of how they can coexist in the same habitat. We studied the reproductive output measured by numbers of oothecae of naturally established populations of these two species in an old field during 7 yr (2009–2015) of secondary succession. During the initial herbaceous vegetation-dominated stage, T. angustipennis oothecae were more abundant than those of its congener, but numbers steadily declined, until it had nearly disappeared by 2014. In contrast, numbers of T. a. sinensis oothecae increased from 2007 until 2014, and then sharply declined in 2015. The steady increase in abundance of this species throughout most of the successional development during the study may be owing to greater diversity of plant species used for oviposition. We believe that the most likely reasons for the continuous decline in T. angustipennis were a combination of intraguild predation by the larger T. a. sinensis, and egg parasitism by the wasp Podagrion mantis, which is not able to parasitize oothecae of T. a. sinensis. The later decline in T. a. sinensis may reflect the fact that the site had become dominated by trees, and neither of these species is typically found in forest habitats.

Girdling by the Hispid Cotton Rat as a Significant Source of Mortality in a Loblolly Pine (Pinus Taeda) Successional Forest

Abstract The hispid cotton rat (Sigmodon hispidus) is a common field rodent in the southeastern United States, where volunteer loblolly pine trees (Pinus taeda) rapidly invade open space and accelerate ecological succession from field to pine forest. While using capture-mark-recapture methods in studies of small mammal populations, we observed dead pine trees in late summer, all killed by girdling, prompting us to determine the extent of damage on a 1.26 ha grid in a 5 y old successional pine forest in southeastern Virginia. We recorded damage to 65% of >15,000 trees, of which 2064 were killed by rodents. Areas with a high density of trees 11–40 mm in diameter (3–5 y old) were most actively damaged during late winter and early spring, following autumnal population peaks of cotton rats. Old field habitats undergoing succession to loblolly pine forest may briefly offer optimal habitat for cotton rats, and girdling by cotton rats may regulate the rate of succession depending on densities of trees and rodents.