Experimental tree removal in tallgrass prairie: variable responses of flora and fauna along a woody cover gradient.

Woody plant encroachment is a worldwide phenomenon in grassland and savanna systems whose consequence is often the development of an alternate woodland state. Theoretically, an alternate state may be associated with changes in system state variables (e.g., species composition) or abiotic parameter shifts (e.g., nutrient availability). When state-variable changes are cumulative, such as in woody plant encroachment, the probability of parameter shifts increases as system feedbacks intensify over time. Using a Before-After Control-Impact (BACI) design, we studied eight pairs of grassland sites undergoing various levels of eastern redcedar (Juniperus virginiana) encroachment to determine whether responses of flora and fauna to experimental redcedar removal differed according to the level of pretreatment redcedar cover. In the first year after removal, herbaceous plant species diversity and evenness, woody plant evenness, and invertebrate family richness increased linearly with pretreatment redcedar cover, whereas increases in small-mammal diversity and evenness were described by logarithmic trends. In contrast, increases in woody plant diversity and total biomass of terrestrial invertebrates were accentuated at levels of higher pretreatment cover. Tree removal also shifted small-mammal species composition toward a more grassland-associated assemblage. During the second year postremoval, increases in herbaceous plant diversity followed a polynomial trend, but increases in most other metrics did not vary along the pretreatment cover gradient. These changes were accompanied by extremely high growing-season precipitation, which may have homogenized floral and faunal responses to removal. Our results demonstrate that tree removal increases important community metrics among grassland flora and fauna within two years, with some responses to removal being strongly influenced by the stage of initial encroachment and modulated by climatic variability. Our results underscore the importance of decisive management for reversing the effects of woody plant encroachment in imperiled grassland ecosystems.

Variation in Neotropical river otter (Lontra longicaudis) diet: Effects of an invasive prey species.

Predation is one of the main barriers that exotic species may face in newly colonized areas and may help stop or control the potential negative impacts of invasive species in the environment. We evaluated if the consumption of an invasive prey (armored catfish: Pterygoplichtys sp.) affects the dietary niche breadth and trophic level of a native predator (Neotropical river otter: Lontra longicaudis) in northern Guatemala. We examined otter scats from three rivers: two where the invasive armored catfish occurred and one without the invasive fish. Samples were collected two and seven years after the first report of the catfish in the area. We performed gross scat analysis and stable isotope analyses of nitrogen and carbon of fecal matter. Where the invasive armored catfish occurred, it was the main prey item for L. longicaudis. Particularly in the river outside of protected areas seven years after the first report of the catfish, where it accounted for 49% of the otter diet. Concordance was found between the two techniques to estimate dietary niche breadth and trophic level. The dietary niche breath of otters was narrower seven years after the invasion in comparison to two years after the invasion in both invaded rivers, but the extent of the reduction was lesser inside the protected area. Finally, the trophic level of otters also showed a reduction related to the occurrence of the armored catfish in their diet.

Seasonal changes in the physiology of male Virginia opossums (Didelphis virginiana): signs of the Dasyurid semelparity syndrome?

Semelparity, which is multiplying once in a lifetime, is a rare reproductive strategy among mammals. Several species of the marsupial family Dasyuridae experience 100% male mortality following an intense mating period. We investigated seasonal physiological changes that may be associated with early mortality in the male Virginia opossum (Didelphis virginiana; Didelphidae) and compared these changes with those of semelparous, male dasyurids. Free-ranging male Virginia opossums (n=36) were collected during 2001 at the Oklahoma State University Cross Timbers Experimental Range. Seasonal data were collected on hematological, morphological, and helminth parameters of these individuals. We used one-way ANOVA to determine whether there were seasonal differences among means for each parameter. It appeared that male Virginia opossums experienced some physiological changes similar to those of male dasyurids exhibiting semelparity. All males collected in summer (August) were juveniles of the year. Lack of adult males in August suggests high mortality of this cohort during the breeding season. Opossum characteristics exhibiting the dasyurid semelparity syndrome included packed cell volume, adrenal mass, and helminth numbers. Minor lymphocytopenia, neutrophilia, and testosterone concentrations also were similar to semelparous dasyurids. However, a lack of change in serum cortisol concentration and body mass and dynamics in immunoglobulin protein, serum protein, and testes mass were not consistent with previous reports of semelparous dasyurid physiology. Evolutionary divergence and differences in breeding behavior between dasyurids and didelphids may be responsible for the lack of consistency between the taxa.

Effects of sex, age, and season on plasma steroids in free-ranging Texas horned lizards (Phrynosoma cornutum).

The Texas horned lizard (Phrynosoma cornutum) is protected in several states due to its apparently declining numbers; information on its physiology is therefore of interest from both comparative endocrine and applied perspectives. We collected blood samples from free-ranging P. cornutum in Oklahoma from April to September 2005, spanning their complete active period. We determined plasma concentrations of the steroids, progesterone (P), testosterone (T), and corticosterone (CORT) by radioimmunoassay following chromatographic separation and 17beta-estradiol (E2) by direct radioimmunoassay. T concentrations in breeding males were significantly higher than in non-breeding males. P showed no significant seasonal variation within either sex. CORT was significantly higher during the egg-laying season compared to breeding and non-breeding seasons for adult females and it was marginally higher in breeding than in non-breeding males (P=0.055). CORT concentrations also significantly increased with handling in non-breeding males and egg-laying females. Perhaps most surprisingly, there were no significant sex differences in plasma concentrations of P and E2. Furthermore, with respect to seasonal differences, plasma E2 concentrations were significantly higher in breeding females than in egg-laying or non-breeding females, and they were significantly higher in breeding than in non-breeding males. During the non-breeding season, yearling males exhibited higher E2 concentrations than adult males; no other differences between the steroid concentrations of yearlings and adults were detected. In comparison to other vertebrates, the seasonal steroid profile of P. cornutum exhibited both expected and unexpected patterns, and our results illustrate the value of collecting such baseline data as a springboard for appropriate questions for future research.

Nitrogen outputs from fecal and urine deposition of small mammals: implications for nitrogen cycling.

The contribution of small mammals to nitrogen cycling could have repercussions for the producer community in the maintaining or perhaps magnifying of nitrogen availability. Our objective was to model nitrogen outputs (deposition of feces and urine) of small mammals in an old-field ecosystem and estimate the amount of fecal and urinary nitrogen deposited annually. To address this objective, we used models from laboratory studies and combined these with data from field studies to estimate dietary nitrogen and monthly and annual nitrogen outputs from fecal and urine deposition of five rodent species. The models accounted for monthly fluctuations in density and biomass of small-mammal populations. We estimated that the minimal amount of nitrogen deposited by rodents was 1.0 (0.9-1.1) and 2.7 (2.6-2.9) kg Nha(-1) year(-1) from feces and urine, respectively, for a total contribution of 3.7 (3.5-4.0) kg Nha(-1) year(-1). Hispid cotton rats (Sigmodon hispidus) accounted for >75% of the total nitrogen output by small mammals. Our estimates of annual fecal and urinary nitrogen deposited by rodents were comparable to nitrogen deposits by larger herbivores and other nitrogen fluxes in grassland ecosystems and should be considered when assessing the potential effects of herbivory on terrestrial nitrogen cycles.