Predicting effects of hydrologic alteration and climate change on ecosystem metabolism in a western U.S. river.

We estimated past and future hydrographs and patterns of ecosystem metabolism in a fifth-order river of the western United States, where water use and climate change are both expected to alter hydrology in the immediate future. We first reconstructed the unregulated hydrograph to estimate how the current hydrograph has been altered. Due to consumptive use, 95% as irrigation, current discharge during summer (July-September) was 70% lower than would occur if the river was unregulated. We then predicted a future hydrograph including effects of consumptive use and climate change; the magnitude of flow changes were minor under this regime relative to those already manifested by consumptive uses. We used time-series regression and a six-year continuous record of open-water metabolism to demonstrate that, under the current hydrologic regime, gross primary production (GPP) was dependent on both water temperature and flow and that ecosystem respiration (ER) was most dependent on temperature. Monte Carlo simulations under the three hydrologic regimes and three temperature scenario's (current, +2 degrees C, +4 degrees C) suggested that flow, but not temperature, may have profound effects on the magnitude of metabolism. Linking temporally detailed analyses of ecological function and hydrology may lead to better understanding and management of changes due to basin-scale water use and/or global-scale climate change.

Modelling competition and hybridization between native cutthroat trout and nonnative rainbow and hybrid trout.

Native salmonid fish have been displaced worldwide by nonnatives through hybridization, competition, and predation, but the dynamics of these factors are poorly understood. We apply stochastic Lotka-Volterra models to the displacement of cutthroat trout by rainbow/hybrid trout in the Snake River, Idaho, USA. Cutthroat trout are susceptible to hybridization in the river but are reproductively isolated in tributaries via removal of migratory rainbow/hybrid spawners at weirs. Based on information-theoretic analysis, population data provide evidence that hybridization was the primary mechanism for cutthroat trout displacement in the first 17 years of the invasion. However, under some parameter values, the data provide evidence for a model in which interaction occurs among fish from both river and tributary subpopulations. This situation is likely to occur when tributary-spawned cutthroat trout out-migrate to the river as fry. The resulting competition with rainbow/hybrid trout can result in the extinction of cutthroat trout even when reproductive segregation is maintained.

Three patients with both Hodgkin's lymphoma and Castleman's disease: Clinicopathologic correlations and lack of association with HHV-8.

BACKGROUND: The relationship between Hodgkin's lymphoma (HL) and plasma cell-type Castleman's disease (PCD) has been well documented. There have been over 20 cases reported in the literature and nearly all of them were either diagnosed concurrently, or were initially diagnosed as PCD and upon review were found to have interfollicular HL. Human herpes virus type 8 (HHV-8) is present in about 40% of cases with PCD. It predisposes patients to a much higher risk of other malignancies, including Kaposi's sarcoma and non-Hodgkin's lymphoma. Cases linked to HHV-8 are associated with a different morphology than cases that are not linked to HHV-8. It has been proposed that patients with both HL and CD will have lymph nodes with HHV-8-negative morphology. MATERIALS AND METHODS: We present a series of three cases in a retrospective study where patients had both HL and PCD. Surgical pathology reports, clinical histories, and H and E and various immunohistochemical stains on initial work-up were examined and subsequent immunohistochemical stains for HHV-8 were obtained from the Methodist Hospital. RESULTS: Patient 1 was diagnosed with PCD and interfollicular HL in the same lymph node. Patient 2 was first diagnosed with classic HL and 2 years later returned with enlarged lymph nodes clinically suspected to be recurrent HL. Histology showed angiofollicular hyperplasia and interfollicular plasmacytosis without Reed-Sternberg cells and a diagnosis of PCD was rendered. Patient 3, a male in his third decade, was diagnosed with nodular sclerosing HL in the thymus, and concurrently PCD in the mediastinal lymph nodes. All three cases had architectural features consistent with an HHV-8-negative morphology. Immunohistochemical stains for HHV-8 were done retrospectively and were negative. CONCLUSION: All three of our patients with both HL and CD had HHV-8-negative lymph node morphology and absence of HHV-8 by immunohistochemistry. These patients, therefore, are not at an increased risk for the development of subsequent malignancies, when compared to HHV-8-positive patients. Included in our series is one unique case where the diagnosis of HL preceded CD by 2 years.

The detection and diagnosis of hemoglobin A2' by high-performance liquid chromatography.

Hemoglobin (Hb) A2' is a hematologically silent variant of HbA2 that is detected easily by high-performance liquid chromatography (HPLC), where it elutes in the S window. Our purposes were to define diagnostic criteria for the HbA2' trait using the Variant II (Bio-Rad, Hercules, CA) and to determine the prevalence of HbA2' in a metropolitan patient population. All Hb screens (N = 5,862) performed during a 26-month period were reviewed for new hemoglobinopathies. We identified 57 cases of HbA2' trait, making it the fourth most prevalent Hb variant detected in this population after HbS, HbC, and beta-thalassemia minor For HbA2' trait cases, the mean HbA2 level was 1.7% (SD, 0.17%), and the mean HbA2' level was 1.3% (SD, 0.18%). Six possible HbA2'/beta-thalassemia double heterozygotes were identified, for whom the sum of the HbA2 and HbA2' exceeded 4% of total Hb. Hb variants that might interfere with detection of HbA2' include HbS, glycosylated HbC, and HbG2. Diagnostic criteria proposed for the HbA2' trait by HPLC are HbA2 of 2% or less, S window peak of 1% to 2%, no previous diagnosis of HbS, and absence of HbG and HbC.