Showing papers by "University of Tennessee published in 1990"

A set of level 3 basic linear algebra subprograms

Abstract: This paper describes an extension to the set of Basic Linear Algebra Subprograms. The extensions are targeted at matrix-vector operations that should provide for efficient and portable implementations of algorithms for high-performance computers

Stock Returns and Volatility

Abstract: Most asset pricing models postulate a positive relationship between a stock portfolio's expected returns and risk, which is often modeled by the variance of the asset price. This paper uses GARCH in mean models to examine the relationship between mean returns on a stock portfolio and its conditional variance or standard deviation. After estimating a variety of models from daily and monthly portfolio return data, we conclude that any relationship between mean returns and own variance or standard deviation is weak. The results suggest that investors consider some other risk measure to be more important than the variance of portfolio returns.

Performance of various computers using standard linear equations software

Abstract: This report compares the performance of different computer systems in solving dense systems of linear equations. The comparison involves approximately a hundred computers, ranging from a CRAY Y-MP to scientific workstations such as the Apollo and Sun to IBM PCs.

Time since death and decomposition of the human body: variables and observations in case and experimental field studies

Abstract: Much of the difficulty in determining the time since death stems from the lack of systematic observation and research on the decomposition rate of the human body. Continuing studies conducted at the University of Tennessee, Knoxville, provide useful information on the impact of carrion insect activity, ambient temperature, rainfall, clothing, burial and depth, carnivores, bodily trauma, body weight, and the surface with which the body is in contact. This paper reports findings and observations accumulated during eight years of research and case studies that may clarify some of the questions concerning bodily decay.

Adaptation: Statistics and a Null Model for Estimating Phylogenetic Effects

Abstract: -Tests of adaptive explanations are often critically confounded by phylogenetic heritage. In this paper we propose statistics and a null model for estimating phylogenetic effects in comparative data. We apply a model-independent measure of autocorrelation (Moran's I) for estimating whether cross-taxonomic trait variation is related to phylogeny. We develop a phylogenetic correlogram for assessing how autocorrelation varies with patristic distance and for judging the appropriateness and effectiveness of an autoregressive model. We then revise Cheverud et al.'s (1985, Evolution, 39:1335-1351) autocorrelational model to incorporate greater flexibility in the relation between trait variation and phylogenetic distance. Finally, we analyze various comparative data sets (body weight in carnivores, clutch size in birds) and phylogenies (morphological, molecular) to illustrate some of the complications that may arise from using an autoregressive model and to explore the effects of different weighting matrices in adjusting for these complications. Although our approach has limitations, it is both effective in partitioning trait variation into adaptive and phylogenetic components and flexible in adjusting to peculiarities in taxonomic distribution. [Phylogenetic effects; phylogenetic correlation; autoregressive models; comparative methods.] The comparative method is commonly used to investigate adaptation. A researcher examines the attributes of a number of species. Statistical analyses of these data are then used to formulate and test adaptive hypotheses of life history, morphology, physiology, demography, and behavior (e.g., Clutton-Brock and Harvey, 1977; Damuth, 1981; Gittleman and Harvey, 1982; Harvey and Clutton-Brock, 1985; Gittleman, 1986a, b; Huey and Bennett, 1987). If traits are analyzed across a broad range of independently derived taxa, the resulting adaptive explanations may be quite robust (Clutton-Brock and Harvey, 1984; Huey and Bennett, 1986; Gittleman, 1989). If, however, the data reflect a highly structured phylogeny (with little statistical independence), results may be misleading (Felsenstein, 1985). To neglect phylogeny is to invite type I and type II errors (see Fig. 1). A number of techniques have been developed for removing the effects of phylogeny (see reviews in Huey, 1987; Pagel and Harvey, 1988; Gittleman, 1989; Burghardt and Gittleman, 1990). Some of these techniques are better suited for particular variables or certain evolutionary questions, and all possess limitations. Nominal or categorical data (e.g., mating system: monogamy, polygamy) may be analyzed by evaluating the agreement between the variation in a trait and an accepted phylogeny (Dobson, 1985; Greene, 1986) or by using outgroup comparisons to identify evolutionary transitions among traits (Gittleman, 1981; Ridley, 1983). For quantitative data, there are several strategies. One may avoid spurious correlation by averaging over closely related species, thereby reducing the degrees of freedom and significance of the correlation. Alternatively, one may transform the data so that phylogenetically disparate groups appear on a common scale. Even within this general framework there are several methods for evaluating the association between the ordinal or continuous values of a trait and phylogeny: (1) Nested analysis of variance partitions the total variation in a continuous character among various taxonomic levels. By selecting the taxonomic level that accounts for the greatest proportion of the total variance as the appropriate level for analysis, this method attempts to control for bias from low-level clades that are both uniform and speciesrich (Harvey and Mace, 1982; Harvey and

An Investigation of Students' Perspectives on Anxiety and Speaking

Abstract: Language anxiety is becoming an important area of research in our profession. Debilitating language anxiety can have profound consequences on the language learning process. The purpose of this study is to examine anxiety and speaking from the students' perspective. Do speaking activities indeed make students nervous? If so, what is it about speaking that makes students anxious? Which speaking activities make students particularly uncomfortable? Can instructors do anything to reduce anxiety, particularly speaking anxiety, in the classroom? A questionnaire designed to identify sources of anxiety over speaking in the foreign language was administered to 135 university-level beginning Spanish students and 109 high school students. The questionnaire had three sections: the first one asked students to agree or disagree with twenty-four items related to language anxiety; the second section asked students to indicate their level of anxiety regarding certain in-class practices; the third section asked learners to identify instructor characteristics and instructor practices that helped reduce language anxiety. Results of the analysis of data suggest, among other things, that speaking in the foreign language in not exclusively the source of student anxiety, but that speaking in front of the class is. Furthermore, the instructor's relaxed and positive error-correction attitude can greatly reduce language anxiety. Additional suggestions for reducing language anxiety are offered

An adaptive hysteresis-band current control technique of a voltage-fed PWM inverter for machine drive system

Abstract: An adaptive hysteresis-band control method where the band is modulated with the system parameters to maintain the modulation frequency to be nearly constant is described. Although the technique is applicable to general AC motor drives and other types of load, an interior permanent magnet (IPM) synchronous machine load is considered. Systematic analytical expressions of the hysteresis band are derived as functions of system parameters. An IPM machine drive system with a voltage-fed current-controlled PWM (pulse width modulation) inverter has been computer simulated to study the performance of the proposed method. >

LAPACK: a portable linear algebra library for high-performance computers

Abstract: The goal of the LAPACK project is to design and implement a portable linear algebra library for efficient use on a variety of high-performance computers. The library is based on the widely used LINPACK and EISPACK packages for solving linear equations, eigenvalue problems, and linear least-squares problems, but extends their functionality in a number of ways. The major methodology for making the algorithms run faster is to restructure them to perform block matrix operations (e.g., matrix-matrix multiplication) in their inner loops. These block operations may be optimized to exploit the memory hierarchy of a specific architecture. The LAPACK project is also working on new algorithms that yield higher relative accuracy for a variety of linear algebra problems. >

Rapid, Sensitive Bioluminescent Reporter Technology for Naphthalene Exposure and Biodegradation

Abstract: A bioluminescent reporter plasmid for naphthalene catabolism (pUTK21) was developed by transposon (Tn4431) insertion of the lux gene cassette from Vibrio fischeri into a naphthalene catabolic plasmid in Pseudomonas fluorescens The insertion site of the lux transposon was the nahG gene encoding for salicylate hydroxylase Luciferasemediated light production from P fluorescens strains harboring this plasmid was induced on exposure to naphthalene or the regulatory inducer metabolite, salicylate In continuous culture, light induction was rapid (15 minutes) and was highly responsive to dynamic changes in naphthalene exposure Strains harboring pUTK21 were responsive to aromatic hydrocarbon contamination in Manufactured Gas Plant soils and produced sufficient light to serve as biosensors of naphthalene exposure and reporters of naphthalene biodegradative activity The robust and sensitive nature of the bioluminescent reporter technology suggests that new sensing methods can be developed for on-line process monitoring and control in complex environmental matrices

Peer status and aggression in boys' groups: developmental and contextual analyses.

Abstract: The social transactions of popular, rejected, neglected, and average first- and third-grade boys were examined during their initial encounters with peers. 23 groups of 5 or 6 boys each were observed for 45-min free-play sessions conducted on 5 consecutive days, with sociometric interviews following each session. Social preference in the play groups correlated significantly with classroom social preference after the third and subsequent play sessions for the third graders, and after the fourth and subsequent sessions for the first graders. The observational coding system distinguished 4 types of aggressive behavior that were hypothesized to relate to peer status in different ways. The first, rough play, was not related to peer status. However, rejected boys at both ages displayed significantly higher rates of angry reactive aggression and instrumental aggression than average boys. The relation between bullying and peer status varied with the age of the child. Popular first graders engaged in more bullying than average first graders, but popular third graders did not differ from average in bullying. Other questions concerned the temporal relation between play group behaviors and social preference scores within the group. Socially interactive behaviors anteceded high preference by peers, and low preference in turn led to social isolation in subsequent sessions. Language: en

K K molecules

Abstract: We have extended and improved our previous studies of the {ital qq{bar q}}{ital {bar q}} system, paying special attention to the {ital K{bar K}} channels. The new results clarify and reinforce the {ital K{bar K}} molecule interpretation of the {ital S}{sup *} and {delta}. Effective meson-meson potentials extracted from the {ital qq{bar q}}{ital {bar q}} system and couplings to {ital q{bar q}} scalar mesons lead to coupled-channel amplitudes which closely resemble those observed in both the {ital S}{sup *} and {delta} channels. We also show that the {ital K{bar K}} molecules have properties consisent with experimental results on the production of the {ital S}{sup *} and {delta} and of low-mass {ital K{bar K}} pairs.

Estrogen replacement and coronary artery disease. Effect on survival in postmenopausal women.

Abstract: The relationship among postmenopausal estrogen use, coronary stenosis, and survival was examined retrospectively in 2268 women undergoing coronary angiography. The patients were selected for study if their age was 55 years or older at the time of angiography or if they had previously undergone bilateral oophorectomy. Postmenopausal estrogen use in 1178 patients with coronary artery disease (greater than 70% stenosis) and 644 patients with mild to moderate coronary artery disease (5% to 69% stenosis) was compared with 446 control subjects (0% stenosis) using life-table analysis. Over 10 years of follow-up, there was no significant difference in survival among patients initially free of coronary lesions on arteriography who had either never used (377) or ever used (69) estrogens. Among patients with mild to moderate coronary stenosis, 10-year survival of those who had never used estrogens was 85.0% and it was 95.6% among 99 "ever users." Survival was 60.0% among those with more than 70% coronary stenosis who had never used estrogen and it was 97.0% among 70 ever users. The "never users" group were older (65 vs 59 years), had a lower proportion of cigarette smokers (40% vs 57.1%), a higher proportion of subjects with diabetes (21.7% vs 12.9%) and hyperlipidemia (58% vs 44%), and approximately equal numbers of hypertensives (56.0% vs 54.3%). Cox's proportional hazards model was used to estimate survival as a function of multiple covariables. Estrogen use was found to have a significant, independent effect on survival in women. We conclude that estrogen replacement after menopause prolongs survival when coronary artery disease is present, but it has less effect in the absence of coronary artery disease.

A new method for analyzing data from continuous depth-sensing microindentation tests

Abstract: A new parameter, hardness/modulus2 (H/E2), has been derived from the equations used to calculate the hardness and elastic modulus from data taken during continuous depth-sensing microindentation tests. This paper discusses the use of this parameter to treat the data obtained from a sample whose surface roughness was of the same scale as the size of the indents. The resulting data were widely scattered. This scatter was reduced when the data were plotted in terms of H/E2 versus stiffness. The effect of surface roughness on the hardness and elastic modulus results is removed via stiffness measurements, provided single contacts are made between the indenter and the specimen. The function relating the cross-sectional area of the indenter versus the distance from its point is not required for calculation of H/E2, but the hardness and modulus cannot be determined separately. The parameter H/E2 indicates resistance to plastic penetration in this case.

An approach to automated detection of tumors in mammograms

Abstract: An automated system for detecting and classifying particular types of tumors in digitized mammograms is described The analysis of mammograms is performed in two stages First, the system identifies pixel groupings that may correspond to tumors Next, detected pixel groupings are subjected to classification The essence of the first processing stage is multiresolution image processing based on fuzzy pyramid linking The second stage uses a classification hierarchy to identify benign and malignant tumors Each level of the hierarchy uses deterministic or Bayes classifiers and a particular measurement The measurements pertain to shape and intensity characteristics of particular types of tumors The classification hierarchy is organized in such a way that the simplest measurements are used at the top, with the system stepping through the hierarchy only when it cannot classify the detected pixel groupings with certainty >

Communities as assembled structures: Do rules govern pattern?

Abstract: Because of its unwieldy complexity, understanding organization at the community level has been an elusive, often frustrating enterprise. This survey concentrates on the assembly process itself, and examines whether there are mechanics to community assembly that provide the foundation for a general theory of organization at the community level.

Effects of feeds on flavor of red meat: a review.

Abstract: The effects of diet ingredients on red meat flavor are dependent on the type of diet, and, to a large extent, on the species: pork, mutton or lamb and beef Sensory analysis of meat flavor has been used in most of the studies on the effects of feeds on meat flavor In general, high-energy grain diets produced a more acceptable or a more intense flavor in red meats than low-energy forage or grass diets Feeding pigs unsaturated fats increases the unsaturation in pork fat but results in only minor changes in pork flavor Sheep must be fed protected, unsaturated fats in order to increase the unsaturation in their fat to be similar to pork fat This increased unsaturation results in a greater flavor change in lamb or beef than in pork Several dietary ingredients such as fish products, raw soybeans, canola oil and meal, and pasture grasses cause undesirable flavors in red meat Analyses of lamb and beef produced on different diets have shown that type of feed affects the concentration of many flavor volatile compounds However, only a few studies have quantified the volatiles of beef and lamb produced on different diets, and no reports of studies were found in which the volatiles of pork produced on different diets had been investigated Hence, the importance of these changes in volatile concentration to meat flavor has not been determined

Parvalbumin-containing gabaergic interneurons in the rat neostriatum

Abstract: Antibodies to the intracellular calcium binding protein parvalbumin were shown to label specifically a distinct group of neostriatal GABAergic neurons. These neurons corresponded to the intensely staining subclass of neostriatal GABAergic neurons that have previously been shown to be a class of aspiny interneurons in the neostriatum. The parvalbumin neurons were aspiny neurons with varicose dendrites distributed throughout the neostriatum in a pattern identical to the intensely stained GABA neurons, and both populations of neurons showed increased numbers in the lateral part of the neostriatum. Double labeling of single neurons with both the GABA and parvalbumin antisera showed that all parvalbumin neurons were positive for GABA, but some GABA labelled neurons were not immunoreactive for parvalbumin. These parvalbumin-negative GABAergic neurons were morphologically similar to the spiny projection neurons, which are GABAergic but usually are not so heavily stained. The relationship of the GABA-containing parvalbumin neurons to the striatal mosaic organization was determined by using immunocytochemistry for another calcium binding protein, calbindin D28K, to label the matrix compartment of the striatum. The distribution of parvalbumin-positive neurons relative to the calbindin-positive matrix and calbindin-poor patches was determined by using pairs of adjacent sections stained with the calbindin and parvalbumin antisera. This analysis showed that the somata of the parvalbumin neurons were present in both patch and matrix compartments, and their axons and dendrites crossed the boundaries between compartments. A quantitative analysis of the number of neurons in each compartment revealed that the neurons showed no preferential distribution in either compartment, but instead were present according to the area occupied by that compartment. Approximately 10% of parvalbumin neurons were in the patch compartment, and in these same sections, the patch compartment occupied approximately 10% of the area of those sections. Staining with parvalbumin antibodies can therefore be used to identify a single class of GABAergic aspiny interneurons that is present in both patch and matrix compartments, and whose processes cross the borders between these compartments.

Ferritin is a component of the neuritic (senile) plaque in Alzheimer dementia.

Abstract: A strong immunoreactivity for ferritin was observed in the neuritic (senile) plaques in Alzheimer's disease hippocampus. The ferritin accumulation was almost exclusively associated with the microglia, which appeared to have proliferated greatly. These cells were also positive for HLA-DR, a putative marker for reactive microglia. In contrast, in the diffuse plaques, which were without neuritic pathology, the ferritin-stained microglia appeared to be normal. Microglia were seen frequently in contact with neurons undergoing neurofibrillary changes but only the tangles in the extracellular space were ferritin positive. No ferritin was detected, by Western blots, in paired helical filaments isolated from Alzheimer's disease brain, suggesting that ferritin was most likely weakly associated with and was not a constituent of these fibrils. No correlation between increased ferritin/microglia activity and blood-brain barrier leakage was detected. Ferritin, an iron-storage protein, might have a role in the formation of amyloid through the action of free radicals generated during the release of iron from the ferritin molecule. Alternatively, the ferritin/microglia system might be secondarily involved in the removal and processing of the amyloid.

Dynamic voting algorithms for maintaining the consistency of a replicated database

Abstract: There are several replica control algorithms for managing replicated files in the face of network partitioning due to site or communication link failures. Pessimistic algorithms ensure consistency at the price of reduced availability; they permit at most one (distinguished) partition to process updates at any given time. The best known pessimistic algorithm, voting, is a “static” algorithm, meaning that all potential distinguished partitions can be listed in advance. We present a dynamic extension of voting called dynamic voting. This algorithm permits updates in a partition provided it contains more than half of the up-to-date copies of the replicated file. We also present an extension of dynamic voting called dynamic voting with linearly ordered copies (abbreviated as dynamic-linear). These algorithms are dynamic because the order in which past distinguished partitions were created plays a role in the selection of the next distinguished partition. Our algorithms have all the virtues of ordinary voting, including its simplicity, and provide improved availability as well. We provide two stochastic models to support the latter claim. In the first (site) model, sites may fail but communication links are infallible; in the second (link) model the reverse is true. We prove that under the site model, dynamic-linear has greater availability than any static algorithm, including weighted voting, if there are four or more sites in the network. In the link model, we consider all biconnected five-site networks and a wide variety of failure and repair rates. In all cases considered, dynamic-linear had greater availability than any static algorithm.

Calculations of Mott scattering cross section

Abstract: Calculations of Mott elastic scattering cross section of electrons for most elements of the periodic table up to element number 94 in the energy range 20 eV–20 keV have been performed. The Dirac equation transformed to a first‐order differential equation was solved numerically. The influence of the choice of atomic potential on the scattering factor was studied in comparison to a simple muffin‐tin approximation of the atomic potential in solids. The application of calculated cross sections to a conventional Monte Carlo model for electron scattering using modified Bethe equation is described and results concerning the electron backscattering for different atomic potentials are compared.

Assessment and Modification of Cognitive Distortions in Sex Offenders

Abstract: Sexual aggression against children and adults is being recognized both by professionals and by the public as a significant problem for society (Badgley, 1984; Koss, Gidycz, & Wisniewski, 1987; Russell, 1982). In an attempt to address the problem, there have been attempts to more fully understand the individuals who perpetuate such abuse and to determine relevant target areas for treatment. This literature has produced numerous descriptive studies, personality assessment studies, and classification systems which have been well-reviewed previously (Knight, Rosenberg, & Schneider, 1985; Levin & Stava, 1987; Murphy & Stalgaitis, 1987; Prentky, Cohen, & Seghorn, 1985; Quinsey, 1984, 1986). All of these approaches provide some understanding of sex offenders, although they do not always clearly translate into specific treatment programs.

The Role of Handwashing in Prevention of Endemic Intensive Care Unit Infections

Abstract: Handwashing is believed to be the most important means of preventing nosocomial infections. Previous studies of healthcare workers (HCWs) have shown that handwashing practices are poor. No one has shown that handwashing practices can be easily improved and that this prevents endemic (nonepidemic) nosocomial infection. Handwashing and infection rates were studied in two intensive care units (ICUs) of a community teaching hospital. Handwashing rates were monitored secretly throughout the study. After six months of observation, we started interventions to increase handwashing. Handwashing increased gradually, but overall rates before (22.0%) and after (29.9%) interventions were not significantly different (p = .071). Handwashing never occurred before intravenous care, whereas it occurred 67.5% for all other indications (p less than .0001). When questioned, nurses felt they were washing appropriately nearly 90% of the time. Infection rates seemed unrelated to handwashing throughout the study, and no clusters of infection were detected. We conclude that handwashing rates, when measured against arbitrary but reasonable standards, are suboptimal, difficult to change and not closely related to evidence of cross-infection. Further, nurses wash hands selectively, depending on the indication for handwashing, and generally believe they are washing much more frequently than an objective observer believes they are.

Algorithm 679: A set of level 3 basic linear algebra subprograms: model implementation and test programs

Abstract: This paper describes a model implementation and test software for the Level 3 Basic Linear Algebra Subprograms (Level3 BLAS). The Level3 BLAS are targeted at matrix-matrix operations with the aim of providing more efficient, but portable, implementations of algorithms on high-performance computers. The model implementation provides a portable set of Fortran 77 Level 3 BLAS for machines where specialized implementations do not exist or are not required. The test software aims to verify that specialized implementations meet the specification of the Level 3 BLAS and that implementations are correctly installed.

Lipid composition is important for highly efficient target binding and retention of immunoliposomes.

Abstract: By taking advantage of a monoclonal IgG antibody, 34A, which is highly specific to pulmonary endothelial cells, we have prepared liposomes containing various amounts of antibody molecules (immunoliposomes). These immunoliposomes accumulate specifically in the lung when injected i.v. Two lipid compositions were used: phosphatidylcholine/cholesterol/phosphatidylserine (PS), 10:5:1 (mol/mol), a composition that allows liposomes to be readily taken up by the reticuloendothelial system (RES) (liver and spleen), and phosphatidylcholine/cholesterol/ganglioside GM1, 10:5:1 (mol/mol), a composition that allows liposomes to avoid or delay the RES uptake (the so-called stealth liposomes). Although an increase in the number of antibody molecules per liposome was accompanied by an increased level of lung binding of the immunoliposomes, differences due to the lipid composition were more profound. For example, stealth immunoliposomes containing an antibody/lipid ratio = 1:37 (wt/wt) accumulated in lung to a level of 60% of the injected dose, whereas PS-containing immunoliposomes with a higher antibody/lipid ratio (1:8) only accumulated 50% of the injected dose in the lung. Conjugation of antibody to the stealth liposome did not increase the rate of liposome uptake by liver; this rate was approximately 10-fold lower than that of the PS-containing liposomes without antibody. Stealth immunoliposomes with high antibody content also showed long retention in the lung. The t1/2 of lung residence for the stealth immunoliposomes with an antibody/lipid ratio = 1:11 (wt/wt) was approximately 24 hr. The fact that stealth immunoliposomes showed a longer retention time in the lung than the PS-containing immunoliposomes of similar antibody content suggests that macrophages may play a role in the removal of the bound immunoliposomes from the pulmonary endothelium. Alternatively, dissociated stealth immunoliposomes may reenter the circulation and rebind to the lung target, causing an apparent slow overall dissociation rate. These results can be understood on the basis of two competing kinetic processes: lung binding whose rate is directly proportional to the antibody content of the immunoliposomes and uptake by RES whose rate is significantly reduced in the case of the stealth liposomes. Even for a modest level of antibody content, the half-life for target binding of immunoliposomes was significantly shorter than the half-life of liver uptake of the liposomes, resulting in a favorable target binding. Significant immunoliposome binding to the lung is not due to the fact that tail vein-injected liposomes flow through the lung capillary bed before they encounter the liver, because portal vein-injected immunoliposomes showed the same rate and extent of target binding as the tail vein-injected ones.

Phytoplankton growth and new production in the Weddell Sea marginal ice zone in the austral spring and autumn

Abstract: Rates of primary production, N03- uptake, and NH,l uptake by phytoplankton were measured in the marginal ice zone of the Weddell Sea in austral spring 1983 and autumn 1986. In spring Chl concentrations averaged 3 pg liter-‘, primary productivity 490 mg C m-2 d-l, and surface phytoplankton growth rates 0.30 doublings d-l. In autumn these rates were all much lower, averaging 0.14 pg liter- l, 126 mg C m-2 d-l, and 0.14 doublings d-l. During both seasons NH,+ was consistently the preferred source of N, but because of the much greater availability of N03- in the euphotic zone (21-28 PM NO,- vs. -0.4 yM NH,+ during both seasons), N03- uptake rates generally equaled or exceeded those of NH, +. Vertically integrated&ratios (the ratio of NO3- uptake to the sum of NO,- plus NH,+ uptake) averaged 0.52 (range from 0.35 to 0.70) in spring and 0.72 (range from 0.60 to 0.84) in autumn. In 1983 we observed sharply elevated relative preference indices for NH,+ whenever the ambient NH,+ concentration was ~0.3 PM, which we interpret as evidence that Antarctic phytoplankton growing at NH.,+ concentrations CO.3 PM have the ability to increase their rate of NH,+ uptake rapidly in response to increased availability. We estimate the new production of the marginal ice zone to be at least 49 g C m-2 yr-1 for the NovemberMarch period. This value supports previous findings that the ice edge is the main site of production in the Southern Ocean and that it is likely to dominate the annual cycle of organic matter flux from the euphotic zone.