Latent variable models for neural data analysis

The brain is perhaps the most complex system to have ever been subjected to rigorous scientific investigation. The scale is staggering: over 10^11 neurons, each making an average of 10^3 synapses, with computation occurring on scales ranging from a single dendritic spine, to an entire cortical area. Slowly, we are beginning to acquire experimental tools that can gather the massive amounts of data needed to characterize this system. However, to understand and interpret these data will also require substantial strides in inferential and statistical techniques. This dissertation attempts to meet this need, extending and applying the modern tools of latent variable modeling to problems in neural data analysis.

It is divided into two parts. The first begins with an exposition of the general techniques of latent variable modeling. A new, extremely general, optimization algorithm is proposed - called Relaxation Expectation Maximization (REM) - that may be used to learn the optimal parameter values of arbitrary latent variable models. This algorithm appears to alleviate the common problem of convergence to local, sub-optimal, likelihood maxima. REM leads to a natural framework for model size selection; in combination with standard model selection techniques the quality of fits may be further improved, while the appropriate model size is automatically and efficiently determined. Next, a new latent variable model, the mixture of sparse hidden Markov models, is introduced, and approximate inference and learning algorithms are derived for it. This model is applied in the second part of the thesis.

The second part brings the technology of part I to bear on two important problems in experimental neuroscience. The first is known as spike sorting; this is the problem of separating the spikes from different neurons embedded within an extracellular recording. The dissertation offers the first thorough statistical analysis of this problem, which then yields the first powerful probabilistic solution. The second problem addressed is that of characterizing the distribution of spike trains recorded from the same neuron under identical experimental conditions. A latent variable model is proposed. Inference and learning in this model leads to new principled algorithms for smoothing and clustering of spike data.

Coastal erosion as a natural resource management problem: An economic perspective

Unremitting waves and occasional storms bring dynamic forces to bear on the coast. Sediment flux results in various patterns of erosion and accretion, with an overwhelming majority (80 to 90 percent) of coastline in the eastern U.S. exhibiting net erosion in recent decades. Climate change threatens to increase the intensity of storms and raise sea level 18 to 59 centimeters over the next century. Following a lengthy tradition of economic models for natural resource management, this paper provides a dynamic optimization model for managing coastal erosion and explores the types of data necessary to employ the model for normative policy analysis. The model conceptualizes benefits of beach and dune sediments as service flows accruing to nearby residential property owners, local businesses, recreational beach users, and perhaps others. Benefits can also include improvements in habitat for beach- and dune-dependent plant and animal species. The costs of maintaining beach sediment in the presence of coastal erosion include expenditures on dredging, pumping, and placing sand on the beach to maintain width and height. Other costs can include negative impacts on the nearshore environment. Employing these constructs, an optimal control model is specified that provides a framework for identifying the conditions under which beach replenishment enhances economic welfare and an optimal schedule for replenishment can be derived under a constant sea level and erosion rate (short term) as well as an increasing sea level and erosion rate (long term). Under some simplifying assumptions, the conceptual framework can examine the time horizon of management responses under sea level rise, identifying the timing of shift to passive management (shoreline retreat) and exploring factors that influence this potential shift. (PDF contains 4 pages)

Southeast White Oak River shellfish restoration project

The 42-mile-long White Oak River is one of the last relatively unblemished watery jewels of the N.C. coast. The predominantly black water river meanders through Jones, Carteret and Onslow counties along the central N.C. coast, gradually widening as it flows past Swansboro and into the Atlantic Ocean. It drains almost 12,000 acres of estuaries -- saltwater marshes lined with cordgrass, narrow and impenetrable hardwood swamps and rare stands of red cedar that are flooded with wind tides. The lower portion of the river was so renowned for fat oysters and clams that in times past competing watermen came to blows over its bounty at places that now bear names like Battleground Rock. The lower river is also a designated primary nursery area for such commercially important species as shrimp, spot, Atlantic croaker, blue crabs, weakfish and southern flounder. But the river has been discovered. The permanent population along the lower White Oak increased by almost a third since 1990, and the amount of developed land increased 82 percent during the same period. With the growth have come bacteria. Since the late 1990s, much of the lower White Oak has been added to North Carolina’s list of impaired waters because of bacterial pollution. Forty-two percent of the rivers’ oyster and clam beds are permanently closed to shellfishing because of high bacteria levels. Fully two-thirds of the river’s shellfish beds are now permanently off limits or close temporarily after a moderate rain. State monitoring indicates that increased runoff from urbanization is the probable cause of the bacterial pollution. (PDF contains 4 pages)

Aerosol-cloud-precipitation interactions in marine stratocumulus clouds

Marine stratocumulus clouds are generally optically thick and shallow, exerting a net cooling influence on climate. Changes in atmospheric aerosol levels alter cloud microphysics (e.g., droplet size) and cloud macrophysics (e.g., liquid water path, cloud thickness), thereby affecting cloud albedo and Earth’s radiative balance. To understand the aerosol-cloud-precipitation interactions and to explore the dynamical effects, three-dimensional large-eddy simulations (LES) with detailed bin-resolved microphysics are performed to explore the diurnal variation of marine stratocumulus clouds under different aerosol levels and environmental conditions. It is shown that the marine stratocumulus cloud albedo is sensitive to aerosol perturbation under clean background conditions, and to environmental conditions such as large-scale divergence rate and free tropospheric humidity.

Based on the in-situ Eastern Pacific Emitted Aerosol Cloud Experiment (E-PEACE) during Jul. and Aug. 2011, and A-Train satellite observation of 589 individual ship tracks during Jun. 2006-Dec. 2009, an analysis of cloud albedo responses in ship tracks is presented. It is found that the albedo response in ship tracks depends on the mesoscale cloud structure, the free tropospheric humidity, and cloud top height. Under closed cell structure (i.e., cloud cells ringed by a perimeter of clear air), with sufficiently dry air above cloud tops and/or higher cloud top heights, the cloud albedo can become lower in ship tracks. Based on the satellite data, nearly 25% of ship tracks exhibited a decreased albedo. The cloud macrophysical responses are crucial in determining both the strength and the sign of the cloud albedo response to aerosols.

To understand the aerosol indirect effects on global marine warm clouds, multisensory satellite observations, including CloudSat, MODIS, CALIPSO, AMSR-E, ECMWF, CERES, and NCEP, have been applied to study the sensitivity of cloud properties to aerosol levels and to large scale environmental conditions. With an estimate of anthropogenic aerosol fraction, the global aerosol indirect radiative forcing has been assessed.

As the coupling among aerosol, cloud, precipitation, and meteorological conditions in the marine boundary layer is complex, the integration of LES modeling, in-situ aircraft measurements, and global multisensory satellite data analyses improves our understanding of this complex system.

Fischerei und fischereibiologische Untersuchungen im Nordost-Atlantik im 1. Halbjahr 1996

Biological investigations were carried out onboard the commercial freezing trawler "KIEL" off the Norwegian coast in February/March and in the Bear Island area in May 1996. Data will be used as German contribution to the assessments of the ICES "Arctie Fisheries Working Group". Informations about the fishery and the concentrations of cod, haddock, saithe and redfish in the areas between Röst-, Nordvest-, Fuglöy-Bank and Bear Island are given. Length - and age distributions as well as preliminary results of stomach- and gonad investigations are represented.

Analyse des Fluchtverhaltens von Fischen auf Schiffsgeräusche mit hydroakustischen Methoden

Investigations on the avoidance reactions of pelagic schooling fish (herring and sprat) released by an approaching fishery vessel were carried out during the 378th cruise of FRC "Solea" from 25 September to 3 October 1995 in the Arkona Sea, southern Baltic. An echosounder system EK 500/BI500 with a 38 kHz transducer mounted on a towed body as weIl as a 120 kHz hull mounted transducer were used. Fish densities were measured synchronously as well as under the ship as at a laterally distances from the ship by the transducer of the towed body. By these means the variation of fish densities up to a certain distance from the ship is possible. The advantage of using an echo integrating system for these measurements is, that it works also for not schooling fish and under conditions where schooling fish disperse (e.g. at night).

Gute Kabeljau- und Schellfischvorkommen bei der Bären Insel

Biological investigations were carried out onboard the commercial freezing trawler "Kiel" in the Bear Island area in May/June 1995. Data will be used as German contribution to the assessments of the ICES "Artic Fisheries Working Group". Concentrations of cod were found southwest of Bear Island. Catches ammounted to 5-60 t/haul. Relatively good catches (5-25 t) with high proportions of haddock (40-80 %) were observed to the northwest of the island. Length- and age distributions as well as preliminary results of stomach- and gonad- investigations are presented.

Geological evolution of two crustal scale shear zones. Part I. The Rand thrust complex, northwestern Mojave Desert, California. Part II. The Magdalena metamorphic core complex, north central Sonora, Mexico

The geology and structure of two crustal scale shear zones were studied to understand the partitioning of strain within intracontinental orogenic belts. Movement histories and regional tectonic implications are deduced from observational data. The two widely separated study areas bear the imprint of intense Late Mesozoic through Middle Cenozoic tectonic activity. A regional transition from Late Cretaceous-Early Tertiary plutonism, metamorphism, and shortening strain to Middle Tertiary extension and magmatism is preserved in each area, with contrasting environments and mechanisms. Compressional phases of this tectonic history are better displayed in the Rand Mountains, whereas younger extensional structures dominate rock fabrics in the Magdalena area.

In the northwestern Mojave desert, the Rand Thrust Complex reveals a stack of four distinctive tectonic plates offset along the Garlock Fault. The lowermost plate, Rand Schist, is composed of greenschist facies metagraywacke, metachert, and metabasalt. Rand Schist is structurally overlain by Johannesburg Gneiss (= garnet-amphibolite grade orthogneisses, marbles and quartzites), which in turn is overlain by a Late Cretaceous hornblende-biotite granodiorite. Biotite granite forms the fourth and highest plate. Initial assembly of the tectonic stack involved a Late Cretaceous? south or southwest vergent overthrusting event in which Johannesburg Gneiss was imbricated and attenuated between Rand Schist and hornblende-biotite granodiorite. Thrusting postdated metamorphism and deformation of the lower two plates in separate environments. A post-kinematic stock, the Late Cretaceous Randsburg Granodiorite, intrudes deep levels of the complex and contains xenoliths of both Rand Schist and mylonitized Johannesburg? gneiss. Minimum shortening implied by the map patterns is 20 kilometers.

Some low angle faults of the Rand Thrust Complex formed or were reactivated between Late Cretaceous and Early Miocene time. South-southwest directed mylonites derived from Johannesburg Gneiss are commonly overprinted by less penetrative north-northeast vergent structures. Available kinematic information at shallower structural levels indicates that late disturbance(s) culminated in northward transport of the uppermost plate. Persistence of brittle fabrics along certain structural horizons suggests a possible association of late movement(s) with regionally known detachment faults. The four plates were juxtaposed and significant intraplate movements had ceased prior to Early Miocene emplacement of rhyolite porphyry dikes.

In the Magdalena region of north central Sonora, components of a pre-Middle Cretaceous stratigraphy are used as strain markers in tracking the evolution of a long lived orogenic belt. Important elements of the tectonic history include: (1) Compression during the Late Cretaceous and Early Tertiary, accompanied by plutonism, metamorphism, and ductile strain at depth, and thrust driven? syntectonic sedimentation at the surface. (2) Middle Tertiary transition to crustal extension, initially recorded by intrusion of leucogranites, inflation of the previously shortened middle and upper crustal section, and surface volcanism. (3) Gravity induced development of a normal sense ductile shear zone at mid crustal levels, with eventual detachment and southwestward displacement of the upper crustal stratigraphy by Early Miocene time.

Elucidation of the metamorphic core complex evolution just described was facilitated by fortuitous preservation of a unique assemblage of rocks and structures. The "type" stratigraphy utilized for regional correlation and strain analysis includes a Jurassic volcanic arc assemblage overlain by an Upper Jurassic-Lower Cretaceous quartz pebble conglomerate, in turn overlain by marine strata with fossiliferous Aptian-Albian limestones. The Jurassic strata, comprised of (a) rhyolite porphyries interstratified with quartz arenites, (b) rhyolite cobble conglomerate, and (c) intrusive granite porphyries, are known to rest on Precambrian basement north and east of the study area. The quartz pebble conglomerate is correlated with the Glance Conglomerate of southeastern Arizona and northeastern Sonora. The marine sequence represents part of an isolated arm? of the Bisbee Basin.

Crosscutting structural relationships between the pre-Middle Cretaceous supracrustal section, younger plutons, and deformational fabrics allow the tectonic sequence to be determined. Earliest phases of a Late Cretaceous-Early Tertiary orogeny are marked by emplacement of the 78 ± 3 Ma Guacomea Granodiorite (U/Pb zircon, Anderson et al., 1980) as a sill into deep levels of the layered Jurassic series. Subsequent regional metamorphism and ductile strain is recorded by a penetrative schistosity and lineation, and east-west trending folds. These fabrics are intruded by post-kinematic Early Tertiary? two mica granites. At shallower crustal levels, the orogeny is represented by north directed thrust faulting, formation of a large intermontane basin, and development of a pronounced unconformity. A second important phase of ductile strain followed Middle Tertiary? emplacement of leucogranites as sills and northwest trending dikes into intermediate levels of the deformed section (surficial volcanism was also active during this transitional period to regional extension). Gravitational instabilities resulting from crustal swelling via intrusion and thermal expansion led to development of a ductile shear zone within the stratigraphic horizon occupied by a laterally extensive leucogranite sill. With continued extension, upper crustal brittle normal faults (detachment faults) enhanced the uplift and tectonic denudation of this mylonite zone, ultimately resulting in southwestward displacement of the upper crustal stratigraphy.

Strains associated with the two ductile deformation events have been successfully partitioned through a multifaceted analysis. R_f/Ø measurements on various markers from the "type" stratigraphy allow a gradient representing cumulative strain since Middle Cretaceous time to be determined. From this gradient, noncoaxial strains accrued since emplacement of the leucogranites may be removed. Irrotational components of the postleucogranite strain are measured from quartz grain shapes in deformed granites; rotational components (shear strains) are determined from S-C fabrics and from restoration of rotated dike and vein networks. Structural observations and strain data are compatable with a deformation path of: (1) coaxial strain (pure shear?), followed by (2) injection of leucogranites as dikes (perpendicular to the minimum principle stress) and sills (parallel to the minimum principle stress), then (3) southwest directed simple shear. Modeling the late strain gradient as a simple shear zone permits a minimum displacement of 10 kilometers on the Magdalena mylonite zone/detachment fault system. Removal of the Middle Tertiary noncoaxial strains yields a residual (or pre-existing) strain gradient representative of the Late Cretaceous-Early Tertiary deformation. Several partially destrained cross sections, restored to the time of leucogranite emplacement, illustrate the idea that the upper plate of the core complex bas been detached from a region of significant topographic relief. 50% to 100% bulk extension across a 50 kilometer wide corridor is demonstrated.

Late Cenozoic tectonics of the Magdalena region are dominated by Basin and Range style faulting. Northeast and north-northwest trending high angle normal faults have interacted to extend the crust in an east-west direction. Net extension for this period is minor (10% to 15%) in comparison to the Middle Tertiary detachment related extensional episode.

Perspectiva social de la infancia en algunas comunidades del País Vasco: Opiniones de las y los adolescentes de Santurtzi sobre sexismo

Resumen: El presente estudio ha intentado conocer la perspectiva que tienen las y los adolescentes de Santurtzi sobre el sexismo, eligiendo la adolescencia por su carácter fundamental en el desarrollo vital. Partimos de una investigación cualitativa llevada a cabo con 17 adolescentes a través de grupos de discusión sobre sexismo, patriarcado y alternativas de construcción personal, como son el movimiento feminista y el movimiento de nuevas masculinidades. Hemos podido corroborar el desconocimiento que todavía existe en materia de igualdad y la necesidad de intervenciones socioeducativas que de ello se desprenden, para así continuar con la lucha contra el patriarcado y contra la discriminación hacia la mujer, desde la perspectiva que nos aporta el sexismo ambivalente.

Ancillary ligand effects : from zirconium(IV)-catalyzed homogeneous propylene polymerization to platinum(II)-mediated C-H bond activation

A series of C_s- and C₁-symmetric doubly-linked ansa-metallocenes of the general formula {1,1'-SiMe₂-2,2'-E-('ƞ⁵-C₅H₂-4-R¹)-(ƞ⁵-C₅H-3',5'-(CHMe₂)₂)}ZrC₂ (E = SiMe₂ (1), SiPh₂ (2), SiMe₂ -SiMe₂ (3); R¹ = H, CHMe₂, C₅H₉, C₆H₁₁, C₆H₅) has been prepared. When activated by methylaluminoxane, these are active propylene polymerization catalysts. 1 and 2 produce syndiotactic polypropylenes, and 3 produces isotactic polypropylenes. Site epimerization is the major pathway for stereoerror formation for 1 and 2. In addition, the polymer chain has slightly stronger steric interaction with the diphenylsilylene linker than with the dimethylsilylene linker. This results in more frequent site epimerization and reduced syndiospecificity for 2 compared to 1.

C₁-Symmetric ansa-zirconocenes [1,1 '-SiMe₂-(C₅H₄)-(3-R-C₅H₃)]ZrCl₂ (4), [1,1 '-SiMe₂-(C₅H₄)-(2,4-R₂-C₅H₂)]ZrCl₂ (5) and [1,1 '-SiMe₂-2,2 '-(SiMe₂-SiMe₂)-(C₅H₃)-( 4-R-C₅H₂)]ZrCl₂ (6) have been prepared to probe the origin of isospecificity in 3. While 4 and 3 produce polymers with similar isospecificity, 5 and 6 give mostly hemi-isotactic-like polymers. It is proposed that the facile site epimerization via an associative pathway allows rapid equilibration of the polymer chain between the isospecific and aspecific insertion sites. This results in more frequent insertion from the isospecific site, which has a lower kinetic barrier for chain propagation. On the other hand, site epimerization for 5 and 6 is slow. This leads to mostly alternating insertion from the isospecific and aspecific sites, and consequently, a hemi-isotactic-like polymers. In comparison, site epimerization is even slower for 3, but enchainment from the aspecific site has an extremely high kinetic barrier for monomer coordination. Therefore, enchainment occurs preferentially from the isospecific site to produce isotactic polymers.

A series of cationic complexes [(ArN=CR-CR=NAr)PtMe(L)]⁺[BF₄]⁺ (Ar = aryl; R = H, CH₃; L = water, trifluoroethanol) has been prepared. They react smoothly with benzene at approximately room temperature in trifluoroethanol solvent to yield methane and the corresponding phenyl Pt(II) cations, via Pt(IV)-methyl-phenyl-hydride intermediates. The reaction products of methyl-substituted benzenes suggest an inherent reactivity preference for aromatic over benzylic C-H bond activation, which can however be overridden by steric effects. For the reaction of benzene with cationic Pt(II) complexes, in which the diimine ligands bear 3,5-disubstituted aryl groups at the nitrogen atoms, the rate-determining step is C-H bond activation. For the more sterically crowded analogs with 2,6-dimethyl-substituted aryl groups, benzene coordination becomes rate-determining. The more electron-rich the ligand, as reflected by the CO stretching frequency in the IR spectrum of the corresponding cationic carbonyl complex, the faster the rate of C-H bond activation. This finding, however, does not reflect the actual C-H bond activation process, but rather reflects only the relative ease of solvent molecules displacing water molecules to initiate the reaction. That is, the change in rates is mostly due to a ground state effect. Several lines of evidence suggest that associative substitution pathways operate to get the hydrocarbon substrate into, and out of, the coordination sphere; i.e., that benzene substitution proceeds by a solvent- (TFE-) assisted associative pathway.

Dynamic rupture simulation integrated with earthquake observations

Dynamic rupture simulations are unique in their contributions to the study of earthquake physics. The current rapid development of dynamic rupture simulations poses several new questions: Do the simulations reflect the real world? Do the simulations have predictive power? Which one should we believe when the simulations disagree? This thesis illustrates how integration with observations can help address these questions and reduce the effects of non-uniqueness of both dynamic rupture simulations and kinematic inversion problems. Dynamic rupture simulations with observational constraints can effectively identify non-physical features inferred from observations. Moreover, the integrative technique can also provide more physical insights into the mechanisms of earthquakes. This thesis demonstrates two examples of such kinds of integration: dynamic rupture simulations of the M_w 9.0 2011 Tohoku-Oki earthquake and of earthquake ruptures in damaged fault zones:

(1) We develop simulations of the Tohoku-Oki earthquake based on a variety of observations and minimum assumptions of model parameters. The simulations provide realistic estimations of stress drop and fracture energy of the region and explain the physical mechanisms of high-frequency radiation in the deep region. We also find that the overridding subduction wedge contributes significantly to the up-dip rupture propagation and large final slip in the shallow region. Such findings are also applicable to other megathrust earthquakes.

(2) Damaged fault zones are usually found around natural faults, but their effects on earthquake ruptures have been largely unknown. We simulate earthquake ruptures in damaged fault zones with material properties constrained by seismic and geological observations. We show that reflected waves in fault zones are effective at generating pulse-like ruptures and head waves tend to accelerate and decelerate rupture speeds. These mechanisms are robust in natural fault zones with large attenuation and off-fault plasticity. Moreover, earthquakes in damaged fault zones can propagate at super-Rayleigh speeds that are unstable in homogeneous media. Supershear transitions in fault zones do not require large fault stresses. In the end, we present observations in the Big Bear region, where variability of rupture speeds of small earthquakes correlates with the laterally variable materials in a damaged fault zone.

Heavy metal pollution in sediments at shipbreaking area of Bangladesh

The research was carried out to assess the trace metal concentration in sediments of ship breaking area in Bangladesh. The study areas were separated into Ship breaking Zone and Reference Site for comparative analysis. Metals like Iron ( Fe) was found at 11932 to 41361.71µg.g-1 in the affected site and 3393.37 µg.g-1 in the control site. Manganese (Mn) varied from 2.32 to 8.25 µg.g-1 in the affected site where as it was recorded as 1.8 µg.g-1 in the control area. Chromium(Cr), Nickel (Ni), Zinc(Zn) and Lead (Pb) were also varied from 22.89 to 86.72 µg.g-1; 23.12 to 48.6;83.78 to 142.85 and 36.78 to 147.83 µg.g-1 respectively in the affected site whereas these were recorded as 19; 3.98; 22.22 and 8.82 µg.g-1 in the control site. Copper (Cu); Cadmium (Cd) and Mercury (Hg) concentration were varied from 21.05 to 39.85; 0.57 to 0.94 and 0.05 to 0.11 µg.g-1 in the affected site and 33.0; 0.115 and 0.01 µg.g-1 in the control site. It may conclude that heavy metal pollution in sediments at ship breaking area of Bangladesh is at alarming stage.

Valley evolution by meandering rivers

Fluvial systems form landscapes and sedimentary deposits with a rich hierarchy of structures that extend from grain- to valley scale. Large-scale pattern formation in fluvial systems is commonly attributed to forcing by external factors, including climate change, tectonic uplift, and sea-level change. Yet over geologic timescales, rivers may also develop large-scale erosional and depositional patterns that do not bear on environmental history. This dissertation uses a combination of numerical modeling and topographic analysis to identify and quantify patterns in river valleys that form as a consequence of river meandering alone, under constant external forcing. Chapter 2 identifies a numerical artifact in existing, grid-based models that represent the co-evolution of river channel migration and bank strength over geologic timescales. A new, vector-based technique for bank-material tracking is shown to improve predictions for the evolution of meander belts, floodplains, sedimentary deposits formed by aggrading channels, and bedrock river valleys, particularly when spatial contrasts in bank strength are strong. Chapters 3 and 4 apply this numerical technique to establishing valley topography formed by a vertically incising, meandering river subject to constant external forcing—which should serve as the null hypothesis for valley evolution. In Chapter 3, this scenario is shown to explain a variety of common bedrock river valley types and smaller-scale features within them—including entrenched channels, long-wavelength, arcuate scars in valley walls, and bedrock-cored river terraces. Chapter 4 describes the age and geometric statistics of river terraces formed by meandering with constant external forcing, and compares them to terraces in natural river valleys. The frequency of intrinsic terrace formation by meandering is shown to reflect a characteristic relief-generation timescale, and terrace length is identified as a key criterion for distinguishing these terraces from terraces formed by externally forced pulses of vertical incision. In a separate study, Chapter 5 utilizes image and topographic data from the Mars Reconnaissance Orbiter to quantitatively identify spatial structures in the polar layered deposits of Mars, and identifies sequences of beds, consistently 1-2 meters thick, that have accumulated hundreds of kilometers apart in the north polar layered deposits.

Feira de São Cristóvão: uma incursão sobre os estudos na área e a tentativa de aproximação à relação global-local

O presente trabalho se propõe fazer uma aproximação aos estudos sobre a tradicionalmente chamada Feira de São Cristóvão, levando em consideração a relação entre os agentes partícipes e a dinâmica no espaço. O objetivo é compreender melhor a força que move os atores no sentido da mudança da forma da feira para a forma do Centro Luiz Gonzaga de Tradições Nordestinas (CLGTN), e assim me aproximar da seguinte questão: Por que uma forma de comércio como a modelada na Feira de São Cristóvão, popularmente conhecida como feira dos paraíbas, sobrevive à força das mudanças operadas no ramo, dando lugar aos sofistificados shopping-centers? Para isso, faço uma aproximação ao estado de arte da pesquisa na área com base nas seguintes categorias de análise: a Feira como forma de produção do espaço e a relação Forma-Conteúdo, a relação espaço-identidade na feira e a relação entre agentes internos e externos no Espaço. Vendo a migração como movimento responsável pela constituição de um espaço social e territorialmente diferenciado do nordestino na cidade do Rio de Janeiro, é por ela que inicio a exposição. Esta se continua com a análise dos trabalhos escolhidos para estudo, via as categorias indicadas e, com um capítulo no qual procuro repensar a feira a partir da relação global-local que vem referenciando as reflexões acumuladas na área do urbano a partir da segunda metade do século XX. Por fim, concluo observando que a permanência da feira se deve a um duplo movimento de resistência-absorção que, se redefinindo no tempo e no espaço, não consegue, entretanto, apagar os traços da tradição que lhe deu origem.

Selective plankton feeding by the coregonid ”bondella” (Coregonus sp.) of Lake Maggiore [Translation from: Translation from: Memorie dell'Istituto Italiano di Idrobiologia Dott.Marco de Marchi 31 181-203, 1974]

In recent times, some specialists have begun to study the subject of food selection of fish in an organized way, thus highlighting its importance. In the present work, the author intends to evaluate the impact of predation on the eupelagic zooplanktonic biocenosis of Lake Maggiore, producing, in this way, as far as possible, a basis for a better understanding of the population dynamics of the species of zooplankton directly involved. Another aspect which has been studied is that related to the mechanism of selective capture set in action by the predator. To this end the study tries to bear in mind that the subject should be interpreted as a function of numerous factors acting contemporaneously, that is as the interaction of characters peculiar to the predator and to the prey. The species studied, locally called ”bondella”, belongs to the family Salmonidae, subfamily Coregoninae and was introduced into Lake Maggiore in 1950.