Showing papers by "Durham University published in 1990"

Early and long-term effects of child sexual abuse: An update.

Abstract: Revue de questions sur le probleme de l'influence des agressions sexuelles sur les enfants. Cette revue complete et actualise une revue de question anterieure (1986) et permet d'identifier de nouveaux domaines de recherche

Intertrochanteric femoral fractures. Mechanical failure after internal fixation

Abstract: In a prospective study we assessed the causes of mechanical failure in a series of 230 intertrochanteric femoral fractures which had been internally fixed with either a sliding hip screw or a Kuntscher Y-nail. The overall rate of mechanical failure was 16.5%; cutting-out of the implant from the femoral head was the cause in three-quarters of the instances. Implants placed posteriorly in the femoral head cut out more often (27%) than those placed centrally (7%). The cut-out rate was also determined by the quality of the fracture reduction, but age, walking ability and bone density (assessed by the Singh grade and metacarpal indices) had no significant influence. We conclude that these fractures should be reduced as accurately as possible and it is imperative that the implant is placed centrally within the femoral head.

The 60-μ and far-infrared luminosity functions of IRAS galaxies

Abstract: Une etude d'un echantillon de 2818 galaxies IRAS est realisee. La fonction de luminosite a 60 μm independante des variations de flux, et la distribution de densite radiale, independante de la fonction de luminosite, sont determinees avec une methode du maximum de vraisemblance

Parallel long-term trends across four marine trophic levels and weather

Abstract: THE temporal changes revealed by long-term studies of components of the marine ecosystem1–4 are of particular relevance given the growing concern about trace-gas-induced climatic change5. The extent to which they reflect the signal of weather relates directly to the validity of using them for detecting climatic change6–8. We report here on remarkable similarities in trends among four series of marine data and one of climatic data. The series overlap in time over 33 years (1955–1987), and in space over the northwestern quadrant of the North Sea. They relate to the abundances of phytoplankton, zooplankton and herring (Clupea harengus), to the breeding performance of the kittiwake gull (Rissa tridactyla), and to variations in weather pattern. The mechanisms behind the parallelism in trends remain unclear; a major challenge is to construct food-web models capable of reproducing the pattern and thereby of use in interpreting and detecting global climatic change.

Twistor Geometry and Field Theory

Abstract: Part I. Geometry: 1. Klein correspondence 2. Fibre bundles 3. Differential geometry 4. Integral geometry Part II. Field Theory: 5. Linear field theory 6. Gauge theory 7. General relativity Part III. The Penrose Transform: 8. Massless free fields 9. Self-dual gauge fields 10. Self-dual space-times 11. General gauge fields 12. Stationary axisymmetric space-times Special topics.

Large-scale distribution of galaxies at the Galactic poles

Abstract: GALAXIES, mapped in two or three dimensions, are not distributed randomly but are clustered on small scales ( 0.2. Similarly deep surveys with greater angular spread are needed to verify our results and to determine the implications for cosmology.

Method of preparing silicon carbide surfaces for crystal growth

Abstract: The invention is a method of forming a substantially planar surface on a monocrystalline silicon carbide crystal by exposing the substantially planar surface to an etching plasma until any surface or subsurface damage caused by any mechanical preparation of the surface is substantially removed. The etch is limited, however, to a time period less than that over which the plasma etch will develop new defects in the surface or aggravate existing ones, and while using a plasma gas and electrode system that do not themselves aggravate or cause substantial defects in the surface.

CMAS: an associative neural network alternative to backpropagation

Abstract: The CMAC (cerebellar model arithmetic computer) neural network, an alternative to backpropagated multilayer networks, is described. The following advantages of CMAC are discussed: local generalization, rapid algorithmic computation based on LMS (least-mean-square) training, incremental training, functional representation, output superposition, and a fast practical hardware realization. A geometrical explanation of how CMAC works is provided, and applications in robot control, pattern recognition, and signal processing are briefly described. Possible disadvantages of CMAC are that it does not have global generalization and that it can have noise due to hash coding. Care must be exercised (as with all neural networks) to assure that a low error solution will be learned. >

Method of improving mechanically prepared substrate surfaces of alpha silicon carbide for deposition of beta silicon carbide thereon and resulting product

Abstract: The invention is a method of improving a mechanically prepared surface of alpha silicon carbide for increasing the polytype purity of an epitaxial layer of beta silicon carbide grown thereon. The method comprises decreasing the frequency of exposed higher energy lattice positions along a mechanically prepared {0001} surface of a 6H alpha silicon carbide substrate by adding atoms to vacant lattice positions and by increasing the average height and separation between steps on a mechanically prepared {0001} surface of 6H silicon carbide.

Real-time dynamic control of an industrial manipulator using a neural network-based learning controller

Abstract: A learning control technique that uses an extension of the cerebellar model articulation control network developed by J.S. Albus (1975) is discussed, and results of real-time control experiments that involved learning the dynamics of a five-axis industrial robot (General Electric P-5) during high-speed movements are presented. During each control cycle, a training scheme was used to adjust the weights in the network in order to form an approximate dynamic model of the robot in appropriate regions of the control space. Simultaneously, the network was used during each control cycle to predict the actuator drives required to follow a desired trajectory, and these drives were used as feedforward terms in parallel to a fixed-gain linear feedback controller. Trajectory tracking errors were found to converge to low values within a few training trials, and to be relatively insensitive to the choice of control system gains. The effects of network memory size and trajectory characteristics on learning system performance were investigated. >

Motives for modular forms.

Abstract: Let M be a pure motive over a number field F of rank n with coefficients in T ⊂ C; M may be thought of as a direct factor of the cohomology of a smooth projective variety X over F cut out by an algebraic correspondence defined over T . It was Langlands who propounded that there should be a bijective correspondence between (resp. absoultey irreducible) motives M and algebraic (resp. cuspidal) automorphic representations Π of GLn(AF ), in such a way that their invariants, ‘L-functions in s ∈ C’, L(M, s) and L(Π, s) match up. Until twenty years ago, most of positive results had been about the direction Π 7→ M = MΠ inspired by Deligne’s insights in [4]; but there have been some spectacular development in the opposite ‘harder’ direction M 7→ Π = ΠM (commonly knowns as ‘modularity of Galois representations’) in Wiles/TaylorWiles on modularity of semi-stable elliptic curves over Q and Breuil-Conrad-Diamond-Taylor on the Shimura-Taniyama-Weil conjecture, culminating in Kisin and Emerton (independently) on the FontaineMazur conjecture for GL2 over Q. One can even formulate and prove some cases of mod p and local p-adic analogues of ‘Langlands correspondences’, while the research currently led by Calegari and Geraghty seems very promising to improve considerably our understanding of significantly many new cases of bijections between M ’s and Π’s. With that spectacular development aside, the goal of this seminar is to unravel the correspondence Π 7→M in the case n = 2 and F is a totally real field; more precisely, we shall read [4] and [8] for F = Q, and [1] for general F .

Growing up and growing apart: a developmental meta-analysis of twin studies.

Abstract: Developmental change in twin similarity was examined with age contrasts in a meta-analysis of twin studies from 1967 through 1985. Intraclass rs were coded from 103 papers that included data for monozygotic or dizygotic twins, or for both, on personality or intelligence variables. Analyses indicated that there was a general tendency for some intraclass rs to decrease with age. In other words, as twins grow up, they grow apart. There were also developmental differences associated with components of variance for heritability, the shared environment, and the nonshared environment. Mechanisms through which the nonshared environment may operate are discussed.

Earthquake mechanisms in the Hellenic Trench near Crete

Abstract: SUMMARY In the Hellenic Trench south of Crete convergence between the southern Aegean Sea and Africa occurs at a rate of at least 60 mm yr-1. Previously published first motion fault plane solutions show a variety of different fault orientations and types, but are not well constrained. Furthermore, the lack of reliable focal depths for these earthquakes has obscured any simple pattern of deformation that might exist. Nonetheless, the mechanisms of these earthquakes have strongly influenced views of the tectonics in the Hellenic Trench. We have improved estimates of the fault parameters and focal depths for 14 of these earthquakes, using long-period P- and SH-waveforms. The earthquake mechanisms fall into four groups: (a) normal faults with a N-S strike in the over-riding material above the subduction zone; (b) low-angle thrusts with an E-W strike at a depth of about 40 km; (c) high-angle reverse faults with the same strike but shallower focal depths than (b); (d) events within the suducting lithosphere with approximately E-W P axes. The thrusting in groups (b) and (c) is probably the mechanism by which the sediments of the Mediterranean Sea underplate and uplift Crete. These events have slip vectors in the direction 025 ± 12° which represents the convergence direction between Crete and Africa along the SW-facing boundary of the Hellenic Trench. One of the events in group (d) occurred beneath the Mediterranean Ridge and involved high-angle reverse faulting with a WNW-ESE P axis: almost perpendicular to the direction of shortening deduced from folds at the surface. The Mediterranean Sea floor in this region appears to be in a state of E-W compression, for reasons that are not clear.

A new mechanism of granite emplacement: intrusion in active extensional shear zones

Abstract: ONE of the principal ways in which continental crust grows is by the incorporation, from deeper within the Earth, of large volumes of granitic magma A 'space problem' exists1 as to how these magmas are accommodated in the crust Traditionally two main emplacement mechanisms have been emphasized: 'forceful' intrusion, whereby buoyancy-driven magmas physically push the crust aside, creating granitic diapirs and balloons; and 'passive' emplacement characterized by replacive mechanisms such as cauldron subsidence and stoping Although more recent work2,3 has demonstrated that space for granites may be created within bends and offsets of large transcurrent faults, the simple view of either forceful or passive still fails to account for the intrusive mechanisms of many granites Here we report the discovery of a new igneous intrusion mechanism In the superbly exposed Proterozoic continental crust of South Greenland we have observed that rapakivi granite was intruded as large-scale sheets along ductile extensional shear zones that were active during emplacement In such a process the space problem seems to be simply resolved

The use of heuristic strategies in the interpretation of pronouns.

Abstract: The aim of the two experiments reported here was to distinguish between two heuristic strategies that have been proposed to account for the assignment of pronouns: the subject assignment strategy and the parallel function strategy. According to the subject assignment strategy, a pronoun is assigned to a preceding subject noun phrase, whereas according to the parallel function strategy, a pronoun is assigned to a previous noun phrase in the same grammatical position as the pronoun. These two strategies were tested by examining the interpretation of single object pronouns, first in a reading task and second in an assignment task. In both experiments, there was a strong preference for assigning an object pronoun to the preceding subject noun phrase, thus supporting the subject assignment strategy. However, this was only the case for pronouns that were linguistically ambiguous. When assignment was constrained by gender, there was no effect of either strategy. It is suggested that heuristic strategies are only used in the absence of other strong cues to assignment.

Repetition priming and face processing: Priming occurs within the system that responds to the identity of a face

Abstract: A familiar stimulus that has recently been recognized will be recognized a second time more quickly and more accurately than if it had not been primed by the earlier encounter. This is the phenomen...

Device for processing biological specimens for analysis of nucleic acids

Abstract: A system is provided for the automated detection of target nucleic acid sequences in which multiple biological samples are individually incorporated into matrices within carriers in a two-dimensional format. The matrix carrier is an inexpensive, disposable unit that represents a closed system separating the specimens. The system integrates sample preparation within the matrix and facilitates biochemical reactions, addition of reagents and washes, removal of waste fluids, temperature control and the automated processing thereof. The carriers are processed in stepwise treatments that expose, amplify, and detect the presence or absence of specific genetic entities in each sample. The nucleic acids or other desired biological components held within the matrices are treated by one or more of the techniques such as amplification, electrophoresis, analyte-receptor binding or hybridization as selected for the desired analysis.

Methane flux from the Amazon River floodplain: Emissions during rising water

Abstract: Methane flux data obtained during a period of high and falling water level in the course of the dry season of 1985 (the Amazon Boundary Layer Experiment, ABLE 2A) and a period of moderate and rising water during the wet season of 1987 (ABLE 2B) were used to characterize the influence of seasonal variations in the vegetation, water column depth, and chemistry, as well as atmospheric dynamics, on the methane flux from the Amazon River floodplain. It was found that the annual estimate of methane from wetlands is identical to the annual estimate made by Matthews and Fung (1987) (both at 111 Tg). However, it was found that peatlands between 50 and 70 N contribute 39 Tg, with the large areas of forested and nonforested bogs making up 37 Tg of this figure, while the figures of Matthews and Fung were 63 and 62 Tg, respectively.

Truncating the Schwinger-Dyson equations: How multiplicative renormalizability and the Ward identity restrict the three point vertex in QED

Abstract: Nonperturbative studies of field theory require the Schwinger-Dyson equations to be truncated to make them tractable. Thus, when investigating the behavior of the fermion propagator, for example, an Ansatz for the three-point vertex has to be made. While the well-known Ward identity determines the longitudinal part of this vertex in terms of the fermion propagator as shown by Ball and Chiu, it leaves the transverse part unconstrained. However, Brown and Dorey have recently emphasized that the requirement of multiplicative renormalizability is not satisfied by arbitrary Ansa$iuml---tze for the vertex. We show how this requirement restricts the form of the transverse part. By considering the example of QED in the quenched approximation, we present a form for the vertex that not only satisfies the Ward identity but is multiplicatively renormalizable to all orders in leading and next-to-leading logarithms in perturbation theory and so provides a suitable Ansatz for the full three-point vertex.

Aerosol chemistry during the wet season in central Amazonia - The influence of long-range transport

Abstract: The temporal variation in the concentration and chemistry of the atmospheric aerosol over central Amazonia, Brazil, during the 1987 wet season is discussed based on ground and aircraft collected data obtained during the NASA GTE ABLE 2B expedition conducted in April/May 1987. It is found that wet-season aerosol concentrations and composition are variable in contrast to the more uniform biogenic aerosol observed during the 1985 dry season; four distinct intervals of enhanced aerosol concentration coincided with short periods (3 to 5 d) of extensive rainfall. It is hypothesized that aerosol chemistry in Amazonia during the wet season is strongly influenced by long-range transport of soil dust, marine aerosol, and possibly biomass combustion products advected into the central Basin by large-scale tropospheric circulation, producing periodic pulses of material input to local boundary layer air. The resultant wet-season aerosol regime is dynamic, in contrast to the uniformity of natural biogenic aerosols during the dry season.

A method for correcting geographically separated remanence directions for the purpose of archaeomagnetic dating

Abstract: SUMMARY Spatial variations in the geomagnetic field must be taken into account if secular variation master curves and directional magnetic dates are to be optimized. Two methods for relocating remanence vectors have been proposed and in this paper their relative accuracies are compared using a numerical model based on the present-day field. A method which converts archaeomagnetic directions via a virtual geomagnetic pole is shown to be the more efficient transformation. For an‘archaeomagnetic region’the size of the British Isles, (900 km radius), the maximum error in relocating vectors to a central location is predicted to be of the order of 1.2°.

European post‐glacial forests: compositional changes in response to climatic change

Abstract: . Multivariate analysis of an extensive palyno-logical database for Europe has enabled reconstruction of broad-scale vegetation history. Whereas many major features of present vegetation patterns were established early in the Holocene, floristic composition of the forests has changed continuously up to the present day. For example, although ‘mixed deciduous forests’ had reached approximately their present extent in northwest Europe by 8000 B.P., Tilia peaked in abundance in these forests during the middle post glacial, whereas Pinus was most abundant in these forests during the early post-glacial and Fagus increased in abundance only in recent millennia. Pollen-climate response surfaces for major pollen taxa show how their distribution and abundance patterns relate to contemporary climate. Past forest-compositional changes were responses to climatic changes, the nature of which can be inferred from pollen-climate response surfaces. Post-glacial climate changes have been different in magnitude and direction in different regions of Europe. For example, in recent millennia the vegetation changes indicate decreasing summer temperatures in northern Europe but increasing summer temperatures in the Mediterranean region. The way in which vegetation responded to past climatic changes gives insight into the likely response of vegetation to future climate changes induced by the ‘greenhouse effect’.

A comparison between CMAC neural network control and two traditional adaptive control systems

Abstract: A comparison is made of a neural-network-based controller similar to the cerebellar model articulation controller (CMAC) and two traditional adaptive controllers, a self-tuning regulator (STR) and a Lyapunov-based model reference adaptive controller (MRAC). The three systems are compared conceptually and through simulation studies on the same low-order control problem. Results are obtained for the case where the system is linear and noise-free, for the case where noise is added to the system, and for the case where a nonlinear system is controlled. Comparisons are made with respect to closed-loop system stability, speed of adaptation, noise rejection, the number of required calculations, system tracking performance, and the degree of theoretical development. The results indicate that the neural network approach functions well in noise, works for linear and nonlinear systems, and can be implemented very efficiently for large-scale systems. >