Single Feature Set Research Articles

User-perceived quality assessment of streaming media using reduced feature sets

While subjective measurements are the most natural for assessing the user-perceived quality of a media stream, there are issues with their scalability and their context accuracy. We explore techniques to select application-layer measurements, collected by an instrumented media player, that most accurately predict the subjective quality rating that a user would assign to a stream. We consider three feature subset selection techniques that reduce the number of features (measurements) under consideration to ones most relevant to user-perceived stream quality. Two of the three techniques mathematically consider stream characteristics when selecting measurements, while the third is based on observation. We apply the reduced feature sets to two nearest-neighbor algorithms for predicting user-perceived stream quality. Our results demonstrate that there are clear strategies for estimating the quality rating that work well in specific circumstances such as video-on-demand services. The results also demonstrate that neither of the mathematically-based feature subset selection techniques identify a single set of features that is unambiguously influential on user-perceived stream quality, but that ultimately a combination of retransmitted and/or lost application-layer packets is most accurate for predicting stream quality.

Application of hybrid image features for fast and non-invasive classification of raisin

A new approach for the non-invasive classification of raisins is presented based on the hybrid image features, namely morphological, color and texture features. A total of 74 features (8 morphological, 30 color, and 36 textural) were extracted from RBG images. Seven kinds of models were established based on different feature sets. They were three kinds of models established based on single feature set, three kinds of models established based on the combination of two feature sets, and one kind of model established based on the combination of all feature sets. Five kinds of classifiers, namely partial least squares (PLS), linear discriminant analysis (LDA), soft independent modeling of class analogy (SIMCA), and least squares support vector machine (LS-SVM) with linear and radial basis function (RBF) kernels were used for the model establishment based on different feature sets. The best correct answer rates (CAR) of 99% was obtained when LDA was used to establish the classification model based on the combination of all feature sets, which was higher than those of the models established based on single feature set or the combination of two feature sets. The results show that the feature combination is helpful to improve the accuracy of raisin classification. It was concluded that the varieties of raisin could be accurately classified based on RGB image features and the combination of morphological, color and texture features was an accurate way to improve the accuracy of classification.

Improving acoustic vehicle classification by information fusion

We present an information fusion approach for ground vehicle classification based on the emitted acoustic signal. Many acoustic factors can contribute to the classification accuracy of working ground vehicles. Classification relying on a single feature set may lose some useful information if its underlying sound production model is not comprehensive. To improve classification accuracy, we consider an information fusion diagram, in which various aspects of an acoustic signature are taken into account and emphasized separately by two different feature extraction methods. The first set of features aims to represent internal sound production, and a number of harmonic components are extracted to characterize the factors related to the vehicle’s resonance. The second set of features is extracted based on a computationally effective discriminatory analysis, and a group of key frequency components are selected by mutual information, accounting for the sound production from the vehicle’s exterior parts. In correspondence with this structure, we further put forward a modified Bayesian fusion algorithm, which takes advantage of matching each specific feature set with its favored classifier. To assess the proposed approach, experiments are carried out based on a data set containing acoustic signals from different types of vehicles. Results indicate that the fusion approach can effectively increase classification accuracy compared to that achieved using each individual features set alone. The Bayesian-based decision level fusion is found to be improved than a feature level fusion approach.

The interplay between comparative concepts and descriptive categories (Reply to Newmeyer)

The interplay between comparative concepts and descriptive categories (Reply to Newmeyer) Martin Haspelmath In his reply to my discussion note ‘Comparative concepts and descriptive categories in crosslinguistic studies’(Haspelmath 2010, henceforth H10), Newmeyer (2010, henceforth N) defends the widespread old idea that there are crosslinguistic categories and that the same kinds of concepts are and should be used both in crosslinguistic studies and in language-particular description. His critique does not amount to a full-blown endorsement of the categorial-universalist position, however, which in its prototypical generative form assumes a single set of categories or features for both purposes. For instance, when N says that ‘it is only by means of working out the interplay between the languageparticular and the language-independent that we can hope to understand either’ (p. 688), he implies that the two are not identical, and that the relationship between them is nontrivial, much as I have emphasized. The bulk of N’s reply is devoted to listing various problems in H10’s examples that he sees as undermining its general claims, especially with respect to the notions of subject, thematic roles, adjectives, clause, and word. In this brief reply, I try to show that none of them are real, and that a closer consideration of these cases actually strengthens my position. N first observes that despite my criticism of the equation of language-particular ‘subject’ categories across languages, I do use ‘subject’ as a comparative concept in some of the generalizations and definitions (in particular, in examples 3 and 13). I did not give the relevant definition in H10 for lack of space, hoping that it could be inferred, but here it is: ‘the agent of a simple transitive clause’ (the same definition is assumed by Greenberg 1963). This is of course a very different concept from ‘Subject in Tagalog’ (defined in terms of case-marking by ang, very often different from the agent NP) and from ‘Subject in English’ (perhaps defined as in N’s ex. 6, comprising expletive there, which surely is not an agent). Thus, this underscores rather than undermines my central point. N’s own concrete proposal for defining ‘subject’ (‘[an entity that manifests] more of the core properties of subjects than of constructs that contrast with subjects’, p. 692) is also a possible comparative concept, provided that the ‘core properties’ can be defined in a universally applicable way. As I noted in H10 (§7.3) in the discussion of Keenan’s (1976) definition of ‘subject’, however, it is unclear how one can draw up such a list of core properties on a principled basis. And most crucially, it is unlikely that such a definition will be found useful in language-particular studies. Descriptive categories are normally defined in terms of specific conditions, not in vague quantitative terms (‘more properties than’). N also criticizes me for using semantic roles such as ‘recipient’ (in 2) and ‘agent’ (in 16) (even though he uses the latter concept himself in ex. 6). But he overlooks that the problems with such semantic roles have never arisen in a typological context. Nobody has ever seen problems with defining the comparative concept ‘ergative’ in terms of ‘agent’. It is only when one tries to describe particular languages with a single universal set of semantic roles that one runs into problems. But we can describe I play the sonata [End Page 696] in purely grammatical terms, without reference to general semantic roles.1 Generative linguists try to use as many crosslinguistic categories in the description of individual languages as possible, and this often leads to insurmountable problems. Typology is happy to limit its generalizations to clear cases of agents, patients, and recipients. Admittedly, we cannot formulate universals over sentences like I play a sonata, but typology does not strive for exhaustiveness of coverage. Comparative linguists can thus simply leave them aside, whereas descriptivists have to find a category for them, and this is difficult if one only wants to use crosslinguistic categories. Let us look at adjectives next. N is of course right that ‘it makes all the difference in the world for generalizations about Adjective-Noun order’ how ‘adjective’ is defined (p. 690). Greenberg and Dryer...

Decision Fusion for Visual Inspection of Textiles

Defect detection of textiles is a necessary requirement for quality control and customer satisfaction. This paper presents a system for decision fusion in order to enhance the accuracy of defect detection in textiles. A multi-classifier decision fusion technique based on majority voting is presented to solve the problems of sensitivity to parameter variation and to make use of the advantages of the individual feature sets for accurate texture characterization. Features based on Gray Level Co-occurrence Matrix (GLCM), Laws Energy (LE) Filter, Local Binary Patterns (LBP), HU Moment invariants, Moment of Inertia (MOI) and Standard Deviation of Gray levels are used to train a set of Learning Vector Quantization (LVQ) classifiers. Detection accuracies of classifiers trained on single-feature sets are compared with those of the majority voting among the individual classifiers. The results obtained from majority voting indicate that the decision fusion technique improves the accuracy and reliability of the detection process. Empirical results indicate the high accuracy of the presented approach. The correct defect detection rate of the proposed approach is 98.64% with an average false acceptance rate of 0.0012.

Combining image, voice, and the patient’s questionnaire data to categorize laryngeal disorders

This paper is concerned with soft computing techniques for categorizing laryngeal disorders based on information extracted from an image of patient's vocal folds, a voice signal, and questionnaire data. Multiple feature sets are exploited to characterize images and voice signals. To characterize colour, texture, and geometry of biological structures seen in colour images of vocal folds, eight feature sets are used. Twelve feature sets are used to obtain a comprehensive characterization of a voice signal (the sustained phonation of the vowel sound /a/). Answers to 14 questions constitute the questionnaire feature set. A committee of support vector machines is designed for categorizing the image, voice, and query data represented by the multiple feature sets into the healthy, nodular and diffuse classes. Five alternatives to aggregate separate SVMs into a committee are explored. Feature selection and classifier design are combined into the same learning process based on genetic search. Data of all the three modalities were available from 240 patients. Among those, 151 patients belong to the nodular class, 64 to the diffuse class and 25 to the healthy class. When using a single feature set to characterize each modality, the test set data classification accuracy of 75.0%, 72.1%, and 85.0% was obtained for the image, voice and questionnaire data, respectively. The use of multiple feature sets allowed to increase the accuracy to 89.5% and 87.7% for the image and voice data, respectively. The test set data classification accuracy of over 98.0% was obtained from a committee exploiting multiple feature sets from all the three modalities. The highest classification accuracy was achieved when using the SVM-based aggregation with hyper parameters of the SVM determined by genetic search. Bearing in mind the difficulty of the task, the obtained classification accuracy is rather encouraging. Combination of both multiple feature sets characterizing a single modality and the three modalities allowed to substantially improve the classification accuracy if compared to the highest accuracy obtained from a single feature set and a single modality. In spite of the unbalanced data sets used, the error rates obtained for the three classes were rather similar.

Creating diverse nearest-neighbour ensembles using simultaneous metaheuristic feature selection

The nearest-neighbour (1NN) classifier has long been used in pattern recognition, exploratory data analysis, and data mining problems. A vital consideration in obtaining good results with this technique is the choice of distance function, and correspondingly which features to consider when computing distances between samples. In recent years there has been an increasing interest in creating ensembles of classifiers in order to improve classification accuracy. This paper proposes a new ensemble technique which combines multiple 1NN classifiers, each using a different distance function, and potentially a different set of features (feature vector). These feature vectors are determined for each distance metric simultaneously using Tabu Search to minimise the ensemble error rate. We show that this approach implicitly selects for a diverse set of classifiers, and by doing so achieves greater performance improvements than can be achieved by treating the classifiers independently, or using a single feature set. Naturally, optimising the level of ensembles necessitates a much larger solution space, to make this approach tractable, we show how Tabu Search at the ensemble level can be hybridised with local search at the level of individual classifiers. The proposed ensemble classifier with different distance metrics and different feature vectors is evaluated using various benchmark datasets from UCI Machine Learning Repository and a real-world machine-vision application. Results have indicated a significant increase in the performance when compared with various well-known classifiers. Furthermore, the proposed ensemble method is also compared with ensemble classifier using different distance metrics but with same feature vector (with or without feature selection (FS)).

In silico method for systematic analysis of feature importance in microRNA-mRNA interactions

BackgroundMicroRNA (miRNA), which is short non-coding RNA, plays a pivotal role in the regulation of many biological processes and affects the stability and/or translation of mRNA. Recently, machine learning algorithms were developed to predict potential miRNA targets. Most of these methods are robust but are not sensitive to redundant or irrelevant features. Despite their good performance, the relative importance of each feature is still unclear. With increasing experimental data becoming available, research interest has shifted from higher prediction performance to uncovering the mechanism of microRNA-mRNA interactions.ResultsSystematic analysis of sequence, structural and positional features was carried out for two different data sets. The dominant functional features were distinguished from uninformative features in single and hybrid feature sets. Models were developed using only statistically significant sequence, structural and positional features, resulting in area under the receiver operating curves (AUC) values of 0.919, 0.927 and 0.969 for one data set and of 0.926, 0.874 and 0.954 for another data set, respectively. Hybrid models were developed by combining various features and achieved AUC of 0.978 and 0.970 for two different data sets. Functional miRNA information is well reflected in these features, which are expected to be valuable in understanding the mechanism of microRNA-mRNA interactions and in designing experiments.ConclusionsDiffering from previous approaches, this study focused on systematic analysis of all types of features. Statistically significant features were identified and used to construct models that yield similar accuracy to previous studies in a shorter computation time.

Combining magnitude and shape features for hyperspectral classification

The spectral angle mapper (SAM) and maximum likelihood classification (MLC) are two traditional classifiers for hyperspectral classification. This paper presents two methods to combine magnitude and shape features, one for each classifier. As the magnitude and shape features are complementary, combining both features can improve the classification accuracy. First, magnitude features are represented by the spectral radiance vector, whereas shape features are represented by the spectral gradient vector. Then, in SAM, each feature vector generates a spectral angle for each class. The two generated angles are added together to obtain a single similarity, which is used for the final classification. Similarly, in MLC, after the dimensionality reduction using Fisher's linear discriminant (FLD), each feature vector in the new feature space generates a likelihood. The two generated likelihoods are multiplied to obtain a single value, which is adopted for the final classification. Experimental results on an Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data set demonstrate that the proposed methods outperform the methods with a single feature set.

Automated classification of lung bronchovascular anatomy in CT using AdaBoost.

Lung CAD systems require the ability to classify a variety of pulmonary structures as part of the diagnostic process. The purpose of this work was to develop a methodology for fully automated voxel-by-voxel classification of airways, fissures, nodules, and vessels from chest CT images using a single feature set and classification method. Twenty-nine thin section CT scans were obtained from the Lung Image Database Consortium (LIDC). Multiple radiologists labeled voxels corresponding to the following structures: airways (trachea to 6th generation), major and minor lobar fissures, nodules, and vessels (hilum to peripheral), and normal lung parenchyma. The labeled data was used in conjunction with a supervised machine learning approach (AdaBoost) to train a set of ensemble classifiers. Each ensemble classifier was trained to detect voxels part of a specific structure (either airway, fissure, nodule, vessel, or parenchyma). The feature set consisted of voxel attenuation and a small number of features based on the eigenvalues of the Hessian matrix (used to differentiate structures by shape). When each ensemble classifier was composed of 20 weak classifiers, the AUC values for the airway, fissure, nodule, vessel, and parenchyma classifiers were 0.984+/-0.011, 0.949+/-0.009, 0.945+/-0.018, 0.953+/-0.016, and 0.931+/-0.015, respectively. The strong results suggest that this could be an effective input to higher-level anatomical based segmentation models with the potential to improve CAD performance.

Feature Extraction for Systolic Heart Murmur Classification

Heart murmurs are often the first signs of pathological changes of the heart valves, and they are usually found during auscultation in the primary health care. Distinguishing a pathological murmur from a physiological murmur is however difficult, why an "intelligent stethoscope" with decision support abilities would be of great value. Phonocardiographic signals were acquired from 36 patients with aortic valve stenosis, mitral insufficiency or physiological murmurs, and the data were analyzed with the aim to find a suitable feature subset for automatic classification of heart murmurs. Techniques such as Shannon energy, wavelets, fractal dimensions and recurrence quantification analysis were used to extract 207 features. 157 of these features have not previously been used in heart murmur classification. A multi-domain subset consisting of 14, both old and new, features was derived using Pudil's sequential floating forward selection (SFFS) method. This subset was compared with several single domain feature sets. Using neural network classification, the selected multi-domain subset gave the best results; 86% correct classifications compared to 68% for the first runner-up. In conclusion, the derived feature set was superior to the comparative sets, and seems rather robust to noisy data.

Ab initio phasing based on topological restraints: automated determination of the space group and the number of molecules in the unit cell.

The connectivity-based phasing method has been demonstrated to be capable of finding molecular packing and envelopes even for difficult cases of structure determination, as well as of identifying, in favorable cases, secondary-structure elements of protein molecules in the crystal. This method uses a single set of structure factor magnitudes and general topological features of a crystallographic image of the macromolecule under study. This information is expressed through a number of parameters. Most of these parameters are easy to estimate, and the results of phasing are practically independent of these parameters when they are chosen within reasonable limits. By contrast, the correct choice for such parameters as the expected number of connected regions in the unit cell is sometimes ambiguous. To study these dependencies, numerous tests were performed with simulated data, experimental data and mixed data sets, where several reflections missed in the experiment were completed by computed data. This paper demonstrates that the procedure is able to control this choice automatically and helps in difficult cases to identify the correct number of molecules in the asymmetric unit. In addition, the procedure behaves abnormally if the space group is defined incorrectly and therefore may distinguish between the rotation and screw axes even when high-resolution data are not available.

Flexent® IMS-the convergence of circuit and packet core networks

As the Internet model created an environment for creativity and innovation, so, too, will the new Internet protocol multimedia subsystem (IMS) based on the session-initiation protocol (SIP) and Internet protocol (IP) mobile networks. This paper describes Lucent Technologies' innovative architecture that unifies the circuit and IMS domains by providing a single control plane for both. The Flexent® IMS architecture is specifically designed to accomplish this by the deployment of a hardware platform based on the wireless Lucent Softswitch (LSS) that initially supports circuit mobiles and then evolves through software upgrades to support both packet and circuit mobiles off of a common SIP call processing architecture. This evolutionary approach bridges not only the circuit and packet worlds, but brings a common core network architecture that is independent of the access technology, whether it is wireless (CDMA2000 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∗</sup> or Universal Mobile Telecommunications System [UMTS]) or wireline (digital subscriber line [DSL], cable modem, or analog modem). This approach truly brings a single set of features and services that can be accessed from anywhere, at anytime, via any access.

Texture classification by multi-model feature integration using Bayesian networks

In this paper, a texture classification method based on multi-model feature integration by Bayesian networks is proposed. Considering that many image textures exhibit both structural and statistical properties, two feature sets based on two texture models––the Gabor model and the Gaussian Markov random field model are used to describe the image properties in both structure and statistics. A Bayesian network classifier is then used to combine these two sets of features along with their individual confidence measures for texture classification. Seventy eight Brodatz textures were used to evaluate the classification performance. The results show that the proposed method is better than that using a single set of features from either model for texture classification.

The electronic head: a virtual quality instrument

This paper describes a new approach to virtual instrumentation, qualitative estimation, and decision making of a dynamically changing quality assessment. The authors illustrate this approach in an electronic head concept, in which they combine the amount of information received, and apply feature extraction analysis and a fuzzy clustering technique to assess the quality as acquired from a human expert. By combining data from different artificial sensor systems into a single set of meaningful features, they obtain information that is of greater human benefit than the aggregate of its contributing sensors. The combination of sensor data by fuzzy rules has the aim of performing human-like inferences that may be impossible by the single artificial sensors. The entire sensor system acts as a virtual instrument for dynamic industrial process monitoring. This virtual instrument allows easy sensor observation and learning interaction with a human operator.

Multi-resolution sub-band features and models for HMM-based phonetic modelling

HMM acoustic models are typically trained on a single set of cepstral features extracted over the full bandwidth of mel-spaced filterbank energies. In this paper, multi-resolution sub-band transformations of the log energy spectra are introduced based on the conjecture that additional cues for phonetic discrimination may exist in the local spectral correlates not captured by the full-band analysis. In this approach the discriminative contribution from sub-band features is considered to supplement rather than substitute for full-band features. HMMs trained on concatenated multi-resolution cepstral features are investigated, along with models based on linearly combined independent multi-resolution streams, in which the sub-band and full-band streams represent different resolutions of the same signal. For the stream-based models, discriminative training of the linear combination weights to a minimum classification error criteria is also applied. Both the concatenated feature and the independent stream modelling configurations are demonstrated to outperform traditional full-band cepstra for HMM-based acoustic phonetic modelling on the TIMIT database. Experiments on context-independent modelling achieve a best increase on the core test set from an accuracy of 62.3% for full-band models to a 67.5% accuracy for discriminately weighted multi-resolution sub-band modelling. A triphone accuracy of 73.9% achieved on the core test set improves notably on full-band cepstra and compares well with results previously published on this task.

Data mining for features using scale-sensitive gated experts

Introduces a tool for exploratory data analysis and data mining called scale-sensitive gated experts (SSGE) which can partition a complex nonlinear regression surface into a set of simpler surfaces (which we call features). The set of simpler surfaces has the property that each element of the set can be efficiently modeled by a single feedforward neural network. The degree to which the regression surface is partitioned is controlled by an external scale parameter. The SSGE consists of a nonlinear gating network and several competing nonlinear experts. Although SSGE is similar to the mixture of experts model of Jacobs et al. (1991) the mixture of experts model gives only one partitioning of the input-output space, and thus a single set of features, whereas the SSGE gives the user the capability to discover families of features. One obtains a new member of the family of features for each setting of the scale parameter. We derive the scale-sensitive gated experts and demonstrate its performance on a time series segmentation problem. The main results are: (1) the scale parameter controls the granularity of the features of the regression surface, (2) similar features are modeled by the same expert and different kinds of features are modeled by different experts, and (3) for the time series problem, the SSGE finds different regimes of behavior, each with a specific and interesting interpretation.

Feature-based fixturing analysis for machining parts represented with multiple sets of features

In a feature-based model, a typical approach is to represent a part with a single set of features. However, in a part with feature interactions, there may exist multiple sets of features to represent the same part. The multiple sets of features represent different ways of machining the part. It is necessary to analyse and evaluate the various sets of features to measure their performance when using different machining methods. In a machining process, different fixturing parameters are required when different sets of features are used for machining. The purpose of this research is to determine a 'good' set of features that is more suitable from the fixturing point of view. A feature-based fixturing analysis method is developed based on the 3-2-1 and the 4-2-1 fixture design principles to analyse the supporting, locating, and clamping parameters. First, a set of fixturing parameters is determined for machining one set of machining features. Next, multiple sets of fixturing parameters are determined for the multiple sets of features. Finally, the multiple sets of fixturing parameters are evaluated based on the number of different fixturing setups and the number of changes in fixturing setups. A feature set that can be machined with the minimal number of changes in fixturing setups is considered a 'good' set. The workpieces considered in this research are prismatic machining parts and prismatic machining features represented in boundary representation. The result is to provide a way of analysing the different fixturing setups when different sets of features are used for machining. From a process planning point of view, a good or a more suitable way for producing a part can be evaluated and determined for the better integration of CAD and CAM.

Landed and Literary: Hamlin Garland, Sarah Orne Jewett, and the Production of Regional Literatures

In 1894 there appeared in Chicago a little book on literary topics, the manner of whose manufacture bore the marks of a dilettante taste in book-making.... The book should have been printed on birch bark and bound in butternut home-spun, and should have had for cover design a dynamite bomb, say, with sputtering fire-tipped fuse: for the essays which it contained were so many explosions of literary Jingoism and anarchy.(1) To readers in 1895 familiar with Garland's work and its reception, reviewer C. M. Thompson's analogy between Hamlin Garland's Crumbling Idols (1894) and a dynamite bomb reminiscent of the Haymarket Riot would have seemed like standard critical fare. But if most reviewers of the early Garland found his work too radical and artistically compromised by its focus on social conflict, no one familiar with his writing would have questioned his commitment to the nation and its foundational ideals. For Garland, the exemplary regional writer was above all a devoted American, attempting to realize the principles of democracy and decentralization within broad national frameworks, both literary and political. In this respect, his work was much like that of other local writers dedicated to celebrating rural communities as sites of national promise, where the country would achieve its ultimate fulfillment as a union of diverse cultures. Yet while Garland's regionalism could be likened to a dynamite bomb, most regional writing was described by critics as quaint and picturesque. Such discrepancies in reception suggest that a common embrace of cultural pluralism among local writers and their readers--and a shared belief in the region as redeemer of the nation--did not always point to the same discourse of nationalism. Regional writing could be adapted to alternative nationalisms; an adequate assessment of the category therefore requires consideration not only of regionalism's coherence, but of its political and aesthetic inconsistencies. This essay juxtaposes several of Garland's early stories with Sarah Orne Jewett's Country of the Pointed Firs in order to explore the nature of an ideological conflict within literary regionalism and its effect on the formal conventions that writers employed and rejected. To affirm more than one variety of regionalism is to grant the author more agency than she or he might have if such writing were limited to a single set of formal features or a single social logic. My argument assumes that writers like Jewett and Garland made important ethical choices in determining which forms of local solidarity to embrace and which to reject.(2) If they conformed to the generic limits placed on them by the major monthly magazines, they often did so knowingly, providing the sub-national communities that audiences desired while recognizing the possibility of rethinking and rewriting them.(3) Even more significantly, a writer such as Garland avoided the major markets for long stretches of time, publishing his work in alternative venues like The Arena, where the representation of rural communities was not limited to a literary tourism in the service of national incorporation and reunion. Garland's work reveals a literary marketplace that never succeeds completely in homogenizing either its readers or its writers.(4) His support for radical social reform leads him to challenge some of the crucial conventions of late-nineteenth-century realism; in particular, he often refuses to draw any clear distinction between the world that his characters occupy and the world of artistic consumption. It is the merging of these two worlds that Garland sees as a necessary precursor to the realization of his national vision. In the teleology implied by Jewett's Country, on the other hand, class stratification poses no obstacle to the fulfillment of a national destiny. At the same time, Jewett shares with Garland a common approach toward the question of genealogical inheritance and the role that it is to play in the nation's ultimate redemption. …

The role of time versus climate in the formation of deep soils of the Apennine fringe of the Po Valley, Italy

Two very deep, red soils on a loess-mantled fluvial terrace of the Trebbia River in northern Italy were sampled and analyzed to determine their history and pedologic significance. Magnetostratigraphy confirms that the upper part of the terrace gravel and all of the 4-m-thick loess cover is normally magnetized and therefore less than about 780,000 yr. Artifacts collected from the oldest loess suggest an age of about 400,000 yr for the inception of loess deposition in this area. The soil horizons in loess dominantly have argillic or albic character, with redoximorphic features that include mottles, glossic tongues, and zones of ferromanganese concretions. Soil colors become progressively redder with depth in the loess cover and are most red, a 2.5 YR color hue, in the top of the terrace gravel. The upper horizons in gravel have abundant ghosts of completely weathered marl and sandstone. Rubification and weathering of stones diminishes with depth to fresh, slightly carbonate enriched parent material at 12 m. The heavy mineral composition of the sand fraction suggests minor discontinuities in loess-derived horizons. Abundance of opaque heavy minerals and picotite, which are resistant to weathering, increases from the soil surface to the base of the loess, is highest in the upper part of the terrace gravel, and decreases from there to the C horizon in gravel. Depth functions of major elements to Si in the medium silt fraction (10–30 μm) indicate a downward increase in degree of decalcification and of dissolution of primary minerals. Micromorphology adds detail to the interpretation of these complex soils: yellow complex ferriargillans are present in almost all horizons; they are rare at the top of the sequence in loess and become progressively more abundant with depth, reaching a maximum in the upper 3 m of the gravel and then decreasing with further depth to 12 m. We interpret this peak in abundance of yellow cutans to result from long continued clay illuviation to this depth from a succession of eluvial zones in the aggrading loessial sequence above. Distinctive red, hematite-rich argillans are abundant only in the upper horizons in the gravel. The red argillans relate to initial development of the soil in gravel, because they are included as pedorelicts in the deepest loess layer, where they are embedded in the yellow argillans. Stratigraphic and pedologic evidence indicates that these soils began forming in terrace gravel for perhaps 300,000 yr during the early middle Pleistocene. From this relict core these soils continued developing, both downward by continuing deep decalcification and argilluviation and upward as a sequence of wet Alfisols developed in a succession of loess layers that began aggrading about 400,000 yr ago. A paleoclimatic record can be perceived weakly in these soils, mainly by thick horizons of FeMn nodules that mark boundaries between episodes of loess deposition that took place during glacial cold phases. Paleofeatures from one unit to another are similar in kind and intensity, indicating a degree of continuity of a basic set of pedogenic processes, principally declcification, argilluviation, and redoximorphism. These soils can be regarded as Vetusols because a single dominant set of pedogenic features are represented in all parts of these soils and deeper parts of these aggrading super profiles have never been isolated from continuing pedogenic change to the present.