The America intelligence community in 1950, unprepared to perform its missions, failed to provide adequate indications and warning to U.S. national leaders and to the Commander, Far East Command (FEC), about the North Korean invasion of South Korea and Red Chinese intervention in the Korean War. Post-World War II policies that reduced the size of the military, cut systems and training, and reorganized intelligence services are responsible for that failure. Training deficiencies meant that intelligence soldiers deployed to Korea without required skills. The military trained analysts to assess enemy capabilities rather than intentions, contributing to poor predictive analysis. Shortages of analysts, photo interpreters, and linguists further plagued the intelligence community and degraded intelligence collection and production. The post-war political climate focused on the Soviet threat. American estimates saw the Soviet Union as the center of control for other communist states, such as Red China and that paradigm framed analysis of Chinese intentions. Assessments by General MacArthur, head of the FEC and Supreme Commander of United Nations Forces in Korea, of Chinese intentions proved decisive in shaping the course of the war in the fall of 1950. The Far Eastern Command (FEC) G-2 shared MacArthur's view that the Chinese would not intervene and spread that appraisal throughout the FEC intelligence community. In MacArthur opinion, the Chinese would not attack late in 1950 because the opportune time to do so had passed; furthermore, he thought since the Red Chinese lacked an Air Force, they would be annihilated by U.S. airpower. Analysts at all levels underestimated the Peoples Republic of China (PRC) and failed to understand Chinese operational art and tactics. Consequently, they did not recognize the Chinese first phase offensive in North Korea, and erroneously concluded that the Chinese would withdraw and defend its border. The intelligence community's poor readiness and lack of capability to provide indications and warning resulted in the enemy's achieving surprise. The first surprise led to the deployment, and near defeat, of Task Force Smith at Pusan. The second surprise led to the withdrawal and rout of U.N. forces, which led to a prolonged war of attrition.

This study examines variations in the performance of speech acts and additional discourse features in situational speech patterns of Peninsular Spanish. Based on studies by Blum-Kulka, et al. (1989), nine situations were created, varying from less severe to more severe, to elicit natural responses for apologies and requests. Forty participants from Castile and Andalusia were interviewed, and the data were coded to examine the differences in speech act realizations and the use of specific discourse features. The participants responses were classified by regional, gender, and age differences for the data set. Sociolinguistic differences in the use of additional discourse features were also compared, examining the use of alerters, personal address items, intensifiers, polite markers, hedges, accepting responsibility, offering repair, and the expression of need. In part three of this study, native speaker judgments and metalinguistic discussions were conducted to test the speech act data of participants from Castile and Andalusia and to verify the acceptability of the responses. Results show that in the performance of speech acts, little variation is seen for region, gender, and age. The speech acts for apologies and requests are formulaic in nature, and only change with situational variation. Likewise, even though some differences exist, there is no statistical significance in the use of additional discourse features according to region, gender, and age. The use of these discourse features provides information for Spanish language variation and in the areas of linguistic politeness and language and gender. Results from the metalinguistic discussions provide qualitative data, supporting the findings of speech act realizations. Further investigation is suggested modifying targeted speech acts and situational contexts.

Nanoscale multilayer coatings have high potential for numerous engineering applications because they can exhibit enhanced properties due to nanoscale effects and combine different properties from individual components. At present, scale effects on the mechanical behavior of multilayers are not well understood. Three multilayer nanocomposite systems, namely Al/Al2O3, Ti/TiN, and Cr/a-C, have been synthesized by using a dual-gun e-beam physical vapor deposition, to investigate the effect of layer thickness, the nature of components and their microstructures on the mechanical behavior. The deposited Al and Ti nanolayers were found to have polycrystalline fcc and hcp structure, respectively, the Cr and TiN layers had fine columnar bcc and fcc structure, respectively, and the Al2O3 and C layers were amorphous. Nanoscale effects were observed in all three systems with the metal layer thickness affecting significantly the mechanical behavior. The hardness response of the present systems can be described as a function of the metal layer thickness by a Hall-Petch relationship. A critical Al layer thickness of 40 nm, below which there was no further hardness enhancement, was found for the Al/Al2O3 multilayers. The critical Al layer thickness could be predicted by previous theoretical models. A hardness increase was observed down to a Ti layer thickness of 5 nm for the Ti/TiN system. The strengthening of the Ti/TiN multilayers was consistent with the macroyield maps based on a confined layer slip model. Hardness in the Cr/a-C system showed a continuous increase down to a Cr layer thickness of 20 nm. The fracture toughness of the monolithic ceramic phase was significantly improved by introducing a metal/ceramic multilayered structure. The wear behavior of the present multilayers was mainly controlled by the ceramic phase. The Cr/a-C multilayers achieved a low friction coefficient (~0.1) and low wear rate (~10-5 mm3/N m). The present research shows that properties can be tailored by appropriate selection of layer thickness and nature of individual components.

This dissertation traces the development of the supernatural from the late eighteenth to the late nineteenth-century. Since supernatural elements are unknown and unfamiliar, they easily arouse anxiety, fear, and even result in terror. As such they produce the effect of the uncanny and introduce the psychological component into the selected literary corpus taken from the English Gothic novel, the German Schauerroman, and the French littérature fantastique. The analysis of the selected material is based on a psychoanalytical approach using Sigmund Freuds understanding of the uncanny, his dream analysis, and his view of the conscious and unconscious, but also considers Carl Gustav Jungs perception of dreams and of the unconscious. In doing so, man descends into his psyche, the place where he confronts something unfamiliar, something unheimlich. In stressing literatures psychological component and in focusing on the literary formation of the uncanny, I elaborate the development of a genre, which has always existed but so far never been defined: the Literature of the uncanny; a genre comprising Gothic fiction, Schauerroman, and la littérature fantastique. Within this comparative project, I do not only attempt to erase the long-erroneous apprehension that the three genres just mentioned are culturally and temporally independent from each other, but I demonstrate that these genres are rather building blocks than independent factors of uncanny literary fiction. This project will illustrate that the uncanny has always been an important characteristic of the genre, but that, over time, its psychological connotation has architecturally changed from the once gothic setting of an old mysterious castle to the human mind, encompassing the Freudian ego, Id, and super-ego on the one hand, and / or the Jungian personal and collective unconscious on the other hand.

Daily and monthly output from seven evapotranspiration models (FAO-24 Radiation, FAO-24 Blaney-Criddle, Hargreaves-Samani, Priestly-Taylor, Makkink, and Turc) have been tested against reference evapotranspiration data computed by the FAO-56 Penman-Monteith model to assess the accuracy of each model in estimating grass-reference evapotranspiration in Louisiana. Models were compared at eight stations of the Louisiana Agriclimate Information System using data from December 2002 to November 2003. Comparisons were also made using three composite regions: statewide, inland, and coastal. A pan evaporation to reference evapotranspiration model (FAO-24 Pan Evaporation) was also tested against daily grass reference evapotranspiration from the FAO-56 Penman-Monteith model using data from two pan evaporation sites. Statewide and in the coastal region the Turc model was the most accurate daily model with a mean absolute error (MAE) of 0.26 mm/day and 0.27 mm/day, respectively. Inland the FAO-24 Blaney-Criddle performed best with a MAE of 0.31 mm/day. On a monthly basis, Turc again performed the best statewide and in the coastal region (MAE 0.17 and 0.27 mm/day respectively). Inland, FAO-24 Blaney-Criddle and Makkink tied for the most accurate model, although this may change with a longer dataset. Pan evaporation at both stations performed poorly with MAE values over 1.0 mm/day. It is possible that the equation for calculating the pan coefficient that is suggested by the FAO in the FAO-56 manual may not be a suitable equation for use in Louisiana. These results will assist agricultural and environmental planners in assessing the available water resources in Louisiana.

The availability of large amounts of genetic data from the mitochondrial DNA of species has created an unprecedented opportunity for the study of evolutionary processes. Being our closest relatives on the evolutionary tree the primates are a prime candidate for the study of evolutionary processes. The availability of large amounts of genetic data from the primates allows us to study and compare results from different phylogenetic reconstruction methods and to study and trace rudimentary evolutionary processes within the primate lineage. The evolutionary process studied here is the response of the nucleotide frequency ratios to single-strandedness of sites during mitochondrial DNA replication. This response curve is shown to be linear where the slope and intercept of the curve are related to the efficacy of the replication mechanisms and the binding capacity of the gamma-polymerase responsible for mitochondrial DNA replication. A Bayesian analysis of the response curves of the species is conducted and clustering schemes are developed to partition the species based on their response curves. These partitions are then mapped on the phylogenetic tree of the species to trace the evolution of the response curve within the primates.

The U.S. Army in 1940 was in the early stages of developing an airborne capability to exploit the vertical envelopment concept. That concept became reality in 1942 with the first airborne operation in North Africa. Although the first parachute drop contributed virtually nothing to the overall success of the mission, it was the beginning of an important capability. In 1943, the War Department authorized five airborne divisions despite a lack of experience and doctrine to direct the new organizational structure. The airborne initiative expanded much more quickly than did the doctrine, training, or employment principles. The first attempts of conducting large-scale airborne operations in combat during the Sicilian Campaign that year proved to be disastrous. Because of these failures, the airborne division, as well as the vertical envelopment concept itself, were in jeopardy. Army Chief of Staff George C. Marshall appointed a special board to investigate the causes of the disasters and make recommendations as to the soundness of the airborne division. While the board was meeting, half-way around the world in the South West Pacific Area, a successful airborne operation occurred when the 503d Parachute Infantry Regiment executed a drop at the Nadzab Emergency Landing Strip that allowed the capture of the essential port of Lae, New Guinea. This operation had a broader impact than just the tactical objectives that it achieved. This was the first unqualified successful American airborne operation of World War II and it allowed the airborne advocates to make a case for the soundness of the vertical envelopment concept, as well as that of the airborne division. Had it not been for this parachute drop, the U.S. Army might have abandoned the whole initiative just when it was planning to employ two airborne divisions during Operation NEPTUNE, the airborne portion of Operation OVERLORD.

Rapid intensity changes of Hurricane Lili in the Gulf of Mexico (GOM) were studied in three distinct ways: climatology, satellite remote sensing, and surface meteorological and oceanographic measurements. Each research methodology provided insight about Hurricane Lili's intensity behavior. A climatology of rapid and explosive intensifications of hurricanes was developed using minimum central pressure observations for the Atlantic tropical cyclone record. Results showed these events were frequent, especially in the GOM. The majority of intensification events occurred ≤ 24 h before landfall, with a third to one-half ≤ 12 h. Lili emerged anomalous as the only hurricane to weaken at a greater rate (+17 hPa over 6 h or +2.83 hPa h^-1) than its rapid intensification event rate (-13 hPa over 6 h or -2.16 hPa h^-1). GOES-8 satellite water vapor brightness temperature data were investigated using a -24°C vapor front to delineate a dry air mass west of Lili. Drier air was shown to affect Lili during a rapid weakening phase after the two features were less than 250 km mean or 215 km minimum distance apart. These critical distances are offered as a criterion for a relationship between tropical cyclone weakening and dry air advection. During the time periods where Lili was intensifying or maintaining intensity, this vapor front exhibited more complex signatures of definitive breaks, shape changes, and protrusions. During the rapid weakening phase and when the two features were the greatest distance apart, the vapor front resembled a smooth, strong boundary line. The dry air mass was shown to have the greatest effect upon Lili after the rapid intensification phases. SST and heat flux calculations illustrated that oceanic contributions to hurricane intensity were more significant during rapid intensification periods. Ocean heat content directly under the center was more vital as seen in Lili's second rapid intensification phase occurring directly over the GOM Loop Current. Although Tropical Storm Isidore cooled surface water temperatures by around 1°C, pre-Isidore water temperature levels were shown to increase latent heat flux values by > 100 W m^-2 (+ 40%) over observed values during Lili's weakening phase.

The Adjunct Island Constraint proposed by Ross (1967) together with Chomsky’s Barriers (1986) are not sufficient to account for the ungrammaticality of wh-extraction out of adjunct clauses, nor do they address the instances of grammatical extraction out of such constructions. Extraction out of Adjunct Islands is now completely predictable using a combination of Chomsky’s Minimalist Program (1995) and Kehler’s Coherence Theory (2002). The combination of these two theories gives an account of both the grammatical and ungrammatical instances of wh- extraction out of adjuncts. The principles of the Minimalist Program together with evidence from Old English adverbial clauses determines the necessary structure for grammatical extraction, while an extension of the Coherence Theory mandates the required semantic relation between the matrix and adjunct clauses. This thesis proves that the possibility of wh-extraction out of adjunct islands is dependent upon the seamless integration of syntax and semantics.

Accurate hydrologic models are needed to aid engineers and researchers design, install and evaluate efficient and cost-effective agricultural water management systems to reduce risks associated with food production, and to reduce soil erosion and water pollution. One model used for the alluvial soils of Louisiana is DRAINMOD. This model does not accurately predict infiltration and runoff for the crusting-prone alluvial soils of Louisiana. The main goal of this study was to modify the current DRAINMOD model to incorporate the effects of rainfall intensity and deep chiseling to improve its estimation of infiltration and surface runoff. The second goal was to use information gained from the modified DRAINMOD model to assess how long farmers and environmentalists benefit from a particular deep chiseling operation and determine optimum deep chiseling frequency for given climatic conditions. A methodology for using a five-minute rainfall time increment subroutine within DRAINMOD was developed. Field experiments yielded an initial vertical saturated hydraulic conductivity of 2.0 cm/hr, a final vertical saturated hydraulic conductivity of 0.50 cm/hr and exponent of 0.03 cm-1 for model calibration. Deep chiseling modifications resulted in the DRAINMOD-STMAX, DRAINMOD-Ks and the combined DRAINMOD-Ks-STMAX models. DRAINMOD-STMAX, DRAINMOD-Ks and DRAINMOD-Ks-STMAX model improved surface runoff prediction by 57%, 73%, and 82% respectively in1995/96 season and by 27%, 45%, and 62% respectively in 1996/7 season. Using DRAINMOD-Ks-STMAX model, deep chiseling a Commerce silt loam soil increased infiltration by 9.4% and reduced runoff by 19.7% in 1995/96 season and by 5.7% and 19.2% respectively in 1996/97 season. All benefits resulting from deep chiseling were lost after 115 cm of rainfall since deep chiseling. Farmers should deep chisel once every year when annual rainfall is greater than 100 cm and once every two to three years when annual rainfall is less than 100 cm. Sixty percent or more of the maximum deep chiseling benefits had been lost by planting time; therefore, farmers need to deep chisel their fields just before planting. Further work is needed in the field to determine other factors affecting variation of Ks, to validate the DRAINMOD-STMAX, DRAINMOD-Ks and DRAINMOD-Ks-STMAX models, and to incorporate rainfall intensity subroutine.