Thesis (Ed.D.)--Boston University. Note: Missing page 366.

Body size affects rates of most biological and ecological processes, from individual performance to ecosystem function. Within species, emergent body size patterns have been formalised into prominent biogeographical and biological rules, including James' Rule (larger individuals are found at higher, colder latitudes), and the Temperature-Size Rule (individuals reared in warmer conditions grow to a smaller adult size). Body size also varies seasonally and with altitude. Yet, the patterns and drivers of these size gradients, and the degree to which they co-vary and share explanatory mechanisms, have never been systematically evaluated. We undertake the most comprehensive metaanalyses to date of temperature- and biogeographical-size clines within arthropod species. Aquatic species show greater reductions in body size with warming and decreasing latitude compared to terrestrial species, likely an adaptive response to deal with increased metabolic demand in the warm and the greater difficulty to uptake oxygen in water than in air. Voltinism explains variation in laboratory temperature- and latitudinal-size clines in terrestrial species. While size decreases with warming and with decreasing latitude in multivoltine terrestrial arthropods, size increases on average in univoltine species, consistent with predictions from size vs. season-length trade-offs. In the globally distributed sub-class Copepoda, seasonal temperature-size (T-S) gradients differ between current-feeding calanoids and ambush-feeding cyclopoids, suggesting that differences in the size- and temperature-dependence of alternative feeding strategies may influence the T-S response. Finally, through experimentation, we explore the progression of the T-S response of Copepoda during ontogeny. The T-S response is more strongly generated in particular life stages, and even reduced in some periods, providing evidence that the temperature-dependence of growth and developmental rates is modified during ontogeny. Ultimately, close similarities between T-S responses measured in controlled laboratory conditions, and seasonal and biogeographical size clines in the field across different arthropod taxa, suggests that these patterns share similar selective pressures.

Thesis (M.S.)--University of Missouri--Rolla, 2007. / Vita. The entire thesis text is included in file. Title from title screen of thesis/dissertation PDF file (viewed December 11, 2007) Includes bibliographical references (p. 50-51).

Thesis (M. Ed.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

One of the most popular methods for achieving control of the Red Imported Fire Ant, Solenopsis invicta Buren, is through the use of broadcast baits. Several factors contribute to bait efficacy, one of which may be particle size. The goals of these laboratory studies were to determine particle size and bait preference using Advance Select Granular Ant Bait and Advance Carpenter Ant Scatter Bait, determine the effect of starvation on bait removal and recruitment to Carpenter Ant Scatter Bait, and determine if any correlation existed between head capsule width and particle size selected. Experimental colonies removed significantly more 1400-2000 μm particles of Select Granular Ant Bait, while ants foraging on Carpenter Ant Scatter Bait preferred 1000-1400 μm particles. Mean number of ants present at bait mirrored results from bait removal test. Ants displayed a preference for Carpenter Ant Scatter Bait based on mean number of ants present at bait for the 10-d foraging period. For starvation assays, significant differences in bait removal and number of ants present occurred in the 0-d group. Ants starved for 5 d removed significantly more bait of all particle sizes, and removed greater amounts of 1000-1400 μm Carpenter Ant Scatter Bait than other sizes. Head width reliably predicted particle size selected, but the linear model explained very little of the observed variation for ants foraging on Select Granular Ant Bait (R2 = 0.043) or Carpenter Ant Scatter Bait (R2 = 0.047). This study supported the significant role of bait size and starvation period in S. invicta bait preference, and demonstrated how size preference may vary depending on bait type.

Semiconductor Electroabosortion Modualtor (EAM) has become an important element in optical fiber communications because of its capability to integrate with other semiconductor devices, high-speed and low driving voltage. However, high optical insertion loss and low tolerance in optical power coupling are main general problems to be solved in order to get high electro-optical (EO) efficiency. Monolithically integrating EAM with optical spot-size converter (SSC) can lead to high-efficiency single-mode fiber coupling, but the price is on the complex fabrication methods. In this paper, based on previous work, the selective undercut etching active region (UEAR) and the whole wet-etching techniques are employed to fabricate the integration of laterally tapered SSC and EAM. Also, by applying the ion-implantation in SSC region, the reliable transfer efficiency and also high-speed performance are obtained based on the high resistance and low parasitic capacitance in SSC. The active region containing 10 strain compensated multiple-quantum-wells (MQWs) sandwiched by n-InP (bottom) and p-InP (top) for the electroabsorption region of EAM and also the top region of lateral tapered SSC. The converted waveguide in SSC consists of alternating InGaAsP and InP layers. An HBr-base etching solution is first used to define the top p-cladding with the widths of from 6um to 8um. An H2O2-base solution is then utilized to selectively undercut-etch the MQWs from InP material. The active waveguide p-cladding in EAM is set as 8um. After defining EAM and SSC, the converted waveguide is fabricated by aligning the top SSC and then wet-etched. By using an e-beam evaporator, Ti/Pt/Au and Ni/AuGe/Ni/Au are deposited as p- and n-type metallization, respectively. PMGI is spun serving as the passivation, planarization and bridging. The microwave coplanar waveguide (CPW) line is finally defined by depositing Ti/Au for microwave load- and feed- lines and connecting EAM. The length of SSC is 350um. The Spot-Size Converter monolithically integrated with Electroabsortion Modulator using whole wet-etching technique is demonstrated. ¡V12.5dB of fiber-to-fiber insertion loss and 10dB (TE) 10dB(TM) extinction ration in 1V(1570nm excitation) is obtained in this device. Using Fabry-Perot method, the average optical transfer loss in SSC is extracted to be 2dB, quite consistent with simulation results. By applying ion-implantation on SSC, the broadband EO performance 45GHz of ¡V3dB bandwidth is achieved for 100um long device due to the low capacitance and the high resistivity in SSC.

We present the fabrication of InGaAsP/InP buried heterostructure spot-size converter lasers. In the lateral conversion, we use photolithography to make tapered ridge waveguides. In the vertical conversion, we use a pair of step-index passive waveguides, namely guard waveguides (GWs), in the two sides of the step-index active waveguide region to increase optical-field profile. In order to decrease leakage current, we use a p-n-p current blocking structure by MOCVD regrowth. From numerical simulations, the far-field divergence is 21x21. The step-index GW structure shows an internal efficiency of 63%. However, the BH lasers did not lase from our fabrication processes. From the I-V characteristics, a large leakage current has bypassed through the blocking structure. The reason may relate to the high background doping concentration of our MOCVD growth.

Abstract Optical spot-size converter (SSC) is an essential element in the opto-electronic integrated circuit because of its direct coupling to single-mode optical fiber, low-cost from the misalignment issue in package. By the tapered waveguide structure, SSC offers a capability to transfer an elliptical optical mode of optical waveguide to single-mode fiber matched mode, allowing the independence of device design from the coupling issue. However, the conversion efficiency of SSC is strongly reliant on tapered waveguide structure, material index, and also excitation wavelengths, restricting design and fabrication of SSC. In order to enhance the reliability on SSC fabrication, a voltage-controllable SSC integrated with electroabsorption modulator (EAM) is designed and fabricated. The tapered waveguide processing is based on the selective undercut wet-etching on InGaAsP-material system. The active waveguide width of EAM is 3.5£gm, where the tapered waveguide widths is ranged from 1.7£gm to 3.5£gm. The transferred optical waveguide (passive waveguide) is 8£gm wide. By adjusting the voltage of tapered waveguide, coupling efficiency is enhanced by 2dB from 1560nm to 1570nm. A reliable SSC-integrated EAM with -12dB of insertion loss, extinction ratio of 22dB is demonstrated. By extracting the index change from Fabry-Perot optical mode of waveguide, index variation from reverse bias is in the order of 10-3, consistent with the calculation based on Kramer-Kronig model.

Thesis--University of Oklahoma. / Includes bibliographical references (leaves 45-46).

The three-level multivariate multilevel model (MVMM) is a multivariate extension of the conventional univariate two-level hierarchical linear model (HLM) and is used for estimating and testing the effects of explanatory variables on a set of correlated continuous outcome measures. Two simulation studies were conducted to investigate the sample size requirements for restricted maximum likelihood (REML) estimation of three-level MVMMs, the effects of sample sizes and other design characteristics on estimation, and the performance of the MVMMs compared to corresponding two-level HLMs. The model for the first study was a random-intercept MVMM, and the model for the second study was a fully-conditional MVMM. Study conditions included number of clusters, cluster size, intraclass correlation coefficient, number of outcomes, and correlations between pairs of outcomes. The accuracy and precision of estimates were assessed with parameter bias, relative parameter bias, relative standard error bias, and 95% confidence interval coverage. Empirical power and type I error rates were also calculated. Implications of the results for applied researchers and suggestions for future methodological studies are discussed. / text