Proceedings of the 27th Southern Forest Tree Improvement Conference

Browse

Recent Submissions

  • Publication
    Molecular Pathology of Pitch Canker Disease
    (2003) Smith, K. E.; Morse, A. M.; Kayihan, G.; Huber, D. A.; White, T. L.; Nelson, C. D.; Davis, J. M.
    The inciting agent of pitch canker disease in Pinus species is the necrotrophic fungal pathogen Fusarium circinatum (teleomorph Gibberella circinata). Pitch canker disease has been identified in the southeastern United States, as well as in California, Mexico, Japan and South Africa. The disease is episodic in nature and can reach epidemic proportions with potentially devastating consequences for both managed and natural forests. Symptoms of pitch canker disease include discolored lesions (cankers) on stems and branches that generate profuse amounts of resin (pitch). Infected shoots eventually desiccate due to reduced water transport caused by pathogen development in vascular tissues. As with most diseases of pine, pitch canker disease is not well understood with respect to mechanisms of fungal pathogenicity, disease development, and disease resistance. To gain insight into the processes associated with disease development, we used a method called differential display of messenger RNA. This method was coupled with gene expression array analysis to identify genes from the pine host and from the fungal pathogen that are regulated differently during the disease state compared to the freeliving states of host and pathogen. The functions of these genes appear to be associated with plant defense and desiccation, however many genes of unknown function were also identified. The next goal of these studies is to identify genes that are associated with resistance, and to compare and contrast the gene expression programs that are associated with resistance vs. susceptibility. The large-scale screening of loblolly pine clones for resistance to pitch canker disease, which was carried out as part of the ADEPT project (Allele Discovery of Economically-important Pine Traits), is a major step toward this goal. Analysis of the screening study data revealed that resistance was heritable on the clone mean basis. A mixed linear model was used to predict genotypes that are highly resistant or highly susceptible to pitch canker disease. Ultimately it is hoped that association genetics approaches - such as those utilized in the ADEPT project - will identify host genes and alleles that condition pitch canker disease resistance. The genes identified by differential display serve as candidate genes for evaluating this approach.
  • Publication
    Representatives
    (2003)
  • Publication
    Registration List
    (2003)
  • Publication
    Loblolly Pine Karyotype Using FISH and DAPI Positive Banding
    (2003) Islam-Faridi, M. N.; Nelson, C. D.; Kubisiak, T. L.; Gullirmo, M. V.; McNamara, V. H.; Price, H. J.; Stelly, D. M.
    A loblolly pine (Pinus taeda L.) karyotype has been developed based on fluorescent insitu hybridization (FISH) using cyto-molecular landmarks including plant telomere repeat, 18S-28S rDNA and 5S rDNA probes and DAPI positive bands. Somatic chromosome spreads of loblolly pine root tips were prepared using a modified enzymatic digestion technique. We observed ten pairs of long metacentric, one pair of long submetacentric and one pair of short sub-metacentric chromosomes. All the chromosomes showed characteristic DAPI positive bands (A-T rich regions) near and/or around the centromeres. At least one DAPI positive band was also observed in intercalary positions on all chromosome arms. Plant telomere FISH signals were observed towards the end of each chromosomal arm as expected. In addition, most of the chromosomes showed telomeric sites near and/or around the centromeres except for one or possibly two chromosomes. A total of seventeen 18S-28S rDNA sites were identified per haploid genome. Eight of these were located near and/or around the centromeres and seven were at intercalary positions. One major 5S rDNA site was observed in an intercalary region of a metacentric chromosome that lacked 18S-28S rDNA sites. One or possibly two minor 5S rDNA sites were observed near the ends of two different chromosomes. We are also developing a slash pine karyotype for direct comparison with loblolly as well as a comparison with a previously published slash karyotype (Doudrick et al. 1995, Journal of Heredity 86:289-296). Finally, we will provide an update on our progress toward using BAC clones as FISH probes on pine chromosomes.
  • Publication
    Evaluation of an Open-Nucleus Model for Forest Tree Breeding
    (2003) Lstiburek, M.; Mullin, T. J.
    Open-nucleus breeding was evaluated by stochastic simulation. Methodology was developed for unrestricted migration rate between two tiers (main and elite). Genetic progress in the breeding population was iteratively maximized for a wide range of restrictions on diversity varying from strong family selection to balanced within-family selection. The model assumed genetic parameters typical for growth traits in conifers and accounted for inbreeding depression. Comparison was made with a single population without hierarchical structure assuming constant testing effort. A seed orchard was established in each breeding cycle as a selected subset of the breeding population. The extra gain achieved by assigning better mates into the elite population was counteracted by increased group coancestry (relatedness) among seed orchard selections. The size of elite tier was found to have little importance in this study. When more effort was concentrated into elite crosses, potential for inbreeding in the seed orchard crop increased.
  • Publication
    Functional Genomics Pipeline in Loblolly Pine and Eastern Cottonwood
    (2003) Kodrzycki, B.; Sanders, P.; Grigor, M.; Roberts, S.; Winkeler, K.; Foutz, K.; Holley, H.; Huetteman, C.
    Advances in transformation technology allow functional genomics techniques that are commonly applied to Arabidopsis thaliana to be considered for tree species. ArborGen is applying the principles of high throughput functional genomics originally developed for Arabidopsis to both hardwood and conifer trees as a means to screening genes that affect wood quality traits and productivity. The ability to demonstrate gene function in commercially important tree species is an essential technology for developing improved tree products based on gene transfer. Starting with a large database of ESTs isolated from Pinus radiata and Eucalyptus grandis, a systematic approach to uncovering gene function is in progress. This integrated functional screen consists of bioinformatics characterization, cell-based assays, and Arabidopsis thaliana screens to identify candidates for high throughput functional testing in two commercially important species: Populus deltoides (Eastern cottonwood) and Pinus taeda (Loblolly pine). Efficient gene transfer methods are being used to introduce large numbers of genes into these tree species and methods for early detection of transgene function are being developed based on phenotypic and chemical composition screens. These methods will enable ArborGen to functionally test several hundred genes per year in commercial tree species. These functional screens in trees are being used to identify candidate genes that are expected to affect key commercial traits in plantation forestry. This integrated system for characterizing tree genes including bioinformatics, cell-based assays, Arabidopsis screens, and high throughput Pine and Populus screening systems will be described.
  • Publication
    Field Selection of American Sweetgum Transformed for Herbicide Resistance
    (2003) Huetteman, C. A.; Zhao, Y.; Gause, K. C.; Wilde, H. D.; Parks, B. T.
    One American sweetgum clone from an open-pollinated parent was transformed with Agrobacterium tumefaciens containing the gene for acetolactate synthase (ALS), and regenerated through organogenesis. Seventy independently transformed lines were selected in vitro in the presence of an ALS-targeting herbicide. Containerized ramets of the 70 lines were established at an irrigated fiber farm in South Carolina, in May, 2002. Over 1,000 trees were planted in a completely randomized design with up to 15 ramets per transline. Establishment survival in June was near 100 percent. Two months after planting, the actively growing trees received one over-the-top application of a tank mix of two ALS-targeting herbicides to evaluate resistance in each transformed line. Thirty days following application, over 85 percent of the planted ramets from three lines displayed no damage symptoms. Sixteen additional lines had a mean damage rating less than �slight�. Twelve of these nineteen lines had first-year heights that were not significantly different from the non-sprayed control line. One transformed line was significantly taller than the control clone (103 cm vs 91 cm mean height, respectively). At the end of the year, only 11 of 1,027 individual trees (1 ramet from each of 9 lines and 2 ramets of another line) died due to herbicide damage. Although the test was terminated after only one growing season, at least four lines were sufficiently resistant to be considered for further plantation development.
  • Publication
    Characterization of a Bacterial Contaminant in Loblolly Pine Tissue Culture
    (2003) Cook, M. J.; Clark, J. J.; Anderson, J. W.; Junkins, A. D.; Karr, C. K.; Becwar, M. R.
    Bacterial contamination of immature seed explants can decrease the efficiency in initiating and establishing pine embryogenic cultures. A type of contaminant commonly observed surrounding the megagametophyte explant of immature loblolly pine (Pinus taeda L.) appears as a white to creamy halo on the surface or within the culture medium. This contaminant appears within 24-48 hours after culturing the immature seed. The severity of the contamination varies by year and family. The pattern of contamination among seeds pooled from several cones did not suggest operator technique as the source of contamination. Tests with seeds separated by cone during sterilization and culture suggest that the source of the contamination was from seeds of specific cones. To better understand the causal bacterium and provide insight to help prevent or decrease the occurrence of contamination, several isolates taken from different contaminated explants were identified. This identification was done by PCR amplification of a 500bp sequence of a 16S ribosomal RNA gene. The closest match in terms of sequence similarity was Erwinia amylovora. Erwinia species are the causal agents of several economically important diseases. For example, fire blight (E. amylovora) in the Roseaceae family that affects apple and pear trees, and soft rot (E. carotovora) in several crops including potato. Our results suggest that Erwinia may also infect immature loblolly pine seeds and is a probable causal agent of this type of bacterial contamination in pine tissue culture. We have been able to significantly reduce the negative impact of this contamination and increase culture establishment efficiency by separating seeds by cone for sterilization and culture.
  • Publication
    Microprojectile-Mediated Genetic Transformation of Embryogenic Sweetgum Cultures
    (2003) Walsh, M. J.; Merkle S. A.
    With its rapid biomass accumulation, sweetgum (Liquidambar styraciflua) may be particularly suitable for phytoremediation purposes, especially if it can be engineered with genes that enhance tolerance and/or accumulation of heavy metals. Sweetgum has been transformed using Agrobacterium tumefaciens-mediated transformation of shoot explants or nodule cultures, followed by adventitious shoot regeneration, and by microprojectile bombardment of nodule cultures. We tested microprojectile bombardment of embryogenic cultures as an alternative means to produce transgenic sweetgum trees, with the long-term goal of engineering the species with genes conferring heavy metal resistance. Embryogenic cultures were obtained by culturing immature seeds on an induction/maintenance medium (IMM) with 2,4-D and BA. Cultures proliferated as proembryogenic masses (PEMs) with monthly transfer to fresh medium and formed suspensions following transfer to liquid IMM. Bombardment parameters were optimized using transient GUS expression. Different osmotic conditioning and selection agents and concentrations were tested to determine suitable levels of these agents for sweetgum PEMS. PEMs representing 4 different genotypes were sizefractionated and bombarded with pBI426, which contains a translational fusion between the GUS and the NPTII coding region, under the control of a double 35S promoter. Following selection on proliferation medium supplemented with 50 mg/l of kanamycin, 8 kanamycin-resistant sweetgum lines were recovered, all from one of the bombarded genotypes. While all lines remaining kanamycin-resistant were GUS- and PCR-positive, and Southern analysis indicated stable integration of the transgenes, neither the transclones nor the untransformed control line from which they were derived were capable of producing somatic embryos, having apparently lost this potential over time. Sweetgum transclones cryostored for several months maintained GUS expression.
  • Publication
    Evaluation of the Inheritance of Tip Moth Susceptibility Using Pine Hybrids Planted in Southeast Mississippi
    (2003) Highsmith, M. T.; Lott, L. H.; Gwaze, D. P.; Lominy, D.; Nelson, C. D.
    Nineteen families of susceptible and resistant pine parent species and their interspecific hybrid progenies were quantitatively assessed for Nantucket pine tip moth (Rhyacionia frustrana Comstock) damage in a study in southeast Mississippi. The seeds for this study were leftover from previously performed experiments and the collection did not provide for a balanced set of families. However, it did include seven slash pine families, two shortleaf families, two loblolly families, three three-way hybrid families, one testcross hybrid family, and two F1 hybrid families of susceptible and resistant parents. The loblolly (L1 and L2) and shortleaf (Sf1 and Sf2) pine parents were susceptible, while the slash (S1 to S7) parents were resistant. The hybrids were produced from controlled pollinations that included eight additional parent trees along with the S1 slash pine parent and both of the shortleaf (Sf1 and Sf2) parents. The eight additional parent species included two F1 hybrid parents; one of which was a slash x longleaf, combining two resistant parent species and the other was a longleaf x shortleaf, combining a resistant and a susceptible parent species. The F1 hybrids in this study were the progeny of the S1 slash parent and a loblolly parent. The three-way hybrids were the progeny of the slash x longleaf F1 hybrid parent and the Sf1 and Sf2 plus one additional shortleaf parent, while the testcross hybrid was produced from the longleaf x shortleaf F1 hybrid parent and a slash pine parent. The F1 hybrid families and the three-way hybrid families were susceptible to tip moth attack like their susceptible loblolly and shortleaf parents, but tip moth damage on all of these families was significantly higher than that on the slash pine parent and testcross hybrid families. The phenotypes of the pure species and hybrid families supports a dominant mode of inheritance for susceptibility to tip moth in three different ways: (1) the phenotype of the F1 hybrids expressed the phenotype of the susceptible parents, (2) the phenotype of the three-way hybrids expressed the phenotype of the susceptible shortleaf parents when those parents were crossed with an interspecific F1 hybrid of two resistant parents, and (3) the testcross hybrid expressed a level of susceptibility that was intermediate to that of the susceptible and resistant parents when the longleaf x shortleaf F1 hybrid parent was backcrossed to a resistant slash pine parent.
  • Publication
    Geographic Information Systems(GIS) and Virtual Reality Models (VRM) to Improve the Analysis of Genetic and Silvicultural Trials
    (2003) Rubilar, R.; McKeand, S.; Allen, H. L.
    GIS systems have become core tools for mapping needs in forestry. During the last decade software platforms have expanded the basic capabilities of data storage and retrieval in map formats. Complex overlay procedures, terrain analysis, 3D modeling, spatial and geostatistical tools, and remote sensing integration have increased the power of space related information. In addition, new VRM have emerged as improved tools for visualization, simulation and teaching. Research field trials in forestry are usually established to minimize spatial environmental variation, however exploring this assumption �ex ante� or �ex post� has been always tedious and uncertain and has lacked the power of visualization and analysis. Powerful spatial statistical analyses and interpolation analyses may be integrated to visualize site variability, remove environmental trends or integrate those to conventional statistical analyses. We investigated the �ex post� analysis of a research trial using ARCMAP/GIS and ARCScene VRM tools (ESRI, Inc) in order to explore their utility for trial analysis. Detailed sampling activities investigating specific physiological or ecological process may take full advantage of GIS and VRM tools capabilities to understand site variability and locate highly representative sampling points. That information may be used for modeling based on the same spatial information.
  • Publication
    Rooting Cuttings of Northern Red Oak (Quercus rubra L.)
    (2003) Gocke, M. H.; Goldfarb, B.; Robinson, D. J.
    Several techniques have been used experimentally to vegetatively propagate northern red oak (Quercus rubra L.), including: 1) rooting juvenile softwood cuttings in intermittent mist, 2) rooting shoots originating from mature buds grafted onto juvenile root stocks, and 3) in vitro shoot proliferation of juvenile or mature shoots followed by in vitro rooting. Of these techniques, rooting juvenile softwood cuttings has provided the most consistent results for northern red oak (NRO). Juvenility (or at least the associated ability to form adventitious roots) disappears rapidly among progressive flushes of growth in NRO seedlings. Decreased rooting has been reported for NRO shoots obtained from progressive flushes of growth produced within a growing season, as well as shoots representing flushes obtained from successive seasons of growth. However, as with many other tree species, the process of maturation in NRO can be slowed by pruning to encourage juvenile shoot production. Optimizing the number of juvenile cuttings produced from each stock plant is necessary for efficient rooted cutting production systems. In addition, rooting conditions must be determined for the shoots produced under these pruning regimes. Two NRO rooted cutting studies are currently being conducted at NCSU. The objective of the first study is to evaluate the effects of stock plant pruning location, diameter, and age on new shoot production. Treatments include pruning first-year seedlings, as well as one-, two-, and three-year-old seedlings to the base of the first, second, third, or fourth flush of growth produced during the first growing season. The objective of the second study is to evaluate the ability of the shoots produced in the first study to form adventitious roots. Treatments include three rooting hormone levels (0.5%, 1%, and 1.5% IBA) and a control (45% EtOH). Preliminary results from both studies will be presented.
  • Publication
    Progress with American Chestnut Somatic Embryogenesis
    (2003) Andrade, G. M.; Merkle, S. A.
    American chestnut (Castanea dentata), once the dominant forest species of the Appalachian Mountains in the eastern United States, was devastated during the early twentieth century by the introduction of the chestnut blight fungus. As part of an effort to restore the species to the forest, we have been working with embryogenic cultures of the species, aiming to establish a reliable somatic embryogenesis system for mass clonal propagation, as well as for genetic transformation with potential anti-fungal genes. While initiation of embryogenic cultures from immature ovules of American chestnut has become routine, bottlenecks still remain for embryo maturation, germination and conversion. Effects of cold stratification, gellan gum concentration and activated charcoal on somatic seedling production were investigated. Studies of other variables, such as the effects of light quality on germination and conversion, are underway. Using five genotypes, clusters of proembryogenic masses maintained on WPM with 2 mg/l 2,4-D were transferred to basal WPM for somatic embryo development. Individual cotyledonary-stage embryos (2-4 mm) were cultured for 10 days on basal medium prior to storage at 4o C for 12 weeks in the dark. These embryos germinated at an average frequency of 23% following transfer to WPM basal medium in GA7 vessels in the light. Embryos that did not receive cold treatment or were stored for only 6 weeks failed to germinate. Embryos on WPM with 5 g/l activated charcoal (AC) germinated at the same frequencies as those cultured without AC, but AC prevented darkening of taproots. Embryos cultured on 3 or 6 g/l Phytagel germinated at higher frequencies following 12 weeks cold storage than did those cultured on 10 g/l Phytagel. To date, over 30 somatic seedlings representing 3 genotypes have been transferred successfully to greenhouse conditions.
  • Publication
    Volume Gains of Rooted Loblolly Pine Clones at Age 10 in Florida and Alabama
    (2003) McCall, E.; Isik, F.
    An estimate of the growth potential of clonal lines produced from elite crosses compared to trees grown from seed is needed to justify clonal forestry programs. A series of age 10 loblolly pine (Pinus taeda L.) rooted cutting tests planted in Nassau County, Florida and Monroe County, Alabama are some of the oldest clonally replicated studies in existence. A 3 X 3 factorial produced 9 cross families from which 4-6 clones were produced as rooted cuttings from seedling hedges. Clones and seedlings were planted in a randomized complete block design with split-plots for seedlings and rooted cuttings. Clone genetic values were estimated by Best Linear Unbiased Prediction method and genetic gains were estimated for various clone selection scenarios. Average volume gain over the family mean estimate by choosing the best clones from each family was 12.6%. The top clone of the 45 tested clones yielded 39.8% more volume than the grand mean. The top five and the top 10 clones had 30% and 23% more volume gain than the grand mean, respectively.
  • Publication
    High Efficiency Transformation of Loblolly Pine (Pinus taeda L.) Using Green Fluorescent Protein as a Vital Screenable Marker
    (2003) Tang, W.; Newton, R. J.
    An engineered green fluorescent protein (m-gfp5-ER) gene under the control of the 35S Cauliflower Mosaic Virus promoter was used to develop a facile and rapid loblolly pine (Pinus taeda L.) transformation system via Agrobacterium tumefaciens-mediated transformation of mature zygotic embryos. Green fluorescent protein has been introduced into three different loblolly pine families that are considered recalcitrant to transformation. The m-gfp5-ER gene produced bright-green fluorescence easily detectable and screenable in loblolly pine tissue 3-30 days after explants were cocultivated with Agrobacterium. A high-level of GFP expression was detected in transgenic cells, tissues, and plants, and was localized in specific cells derived from cotyledons, hypocotyls, and radicles of mature zygotic embryos. Furthermore, in vitro and in vivo monitoring of GFP expression permitted a rapid and easy discrimination of transgenic shoots, and drastically reduced the quantity of tissue to be handled and the time required for the recovery of transformed plants. Integration of the m-gfp5-ER was confirmed by polymerase chain reaction (PCR), by Southern and northern blot analysis, and by junction DNA sequence analysis. Molecular analysis of Agrobacterium T-DNA loci in transgenic loblolly pine demonstrated that most of transgenic plants were derived from single transformation events. GFP-expressing shoots were also observed in loblolly pine explants co-cultivated with Agrobacterium but cultured in a medium without the selective agent kanamycin. This provides the opportunity to regenerate transgenic plants without using selectable-marker antibiotic-resistance genes, which will enhance the commercialization of transgenic plants.
  • Publication
    USDA Forest Service Forest Health Protection Resistance Screening Center
    (2003) Young, C. H.
    The Resistance Screening Center (RSC) is operated by the Forest Health Protection unit of the USDA Forest Service, Southern Region, State and Private Forestry. The Center is located at the Bent Creek Experimental Forest near Asheville, NC. The Center evaluates seedlings for resistance to disease, primarily fusiform rust (caused by Cronartium quercuum F. sp. fusiforme) and pitch canker (caused by Fusarium circinatum) as a service to tree improvement specialists, seed orchard managers, scientists, government agencies, research institutions, universities, and private industry. Testing enables clients to obtain information on the relative resistance of their materials in much less time than is possible in field progeny tests. The RSC has the flexibility to modify current screening procedures to accommodate specialized requests. This allows researchers to use the RSC as an additional experimental tool. By using information from the Resistance Screening Center tests, trees producing resistant progeny can be identified or questions may be answered concerning such things as the nature of variation in the rust fungus or the effectiveness of fungicides.
  • Publication
    Antibiotics Effects on the Elimination of Agrobacterium tumefaciens from Loblolly Pine (Pinus taeda L.) Zygotic Embryo Explants and on Transgenic Plant Regeneration
    (2003) Tang, W.; Newton, R. J.
    Three antibiotics were evaluated for their effects on the elimination of Agrobacterium tumefaciens during the genetic transformation of loblolly pine (Pinus taeda L.) using mature zygotic embryos as targets. Agrobacterium tumefaciens strains, EHA105, GV3101, and LBA 4404, all harboring the plasmid pCAMBIA1301, which carries the selectable marker gene, hygromycin phosphotransferase (hpt) controlled by the cauliflower mosaic virus 35S promoter and terminator, and the uidA reporter gene (GUS) driven by the cauliflower mosaic virus 35S promoter and the terminator of nopaline synthase gene, were used in this study. Exposure to 350 mg/l carbenicillin, claforan, and timentin respectively for up to 6 weeks did not eliminate the Agrobacteria, while antibiotics at 500 mg/l eradicated them from the co-cultivated zygotic embryos. All three antibiotics increased callus growth and shoot regeneration at 350 mg/l and 500 mg/l each, but reduced callus growth and shoot regeneration at 650 mg/l when compared with controls. Putative transgenic calli were selected for continued proliferation and differentiation on 4.5 mg/l hygromycin-containing medium. Transformed calli and transgenic plants produced on a selection medium containing 4.5mg/l hygromycin were confirmed by GUS histochemical assays, by polymerase chain reaction (PCR), and by Southern blot analysis. These results are useful for future studies on optimizing genetic transformation procedures in loblolly pine.
  • Publication
    Fourth-Year Results from a Clonal Test of Loblolly Pine
    (2003) Goldfarb, B.; Isik, F.; LeBude, A.; McKeand, S.; Li, B.
    In November and December 1998, two experimental plantings were established using rooted cuttings from 450 clones of eight unrelated full-sib families of loblolly pine (Pinus taeda L.). Clones from four of the families (282 clones) were planted in South Carolina and clones from the other four families in Florida (168 clones). Both tests were laid out as randomized complete blocks with nine blocks and one ramet/clone/block. Height, survival and rust resistance were measured annually and diameter was measured at age four. Best Linear Unbiased Prediction of clone genetic values were estimated for height and volume at age four. Estimated genetic gains from various clone selection strategies, the effect of increasing or decreasing the number of ramets for testing on genetic gain, height age-age and trait-trait genetic correlations were estimated. Estimated genetic gain was highly sensitive to the intensity of clonal selection. Selecting the single best clone from each test resulted in an estimated gain of 13% (SC) and 14% (FL) in height at age four over the test average (all the clones). The single best clone from each family at each site (four in total) resulted in an estimated gain of 10% (SC) and 10% (FL). However, if six clones were selected from each family (twenty-four in total), gain was reduced to 8% (SC) and 6% (FL) in height. The genetic correlations between height at age one and height at age four were low (0.60 at SC, 0.58 at FL), but increased to 0.96 (SC) and 0.97 (FL) between heights at ages three and four. Simulations using test parameters showed that estimated gain increased with the number of ramets tested up until ten ramets per clone, but did not increase appreciably with ramet numbers above ten. Moreover, approximately 90% of the gain could be obtained using only six ramets. These data, additional details and the implications of these results will be discussed.
  • Publication
    Light Quality Effects on Germination and Conversion of Southern Pine Somatic Embryos
    (2003) Merkle, S. A.; Upchurch, B. L.; Montello, P. M.; Xia, X.; Smith, D. R.