Good agronomic research is often characterized by experiments designed with more than one factor identified as a fixed effect. However, many researchers do not recognize and report the maximum impact of their research because they do not appropriately analyze and interpret the well-designed interactions between factors. One purpose of this chapter is to provide guidance on properly analyzing and interpreting interactions between fixed effects. We demonstrate why it is crucial to consider all interactions of all factors and optimize your model to reach proper conclusions and make correct recommendations that maximize the impact of your research. We provide examples of the analysis and interpretation of interactions, including single degree of freedom contrasts of linear, quadratic, and cubic responses when a quantitative fixed effect is one component of an interaction. A second purpose of this chapter is to discuss considerations that will help researchers determine if effects should be analyzed as random or fixed and to clarify how this affects inferences. The discussion on random effects is based on examples using 2 and 10 yr of data. Key concepts learned by the reader in this chapter and its appendices include how to use the GLIMMIX Procedure in SAS to calculate a LSD, how to develop contrast statements involving regression responses, how to interpret interactions involving regression responses, and precautions necessary in interpreting traditionally random effects such as years or locations if it is prudent to analyze them as fixed effects.

Using eight examples, each posing some specific challenges, we show the basics and possible extensions of the simple and multiple linear regression models. We discuss similarities and differences if both the regressor and the regressand are random variables and if only the regressand is a random variable. Several methods to evaluate model assumptions are illustrated. Transformations are applied where needed to achieve linearity between variables or to resolve issues of variance heterogeneity. Further possibilities to allow for variance heterogeneity are shown in the context of the mixed linear model. The integration of fixed or random classification variables (treatment factor or disturbance factor) into regression models are discussed. In two examples, we show that linear regression is sometimes only a first approach, and for such cases, we provide guidance on completing the analyses with nonlinear models or models with covariances between the residuals.

The addition of organic residues to sand soils can improve soil properties and sugarcane productivity. However, biochar use can have variable effects on crops, and few studies have evaluated the effect of mill ash applications on sand soils. This study aimed to determine the effect of mill ash, and three biochars on yields of sugarcane grown on sand soils of South Florida near the Everglades Agricultural Area (EAA). Nine treatments and a control were evaluated. Treatments consisted of mill ash (AS) and biochars produced from hardwood yard waste (HY), horse barn shavings (HM), and rice hulls (RH) incorporated at 1% and 2% (by weight) prior to the plant-cane crop of a lysimeter experiment that extended for two crop cycles. A standard practice treatment of mill ash applied at 6% was also included. Results show that mill ash applied at 6% and 2% (AS6 and AS2), and rice hulls biochar applied at 2% (RH2) produced significantly greater biomass and sucrose yields compared with the control in the plant-cane and the first-ratoon crops. Soil silicon concentrations increased with AS and RH in the plant-cane crop, which may have contributed to improved yields. Treatments AS6 and RH2 may have also increased sugarcane yields through improved soil properties. Our results show that mill ash and rice hulls biochar amendments can potentially improve sugarcane yields on sand soils near the EAA.

‘CP 06‐2042’ (Reg. No. CV‐169, PI 675156) sugarcane (a complex hybrid of Saccharum spp.) was developed through cooperative research conducted by the USDA‐ARS, the University of Florida, and the Florida Sugarcane League and released to growers for organic (muck) and mineral (sand) soils in Florida in September 2014. CP 06‐2042 was selected from a cross of ‘CP 96‐1252’ × 01 P 04 made at Canal Point, FL, in December 2001. The female parent, CP 96‐1252, is a commercial cultivar. The male parent is unknown, as it is one of 19 sugarcane cultivars in the polycross. Cane yield of CP 06‐2042 on muck soils, averaged across 18 harvests through three crop cycles (plant cane, first ratoon, and second ratoon), was 26.1% higher (P < 0.1) than that of ‘CP 89‐2143’, a commercial check for muck soils. Although CP 06‐2042 had 1.8% lower commercial recoverable sucrose than CP 89‐2143, high cane yield resulted in 22.8% higher (P < 0.1) sucrose yield than CP 89‐2143. Averaged across nine harvests through three crop cycles on sand soils, CP 06‐2042 had 12.1% (P < 0.1) higher cane yield, 2.9% higher commercial recoverable sucrose, and 13.1% (P < 0.1) higher sucrose yield than ‘CP 78‐1628’, a commercial check for sand soils. CP 06‐2042 was released because of its high cane and sucrose yields on both muck and sand soils and its acceptable levels of resistance to brown rust, leaf scald, Sugarcane mosaic virus strain E (mosaic), and ratoon stunt. CP 06‐2042 is susceptible to orange rust and has poor freeze tolerance.

Rising energy costs, decreasing soil depth, and best management practices that require on-farm water retention have led to interest in flood-tolerant bioenergy crops for production in the Everglades Agricultural Area (EAA) of Florida. Information on the effects of high water tables on sugarcane (Saccharum spp. L.) is available; however, little is known about the bioenergy crops elephant grass (Pennisetum purpureum Schumach.), energycane (Saccharum spp. L.), and giant reed (Arundo donax L.). The objective of this study was to evaluate the role of water-table depth on the yields, morphology, physiology, and early season growth of these three bioenergy species in relation to sugarcane. The species were grown in a greenhouse and subjected to three water management strategies, −40 or −16 cm constant water-table depths, and bi-weekly flooding with drainage to −40 cm. The experiment was conducted in plant-cane, first-ratoon, and successive plant-cane crop cycles.       Periodic flooding relative to a −40 cm constant water table significantly reduced dry weights and leaf area index and increased aerenchyma development. Averaged among crops, dry weight declined by 32, 49, 37, and 52% while leaf area index (LAI) decreased by 31, 50, 46, and 57% for giant reed, elephant grass, energycane, and sugarcane, respectively, in response to flooding. Aerenchyma proportional area of the stalk increased by 52, 90, and 95% for elephant grass, energycane, and sugarcane, respectively, but aerenchyma in giant reed decreased by 11%. High yields and stalk populations for all water tables were observed in energycane and elephant grass. These two species show potential for bioenergy production in the EAA. Field trials are recommended to confirm production scale yield.

Long-term improvement of sugarcane and energy cane (complex hybrids of Saccharum spp) cultivars can be enhanced by breeding with the type of diverse germplasm available at the World Collection of Sugarcane and Related Grasses (WCSRG) maintained in Miami, Florida. To evaluate germplasm in the WCSRG for breeding purposes, a diversity panel was selected with approximately 300 accessions and planted at Canal Point, FL in three replications. These accessions were measured for stalk height and stalk number multiple times throughout the plant-crop growing season and for Brix and fresh biomass during the 2013 harvest. First-ratoon stalk height, stalk number, stalk diameter, internode length, Brix, and fresh and dry biomass were evaluated in 2014. The highest correlations were found between early season measurements and harvest traits. Hybrids had higher fresh weight and Brix while Saccharum spontaneum had higher stalk number and dry mass. According to the principal component analysis, the diversity panel was divided into two groups. One group had accessions with high stalk number and high dry biomass like S. spontaneum and the other had accessions with higher Brix and fresh biomass such as S. officinarum. In first ratoon, there were 110 accessions not significantly different in Brix from the sugarcane commercial standards, including 10 S. spontaneum accessions, and 17 and six accessions that were higher than commercial standards in dry and fresh mass, respectively. This study shows the variability in traits of interest and the breeding potential of accessions within the WCSRG for sugar-and energy-cane cultivar development.

‘CPCL 05‐1201’ (Reg. No. CV‐163, PI 672485) is a sugarcane (Saccharum spp.) hybrid cultivar bred by United States Sugar Corporation, tested in the breeding program led by the USDA–ARS, University of Florida, and Florida Sugarcane League, and released to Florida growers in October 2012. CPCL 05‐1201 combines the high‐sucrose genes of its female parent (CL 87‐2882) with the high tonnage genes of its male parent (CL 93‐2679). In the final selection stage, CPCL 05‐1201 was compared with 21 genotypes for 3 yr (2008–2012) on organic and sand soils in south Florida and the freeze tolerance tested for 2 yr (2011–2013) in north Florida. CPCL 05‐1201 is recommended for planting on muck and sand soils and for the successive cropping system. On muck, CPCL 05‐1201 yielded 5 to 10% higher sucrose than reference cultivars ‘CP 78‐1628’ and ‘CP 89‐2143’. On sand, CPCL 05‐1201 yielded sucrose comparable to CP 89‐2143 and 4% more than CP 78‐1628. Owing to high tonnage and moderate level of sucrose, CPCL 05‐1201 was 11% more profitable on muck and 26% more profitable on sand than CP 78‐1628 or CP 89‐2143. No yield decline was observed with CPCL 05‐1201 under the successive system. In addition, CPCL 05‐1201 is resistant to the most important diseases in Florida, except for Sugarcane yellow leaf virus. CPCL 05‐1201 has the Bru1 gene in its genome and is resistant to brown rust (caused by Puccinia melanocephala H. & P. Sydow) and orange rust [caused by P. kuehnii (Kruger) E. Butler]. CPCL 05‐1201 has comparable freeze tolerance to CP89‐2143 to mild and severe freeze conditions.

There are three kinds of lies; lies, damned lies and statistics. Statistics is a body of methods for making wise decisions in the face of uncertainty. In order to shed light on the use and misuse of statistics in the agricultural and natural resource sciences, Agronomy Journal set about to provide its authors and readers with useful information on statistical approaches and strategies for conducting and publishing scientific information. As readers, researchers, reviewers, or as editors of scientific literature, we may have come across situations or articles where we had serious doubts whether the most appropriate method to analyze an experiment or set of experiments was used. Furthermore, ever-increasing computing power has enabled researchers to use more intensive and complex statistical methods with the help of new and updated general and special purpose statistical software programs always becoming available. With these more complex computations, we now also have the ability to design more advanced and efficient experiments. We as biological scientists may or may not enjoy or have the time or training for learning to properly design, program, and interpret these sophisticated tools, a set of activities that is itself a continually dynamic field of study. As scientific researchers, it is our charge to produce forward-thinking and meaningful experimentation with the most appropriate and accurate methodology available. With these thoughts in mind, Brent Godshalk, then Editor of the Agronomy Journal, and his Editorial Board recognized the need for communications on these topics at the 2011 Agronomy Journal Board meeting and chartered the Statistics Committee. The Agronomy Journal Statistics Committee set a goal to provide clear guidance on the appropriate use of effective statistical tools and procedures to Agronomy Journal readers and contributors; thereby allowing them to focus on and support their science with proper statistical approaches. To this end, the committee and the ASA Biometry and Statistical Computing Section co-sponsored the Symposium, “Statistical Concepts and Tools to Aid in Publishing Proper Research Conclusions” at the 2012 ASA, CSSA, and SSSA International Annual Meetings. Our second and major output is this series of articles in the Agronomy Journal. The articles will also be part of the “Special Sections” webpage (https://www.agronomy.org/publications/aj/special-sections). All of the software code that was presented in appendices in separate articles is also available in one file that can be assessed at https://www.agronomy.org/publications/aj/special-sections. The software code is also a supplemental file with this introduction article. These articles and the additional output of our committee are targeted to support scientists in key basic concepts and methodologies in statistics and biometry. We believe that many of us need support in these basic concepts as an effective means of complementing the trend to focus on supporting scientists to learn exceedingly more complex and advanced statistical tools and methodologies. Neither the committee nor the contributing authors pretend that this effort is comprehensive or will solve all the challenges of data analysis and interpretation facing the agricultural and natural resources research communities. Our goal is that Agronomy Journal contributors and others in the scientific community strive to learn and use the newer, more advanced statistical methodologies—but first are sure to understand and use the basic methodologies when these are more appropriate. We hope that each of you agree with the goal of striving to conduct, interpret, and communicate your excellent science in the best possible way. Each article in this series received robust reviews from several reviewers. More than 30 anonymous reviewers in all donated their time, and we thank them for the substantial contribution they have made to this series. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Sugarcane (Saccharum L. spp. hybrids) growers depend on breeding programs for new, high-yielding cultivars that have resistance to abiotic and biotic stresses, so breeders continually seek out widely adapted, high yielding germplasm to be used as parents for their programs. Cultivars are sometimes used for this purpose, but their use may be minimized to prevent genetic diversity erosion. The purpose of this study was to determine the importance of cultivars as parents in three USA (one in Florida and two in Louisiana) sugarcane breeding programs by quantifying the percentage of cultivars that had these parental groupings based on published registrations and crossing records. The percentage of cultivars with at least one commercial parent for each program was 81.8%, 77.5%, and 64.3% for the Houma (Ho), Louisiana, Canal Point (CP), Florida and Louisiana State University (LSU) programs, respectively, but cultivars were recently used as parents in only 11.8% (Ho), 16.39% (CP), and 34.3% (LSU) of crosses. The results indicate that the CP and Ho programs should consider increasing the use of cultivars as parents in their breeding programs to increase the probability of selecting potential commercial genotypes, but this should be balanced with high diversity crosses to avoid the loss of diversity.

Agronomic research results include Type 1 (α) and Type 2 (β) errors. Results are often reported using α ≤ 0.05 while β is ignored. Our objective was to discuss whether a false positive was more serious than a false negative in agronomic research. For comparison, current statistical methods used in Agronomy Journal were tabulated. Most papers used null hypothesis tests with α ≤ 0.05, reporting results based on the LSD among all treatment pairs. Current practices do not account for the relative costs of false positive vs. false negative errors. A case study from the Washington State Wheat Extension trials was analyzed using mixed models with specific contrasts. While the overall effect for cultivar was significant, the β error rate for the contrast was 40% and additional replications were needed to increase the power of this contrast. A second case study analyzed trials evaluating wheat ( Triticum aestivum L.) resistance to Fusarium crown rot. Optimal α and β error rates were estimated for two to eight replications with the Type1/Type2 error cost ratio set at 1:1 and 1:5. An average error rate (α and β) ≤ 0.05 could be achieved with four replications when a reduction in the β error was critical and α errors could be corrected in future experiments. Effective experimental design requires estimation of the acceptable magnitude and cost ratio of false positive and false negative errors and critical effect sizes. To be truly informative, reports of results should include this information plus observed effect sizes and variances.

Variability in yield among sugarcane (a complex hybrid of Saccharum spp.) genotypes grown with and without mill mud application on sand soils in Florida has been documented, but little is known about what causes yield differences and if there are any relationships between yield components and physiological parameters. This study determined responses of leaf relative chlorophyll level (SPAD readings), photochemical efficiency of photosystem II in dark‐adapted leaves ( F v/ F m), leaf stomatal conductance ( g s ), net photosynthetic rate (Pn), transpiration rate ( E ), difference between leaf and air temperatures ( T l– T a), and spectral reflectance in one Erianthus [ Erianthus arundinaceus (Rez.) Jeswiet] and seven sugarcane genotypes and their relationships to yield components across genotypes. Reflectance, F v/ F m, SPAD, g s , Pn, E , and T l– T a were measured on leaves at the top visible dewlap during the first‐ and second‐ratoon crops on a Margate sand soil (siliceous, hyperthermic Mollic Psammaquent) with or without mill mud added before planting. Mill mud application reduced leaf reflectance at wavelengths of 560 and 710 nm and significantly increased leaf SPAD, g s , Pn, and E in the first‐ratoon crop only. Genotypic differences were detected in most physiological traits and yield components. Most physiological traits correlated with cane and sucrose yields when no mill mud was added. Genotypic variation in physiological and yield responses to mill mud application and genotype × mill mud interactions in yields indicated that augmenting yield selection of genotypes with leaf physiological traits will improve sugarcane breeding programs for sand soils.

When reporting on well‐conducted research, a characteristic of a complete and proper manuscript is one that includes analyses and interpretations of all interactions. Our purpose is to show how to analyze and interpret interactions in agronomy and breeding research by means of three data sets comprising random and fixed effects. Experiment 1 tested wheat (Triticum aestivumL.) at two N and four P fertilizer rates in two soil types. For this data set, we used a fixed‐effect linear model with the highest order (three‐way) interaction considered first and then worked down through the lower order interactions and main effects to illustrate the importance of interactions in data analysis. Experiment 2 evaluated maize (Zea maysL.) hybrids under four rates of N for 3 yr. For this data set, we used a linear mixed model and partitioned the four N rates into orthogonal polynomials. Experiment 3 evaluated genotypes in six environments where the objective was to show how to study genotype × environment interactions. Researchers must analyze all interactions, determine if they are due to changes in rank (crossover) or only to changes in scale, and then judge whether reporting on significant main effects or interactions would best explain the biological responses in their experiments. In an experiment with more than one factor, complete and correct analysis of interactions is essential for reporting and interpreting the research properly.

'CP 06-2400' (Reg. No. CV-159, PI 670018) sugarcane (a complex hybrid of Saccharum spp.) was developed through cooperative research conducted by the USDA-ARS, the University of Florida, and the Florida Sugar Cane League, Inc., and released to growers for organic (muck) soils in Florida in October 2013. CP 06-2400 was selected from a polycross made at Canal Point, FL, on 16 Jan. 2004 with cultivar CP 94-1100 as the female parent. The male parent could be any one of the clones used in the polycross. CP 06-2400 was released because of its high cane and sucrose yields and acceptable commercial recoverable sucrose on muck soils and its acceptable levels of resistance to brown rust (caused by Puccinia melanocephala H. & P. Sydow), orange rust (caused by Puccinia kuehnii E.J. Butler), leaf scald (caused by Xanthomonas albilineans Ashby, Dowson), Sugarcane mosaic virus strain E (mosaic), ratoon stunt (caused by Leifsonia xyli subsp. xyli Evtsuhenko et al.), and smut [caused by Sporisorium scitamineum (Syd.) Piepenbring et al.] in Florida. Based on results of 17 harvests of three crops (plant cane, first ratoon, and second ratoon) in the final-stage replicated yield trials conducted at six locations on muck soils and compared with a reference cultivar CP 89-2143, CP 06-2400 had 38.2% higher cane yield and 26.4% higher sucrose yield (P < 0.01). CP 06-2400 ranked second to third best for freeze tolerance among 21 genotypes field tested for temporal sucrose deterioration.

Yields of sugarcane (a complex hybrid of Saccharum spp.) in FL, USA, are lower on sand soils than on organic (muck) soils. Nitrogen (N) supply may limit sugarcane growth and yields on these sand soils. A 2-year pot study was conducted to determine sugarcane genotypic variation in response to N rate on a sand soil. Treatments included four N rates (0, 75, 150 and 225 kg ha(-1)) and three sugarcane genotypes (CP 80-1743, CP 01-2390 and TCP 87-3388). Nitrogen fertilizer was equally split and applied at about 55 and 125 days after planting (DAP) for each treatment. During the experiment, the number of nodes and length of the primary stalks and tillers were recorded. Leaf relative chlorophyll (soil plant analysis development (SPAD)) and net photosynthetic rate (Pn) were measured biweekly. All plants were harvested at 183 DAP to measure green leaf area (GLA), shoot biomass accumulation and partitioning, and fertilizer N use efficiency (NUE). Genotypes differed significantly in leaf SPAD, Pn, GLA, and shoot biomass accumulation and partitioning. CP 01-2390 had the highest leaf Pn and shoot biomass, and CP 80-1743 had the lowest GLA, shoot biomass and NUE among genotypes. Nitrogen rate affected leaf SPAD, GLA, shoot biomass and NUE, but had much less effect on leaf Pn. Green leaf area and biomass increased with increasing N rates. Our results suggest that a two-pronged approach, selection of genotypes with high NUE while working to optimize N rates and delivery can improve sugarcane yields on sand soils.

Sugarcane (Saccharum spp.) and other members of Saccharum spp. are attractive biofuel feedstocks. One of the two World Collections of Sugarcane and Related Grasses (WCSRG) is in Miami, FL. This WCSRG has 1002 accessions, presumably with valuable alleles for biomass, other important agronomic traits, and stress resistance. However, the WCSRG has not been fully exploited by breeders due to its lack of characterization and unmanageable population. In order to optimize the use of this genetic resource, we aim to 1) genotypically evaluate all the 1002 accessions to understand its genetic diversity and population structure and 2) form a core collection, which captures most of the genetic diversity in the WCSRG. We screened 36 microsatellite markers on 1002 genotypes and recorded 209 alleles. Genetic diversity of the WCSRG ranged from 0 to 0.5 with an average of 0.304. The population structure analysis and principal coordinate analysis revealed three clusters with all S. spontaneum in one cluster, S. officinarum and S. hybrids in the second cluster and mostly non-Saccharum spp. in the third cluster. A core collection of 300 accessions was identified which captured the maximum genetic diversity of the entire WCSRG which can be further exploited for sugarcane and energy cane breeding. Sugarcane and energy cane breeders can effectively utilize this core collection for cultivar improvement. Further, the core collection can provide resources for forming an association panel to evaluate the traits of agronomic and commercial importance.

In Florida, long‐term results for identifying high yielding sugarcane ( Saccharum spp.) cultivars have been better for Histosols (muck soils) than sand soils. We examined whether selection could be improved by comparing genotypes on a sand soil with and without added mill mud (MM) (in Florida, MM is primarily muck soil). One Erianthus and 31 sugarcane genotypes were planted in 2007 with MM at 0 or 1510 m 3 ha –1 in main plots and genotypes as subplots in a 3‐yr field experiment on a Margate sand soil (siliceous, hyperthermic Mollic Psammaquent). Commercial recoverable sucrose (CRS) (g kg –1 ), and cane (CY) and sucrose (SY) yields (Mg ha –1 ) were determined during the next 3 yr. Mill mud reduced CRS from 127 to 111 g kg –1 but increased cane and sucrose yields from 80 to 150 and 10 to 17 Mg ha –1 , respectively. Compared with the check of CP 89‐2143, 2, 10, and 8 genotypes were differentially affected by soil treatment ( P = 0.10) for CRS, CY, and SY, respectively. CP 01‐2390 was the most adapted sand genotype; its CYs on sand with and without MM were 147 and 143 Mg ha –1 , respectively. Compared with previous research, CY responses were well predicted for five genotypes, but poorly predicted for four genotypes in this study. Multiple locations are needed for sugarcane genotype selection on sand soils. Using a sand soil with and without added MM can be a useful supplemental, rather than singular approach for improving sugarcane genotype selection in Florida.

Development of 'CPCL 02-6848' (Reg. No. CV-158, PI 667596), sugarcane (a complex hybrid of Saccharum spp.) was initiated by the United States Sugar Corporation (USSC) and completed by collaborative research of the USDA-ARS, the University of Florida, and the Florida Sugarcane League, Inc. The female parent of CPCL 02-6848 is CL 92-2533. The male parent is not known because the flower of CL 92-2533 was exposed to the pollen of several male flowers (polycross). 'CL 41-223' is the only cultivar in the pedigree of CPCL 02-6848 that was cultivated on a large acreage (87% of total acreage in 1962) in Florida. The major attributes of CPCL 02-6848 include its high yields of cane tonnage in the plant-cane through the second-ratoon crops on both muck and sand soils, its resistance to smut (caused by Ustilago scitaminea Syd. & P. Syd.), brown rust (caused by Puccinia melanocephala Syd. & P. Syd.), and Sugarcane mosaic virus strain E (SCMV), and moderate resistance to leaf scald [caused by Xanthomonas albineans (Ashby) Dowson] and ratoon stunting disease (caused by Leifsonia xyli subsp. xyli Evtsuhenko et al.) in Florida. CPCL 02-6848 was released on 16 Oct. 2012, and it is expected to be cultivated on both muck and sand soils in Florida.

Varietal improvement program at the United States Department of Agriculture-Agricultural Research Service (USDA-ARS), Field Station at Canal Point, Florida consists of 4 stages. Stages I and II consist of unreplicated sugarcane (Saccharum Spp.) genotypes advanced from seedlings while Stages III and IV consist of replicated highly selected genotypes. Although new genotypes are unreplicated in Stage II, checks are replicated in an augmented design. Best Linear Unbiased Predictors (BLUPs), primarily used in animal breeding, could be used in the breeding program to select genotypes having improved characters. The present study was conducted to determine whether selection indices incorporating BLUPs would improve the Stage II selection program. Basic statistics and frequency distributions showed the presence of variability in brix, sucrose, purity, recovery, stalk weight, stalks per plot, commercial recoverable sucrose (CRS), and tons of cane per ha (TCH). However, analysis of variance showed significant differences only for stalk weight and the contrast of checks versus new genotypes. Path analysis revealed highest direct effects for stalk weight and stalks per plot. The selection index comprised of stalk weight, stalks per plot, and CRS. The results showed that although selection indices estimated using Estimated BLUPs (EBLUPs) improved means of the individual genotype selection, values were lower than the means obtained using the selection method used at the USDA-ARS Field Station at Canal Point. However, selection indices calculated using raw data gave lower mean values when compared to EBLUPs. The present study shows promising results with regard to use of BLUPs in the selection index method. Index selection screened genotypes based on multiple characters and were better in performance than the rest of the genotypes. These results could be improved by including an equal number of checks across all blocks for Stage II experiments. Furthermore, culling on the basis of disease ratings should precede the index selection procedure.

The Saccharum L. genus includes important crops that are utilized for sugar and fuel production. The World Collection of Sugarcane and Related Grasses (World Collection) in Miami, FL contains diverse and potentially useful germplasm for this and related genera; however, this collection has been underutilized because little is known about the traits of its accessions. Our objectives were to phenotypically characterize the World Collection and select a representative core collection that could then be studied intensively. In total, eight morphological traits of the World Collection were evaluated three times in 1 year. A core of 300 accessions that included each species in the World Collection was selected by using the Maximization Strategy in MStrat software. The core had a higher diversity rating than random selections of 300 accessions. The Shannon–Weaver Diversity Index scores of the core and whole collection were similar indicating that the majority of the diversity was captured by the core collection. The ranges and medians between the core and World Collection were similar; only two of the trait medians were not significant at P = 0.05 using the non-parametric Wilcoxon method and the coincidence rate (CR % = 96.2) was high (>80) indicating that extreme values were retained. Thus, the phenotypic diversity of these traits in the World Collection was well represented by the core collection. Agronomic studies on the core should be useful for characterizing the World Collection and genes for useful traits should be available in the core collection.