Ante-dependence models can be used to model the covariance structure in problems involving repeated measures through time. They are conditional regression models which generalize Gabriel’s constant-order ante-dependence model. Likelihood-based procedures are presented, together with simple expressions for likelihood ratio test statistics in terms of sum of squares from appropriate analysis of covariance. The estimation of the orders is approached as a model selection problem, and penalized likelihood criteria are suggested. Extensions of all procedures discussed here to situations with a monotone pattern of missing data are presented.

The relationship among live weight, chemical body composition and energy content (at artificial insemination (AI) and three days before parturition), estimated by bioelectrical impedance with fertility rates and the percentage of kits born alive, was studied during the first three AI. The first AI was conducted at 16 weeks of age in 137 rabbit does that weighted 3.91 ± 0.46 kg. Their body chemical composition was 17.4 ± 0.50%, 16.1 ± 2.6%, 1067 ± 219 kJ/100 g body weight, for protein, fat and energy, respectively. An increase in body protein, fat and energy content at the first AI did not affect fertility at the first AI but improved it at the second AI (p ≤ 0.030). Moreover, an increase in body fat and energy content at the second AI improved fertility at the second AI (p ≤ 0.001). Fertility at the third AI was positively influenced by body protein at the third AI and the increase in body protein and fat between the second parturition and the third AI (p ≤ 0.030). The percentage of kits born alive at the first and at the second AI improved with the increase in body protein at the first AI (p ≤ 0.040). In conclusion, a minimal body protein and fat content is required at the first AI to optimize the reproductive performance in young does.

Water quality regulations entail a substantial commitment of resources from governments and private entities. It is important to continually evaluate the effectiveness of these regulations to ensure they are having the intended impact. In this paper, we evaluated nutrient data as indicators of primary productivity and dissolved oxygen (DO) concentrations and pH as response variables to assess historical water quality trends from 55 stations of Puerto Rico. The stations were divided into impaired versus non‐impaired categories based on their historical total phosphorus (TP) mean concentration. Mean TP and total nitrogen (TN) concentrations were significantly higher in the impaired stations relative to the non‐impaired stations. In contrast, DO mean concentrations and mean pH values were significantly lower in the impaired stations. A generalized additive mixed model was used to demonstrate temporal trends. A significant decrease in TP and TN concentrations was observed with time at the impaired stations. This was accompanied by significant increases in DO concentrations and pH. The non‐impaired stations showed a marginal (statistically nonsignificant) decreasing trend with time. The large reductions in nutrient concentrations observed at the impaired stations seem to be related to the closure of several primary wastewater treatment plants (WWTPs) across the island. The conversion of abandoned crop agricultural lands into secondary forest in recent decades has resulted in small but significant decreases in TN (not TP) in receiving streams. We conclude that the Clean Water Act has promoted improvements in water quality in Puerto Rico by advancing upgrades in sanitary infrastructure and the regulation of point sources of pollution.

Introduction. Common beans (Phaseolus vulgaris L.) in Central America and the Caribbean are often produced on low fertility soils which reduces crop yield. Bean breeding programs need to identify genotypes that have superior adaptation to these conditions. Objective. Identify Mesoamerican bean germplasm lines with superior adaptation to low soil fertility. Materials and methods. The performance of twenty-seven Mesoamerican bean (Phaseolus vulgaris L.) lines from the Bean Abiotic Stress Evaluation (BASE) 120 panel were evaluated in an unfertilized oxisol at Isabela, Puerto Rico over five growing seasons (four-year period from 2015-2018). The lines were inoculated with a mixture of Rhizobium etli and R. tropici to promote symbiotic nitrogen fixation (SNF). Results. Four lines produced mean seed yields >1,200 kg ha-1 and had estimates of nitrogen derived from the atmosphere (NDFA) >50 %. Greater nodule number was positively correlated with % NDFA, later maturity and seed yield. The heat and drought tolerant small red cultivar ‘Rojo Chortí’ and the heat tolerant white cultivar ‘Verano’ had among the smallest apparent C isotope discrimination values suggesting greater water use efficiency. Among the elite lines in the trial, root rot damage was minimal and the basal root growth angles were intermediate (40-60 %), which favored the uptake of water and soil nutrients. Conclusion. Mesoamerican bean lines with superior seed yield and enhanced symbiotic nitrogen fixation in a low fertility soil were identified. Many of these lines also possess resistance to other biotic and abiotic factors that limit bean seed yield in Central America and the Caribbean.