Curvilinear Regression Models Research Articles

Effect of weather parameters on pest-disease of okra during summer season in middle Gujarat

To assess the impact of weather on pest-disease of okra, a field experiment was conducted in summer season 2009 in split plot design with five dates of sowing (D1 -15th February, D2 -1st March, D3 -15th March, D4 -1st April and D5 -15th April) and two spraying levels (S0-control, S1-spraying) at Agronomy Farm of B.A. College of Agriculture, Anand Agricultural University, Anand. The results revealed that the infestation of pests and diseases was higher in delayed sowing as compared to early sowing resulting reduction in yield by 44.4%. Different weather parameters were found to have different effect on various pests and diseases. The correlation study revealed that the Tmax had significant influence on jassid population while relative humidity had significant influence on aphid population. The whitefly, percent fruit damage and YVMV disease were significantly influenced by most of the parameters. However, bright sunshine hours (BSS) and pan evaporation (EP) did not have significant influence on any of the pests and disease of okra. Linear simple regression models, multiple regression models and curvilinear regression models developed with various weather parameters. Among all models, the curvilinear regression model using GDD was found well fitted in order to explaining the maximum extent of variability (83** to 98**%) in the population and intensity of most of the pests and disease except whitefly which was well explained by linear regression developed with VPmean (R2 0.88**).

Internationalization and financial performance in the global forest industry

SUMMARY In this study we analysed the relationship between degree of internationalization and corporate financial performance in the forest industry at firm level. Both linear and curvilinear regression models were estimated for the impact of depth and scope of internationalization on corporate performance, using data available for 50 large forest products firms. As a result, we confirmed a U-shaped curvilinear relationship between depth of internationalization (measured by percentage of foreign employees to total employees, FETE) and corporate financial performance (measured by return on capital employed, ROCE). The threshold proportion of foreign employees was found to be 35% in the forest industry. In addition, we also confirmed that the scope of internationalization (measured by number of operating countries, OC) has a positive effect on corporate performance. From a strategic management point of view, our study thus corroborated Michael Porter's dictum: “if you go global, do not get stuck in the midd...

ROOT GROWTH AND DEVELOPMENT OF 'CAS' GUAVA (PSIDIUM FRIEDRICHSTHALIANUM BERG-NIED.) SEEDLINGS DURING NURSERY STAGE

A root growth analysis for 'Cas' guava (Psidium friedrichsthalianum Berg-Nied.) seedlings was established. Seeds were extracted from ripe and healthy fruit harvested at the guava germplasm bank of Centro Fruticola of Zulia, Venezuela (11°00'00 LN, 71°30'00 LW). Seeds were sown in a seedbed that contained sterilized substrate. Seedlings were transferred into bags when they were about eight cm in height. Seedlings were grown in five or ten kilogram polyethylene bags containing three mixtures (1:1, 1:2 and 2:1) of soil and goat manure as substrates. Sampling was done monthly during six months. The variables measured for root growth were main root length, root fresh and dry mass, volume, and area. Treatments were arranged in a factorial, completely randomized design with five replicates. Data were statistically analyzed utilizing the SAS software. SigmaPlot of Jandel Scientific was applied to select the curvilinear regression model that best fitted the results. The model used to adjust the curves was y= a/1+e -(x-xo/b) . The equations that represent root volume growth for the proportions 1:1, 1:2 and 2:1 were: y= 9.310e+0/1+e -(x-1.335e+2/2.404e+1) R 2 =0.94; y= 1.281e+1/1+e -(x-1.379e+2/2,981e+1) R 2 =0.95 and y=9.659e+3/1+e -(x-5.119+2/4.894e+1) R 2 =0.95, respectively. Bag size and soil:goat manure proportions are related with time during nursery stage. Ten kilogram bags and the 1:2 mixtures were appropriate for good root growth and development during six months in the nursery.

Prediction of stature using hand dimensions

In this study, an attempt was made to find out possible correlations between hand dimensions and stature using linear and curvilinear regression models for both genders. The study was conducted amongst 250 medical students (125 male and 125 female) aged 18–30 years. Each student has been studied for measurements of stature, hand length and hand breadth. To explain stature, all the explanatory variables like age, sex, hand length (right and left) and hand breadth (right and left) were included for model generation using SPSS. A general linear regression model was found to be best explanatory in both males and females, however, amongst the curvilinear models; the exponential model emerged as the ‘best’ in explaining stature of the individual. Left hand length alone explained very significantly ( P < 0.001).

Structure‐function relationship in the process of primary open‐angle glaucoma ‐ an OCT study

Abstract Purpose: To evaluate the strength and pattern of the relationship between visual field (VF) sensitivity and retinal nerve fiber layer (RNFL) thickness as measured by StratusOCT, and define the structural decay where glaucomatous VF defects appear. Methods: Normal, preperimetric (PPG) and glaucomatous (POAG) subjects were enrolled in this cross‐sectional study. Linear and non‐linear regression models were used to define the relationship between average RNFL thickness (AVG) and VF sensitivity (mean deviation‐MD and mean sensitivity‐MS) on standard automated perimetry, coefficient of determination (R2) was calculated, their association was described by bivariate Pearson correlation. Results: The correlation of AVG and MD/MS was significant only in POAG eyes (r 0.718/0.733), their relationship fit a curvilinear regression model (R2 0.723/0.694), in normal and PPG groups no correlation and a minimal degree of determination was detected. Receiver operating characteristic (ROC) curves describing the ability of all VF parameters and AVG were evaluated to differenciate PPG from POAG eyes. Repeated analysis with the best performing test parameter, PSD (AUROC=0.937) with cutoff value 1.9dB, showed that regression profiles only in the POAG group with PSD &gt;1.9 dB maintained their strong curvilinear RNFL/VF relationship. Conclusions: Evaluating structure/function relationship in our normal, PPG and POAG subjects, strong curvilinear regression was found in POAG eyes with PSD &gt;1.9 dB and RNFL AVG thickness below 70 μm, which might represent a profound threshold value in glaucomatous structural changes, while no correlation was detectable above these values. This is comparable with histologic data, indicating that 25‐40% of RNFL is lost before VF defects arise.

Estimation of stature by foot length

The aim of this study was to develop a relationship between foot length and stature using linear and curvilinear regression models. Measurements of foot length and stature were taken from 250 medical students (125 males and 125 females) aged 18–30 years. General multiple linear regression model was highly significant ( P < 0.001) and validated with highest values for the coefficients of determination R 2 = 0.769 and multiple correlation coefficient r = 0.877. Right foot length, sex and age explained for about 77% variations in stature.

Relationship between Visual Field Sensitivity and Retinal Nerve Fiber Layer Thickness as Measured by Optical Coherence Tomography

To evaluate the strength and pattern of the relationship between visual field (VF) sensitivity and retinal nerve fiber layer (RNFL) thickness as measured by StratusOCT (Carl Zeiss Meditec, Inc., Dublin, CA). Three hundred eleven subjects--45 normal, 102 with preperimetric glaucoma (PPG), and 164 with primary open-angle glaucoma (POAG)--were enrolled in this cross-sectional study. The relationship between RNFL thickness and VF sensitivity, expressed as mean deviation (MD) and mean sensitivity (MS), were evaluated with linear and nonlinear regression models, and the coefficient of determination (R(2)) was calculated. The association between RNFL/VF was described by bivariate Pearson correlation coefficients. The correlation of RNFL and the VF parameters MS and MD in normal and PPG eyes was not significant. In POAG eyes, RNFL and both MS (r = 0.733) and MD (r = 0.718) correlated significantly. Linear regression plots of MS or MD against RNFL thickness demonstrated a negligible degree of determination in normal (R(2) = 0.0378 and 0.0121, respectively) and PPG groups (R(2) = 0.0215 and 0.0151, respectively), whereas their relationship fit a curvilinear regression model (R(2) = 0.6947 and 0.723) in the POAG group. Receiver operating characteristic (ROC) curves describing the VF parameters and average RNFL thickness (AVG) were evaluated to differentiate PPG from POAG eyes. Repeated analysis with the best-performing test parameter, pattern standard deviation (PSD) (AUROC = 0.937) with a cutoff of 1.9 dB, showed that regression profiles in the POAG group with PSD >1.9 dB maintained a strong curvilinear RNFL/VF relationship, whereas those with PSD <1.9 dB exhibited a relationship almost indistinguishable from the PPG group. Evaluation of the structure-function relationship in normal subjects and those with PPG or POAG showed strong curvilinear regression in POAG eyes with PSD >1.9 dB and RNFL AVG thickness below 70 microm, whereas no correlation was detectable above these values.

Linear and Curvilinear Effects of Inbreeding on Production Traits for Walloon Holstein Cows

The nonlinear effects of inbreeding were studied by comparing linear and curvilinear regression models of phenotypic performances on inbreeding coefficients for production traits (milk, fat, and protein yields) of Holstein cows in their first lactation. Three different regression models (linear, quadratic, and cubic) were introduced separately into a single-trait, single-lactation, random regression test-day model. The significance of the different regression coefficients was studied based on a t-test after estimation of error variances and covariances associated with the different regression coefficients. All of the tested regression coefficients were significantly different from 0. The traditional regression coefficients of milk, fat, and protein yields on inbreeding were, respectively, −22.10, −1.10, and −0.72kg for Holstein cows in their first lactation. However, the estimates of 305-d production losses for various classes of animals based on inbreeding coefficients showed that the effect of inbreeding was not a linear function of the percentage of inbreeding. The 305-d milk yield loss profiles attributable to inbreeding, obtained by the various regression models, were different. However, for inbreeding coefficients between 0 and 10%, these differences were small.

Relationship between silver scurf levels on seed and progeny tubers from successive generations of potato seed

The level of silver scurf on potato seed tubers on successive generations of potato seed tubers and their progeny tubers was investigated during 3 years in the field. The objective was to determine the importance of seed-borne inoculum on silver scurf development on the subsequent progeny tubers. Silver scurf incidence and severity increased with each generation. Coefficients of determination for disease levels among generations were significant and ranged from 0.89 to 0.97, indicating that seed tuber source accounted for a large proportion of silver scurf on progeny seed tubers. Incidence and severity of silver scurf also increased with decreasing time periods between potato crops in the field. In a field near Paterson, WA, where potatoes had not been previously grown, the severity of silver scurf increased on progeny tubers of cvs Russet Norkotah, Ranger Russet, and Shepody as disease severity increased on seed tubers of successive generations. Disease severity index significantly increased as disease incidence increased. The relationship between the two was best described using a curvilinear regression model.

Kinetics of biodegradation of free gossypol by Candida tropicalis in solid-state fermentation

In the present work experiments were carried out to study the effect of free gossypol on the growth of Candida tropicalis ZAU-1, evaluate its ability in biodegrading free gossypol, analyze the time course of solid-state fermentation, and model the microbial growth by determining the kinetics of dry matter weight loss, total carbohydrate concentration and the free gossypol content during solid-state fermentation. Results showed that the biomass in inorganic salts glucose medium were unaffected by free gossypol at 500 and 1000 mg/l levels, compared with the control group ( p > 0.05); degradation of free gossypol reached 95.12% and 94.12%, respectively. A logistic equation ( R 2 = 0.9922), describing the growth model of C. tropicalis ZAU-1 was obtained, with the maximum values of u m and X m at 0.0970 h −1 and 21.8631% of dry matter weight loss, respectively. A good-fit curvilinear regression model was achieved to describe the change pattern of total carbohydrate concentration ( R 2 = 0.9910), and the biodegradation pattern of free gossypol ( R 2 = 0.9825). These models could be used to predict the fermentation course by C. tropicalis ZAU-1 under solid-state fermentation.

Comparative Study of Retinal Nerve Fiber Layer Measurement by StratusOCT and GDx VCC, II: Structure/Function Regression Analysis in Glaucoma

To evaluate the structure/function relationship between visual field sensitivity and retinal nerve fiber layer (RNFL) thickness measured by StratusOCT (Carl Zeiss Meditec, Inc., Dublin, CA) and GDx VCC (Laser Diagnostic Technologies, Inc., San Diego, CA). Eighty-nine subjects (27 who had healthy eyes, 21 who were glaucoma suspect, 41 who had glaucoma) were enrolled in this cross-sectional study. RNFL thickness was measured using the StratusOCT and the GDx VCC, and visual field (VF) was examined using the Humphrey VF analyzer. The relationship between RNFL thickness and VF sensitivity-expressed in terms of mean deviation (MD) in decibel (dB) scale, unlogged 1/lambert (L), and Advanced Glaucoma Intervention Study (AGIS) and Collaborative Initial Glaucoma Treatment Study (CIGTS) VF scores-were evaluated with linear and nonlinear regression models. Coefficient of determination (R(2)) was calculated, and regression models were compared using the Akaike information criterion and the F test. In plotting MD against RNFL thickness, curvilinear regression models demonstrated the best fit, whereas linear regression attained the best associations when VF sensitivity was expressed in 1/L. However, when healthy subjects were excluded from the analyses, the second-order polynomial was better than linear regression in describing the relation between 1/L and GDx VCC-measured RNFL thickness. Regression profiles between AGIS/CIGTS VF scores and RNFL thickness were best described in the linear and the first-order inverse models for GDx VCC and StratusOCT RNFL measurements, respectively. In general, StratusOCT RNFL measurements achieved higher associations with visual function in all the respective regression analyses than did GDx VCC. Description of structure/function relationships in glaucoma depends on the choice of perimetry scale, the type of RNFL measuring device, and the characteristics of the studied groups. The higher association with visual function in StratusOCT RNFL measurements compared with that in GDx VCC suggested optical coherence tomography might be a better approach for evaluating structure/function relationships. Curvilinear regression profiles found between StratusOCT RNFL thickness and MD/VF scores provide an explanation for those longitudinal observations, showing that VFs with higher AGIS/CIGTS VF scores or worse MD at baseline are at higher risk for deterioration. Regression analysis of the structure/function profile could provide important information in the assessment of the trend and pattern of glaucoma progression.

Growth and Curve Stabilization in Girls With Adolescent Idiopathic Scoliosis

A retrospective study of a cohort of 132 girls with adolescent idiopathic scoliosis (AIS). Evaluate the changes in height and scoliosis angle over time by the use of a mathematical model and determine a relationship between height and angle values in patients with AIS. The influence of growth on idiopathic scoliosis (IS) is still not fully understood. Although it has not been completely demonstrated, it is accepted that a relationship exists between height growth and curve progression, as well as that curve stabilization occurs when growth ends, but it has not yet been demonstrated whether both occur at the same time. One hundred thirty-two girls were included in a retrospective study. Inclusion criteria were: adolescent IS, Cobb angle > or =10 degrees , menarche age well documented, and follow-up of at least 2 years in 6-month controls. Main variables were: menarche age, height, Cobb angle, and treatment. Height and angle changes over the time were adjusted by several curvilinear regression models. Calculations were made of the gradient between each consecutive time point (first derivative function). Growth was considered as tending to stabilize when the function gradient changed its sign or was negligible. Height and Cobb angle correlation coefficients for repeated measures were estimated within patients for curves managed with observation and curves managed with a brace. Comparisons among these correlations were based on the Fisher-Z transformation. Height function gradient changed sign at 1 year postmenarche, and Cobb angle function gradient was negligible around menarche. There was a correlation between mean heights and mean angles, being higher for girls managed only with observation. When comparing mean heights in one semester with the mean heights of the previous one, there were statistically significant differences until 2.5 years postmenarche, although after the first year, these differences were clinically irrelevant. A significant increase for angle values was observed in the same period for the group of girls managed with observation and in the 6 months before menarche for the girls managed with a brace. A mathematical model was used to demonstrate when height and angle growths tend to stabilize (1 year after menarche for height values and at the time of menarche for angle values) in AIS. In the absence of a brace effect, a significant correlation between both growth rates was noted up until 2.5 years after menarche.

Long-term ozone exposure and ozone uptake of grapevines in open-top chambers

Ozone uptake of pot-grown grapevines, fruit yield and sugar concentrations in juice were studied over the course of 4 years. In three of the study years, plants were exposed to contrasting ozone levels in open-top chambers. Results showed that juice quality was more sensitive to ozone exposure than grape yield. Ozone-induced reductions in grape yield were best described by the level of ozone exposure in the 2 years prior to harvest, whereas sugar content of juice was mainly affected by current and previous year ozone exposure. Regression models were developed to facilitate quantitative assessments of the effects of ozone on yield and quality. The observed effects on grape yield and on carbohydrate accumulation in fruit were slightly better explained when a threshold of 6 nmol m −2 s −1 was applied to cumulative uptake of ozone (CUO6) compared with AOT40 exposure or other CUO thresholds. Curvilinear regression models were developed which explained up to 75% of variance of the datasets. These models indicate a critical level of 1.1 mmol m −2 ozone (CUO6 from June to September) to protect carbohydrate accumulation in fruit with ozone taken up by the plants over consecutive years. For grape yield the CUO6 threshold was 2.2 mmol m −2 ozone. Exceedance of this threshold would be expected to reduce grape yield and sugar translocation to fruit by 10% with substantial degree of uncertainty (95% confidence interval 1–19%). For individual years, the critical level could be increased to 2.3 mmol m −2 for carbohydrate accumulation and 3.5 mmol m −2 CUO6 for grape yield if the CUO6 of 0 in the subsequent year was not exceeded.

Curvilinear, geometric and phylogenetic modeling of basicranial flexion: is it adaptive, is it constrained?

Prior work has shown that the degree of basicranial flexion among primates is determined by relative brain size, with anatomically modern humans possibly having a less flexed basicranium than expected for their relative brain size. Basicranial flexion has also been suggested to be adaptive in that it maintains a spheroid brain shape, thereby minimizing connections between different parts of the brain. In addition, it has been argued that the degree of flexion might be constrained such that increases in relative brain size beyond that seen in Australopithecus africanus were accommodated by mechanisms other than basicranial flexion. These hypotheses were evaluated by collating an extensive data set on basicranial flexion and relative brain size in primates and other mammals. The data were analyzed using standard least squares regression, geometric and curvilinear modeling, and phylogenetically independent contrasts (PICs). Geometric modeling does not support the hypothesis that flexion is an adaptation that facilitates enlargement of a spheroid brain. Whether humans have a less flexed basicranium than expected for their relative brain size depends on the phylogenetic vantage point from which it is evaluated. They are as flexed as expected for a descendant of the last common ancestor of the Paranthropus–Homo clade, but their degree of flexion cannot be predicted from the basal hominoid node, even if their relative brain size is specified. Humans undoubtedly occupy an unusual part of morphospace in terms of basicranial flexion and relative brain size, but this does not mean that their degree of flexion is or is not constrained. Curvilinear regression models and standard linear regression models describe the relationship between flexion and relative brain size equally well. Hypotheses that the degree of flexion is or is not constrained cannot be discriminated at present. Consideration of recently published ontogenetic data in the context of the interspecific data for adults suggests that much of the variance in basicranial flexion can still be explained as a mechanical consequence of brain enlargement relative to basicranial length.

Influence of slight to moderate risk for birth hypoxia on acquisition of cognitive and language function in the preterm infant: A cross-sectional comparison with preterm-birth controls.

The cognitive and language performance of a group of 26 preterm-birth preschool and early school-age children with slight to moderate risk for perinatal hypoxia was compared with the performance of a preterm-birth comparison group of 26 children. Despite the relatively small discrepancy in degree of risk, the cognitive performance of the 2 groups diverged significantly. When data for children with known perinatal arterial pH were combined, a curvilinear (quadratic) regression model provided the best fit. Increasing acidosis was linearly related to decreases in cognitive skills, with the bend in the curve occurring well within the normal range of pH values. Hence, in the preterm infant, even minor risk for birth hypoxia may result in discernible deviation from the expected developmental trajectory.

Degradation of sodium monofluoroacetate (1080) in buried fox baits

The degradation of 1080 in buried Foxoff® baits was investigated. Baits were exposed to 5 different treatments: shelf-stored controls, prevailing weather, no rainfall, average weekly rainfall and twice average weekly rainfall. Curvilinear regression models were used to model the rates of decay of 1080. These predicted that under normal rainfall conditions, baits would become non-lethal to working dogs at 2.2 weeks and foxes at 2.8 weeks. Management implications of these results are discussed.

Between-group differences in the Iberian dung beetle species-area relationship (Coleoptera: Scarabaeidae)

Abstract Between-group α- and β-diversity differences were derived from species-area relationships fitted to field data. The accuracy of spatial richness variation predictions based on area size was also checked. The log-log model (log S = c + z log A ) was found to be the best-fit linear model, with slopes ( z ) ranging from 0.089 to 0.142. Between-group comparisons of z (slope) and q (intercept) parameters, using the S = q + cA z curvilinear regression model, corroborated early results, indicating a lower β-diversity (slope) for Scarabaeinae than for Geotrupinae and Aphodiinae. The latter group, probably more sensitive to environmental heterogeneity, should contribute more to species richness in large areas. α-Diversity is greater for Aphodiinae, more relevant to local diversity (1 km 2 ), than for Scarabaeinae and considerably greater for these two groups than for Geotrupinae. As earlier results show that the richness of a single dung pat is rather more a function of the Scarabaeinae species pool, richness on dung pat scales is probably due more to the between-dropping mobile Scarabaeinae, while Aphodiinae contribute mainly to local and regional pool richness. Nearly 88 % of the total richness variance is explained by area size. This percentage decreases to 37 % when the spatial structure of area size and species number are extracted. The corresponding figures for Scarabaeinae, Aphodiinae and Geotrupinae are 44, 22 and 31 %, respectively.

Effects of Mental Practice on Performance are Moderated by Cognitive Anxiety as Measured by the Sport Competition Anxiety Test

45 subjects were assessed for cognitive anxiety on the Sport Competition Anxiety Test. Two months later they observed a person performing a new motor task which required high cognitive processing to be performed well. After this observation, 22 subjects were randomly assigned to a Mental Practice and 23 to a Control group. The former performed a cognitive rehearsal of the task, whereas the latter did not. None practiced the task physically before being tested. Analysis of variance showed that both errors and performance time interacted significantly with Mental Practice versus Control group scores and scores on the Sport Competition Anxiety Test. Among subjects who practiced mentally, those scoring low on cognitive anxiety performed significantly better than subjects who scored high. Further, the relationship between test scores of cognitive anxiety and performance for the total sample was analysed by different curvilinear regression models. The cubic model fitted the data better and accounted for a greater percent of variance on error performance explained by anxiety test scores (R = .39) than the linear correlation (r = .25). This cubic model formed a polynomial relationship between cognitive anxiety test scores and error in performance.

Dissociation of [H+] from fatigue in human muscle detected by high time resolution 31P-NMR.

Previous in vivo studies of skeletal muscle fatigue have demonstrated significant relationships between the decline of muscular force and changes in muscle metabolism. However, these studies performed measurements over relatively long time intervals or during steady state exercise, thereby obscuring rapid metabolic changes occurring at the onset of exercise and recovery. To overcome these limitations, fatigue of human calf musculature during sustained isometric foot plantar flexion was quantified continuously as the decline in maximal voluntary contraction force (MVC), while concentrations of phosphocreatine (PCr), inorganic phosphate (Pi), intracellular free hydrogen ion (H+), and monovalent phosphate (H2PO4-) were simultaneously measured at 2-second intervals by 31P nuclear magnetic resonance. The first major finding was that [H+], which has been thought to be a mediator of muscle fatigue, actually declined during the first 10 seconds of exercise when force was declining and rose immediately postexercise, when force partially recovered. Second, the correlations of [H+], [H2PO4-] and Pi with MVC during the first minute of exercise were determined to be curvilinear and not linear as previously suggested. Furthermore, using either a linear or curvilinear regression model, [H2PO4-] and Pi demonstrated a closer correlation to MVC than [H+] during the first minute of exercise. Thus, these results reveal nuances in the relationships of MVC to metabolites previously undetected by low time-resolution measurements. These findings suggest that during sustained isometric exercise, rising [H+] is not likely to be the sole mechanism of muscle fatigue and are consistent with the view that a rise of Pi or [H2PO4-] is a major causation factor in force reduction.

Analysis of Early Mortality of Douglas-Fir Seedlings in Postharvest Plantings in Northwestern California

Abstract A life table study was conducted of injury and mortality of Douglas-fir seedlings in 44 plantations on public and private lands in Humboldt and Del Norte Counties, CA. Field examinations were made three times each year for the first 3 years following planting. Apparent and real mortalities were compiled in life table format for nine time intervals. Seedling mortality per time interval in the different ownerships varied from 0 to 35.0% in real percentage terms, and totalled 14.0 to 70.7%. Highest mean mortality occurred in intervals 2 and 3 (late spring to fall, first year). Trend analyses, using a curvilinear regression model, showed significant differences in rates of mortality among ownerships. The performance of containerized and bareroot seedlings was very similar. Primary conclusions from the life table analyses, using a two-step modeling process, were (1) variations and covariations in mortality associated with intervals 2 and 3 were most important in determining total variation in survival in the 3-year period; intervals 1, 4, 7, and 9 also were critical in some instances; (2) planting- and weather-related factors were most important in intervals 1-4; disease, insects, and animals, alone and jointly, were more important in the second and third years; (3) substantial covariations in mortality among the time intervals indicated carryover effects; covariations among certain categories of causal factors indicated strong positive or negative interactions that either added to or dampened the direct effects of these factors. For. Sci. 37(3):802-826.