Statistics for fixed effectsand their interaction terms were also determined by Waldtype III F test. Post-hoc pair-wise multiple comparisons wereperformed using Tukey’s honestly significant difference (HSD)test at P<0.05 for fixed effects and their interaction terms,while Student’s t test was used to compare only two groupsof means when necessary. These statistical analyses were performedusing the procedure MIXED in the Statistical AnalysisSystem v.9.2 (SAS Institute Inc., Cary, NC). To explain theimpact of glucose concentration on blastospore yield, empiricaldata were fitted to a logistic growth model (sigmoidal with fourparameters) according to the equation: y=y0+(a/(1+exp(−(x–x0)/b))), where y is the concentration (blastospore mL−1), x isthe glucose concentration (g L−1), and a, b, y0, and x0 areconstants estimated by the interactive analysis performed withthe software SigmaPlot 11.0 (Systat Software Inc., San Jose,CA,USA). Proportionmortality data for whitefly nymphs challengedwith strain ESALQ1432 were fitted to a logistic modelin PROBIT procedure of SAS to compute the median lethalconcentration (LC50) for each production media treatment.Mortalities for fungal treatments were automatically correctedfor control mortality in this analysis. The potency ratio test wasused to statistically compare the LC50 values at a 5 % probability(Wheeler et al. 2007). Concentration-mortality nonlinearregression curves from different glucose concentrations werestatistically compared through the sum-of-square reduction test,in which comparisons of nested models were performed by
การแปล กรุณารอสักครู่..
