"Uterine contents were examined to determine the number of implantation sites, resorptions, dead fetuses and live fetuses. Uteri which had no visible implantation sites were stained with ammonium sulfide (10%) to detect very early resorptions (Salewski. 1964)."

"General Linear Models (GLM) procedures were applied for the analyses of variance (ANOVA) of maternal and fetal parameters (SAS Institute,

1989a,b; 1990a,b,c). Prior to GLM analysis, an arcsine-square root transformation was performed on all litter-derived percentage data (Snedecor and Cochran, 1967) and Bartlett's test for homogeneity of variance was performed on all data to be analyzed by ANOVA (Winer, 1962). GLM analysis determined the significance of dose-response relationships and the significance of dose effects, replicate effects and dose x replicate interactions. When ANOVA revealed a significant (p<0.05) dose effect, Williams' and Dunnett's Multiple Comparison Tests (Williams, 1971; 1972; Dunnett, 1955;

1964) compared NAP-exposed to control groups. One-tailed tests were used for all pairwise comparisons except maternal body and organ weights, maternal food and water consumption, fetal body weight and percent males per litter. Nonsignificant (p>0.05) dose x replicate effects on selected fetal parametric measures wire considered justification for pooling data across replicates for nonparametric analysis on related measures. When a significant (p<0.05) dose x replicate interaction occurred, the data for that endpoint and for any related nominal scale data were analyzed separately for dose effects within each replicate in the study, as well as for all replicates combined."

Nonlive Implants / Litter

Endpoint Details

Dataset

Plot