A geometric model for estimating villous surface area in rat small bowel is justified by unbiased estimates obtained using vertical sections.

Sampling schemes developed for use with a geometric model of rat small bowel are tested against a design-based scheme (vertical sectioning with cycloid test lines) which offers unbiased estimates of surface amplifications due to villi. The model-based methods comprise transverse and longitudinal sectioning coupled with putative correction factors. Comparisons are based on proximal, middle and distal segments of six small bowels. Transverse and longitudinal sections through the same segments of each animal were analysed by conventional intersection counting (using straight test lines). Appropriate intersection ratios were multiplied by their respective correction factors in order to calculate surface amplifications. Longitudinal sections were employed further as vertical sections and intersections were counted with cycloid arcs to obtain unbiased estimates of surface amplifications. Both model-based schemes (transverse and longitudinal) gave group mean values similar to those obtained by vertical sectioning. Therefore, the use of a geometric model in past studies on rat small bowel can now be justified on grounds of negligible bias.