QUANTITATIVE ANALYSIS OF SECTIONED BIOLOGICAL MATERIAL

Publisher Website
Google Scholar
Abstract
Some sources of significant error in quantitative cytochemical work are discussed and methods for their control and elimination described. These recommendations are based on a study of distortion and thickness variation in histological sections of embedded biological tissues. To achieve the objectives of the investigation, it proved necessary to manufacture special microtome knives having standardized properties and to develop an optico-mechanical procedure for topographic mapping of section thickness. Altogether 10,500 thickness determinations were made on 140 sections, there being 10 similar sections in each of 14 series which differed with respect to tissue type, embedding substance, microtome drive and feed setting. The results disclosed that, even under ideal sectioning conditions and irrespective of tissue type or embedding substance, considerable variations in thickness must be anticipated, particularly between sections in a series. The mean coefficient of variation for the deviations between and within sections nominally 5 µ thick was 28% and 5% respectively. A method has been devised whereby the thickness of a duplicate of a biological section can be used as an estimate of the mean thickness of that section, thus enabling the magnitude of the intersectional thickness variation in a series to be assessed and corrected for. In this method the microtome knife cuts simultaneously with the section of biological material a plastic duplicate thereof. Such plastic duplicates are optically homogeneous and their thickness can readily be determined in an interference microscope. The method has been adapted for embedded as well as for frozen tissue. Its value has been checked by making a total of 714 thickness determinations on 51 tissue sections and their duplicates. Analysis of regression showed that when the nominal section thickness indicated by the microtome feed setting was used as an estimate of mean section thickness the coefficient of variation was 13% for paraffin sections and 45% for frozen sections with a nominal thickness of 5 µ. The corresponding figures when thickness duplicates were used as estimates of mean section thickness amounted to merely 7% and 14% respectively.

This publication has 0 references indexed in Scilit: