Comparative morphometry of the mammalian brain: estimates of cerebral volumes and cortical surface areas obtained from macroscopic slices

Abstract

Stereological methods for obtaining unbiased estimates of brain volumes and surface areas are illustrated on fixed brains (cerebrum only) from mammals varying in body weight between 2 kg (cat) and 550 kg (ox). Brain sampling was designed so that Cavalieri estimates of volumes (derived via point counting) could be combined with vertical sectioning estimates of surface areas (via intersection counting). Total volumes, cortical volumes and cortical surface areas were calculated. Each cerebral hemisphere was cut into 3 slabs. Each slab was cut further into macroscopic, parallel vertical slices having a uniform random start position. The term vertical as used here signifies orthogonal to the medial aspect of the hemisphere. The direction of vertical slices also had a random start but was varied systematically across the slabs. Test lattices bearing test points and cycloid test lines were superimposed on vertical slices. Fixed volumes and surfaces were corrected for shrinkage effects. A worked example of the calculation sequence is provided. The experimental design was flexible. Brains of different sizes could be analysed by simply altering the distance between slices and the size of the test lattice. Analyses took 30-45 minutes per hemisphere. Whilst volumes and surfaces increased with body weight, specific values declined. Thus, specific surfaces for the cortex fell from 25 cm2/kg (cat) to 2 cm2/kg or less (pig and ox).