Porcine adipose tissue slices were fixed with osmium tetroxide and cells released by treatment with urea. Cell size and number were determined by an instrumental method using a particle counter. Storage of adipose tissue samples as frozen slices or in osmium, in post-osmium saline, in urea or in post-urea Triton all tended to produce less acceptable results than obtained with fresh tissue slices. Various storage conditions either tended to diminish cell size, to produce small particles or to cause aggregation. Repeatabilities of cell number, cell diameter and cell volume from multiple samples obtained from one anatomical location within an animal (perfect repeatability = 1) were .39, .52 and .65, respectively. Repeatabilities of instrument determinations were greater than .98 for cell number, cell diameter and cell volume. Cell number may be estimated from particle counts or indirectly from average size. Particle count number and that calculated from mean cell diameter were similar, whereas cell numbers estimated from mean cell volumes were smaller. Different adipose tissue depots and backfat layers had divergent cell size, making extrapolation to whole-animal cell number complex.