Introduction

Graton & Fraser (1935) studied the relationship between permeability, porosity, and grain packing in sedimentary rocks according to the distribution of voids of spheres on a random plane section. Later Krumbein (1935) experimented with embedding lead shot of the same diameter in sealing wax and grounding to produce a polished surface. An average observed radius was computed to equal 0.763 of the actual radius. Friedman (1958) was the first to determine a sieve size distribution from thin-section data. A plot of thin-section quartile parameters against sieve quartile parameters indicates a linear relationship between the two methods. The mean of a converted thin-section size distribution equals 0.381 + 0.90 times the mean of an observed thin-section size distribution (Textoris, 1971).

For a representative grain size distribution, the thin section should be cut from a rock specimen so as to represent the macroscopic characteristics of the sedimentary rock in outcrop. Generally an appropriate cutting direction is perpendicular to the depositional laminations. According to Friedman (1958), 300 grains should be counted per thin-section grain size analysis. It is also necessary that measurments should be taken along the long-axis direction. Care must be taken so as to discriminate the original particle grain, avoiding interference from grain dissolution or overgrowths due to diagenesis. Generally, the original grain shape can be determined based on residual traces of clay and/or iron oxide particles along boundaries between original grains and secondary growths.