The effect of microscope graticule size and counting rules on the estimation of airborne fibre numbers using the membrane filter technique

The membrane filter method for assessing airborne fibre concentrations is based on the use of a phase-contrast microscope to count the number of fibres on a small proportion of the filter surface. Previous investigations showed that higher estimates of fibre density were obtained as the counting area was reduced to ca 120 μm dia. by the use of an eyepiece graticule, but no further effect was noted below this size. The effect was attributed to 'human error' and could be eliminated by 'rigorous' counting. This paper has examined the graticule effect in more detail when counting 'rigorously', using a wider range of graticule sizes and two counting rules (ARC and NIOSH) for fibres crossing the graticule boundary. The results have confirmed the absence of a significant effect with graticules larger than ca 100 μm dia. but noted an increasing bias when using the ARC rules as the graticule size was reduced from 100 to 20 μm. This bias appears to be caused by a tendency of the observer to count more fibres crossing the upper section of the graticule than the lower. The results suggest that bias can be avoided by using a graticule of dia. 100-120 μm and/or using the NIOSH method of counting each fibre and within the graticule as half a fibre.