Probably the best approach is to replicate, on a small scale, the mechanism that is likely to be present in the preparation of the material for transit. The most common feature is that of forming a repose pile in the loading of containers. Pouring the material through a funnel onto a pile, and comparing a sample scraped from the periphery with one taken from the centre will quickly indicate the extent to which segregation may occur. A more quantitative tets is to rotate a sample in an inclined drum, say length equals diameter at 15 degrees axial inclination. Fill with sufficient material to reach near the rim when rotating. After about 30 rotations a sample should be taken of that quanitiy filling half way down the cylinder and comparing its particles size distribution with the remaining quantity.

the 'User Guide to Segregation', published by the British Materials Handling Board, gives much background and useful tips for dealing with this form of problem. Soe technical publications from www.ajax.co.uk are also available.

Lyn Bates ■