JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS people, probably because of its apparent realism, and the lack of apparatus or other facilities needed. There appear to be two basic ways of conducting drop tests:- Cumulative drop method In this method, each bottle is dropped a number of times, each time from a greater height until a break occurs. The height at which the bottle breaks is recorded. After, say, 50 bottles have been broken, the results are shown on arithmetical probability paper. The cumulative percentage broken up to and including each height is plotted against each height. A straight line drawn through the results will show with fair acccuracy the level at which 50% break and with less accuracy the level at which say, 10% break. Table I shows the results obtained on 50 small PVC bottles. Table I The bottles were dropped from 0.Sin. If they did not break they were dropped from 0.6m, and so on, increasing by 0.Sin at a time until they failed. Only the height at which the bottles failed was recorded. Drop hr. 0.3111 0.6m 0.9m 1.2m 1 x x x x x x x x x x x x x x x x x x x x x x x Totals 3 10 14 13 10 Cumulative % 6% 26% 54% 80% 100% It is obvious that this test makes the assumption that the bottles are undamaged by dropping, until they are dropped from a height sufficient to break them. For bottles made of perfectly elastic, brittle materials this might be believable, since the matehal will recover after dropping, unless the elastic limit has been passed, when it will break. In practice, observation will show that with toughened polystyrene and some PVC compounds at

DROP TESTING OF PLASTIC CONTAINERS 5 least, dropping from heights insufficient to break them will cause defects to appear. Such defects are bruising of the surface, dents, and stress marks. It is felt that this uncertainty outweighs the merits of the technique. The principal merit is that a result is obtained for each bottle. Other tech- niques which only break a proportion of the bottles, do not, on the face of it, make such good use of small numbers of samples. I prefer the "staircase method", described below, for small samples. Single drop method When using a single drop method a variety of statistical techniques can be used. Two of these demonstrate two basically different approaches. Staircase method This is a technique designed to concentrate the testing around the level at which 50}/0 of the samples fail. To conduct a drop test by this method one chooses an initial drop height and an increment. Thus, one may decide to start at lm, and move in steps of «m. This choice is usually influenced by past experience of the type of container being tested. The test is con- ducted using the following rule:- If a sample fails, drop the next sample at that height less the increment if a sample does not fail, drop the next sample at that height plus the increment. Table II is built up as follows:- Table II 2.1m x 1.8m 0 x x x 1.Sm 0 0 x X X 0 X x X 1.2m 0 X X 0 x 0 0 X 0 x 0 0 0.9m 0 X 0 0 0 0 0 0.6m 0 The entries are made from left to right, in order as the test proceeds, O----pass X -•failure In order to calculate the level at which 50%/0 of the samples fail, the results are tabulated as follows. First it is determined whether there were fewer