42 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS films, however, the method •vould require some modification to yield reproducible results. All slow' rate testing is at present performed at extension rates of 12 inches per minute or less, depending upon the type of material tested and its thickness. Ordinary tensile strength and elongation at break measure- ments are made on a tensometer with a strip chart recorder. The measure- ment of ultimate tensile strength and elongation at break is useful when dealing with blown layflat materials, since comparison between machine and transverse direction values gives a good indication of the balance of properties within the film. Slow rate testing of seals can also be made with a tensometer. The method is especially useful when very thick films are sealed together. The procedure for assessing seal strength is very simple. Strips of • to 1 inch width are cut from the test film and similar strips are also cut across seals made with pieces from the test film. The strips are pulled apart at a standard speed in the tensometer and the breaking loads recorded. The seal strength of a seal is then defined as the ratio of the breaking load of the seal strip to the breaking load of the strip from the unsealed films. The coefficient of friction of laminates which is sometimes of importance, can be measured by conventional methods. It is usual to run a standard weighted sled, covered with the laminate, over the required surface at the rate of 6 inches per minute and to measure the horizontal force applied to the sled. (b) Permeability Water Vapour Permeability Rates of films are determined in tile labo- atory by the method given in British Standard 3177: 1959. In this method, a circular disc of the specimen under test is placed in a shallow aluminium dish designed so that the periphery of the disc rests on a shelf. The dish, which has been previously filled with desiccant, normally calcium chloride, is then sealed by running molten wax around a template which is positioned centrally above the test specimen. A number of dishes are so prepared, weighed and stored under controlled conditions of temperature and humidity. At suitable intervals the dishes are weighed and the increase in weight plotted against time. From this curve the water vapour per- meability rate is calculated. A similar arrangement is used for deter- mining permeability of vapours except that in this case the solvent is placed in the container and weight losses observed. Over the normal range of thickness of plastic films the water vapour permeability rate is approxi- mately inversely proportional to the thickness but above 1,000 gauge, i.e. 0.01 inch thickness, this ratio breaks down and it is not possible to forecast rates for thick sections on the results obtained from films.

LABORATORY EVALUATION OF NEW PACKAGING MATERIALS In determining gas permeabilities various methods are in use. One of tlxe simplest is to use the film as a diaphragm in a chamber, one side of which is connected to a gas supply and the other side to a mercury man- ometer and a vacuum pump. A vacuum is drawn to a pressure of about 0.2 mm and the level of mercury in the manometer capillary noted. The gas is admitted to the system at a pressure of 1 atmosphere and the level of mercury noted at intervals. From the increase in pressure, as shown by the drop in height of the mercury the volume of gas passed through the films can be determined. This volume is plotted against time and the gas permeability rate calculated. 2. PHYSICAL TESTING OF CONTAINERS (a) Flexible sachets. Testing of this form of pack first involves heat sealing experiments. Many users appear to carry this out with either a crude hot bar sealer, or one fitted with a simple type of energy regulator where no correlation between thermostat setting and temperature at the sealing jaws has been made. Attempts are even made to seal sachets by a bead seal with a Bunsen burner. There are two machines which may be employed in the laboratory to carry out worthwhile heat sealing experiments one is a simple hot bar sealer and the other is the impulse sealer. The machine operating on the impulse principle is similar in construction and appearance to a hot bar sealer, but differs from it in one important aspect. When the jaws meet, with the material to be sealed in between,. a high temperature is produced by a surge of high current electricity through a wire. The seal made is then cooled, still under the pressure of the jaws. Usually the only variations which can be made are in the relative times of the heating and cooling cycle. In the conventional hot bar sealers, three factors affect the strength of the seal. (1) The temperature of the heated jaw. (2) The pressure which holds the heated jaw on to the material being sealed. (3) The time of dwell of the heated jaw. To be absolutely certain that the best seal has been produced the effect of each of these variables must be considered. In practice we have found that it is useful in the laboratory to maintain an air pressure of about 16 p.s.i.g. on the hot bar and then vary the time of dwell at each temperature. The efficiency of the sea/can be tested with a filled sachet, by applying hydraulic pressure on the sachet until it bursts giving a static bursting strength. Alternatively, it may be tested for resistance to impact forces.