48 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS based. It may, of course, be possible after the process has been running for a period to determine accurately all the statistical data necessary in order to recalculate on that basis. MR. C. Puo}t: Are there any rules for saying what excess xvill be tolerated? ThE LECTURV-R: There are no rules for deciding what excess will be tolerated. The Act merely states that one has to be able to justify over-filling as being necessary in order to prevent under-filling. Presumably a calculation similar to that used for finding the minimum fill would suffice, and I believe that Weights and Measures officers tend to take over-filling much less seriously than under-filling. MR. J. M. BLAX•WA¾: Do you really mean that the check weight we should use in controlling the manufacture and production should be in terms of 50 g, 20 g, 10 g and 5 g weights? Tn• L•CTURER: Yes. The Weights and Measures Act (1963) Chapter 31 Schedule 1 tells you exactly what weights and what measures one is permitted to use, and one has to work to those particular weights, which must have been Board of Trade stamped, and only certain specific denominations may be used. DR. S. M. Bu•: You are correct in assuming that the capacity tolerances which glass container manufacturers adopt are traditional and I certainly feel that they can be adjusted. This has arisen very largely because prior to the introduction of the ¾Veights and Measures Act the only legislation that had to be dealt with was concerned with Customs and Excise, particularly on liquors. In that case it was usual with one single product which went into containers of a variety of shapes to take the worst shape, which gave the largest capacity variation in manufacture, and in effect adopt that. MR. L. GATL•: Supporting this question of bottle tolerances I would like to mention a particular bottle where we had two or more years of elaborate statistical data on its dimensions. We then had a delivery which was completely different and would not take the quantity previously specified, although the bottles were within the maker's quoted tolerances. The maker concerned has now agreed to narrow his tolerances. DR. A. W. Mm•To•: Does the Act actually say that you must check-weigh the goods as you are producing them? Tn• L•CtUR•R: The Act does not put an onus on you to do any check-weighing. Whether you carry out check-weighing is entirely your own responsibility but, if, as in most cases, you do check-weigh then it is laid down in the Act that the correct denominations of Board of Trade stamped weights must be used. The equipment used for weighing, such as balances, scales, etc., must also be to an approved pattern and Board of Trade stamped. This also applies to any equipment used for obtaining tare weights of components in departments other than the filling floor.

Ceramic colour standards The British ceramic industry, in collaboration with the British Ceramic Research Association, the National Physical Laboratory and the Society of Dyers and Colourists has produced a set of 12 coloured tiles for the calibration and control of colour measuring instruments. Such tiles will be o[ considerable value wherever colour measurement is carried out. For example they can be used to indicate how accurate a colorimeter is, i.e. how closely it determines XYZ values or how precise it is, i.e. how reproducible measurements are they can reveal the agreement between different instruments of either the same or different models they can show when an instrument is no longer performing correctly. The choice of the twelve was governed by a number of factors. Perman- ence, temperature insensitivity, and absence of fluorescence in the 300-700 nm region, were essential from the tiles which possessed these qualities, 12 were chosen to cover a reasonably wide region of colour space including three neutral greys and in order to provide the maximum value in diagnosing instrumental faults, the shapes of the reflectance curves were taken into consideration as well as the colour of the tile. One batch of tiles was made for each colour under the most stringent control possible, and 1,000 tiles for general sale were selected from the batch by visual inspection to ensure freedom from significant blemishes. From these 1,000, 50 sets were specially selected for sale as NPL-calibrated standards, the most representative being designated the Master Set. Calibration data of the Master Set will be determined at the National Physical Laboratory as accurately as possible and this will be supplied with each uncalibrated set to indicate approximate values. It is believed that the variations between the selected tiles of each colour will be small enough to be ignored for many purposes. Where greater accuracy is required the NPL-calibrated standards should be used. Uncalibrated sets of ceramic colour standards can be obtained from:- The British Ceramic Tile Council, Federation House, Station Road, Stoke-on-Trent, England. 49