MEASUREMENT AND INTERPRETATION OF DENTIFRICE ABRASIVENESS z[09 mechanical action and partly by ion exchange, will always correspond to the amount that would have entered the slurry in the absence of ion exchange processes. MR. A. I. FREW: What is the effect of the viscosity of the toothpaste sample on the chances of trapping particles between brush fibre and tooth surface, and hence the degree of abrasion recorded? DR. WRmHT: The viscosity of the dentifrice slurry does probably play some part in determining the form of the wear-concentration curves, especially towards the higher concentration levels. In one case, a commercial dentifrice product was observed to show a marked falling off in the level of abrasiveness when the dentifrice slurry con- centration exceeded 80%. This, we believe, was due to the slurry being physically displaced from the dental tissue, thus reducing the concentration of abrasive matter. It is possible that the viscosity of the dentifrice has some effect at lower concentration levels of the dentifrice, but it might be necessary to make use of elastohydrodynamic theory to calculate the magnitude of the effect, since elastic deformation of the soft nylon fibres and the dental tissue cannot be neglected. MR. N. J. VaN ABBg: You have not mentioned the effect of hydration on the brush fibre, which can possibly have quite a marked effect, but on the other hand, you have stressed conditioning of the tooth surface. However, there is no sign that in vivo the equilibrium condition of abrasion on the tooth is ever reached hence it seems to me totally unrealistic to produce the equilibrium surface before testing. I would suggest that it is better to start with an optically flat surface on every brushing run and to resurface between brushings. Another factor that should be taken into account is the aggregate size within the dentifrice, rather than particle size of the polishing agent. I would like to stress that the tracer method in common with weight loss methods only measures average wear on the other hand, a profile method shows the detailed pattern of wear. In Fig. 8 wear is shown by the area under the curve. The average depth of abrasion due to dicalcium phosphate is nearly 50% more than that due to the chalk dentifrice but the primary texture is different and this may be the critical factor with regard to conditioning. If the specimens had been conditioned to equilibrium initially, the relative depths of abrasion might well have been reversed, but this would not necessarily correspond to the natural effects of tooth-brushing. You have mentioned the grittiness test using the nickel coin and glass slide I have tried to replicate some of your results but mine have differed considerably from yours it seems to me that there are nmnerous problems to solve before this test can be obtained as a grittiness standard. DR. WRmHT: We make no great claims for the coin and glass slide technique, although it is a very simple and yet sensitive method for detecting the presence of small concentrations of gritty matter in dentifrices. Despite the fact that it forms the basis of the U.S. Federal Standard Test for Dentifrice Abrasiveness, it is not a satis- factory way to measure the overall abrasivehess of a product. As a grittiness test it clearly works best whenever the gritty material is present as a few large particles, but as the particles diminish in size, it becomes more difficult to decide whether the glass slide has been scratched or finely polished. By its nature the test will be subjective. The present visual criterion meets the requirement that the

410 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS gritty particles shall not disfigure an enamel surface (approx. as hard as soda glass), but not all are agreed upon the levels of disfigurement which are acceptable and, before we can even attempt this, it is necessary to devise some measurement technique whereby one can measure the severity and frequency of the scratch marks. Profilo- meter type instruments can be used to record the nature of the surface damage, but such traces still require interpretation. I agree that the radiotracer technique cannot provide any information about the surface texture of the dental tissue, indeed it has never been claimed to do so. Such information can readily be obtained from a supplementary experiment using either a profilometer or optical reflection technique. One assumes that in designing a new dentifrice formulation, the first requirement that must be met is that it does not cause undue damage to the main dental tissues. When this feature has been settled, one can then proceed to examine the quality of the surface texture that it produces. If both of these factors can be examined simultaneously so much the better. The question of particle aggregation is clearly important and must have some effect on the form of the wear-dentifrice concentration curves. In our theoretical model, the capture probability will now correspond to the aggregated particle size and not to the individual dimensions of the components. This will mean that the wear- concentration curves will tend towards those appropriate to large single particles. On the other hand, the aggregates will break up under the normal and sheer stresses imposed by the loaded fibres, so that the surface finish produced on the dental tissue is likely to be superior to that which would normally be observed with large single particles of abrasive. When considering the problem of the initial surface preparation of the dental tissues, one has to bear in mind that the range of surfaces likely to be encountered in vivo is so wide that one could always argue against the choice of any particular surfacing technique. Whenever one abrades a surface, the topography of which is 'foreign' to that which would be appropriate to the abrasive employed, the wear rate will slowly change until the equilibrium surface texture for that abrasive is obtained. In our paper we refer to this as the conditioning process. If one includes this portion of the wear process in an experimental run, one finds that the average wear is a function of the time of the experiment. This is clearly undesirable unless the wear process closely replicates that which occurs in vivo and this would seem impossible to attain at the present moment. It, therefore, appears better to employ an abrasively conditioned surface and to measure the intrinsic abrasivehess of the dentifrice. Such a measurement is independent of the length of test and enables a range of different measurement techniques, perhaps of different sensitivity, to be employed for evalu- ation purposes. In the case of the tracer technique, this procedure also allows the surface to reach chemical equilibrium with its environment. DR. B. 1•. Puo•: We have examined the effect of "vane" size on brushhead and found that vane dimension affects the concentration curves quite markedly. Do you believe that your hypothesis regarding combinations of small and large particles is therefore only theoretical? DR. WmGI•T: I think you are aware that our equipment is fitted with stirring vanes that closely conform to the internal dimensions of the trough. These stir the dentifrice slurry vigorously and are, we believe, completely effective in preventing sedimen- tation. We have so far only employed two designs, both of which yielded essentially the same results. It is, however, difficult to assess the true efficacy of any vane design as