An original method for fractionating and preparing isolated crystals of homogeneous size was developed. It was demonstrated that enamel apatite crystals are at least 100 micron long. The flexibility of the very long crystallites was demonstrated. Crystal curvatures, accounting for the irregular course of the prisms through the enamel thickness, were visualized and measured. It was shown that in the deep forming enamel layer, lateral branches may grow out of the crystals and crystal fusing often occurs, inducing the crystallites to assume pyramidal shapes with their wide bases pointing toward the dentino-enamel junction and one or two tops toward Tomes' processes. During the maturation process, the two tops of the still immature crystals also fuse so that the mature crystals acquire a rodlike aspect, with parallel faces and steplike graduations along the c axis, allowing a close contact between the crystals. These results support the hypothesis that the crystallites would be continuous from the dentino-enamel junction to the surface.

High resolution electron microscopy has provided precise measurements of human enamel crystallites from the first stages of enamel development to the last stage of maturation, through counting of lattice planes and statistical analysis. Thickness and width measurements of human enamel crystallities in their mature stage are 263 A (21.9 standard error) and 683 A (134 standard error), respectively. The growth curve of enamel crystallites was drawn, the increase in the number of lattice planes in the course of enamel development was demonstrated. Fusions and dislocations related to the maturation process were visualized and discussed. Difficulties linked to the use of ultrastructural techniques were mentioned. The question of the organic stroma of the crystals remained unsettled.

In vitro caries-like lesion formation in F-containing tooth enamel (shark) showed a lower depth of penetration compared to that in human enamel (42 vs. 100%). After longer periods of exposure, the depth of penetration in shark enamel is only 33% relative to that in human enamel. The extent of dissolution in acetate buffer for powdered human enamel and for synthetic F-free calcium-carbonated-apatites was greater than for powdered shark enamel and for synthetic F-containing apatites. These results suggest that the cariostatic action of fluoride may be explained in terms of a combination of two effects: (a) the stabilizing effect against acid dissolution (caries) when fluoride is incorporated in the tooth mineral (apatite), and (b) a greater degree of remineralization (repair) in the presence of F solution.

Identification and indexing of lattice planes through high-resolution transmission electron microscopy provided precise measurements of intertubular crystals of arrested caries in dentine compared to sound dentine. In arrested caries, mean crystal width was 310 Å (SD 44) with a range from 200 to 400 Å. Mean thickness was 86 A (SD 19.3) with a range from 57 to 130 Å. In sound dentine, mean width was not very different at 296 Å (SD 39) with a similar range from 200 to 400Å. Mean thickness was lower: 31.6 Å (SD 5.4) with a range from 24 to 41 Å. A growing process due to thickening of the crystallites was pointed out. The coexistence of whitlockite and apatite within the large crystals of intratubular dentine in arrested caries was shown through high-resolution electron microscopy and micro-electron diffraction. Two stages of arrested caries could be found at the crystal level: firstly, a precipitate of the material dissolved at the very beginning of the carious process, and secondly, remineralization of decalcified dentine by growth of residual crystals (intertubular dentine) and by a maturating process of whitlockite towards apatite (intratubular crystals).

No abstract