278 



A. BOYDE 



they diverge to a greater or lesser degree from this perpendicularity rule. However, 

 they always make a large angle with the former group. Thus changes in orientation 

 of the crystallites (the "prism sheaths") are determined by changes in orientation of 

 the mineralising front. 



The fact that the crystallites in developing enamel tend to be oriented per- 

 pendicular to its mineralising front may be caused by the peculiar growth habit of 

 the crystallites themselves: they grow as very long, thin, flattened hexagons. Mutual 

 interaction between adjacent crystallites might line them up parallel to each other 

 and perpendicular to the mineralising front, or it may just be that only those crystal- 

 lites which begin to grow in the right direction can continue to grow, since any 

 growing at a significant angle to the perpendicular to the surface of the environment 

 in which they are growing must butt up against another crystal. 



Where there are no changes in orientation of the mineralising front all the crystal- 

 lites are parallel and perpendicular to it and there are no prisms. This situation 

 prevails at the commencement of amelogenesis at the enamel-dentine junction before 

 the ameloblasts acquire their Tomes' processes and at the end of amelogenesis when 

 they lose them again during the formation of the true surface-zone enamel. 



A close analysis of the shape of the surface of developing enamel reveals that the 

 cervical and lateral and cuspal regions of the depressions are not equivalent. The cell 

 membrane of Tomes' process must slide past the surface in the latter regions. It seems 

 probable that this is connected with the deficient growth of these surfaces and that it 

 may even cause the deviation of the crystallites which grow in these surfaces from 

 the generalisation that they should be perpendicular to the surface. The crystallites 

 which grow in these cuspal and lateral surfaces belong to the "interprismatic regions" 

 of Pattern 1 and 2 enamels (see Fig. 1) or to the cervical extension or "winged 

 Process" of Pattern 3 prisms. 



Fig. 1. a) Pattern 1 enamel, Cuspal enamel, Cheiroptera, Insectivora, Sirenia, Odontoceti (Cetacea), Lemuroidea 

 (Primates), b) Pattern 2 enamel, Lateral enamel, Ungulata, Marsupialia, Lagomorpha, Cercopithecoidea (Pri- 

 mates) (also in man!), c) Pattern 3 enamel, Carnivora, Proboscidea, Homo sapiens. — Prism cross-sectional 

 outlines: The curved heavier lines represent (sectioned) prism boundary planes of abrupt change in crystallite 

 orientation. There is only a gradual change in crystallite orientation between any two points which can be 

 connected by a line which does not pass through sudi a boundary plane ("prism sheath"). — Secretory terri- 

 tories of ameloblasts: The lighter hexagonal outlines delineate the amounts of enamel produced by separate, 

 single ameloblasts. The dotted areas represent what are conventionally referred to as "prisms"; via this defini- 

 tion it will be seen that "interprismatic regions" can be distinguished in Patterns 1 and 2, and depending on 

 the extent of the prism sheath, perhaps in Pattern 3 



The recent studies of stereo electron-micrographs of replicas of the surface of 

 developing enamel have provided additional data which may help to elucidate two 

 unexplained observations made by Boyde (1964). There it was noted that the crystal- 

 lites in ungulate (Pattern 2) enamel prisms diverge less from the mean axis of the 



