206 E. JOHANSEX 



By a slow exchange phenomenon, cr\'stanites in regions of higher 

 permeabihtv might pick up fluoride from the surrounding fluid and 

 possibly lose carbonate. Such a change in composition could account 

 for the greater resistance to dissolution observed in crystallites in 

 specific regions. Lack of such a "maturation process " in the area 

 might be associated with a more uniform pattern of demineraliza- 

 tion. 



The marked increase in fluoride and the pronounced decrease in 

 carbonate and magnesium observed in carious enamel and dentin, 

 however, suggests more profound chemical changes than can be 

 attributed to a maturation process, and an associated selective de- 

 mineralization. The fact that manv crystallites in all regions ap- 

 peared well preserved lends support to the idea that these crystallites 

 at least in part are products of a recrvstallization process (Johansen 

 and Parks, 1961 ) . By such a mechanism the fluoride liberated from 

 dissolved crystallites could be incorporated into those persisting; 

 also, additional fluoride (Dowse and Jenkins, 1957) and other ions 

 might be available from oral fluids for incorporation into the apatite 

 lattice. The low magnesium and carbonate values, on the other hand, 

 might reflect a low concentration of these ions in the fluid environ- 

 ment during recrystallization as well as preferential loss during de- 

 mineralization. The recrystallization process would be dependent 

 on changes in pH within the lesion and might involve an inter- 

 mediate product of CaHPOi (Neuman and Neuman, 1958) as 

 proposed by Gray and Francis (chapter 8 in this volume). The 

 persistence of cr)'stallites within advanced lesions might thus be 

 explained on the basis of a reduction in solubility resulting from 

 changes in composition. 



The fate of the organic matrix in the carious process has long been 

 a matter of controversv. The electron microscopic findings show that 

 lesions in both enamel and dentin contain abundant organic ma- 

 terial, and the chemical data confirm this observation. It has not, 

 however, been possible to distinguish between extraneous and in- 

 trinsic organic material in lesions of enamel. In dentin, the matrix 

 of collagenous fibrils appears relativeh' intact in areas of advanced 

 demineralization, but its composition and susceptibility to break- 



