THE ANATOMICAL RECORD 226:135-146 (1990)

Structural Characteristics of Staircase-Type Retzius Lines in Human Dental Enamel Analyzed by Scanning Electron Microscopy S.RISNES Department of Anatomy, Dental Faculty, University of Oslo, Blindern, 0316 Oslo 3, Norway

ABSTRACT Based on a scanning electron microscopic (SEM) analysis of adjoining, acid-etched planes cut through human cervical enamel, the structural characteristics of staircase-type Retzius lines have been clarified. Structural features associated with this type of Retzius lines-such as cleftlike defects, decreased dimension of prisms, increased interprism, club-shaped appearance of prisms, oblique ridges, and triangular regions-have been incorporated into a unifying, three-dimensional model. A continuous discontinuity defect, involving both prisms and interprism, is a prominent feature of this type of Retzius lines. Prisms and interprism facing the deep aspect of the cleftlike defect show a n enlarged, flat surface t h a t encroaches on the cervically situated prisms being rebuilt from the same cleft. The initial part of a prism taking off from the cleft is of distorted shape. As the prism reaches the level of the interprism cleft, it abruptly regains its normal size and shape. The relationship between this type of Retzius lines and the carious process is discussed, and it is suggested that the discontinuity defect may retard the carious lesion due to a protective effect of its supposed organic content. The developmental events creating staircase-type Retzius lines are discussed, and it is suggested t h a t Tomes' processes are constricted near their bases with a corresponding increase in interprismatic growth region. Tomes' processes will have to reshape plastically a s they move out of the constricted pits, trailing the parent ameloblasts a s they resume enamel production and move in the direction of the prisms. A hundred and fifty years after the first description of the Retzius lines (Fraenkel, 1835;Linderer and Linderer, 1837; Retzius, 1837), there is still much confusion as to their nature. Although opposing views exist (Warshawsky et al., 1984; Warshawsky, 1985), it has been generally accepted, in accordance with the original interpretation (Linderer and Linderer, 1837), that the Retzius lines are incremental lines reflecting the layered apposition of enamel during amelogenesis. In human enamel Retzius lines are normal features of constant presence, but they show variation in distribution and appearance. Due to geometric considerations, the Retzius lines must cross the prisms (Rimes, 198513, 1987). Their visibility has been attributed to a n altered structure andlor composition of the enamel along the striae: hypomineralization (Gustafson, 1959; Boyde, 19701, hypermineralization (Gustafson, 19591, altered mineral composition (Woltgens et al., 1980; Driessens et al., 1984), altered organic composition (Ducroc and Proust, 1973), increased widthldensity of prism sheaths (Sognnaes, 1949; Jansen and Visser, 1950; Osborn, 1973), increased width of interprism (Gustafson, 19591, reduced width of prisms (Bergman and Engfeldt, 1954; Gustafson, 19591, and change in prism direction, either cervically (Gustafson, 1945; Gustafson, 1959; Hinrichsen and Engel, 1966; Gustafson and Gustafson, 1967; Helmcke and Schulz, 1968; Weber and Ashrafi, 0 1990 WILEY-LISS, INC.

1979) or transversely (Osborn, 1973; Weber and Ashrafi, 1979). It has also been pointed out that variations exist within the same Retzius line (Gustafson, 1959; Klinger et al., 1978). There have been attempts to classify Retzius lines as either normal or pathologic (Asper, 1916; Gustafson, 1959; Wilson and Schroff, 1970; Rose, 1977, 19791, but the criteria have differed. There is clearly a need for a more detailed characterization of various types of Retzius lines in order to improve our understanding of their nature and genesis. Frank (1978) divided the Retzius lines into two main groups, thin and broad. One subgroup of thin Retzius lines is the staircase-, stepped-, or picket fence-type, named according to their appearance in longitudinal sections. They are regularly present in the outer enamel of human permanent teeth, especially in the cervical part. Structural features associated with this type of Retzius lines have been studied with the light microscope (Gustafson, 1959; Gwinnett, 1966a; Wilson and Schroff, 1970; Weber et al., 1974), transmission electron microscope (TEM; Hinrichsen and Engel, 1966; Weber et al., 1974) and scanning electron micro-

Received January 5, 1989; accepted May 17, 1989

136

S.RISNES

Fig. 1. Regularly spaced Retzius lines (arrows) in the outer enamel of a maxillary first premolar visualized in two mutually perpendicular sectionediground planes (Ta, L) and one fractured plane (T).x 100. Fig. 2. Higher magnification of area shown in Figure 1. Large arrow represents general prism direction. Arrows 3 and 4 refer to details in Figures 3 and 4, respectively. x 300. Figs. 3, 4 . Higher magnifications from area shown in Figure 2. Unlabeled arrows refer to details marked 3 and 4 in Figure 2. Oc-

clusal direction toward right. x 1,000.3a,4a:Two Retzius lines (open arrows) in tangential plane made visible through differential etching of prism profiles characterized by rows of large, flat prism profiles (PI, on top of which are built small, irregular, and crested prism profiles (p). Framed areas (10,11) are shown a t higher magnification in Figures 10 and 11. 3b,4b: Adjacent areas in longitudinal plane of areas shown in Figures 3a and 4a. Retzius lines cross vertical prisms at an angle of about 45". Framed areas (5, 6, 8) are shown at higher magnification in Figures 5, 6, and 8.

scope (SEM; Boyde, 1970; Frank, 1978; Weber and Ashrafi, 1979). However, from the evidence accumulated, i t has not been possible to establish a clear understanding of the three-dimensional organization of prisms and interprism in this type of Retzius lines. It was the aim of the present study to clarify and visualize the enamel structure characteristics of the staircase-type Retzius lines by observing sectioned, acid-etched specimens in the SEM.

located under the dissecting microscope (Risnes, 1973). A guiding line was drawn with pencil on the section transversely to the general prism direction in the region. The specimens were glued with cyanoacrylate glue to small specimen stubs and placed in a n apparatus for further controlled mechanical preparation (Risnes, 1985a). The specimens were sectioned andlor ground along the guiding line, resulting in a specimen containing two adjoining planes: longitudinal and oblique tangential. The sectioned planes were finished by grinding with finest grit silicone carbide paper. After rinsing thoroughly under running tap water, the specimens were etched for three periods of 5 seconds each in 1.3% nitric acid (Fosse, 19681, air-dried, and sputter-coated with a n approximately 40-nm layer of gold-palladium. The specimens were placed in a specially designed holder (Risnes, 1982) and studied in a Jeol 50A scanning electron microscope operated a t 15 kV.

MATERIALS AND METHODS

The material consisted of human permanent teeth, mostly premolars and upper central incisors. All teeth were normally developed, unfixed, and dry. About 1mm-thick faciolingual, midcoronal, longitudinal sections of the crowns were cut with a water-cooled diamond wheel (Risnes, 1981). Regions in the cervical enamel showing regularly spaced Retzius lines were

RESULTS

A bbreuiations d IP IPC L

p,P PC T Ta

defect interprism cleft interrupting interprism longitudinal plane prisms cleft interrupting prism transverse plane tangential plane

In SEM specimens of acid-etched enamel distinct Retzius lines were most conspicuous in the outer enamel (Figs. 1, 2). They were regularly spaced a t intervals of about 30-50 pm measured along the prisms (Figs. 1,2,3b, 4b). In the longitudinal plane, they could show a characteristic staircase configuration as they crossed the prisms a t a n angle of about 45" (Figs. 3b, 4b). The horizontal part of the steps in the staircase was formed by a cleftlike defect crossing the prism a t right angles. The severity of these defects varied, but

STRUCTURE OF RETZIUS LINES IN HUMAN ENAMEL

Figs. 3,4

137

138

S.RISNES

judged from stereo-pair micrographs, some seemed to be through-going defects that interrupt the continuity of the involved prisms (Figs. 5-7). The vertical part of the steps was formed by the prism border (prism sheath). In an adjoining tangential plane ground roughly transversely to the long axis of the prisms, the Retzius lines appeared as horizontal (relative to a vertical tooth axis) rows or zones of prisms that were etched differently from the prisms situated between the Retzius lines (Figs. 2,3a, 4a). The prisms within a row marking the position of a Retzius line appeared flat and enlarged compared with the crested prisms situated between the Retzius lines (Figs. 3a, 4a). Immediately cervical to the large, flat prisms, there were one or two rows of small, crested, irregular prisms. The flat-surfaced prism profiles corresponded to the horizontal part of the steps in the staircase configuration and thus represented the end “surface” of prisms that had just reached the cleftlike defect (Figs. 3, 4, 7). Oblique clefts or ridges running in the same direction as the Retzius lines sometimes replaced or were superimposed on the regular staircase configuration (Figs. 3b, 4b). It could be demonstrated that the oblique clefts represented defects of the interprism (Figs. 5, 7, 8). This was in accordance with the fact that the crystals abutting on the oblique clefts were oriented perpendicular to the clefts; i.e., they deviated about 45” cervically relative to the long axis of the prisms. Sometimes the transverse cleft was continuous with an oblique cleft in cervical direction (Figs. 3b, 4b, 8). The cross-striations of prisms may also appear as narrow zones of defective crystal packing, although less marked than the defects associated with staircasetype Retzius lines. Sometimes it was possible to locate regions where cross-striations could be followed into the interprism (Fig. 9). On doing so, the striations bent toward the dentin in a cervical direction, whereby the crystals maintained a perpendicular orientation relative to the striations. Thus, there exists a close resemblance between the cross-striations and the clefts associated with the staircase-type Retzius lines. An oblique ridge often traversed the prism a short distance above the transverse cleft, imposing a triangular appearance on the initial prism segment (Figs. 3b, 4b, 5, 6). In longitudinal sections, some prisms, or rather their associated interprism, were seen to expand in cervical direction as they approached the cleftlike defect, giving them a clubshaped appearance (Figs. 3b, 4b, 8). The dimension of the adjacent, cervically situated prisms decreased correspondingly shortly after their start from the same Retzius line cleft. In the tangential plane, this phenomenon expressed itself as flat-surfaced prisms attaining an increased dimension, especially at the base of the arcade, where the prisms were continuous with the interprism (Figs. 10, 11). In a pattern 3 prism arrangement, this expansion, which was due mainly to an increase in the interprism, encroached on the adjacent, cervically situated prisms, each of which were thus attenuated from both sides, changing their shape from the typical arcade t o a much more narrowed and pointed form. Occasional absence of typical club-shaped prisms along the Retzius lines (Figs. 3-7) may be explained by a variable prism/

sectioned plane-relationship or by a variably expressed expansion of interprism. In a tangential plane ground roughly perpendicular to the long axis of the prisms, cross-cut prism profiles could be observed at various levels relative to a Retzius line (Figs. 3a, 4a, 10, 11). The morphologically expressed reaction toward acid-etching of a single prism (and its associated interprism) as it approaches and leaves a Retzius line could be visualized in the tangential plane by observing the change in appearance of prism profiles along a line starting midway between two Retzius lines and proceeding in cervical direction across and perpendicular to the Retzius line and ending halfway to the next Retzius line. If, in addition, the tangential plane is slightly oblique transversely, intermediate stages can be visualized along the prism rows. Figures 10 and 11 allow this type of analysis. As a prism approached a Retzius line from below, it changed abruptly from a crested to a flat topography, indicating a change in reaction toward acid etching. The interprism that was continuous with the cervical part of the prism attained flatness simultaneously. Both prism and interprism appeared enlarged. The prism being reestablished above the Retzius line ap-

Figs. 5-7. Stereo-pair micrographs of details from Figures 3b and 4b. Singly, the micrographs should be viewed according to the orientation of the lettering, as this allows for a more direct comparison with Figures 3b and 4b. Fig. 5. Higher magnification of framed area (5) in Figure 4b. A prism, P, is interrupted by a transverse cleft, PC, representing the horizontal part of a step in a staircase-type Retzius line. The initial part of the prism, P,, starting off from the cleft, is reduced in size. The original dimension is regained by a sudden increase in width creating a n oblique ridge (unlabeled arrows), which is in register with the oblique cleft, IPC, interrupting the interprism, IP, occlusal to the prism. x 3,000. Fig. 6. Higher magnification of framed area ( 6 ) in Figure 3b. Two prisms, P, are interrupted by transverse clefts, PC, a t a Retzius line. The initial parts of the two prisms taking off from the cleft contains regions of defective enamel, d, Oblique ridges (unlabeled arrows) mark the sudden increase in prism dimension. x 4,000. Fig. 7. Transition between tangential and longitudinal plane. Unlabeled black arrow points a t same detail as unlabeled arrow in Figure 4b. Two large, flat prisms, P, with their associated expanded interprism (white arrows) encroach on three cervically situated prisms, p1-p3. Starting off from the transverse cleft, PC, the prism p1 is a t first of the same size as the flat prism ending at the cleft, whereafter its dimension decreases due to the expansion of the interprism. Above this level the prism expands. The expansion of the two prisms pz and p3 is recognized as ridges (unlabelled, small black arrows). x 3,000. Fig. 8. Higher magnification offramed area (8)in Figures 3b and 4b. As a prism with its associated interprism expands (white unlabeled arrow) on approaching the cleft and attains a club-shaped appearance, the cervically situated prism is diminished accordingly (unlabeled black arrow). In this section a transverse cleft, PC, interrupting the continuity of a prism, is seen to be continuous with a n oblique cleft, IPC, interrupting the continuity of the interprism. x 3,000. Fig. 9. Maxillary central incisor. Cross-striations (arrows) tend to deviate toward the dentin as they continue from the prism, P, into the interprism, IP, thereby maintaining a constant relationship to the crystals. Dentinal direction is toward the bottom, cervical direction is toward right. x 6,000.

STRUCTURE O F RETZIUS LINES IN HUMAN ENAMEL

Figs. 5-9.

139

STRUCTURE OF RETZIUS LINES IN HUMAN ENAMEL

peared crested and somewhat irregular and covered only a small, central part of the prism domain. Gradually, the prism domain was filled out. At the same time, the prism domain was gradually narrowed from side to side, especially near the arcade apex, due to a n overgrowth of the adjacent interprism. The maximum constriction of the prism was reached at the level where the expanded interprism had attained a flat, oblique surface, i.e. where the interprism was interrupted by a cleftlike defect. Above this level, the prism increased abruptly to normal size (Figs. 7, 11). In the longitudinal plane, this sudden expansion of the prism was often demarcated by a n oblique ridge that was in register with the oblique clefts of the Retzius line and that traversed the prism a short distance above the transverse step (Fig. 5). The appearance both in the longitudinal plane (Figs. 3b, 4b, 6) and in the tangential plane (Figs. 1 0 , l l ) indicated that the prism a t the level of its constriction and sudden expansion was of inferior quality, with frequent occurrence of irregular clefts and cavities. Such defects were less frequent in the prisms reaching the Retzius line from below. However, of the two prism surfaces facing a transverse cleft, the “post-cleft” surface often seemed more tightly packed with crystals (Figs. 3b, 4b, 6). The defect of prisms and interprism giving rise to the clefts seen in acid-etched specimens were obviously continuous over considerable distances. If it had been possible to remove the enamel situated outside the cleft of a staircase-type Retzius line, we would be left with a pitted surface in which the pits would be arranged in a pattern identical to the prism pattern in the area (Fig. 12),but the shape of the pits would be altered (Fig. 13). The bottom of the pits would be at a n angle of about 45” to the surface and would be smoothly continuous with the surface in cervical direction. The openings into the pits would not be typically arcade-shaped but would be narrower sideways and more pointed in occlusal/incisal direction. Beyond the opening the pit would widen out gradually toward the bottom, with its typical arcade shape. The first part of the prism starting off from a Retzius line cleft is shaped so that it fits exactly into the narrowed pit (Fig. 13). At the level of the surface, i.e., the interprism, the prism abruptly regains its normal size, the transition being marked by a n oblique ridge on its lateral-incisaUocclusa1 aspect.

Figs. 10, 11. Composite micrographs a t higher magnification of areas 10 and 11 in Figures 3a and 4a. The left and right halves of the micrographs can be viewed in stereo with the micrographs to the left and right, respectively. The events occurring as a prism reach to a Retzius line and continues off from the Retzius line; this can be visualized by observing prism profiles in the direction of the large vector arrow (cervical direction). Intermediate stages can be visualized in the direction of the small vector arrow (transverse direction). Rows of large, flat prism profiles, P, represent prisms that have just reached to the cleftlike defect. These prisms and their associated, expanded interprism (arrows) encroach on the cervically situated prisms, imposing on them a reduced, pointed profile. In Figure 11 the rebuilding of a prism starting off from a cleft can be followed through the prism profiles pI-plo. Due to the effect of the acid, the prism crystals do not fill the whole available prism domain in p1-p3. Although the prism itself seems to increase in size, the prism domain is reduced from p1 to p8. At p9 the sudden expansion has just started resulting in a near full-sized prism at p10. x 3,000.

141

There was a clear tendency that the staircase configuration with its associated clefts became less pronounced in dentinal direction with respect to the same Retzius line and in occlusal/incisal direction with respect to different Retzius lines. DISCUSSION

A discontinuity defect constituted a prominent feature of the staircase-type Retzius lines. In accordance with the findings of Frank (19781, the crystal-deficient cleft was continuous over a considerable distance along the growth plane and encompassed both prisms and interprism. Due to the action of the acid, i t was difficult to decide if the defect was completely devoid of crystals or not. In a TEM study, Weber et al. (1974) found that the defect contained some crystals, but conspicuously fewer than the adjacent enamel. Some investigators have found a n increased width of prism sheaths associated with Retzius lines (Sognnaes, 1949; Jansen and Visser, 1950; Osborn, 1973). This was not corroborated by the present study. The hypomineralization associated with Retzius lines as observed in microradiographs (Gustafson, 1959; Gwinnett, 1966b) may be explained by a combined effect of the discontinuity defect and the inferior quality of prisms starting off from the Retzius line. In the present study, increased interprism, as reported by Gustafson (19451, and decreased dimension of prisms, as reported by Bergman and Engfeldt (1954) and Gustafson (19591, were found to be prominent features associated with the staircase-type Retzius lines. A change in the course of prisms, considered by several authors to be a major feature of Retzius lines (Gustafson, 1945; Gustafson, 1959; Hinrichsen and Engel, 1966, Osborn, 1973; Helmcke and Schulz, 1968; Weber and Ashrafi, 1979), was not responsible for the staircase-type Retzius lines, as also noted by Frank (1978). As seen in the present study, the acid accentuated the structural characteristics associated with the staircase-type Retzius line, especially the cleftlike defect. In accordance with this, surface carious enamel shows distinct steps corresponding to the Retzius lines (Holmen et al., 1985) and has also been found to fracture preferentially along the Retzius lines (Poole and Silverstone, 1969). In fluorosed enamel, which is characterized by increased porosity, the Retzius lines are enhanced (Fejerskov e t al., 1975) and in surface pit lesions fractures seem to occur preferentially along the Retzius lines (Thylstrup and Fejerskov, 1979). Boyde (1970) found that after prolonged extraction with NaOC1, surface enamel fractured preferentially along the Retzius lines. All of these observations stress the planar, three-dimensional extent of the Retzius lines and are in accordance with the notion of a n altered enamel structure along the lines a s observed in the present study. In sections cut through a carious lesion, the Retzius lines often appear accentuated (Bergman and Engfeldt, 1954; Gustafson, 1957; Darling, 1958; Schmidt and Keil, 1971; Kerckaert, 1973; Silverstone, 1973; Simmelink and Nygaard, 1982). The role of the Retzius lines in the carious process is controversial. Darling (1958) proposed that the superficial Retzius lines represent points of entry of the cariogenic agent and that the progress in depth of the lesion was modified by the

142

S.RISNES

12

B

C

Figs. 12, 13.Schematic drawings showing the situation a t the developing enamel surface before (Fig. 12) and a t the time of (Fig. 13) formation of a staircase-type Retzius line, 12A,1 3 A Blocks of enamel sectioned longitudinally (L)and transversely (T). The pitted surfaces (S)show prism domains (P) (bottom of pits) and interprism domains (IP) (top surface). Above is outlined the shape of a prism and its

associated interprism being built from the surface. As the prism approaches the level a t which a Retzius line will form, its base, particularly its associated interprism, expands (Fig. 12, unlabeled arrows). The developing enamel surface attains the topography shown in Figure 13. The expanded interprism encroaches on the cervically situated pits. The pits consequently have a narrowed, pointed entrance and

143

STRUCTURE OF RETZIUS LINES IN HUMAN ENAMEL

13

-R

B

-;-3 C

widen out toward the bottom. The initial part of a prism being rebuilt from the Retzius line decreases in size toward the narrowed pit entrance. At the level of the surface, the prism increases abruptly to normal size, creating an oblique ridge (R) on its incisaliocclusallateral aspects. Above the cleft level, the expanded interprism (unlabeled arrows) decreases to normal size. 12B, 1 3 B Ameloblast/enamel-

relationship when the enamel is sectioned along direction Y (nearly parallel to the L plane) of 12A and 13A. Unlabeled arrow points a t expanded interprism. 12C, 13C: Ameloblast/enamel-relationship when the enamel is sectioned along direction X (parallel to the T plane) of 12A and 13A. Unlabeled arrow points at expanded interprism.

144

S.RISNES

Retzius lines. Crabb (1966, 1968) suggested that the carious lesion spreads from demineralizing centre(s) in a wavelike fashion along the prisms and that the Retzius lines may represent planes of resistance rather than pathways of progress. Judged from their reaction toward acidic demineralization, as observed in the present study, the staircase-type Retzius lines may well exert a certain modifying effect on a surfaceinitiated carious process. The flat-surfaced prism profiles associated with the Retzius lines seemed to be more acid-resistant and better preserved than the irregular prisms building on top of them (Figs. 10, 11). This may explain the apparent incomplete and retarded filling in of prism domains above the Retzius line (Figs. 10, 11). It is possible that the flat-surfaced prisms are protected by organic material present in the discontinuity defect, which may retard the spread of a carious lesion across the Retzius line (Crabb, 1966; Klees and Brabant, 1974). The apparent inferior quality of the initial prism segments being reestablished upon the flat prism profiles may favour a lateral dissolution along the Retzius lines in this zone. Because there are different types of Retzius lines (Gustafson, 1959; Wilson and Schroff, 1970; Frank, 1978; Klinger et al., 1978; Rose, 19791, it seems reasonable to suggest that the interaction between a carious lesion and Retzius lines may vary and give different histological appearances. Klees and Brabant (1974) found indications that the carious process is sometimes accelerated and sometimes retarded along the Retzius lines. Other types of Retzius lines, such as the broader bands (Gustafson, 1959) and the neonatal line (Weber and Eisenman, 1971), also exhibit morphological features related to the staircase configuration. The stepped appearance of Retzius lines as seen in the longitudinal plane can be directly related to the histology of the tooth germ during amelogenesis; longitudinally sectioned developing teeth show a serrated surface with a staircase configuration (Wolf, 1942; Gustafson, 1959; Ronnholm, 1962; Boyde, 1964) (Fig. 12). The ameloblasts stand roughly perpendicular to the enamel surface and are angled about 45" relative to the prisms. The pointed, triangular Tomes' processes occupy the triangular pits of the staircase. This staircase configuration of sectioned developing enamel is consistent with the surface topography of developing enamel (Fig. 12) as visualized by SEM (Boyde, 1967; Thylstrup et al., 1977; Warshawsky et al., 1981; Sasaki and Higashi, 1983). The staircase pattern may vary somewhat, depending on the pattern of pits (prism pattern) in the area and on the direction of the plane of section relative to the pit pattern. Figure 13 shows the hypothetical surface of developing enamel a t the time of formation of a staircase-type Retzius line. The entrance to the pits is constricted due to a n overgrowth of interprism, whereas the floor of the pits is typically arcadeshaped and of somewhat increased size. The prism rebuilt from this surface a t first diminishes in size as it grows from the floor toward the entrance of the pit and then suddenly expands a s it grows past the edge of the expanded interprism, acquiring a n oblique ridge on its lateral aspect. The events taking place during formation of a staircase-type Retzius line may tentatively be described as

follows: when the enamel between the Retzius lines is being formed, the ameloblast/enamel relationship is similar to that shown in Figure 12. As the ameloblasts approach the Retzius line, the Tomes' process becomes progressively constricted near its base on its occlusallateral apects (Fig. 13). As constriction increases, the interprismatic growth region between the Tomes' processes is extended, and more interprism is produced. The overgrowth of interprism will interfere with the movement of the Tomes' process in the direction of the prisms, forcing the Tomes' process to slide along a n increasing overhang diverging from the direction of ameloblast movement. It seems reasonable to assume that the ameloblasts pause as the discontinuity defect is being formed, but the cause of the halt is not known. Perhaps it is related to a physiological rhythmicity in ameloblast function involving for instance a modulation between predominantly secretory and predominantly absorptive phases. On resuming enamel production, the constricted Tomes' process must change its shape plastically in order to be able to pass the narrowed entrance of the pit and reestablish its normal shape above this level as the ameloblasts continue to move along the general prism direction. It thus seems that the ameloblasts move independently of the spatial conditions experienced by the Tomes' process at the mineralizing front. This is in accordance with the finding that prisms in the inner enamel of rat incisors may undergo a n abrupt 90" change in their direction (Risnes, 19791, which means that after the involved ameloblast changed its direction of movement, the long, trailing Tomes' process for a short period of time must slide over a 90" bend a t the mineralizing front. Thus, i t seems that the ameloblasts lead the prisms (Boyde, 1969) and not vice versa (Boyde, 1964; Helmcke, 1964). Boyde (1964,1976) has proposed a hypothesis for the formation of the varicosities of prisms, which would also be applicable to the formation of the staircasetype Retzius lines; due to a diminished rate of matrix production and ameloblast movement, growth of enamel at the pit wall is enhanced, resulting in a n increase in the interprism at the expense of the prisms. However, it has recently been appreciated that the aspect of the Tomes' process facing the pit wall is nonsecretory (Kallenbach, 1977; Skobe et al., 1981; Wakita et al., 1981; Warshawsky et al., 1981; Sasaki and Higashi, 1983; Nanci and Warshawsky, 1984). This indicates that the overgrowth of interprism associated with the staircase-type Retzius lines is related to a n increase in the interprismatic growth region as the Tomes' process is constricted. Another possibility is that the constriction and overgrowth occur after the ameloblasts have stopped moving, and for a while the only active production site is found in relation to a n increasing constriction of a n immobile ameloblast. According to a n alternative view on the origin of Retzius lines offered by Warshawsky (1985), the striae correspond to the borders between circumferential, transverse belts of decussating prisms produced by cohorts of ameloblasts of different ages. The geometric implications of this hypothesis are incompatible with the observed geometry of the Retzius lines (Risnes, 198513, 1987).

STRUCTURE O F RETZIUS L I N E S IN HUMAN ENAMEL

LITERATURE CITED Asper, H. 1916 Uber die “Braune Retzius’sche Parallelstreifung” im Schmelz der menschlichen Zahne. Schweiz. Vierteljahrschr. Zahnheilk., 26t275-314. Bergman, G., and B. Engfeldt 1954 Studies on mineralized dental tissues. 11. Microradiography as a method for studying dental tissues and its application to the study of caries. Acta Odontol. Scand., 12r99-132. Boyde, A. 1964 The structure and development of mammalian enamel. Thesis. University of London, London. Boyde, A. 1967 The development of enamel structure. Proc. R. Soc. Med. Lond., 60t13-18. Boyde, A. 1969 Electron microscopic observations relating to the nature and development of prism decussation in mammalian dental enamel. Bull. Group. Int. Rech. Sci. Stomatol., 12t151-207. Boyde, A. 1970 The contribution of the scanning electron microscope to dental histology. Apex, 4:15-21, 9-16. Boyde, A. 1976 Amelogenesis and the structure of enamel. In: Scientific Foundations of Pentistry. B. Cohen and I.R.H. Kramer, eds. W. Heinemann Medical Books, London, pp. 335-352. Crabb. H.S.M. 1966 Enamel caries. Observations on the histology and pattern of progress of the approximal lesion. Br. Dent. JT; 121: 115-129, 167-174. Crabb, H.S.M. 1968 Structural patterns in human dental enamel revealed by the use of microradiography in conjunction with two dimensional microdensitometry. Caries Res., 2t235-252. Darling, A.I. 1958 Studies of the early lesion of enamel caries. Its nature, mode of spread, and points of entry. Br. Dent. J., 105: 119-135. Driessens, F.C.M., H.J.M. Heijligers, J.M.P.M. Borggreven, and J.H.M. Woltgens 1984 Variations in the mineral composition of human enamel on the level of cross-striations and striae of Retzius. Caries Res., 18t237-241. Ducroc, J., and J.P. Proust 1973 Contribution a l’etude histochimique de la structure organique de la strie de Retzius. J. Biol. Buccale, 1t337-344. Fejerskov, O.,L.M. Silverstone, B. Melsen, and I.J. Mdler 1975 Histological features of fluorosed human dental enamel. Caries Res., 9:190-210. Fosse, G. 1968 A quantitative analysis of the numerical density and the distributional pattern of prisms and ameloblasts in dental enamel and tooth germs. 11. Serial etching of dental enamel. Acta Odontol. Scand., 26:285-314. Fraenkel, M. 1835 De penitiori dentium humanorum structura observationes. Thesis. M. Friedlaender, Vratislavile. Frank, R.M. 1978 Les stries brunes de Retzius en microscopie electronique a balayage. J . Biol. Buccale, 6r139-151. Gustafson, A.-G. 1959 A morphologic investigation of certain variations in the structure and mineralization of human dental enamel. Thesis. Odontol. Tidskr., 67t361-472. Gustafson, G. 1945 The structure of human dental enamel. A histologic study by means of incident light, polarized light, phase contrast microscopy, fluorescence microscopy and microhardness tests. Thesis. Odontol. Tidskr., 53(Suppl.):1-150. Gustafson, G. 1957 The histopathology of caries of human dental enamel. With special reference to the division of the carious lesion into zones. Acta Odontol. Scand., 15:13-55. Gustafson, G., and A.-G. Gustafson 1967 Microanatomy and histochemistry of enamel. In: Structural and Chemical Organization of Teeth. A.E.W. Miles, ed. Academic Press, New York, Vol 1, pp. 75-134. Gwinnett, A.J. 1966a Histology of normal enamel. 111.Phase contrast study. J. Dent. Res., 45:865-869. Gwinnett, A.J. 1966b Histology of normal enamel. IV. Microradiographic study. J. Dent. Res., 45t870-873. Helmcke, J.-G. 1964 Kombination von elektronenmikroskopischen und neuen lichtmikroskopischen Untersuchungsmethoden fur Strukturen des Zahnschmelzes. In: Advances in Fluorine Research and Dental Caries Prevention. J.L. Hardwick, J.-P. Dustin, and H.R. Held, eds. Pergamon Press, Oxford, pp. 127-139. Helmcke, J.-G., and L. Schulz 1968 Elektronenmikroskopische Beobachtungen a n Retziuslinien von gesundem menschlichem Schmelz und im karioesen Bereich. Bull. Group. Int. Rech. Sci. Stomatol., 11:257-278. Hinrichsen, C.F.L., and M.B. Engel 1966 Fine structure of partially demineralized enamel. Arch. Oral Biol., 11t65-93. Holmen, L., A. Thylstrup, J.D.B. Featherstone, L. Fredebo, and M. Shariati 1985 A scanning electron microscopic study of surface changes during development of artificial caries. Caries Res., 19: 11-21. ~

145

Jansen, M.T., and J.B. Visser 1950 Permeable structures in normal enamel. J . Dent. Res., 29:622-632. Kallenbach, E. 1977 Fine structure of secretory ameloblasts in the kitten. Am. J. Anat., 148t479-512. Kerckaert, G.A. 1973 Electron microscopy of human carious dental enamel. Arch. Oral Biol., 18r751-758. Klees, L., and H. Brabant 1974 Sur quelques classifications des lesions carieuses de l’email etablies en vue d’elucider le probleme de l’existence possible de diverses sortes de caries. Bull. Group. Eur. Rech. Sci. Stomatol. Odontol., 17t195-219. Klinger, H.G., C. Rudolph, and E. Gabriel 1978 Quantitative evaluation of the etching pattern in the enamel of a tooth section. Caries Res., 12:231-237. Linderer, C.J., and J. Linderer 1837 Handbuch der Zahnheilkunde. Schlesinger’sche Buch- und Musikhandlung, Berlin, pp. 175-185. Nanci, A., and H. Warshawsky 1984 Characterization of putative secretory sites on ameloblasts of the rat incisor. Am. J . Anat., 171~163-189. Osborn, J.W. 1973 Variations in structure and development of enamel. Oral Sci. Rev., 3r3-83. Poole, D.F.G., and L.M. Silverstone 1969 Observations with the scanning electron microscope on trauma-induced micro-cavities in human enamel. Arch. Oral Biol., 14:1323-1329. Retzius, A. 1837 Bemerkungen uber den innern Bau der Zahne, mit besonderer Riicksicht auf den im Zahnknochen vorkommenden Rohrenbau. Arch. Anat. Physiol., pp. 486-566. Risnes, S. 1973 Three-dimensional features of human enamel as seen with the dissecting microscope. Arch. Oral Biol., 18:647-650. Risnes, S. 1979 A scanning electron microscope study of aberrations in the prism pattern of rat incisor inner enamel. Am. J. Anat., 154:419-436. Risnes, S. 1981 A rotating specimen holder for hard tissue sectioning. Stain Technol., 56t265-266. Risnes, S. 1982 Multiangular viewing of dental enamel in the SEM: a simple specimen holder system. Scand. J . Dent. Res., 90.80-82. Risnes, S. 1985a Multiangular viewing of dental enamel in the SEM. a n apparatus for controlled mechanical specimen preparation. Scand. J. Dent. Res., 93.135-138. Risnes, S. 198513 A scanning electron microscope study of the threedimensional extent of Retzius lines in human dental enamel. Scand. J . Dent. Res., 93:145-152. Risnes, S. 1987 Multiplane sectioning and scanning electron microscopy as a method for studying the three-dimensional structure of mature dental enamel. Scanning Microsc., 1:1893-1902. Ronnholm, E. 1962 The amelogenesis of human teeth as revealed by electron microscopy. 11. The development of the enamel crystallites. J . Ultrastruct. Res., 6t249-303. Rose, J.C. 1977 Defective enamel histology of prehistoric teeth from Illinois. Am. J . Phys. Anthropol., 46t439-446. Rose, J.C. 1979 Morphological variations of enamel prisms within abnormal striae of Retzius. Hum. Biol., 51t139-151. Sasaki, T., and S. Higashi 1983 Scanning and transmission electron microscopy of developing enamel surfaces in the kitten tooth germs. J . Electron Microsc., 32r163-171. Schmidt, W.J., and A. Keil 1971 Polarizing Microscopy of Dental Tissues. Pergamon Press, Oxford, pp. 474-493. Silverstone, L.M. 1973 Structure of carious enamel, including the early lesion. Oral Sci. Rev., 3:100:160. Simmelink, J.W., and V.K. Nygaard 1982 Ultrastructure of striations in carious human enamel. Caries Res., 16:179-188. Skobe, Z., K. Prostak, and D. Stern 1981 Ultrastructure of secretory ameloblasts in a monkey Macaca mulatta. Arch. Oral Biol., 26: 1075-1090. Sognnaes, R.F. 1949 The organic elements of the enamel. 111. The pattern of the organic framework in the region of the neonatal and other incremental lines of the enamel. J . Dent. Res., 28: 558-564. Thylstrup, A., and 0. Fejerskov 1979 A scanning electron microscopic and microradiographic study of pits in fluorosed human enamel. Scand. J . Dent. Res., 87:105-114. Thylstrup, A., P. Skaaring, 0. Fejerskov, and F. Bierring 1977 Surface structure of tooth germs from newborn infants: a light and scanning electron microscopical study. J . Anat., 123.537-547. Wakita, M., H. Tsuchiya, T. Gunji, and S. Kobayashi 1981 Threedimensional structure of Tomes’ processes and enamel prism formation in the kitten. Arch. Histol. Jpn., 44:285-297. Warshawsky, H. 1985 Ultrastructural studies on amelogenesis. In: The Chemistry and Biology of Mineralized Tissues. W.T. Butler, ed. EBSCO Media, Birmingham, pp. 33-44. Warshawsky, H., P. Bai, and A. Nanci. 1984 Lack of evidence for rhythmicity in enamel development. In: Tooth Morphogenesis

146

S.RISNES

microscopic studies of Retzius lines in human cervical enamel. and Differentiation. J.V. Ruch, and A. Belcourt, eds. INSERM, Am. J . Anat., 141t91-104. 125:241-256. Warshawsky, H., K. Josephsen, A. Thylstrup, and 0. Fejerskov. 1981 Wilson, D.F., and F.R. Schroff 1970 The nature of the striae of Retzius as seen with the optical microscope. Aust. Dent. J., 15:162-171. The development of enamel structure in rat incisors as compared Wolf, J . 1942 Der Einfluss der Ameloblastenverschiebungen auf die to the teeth of monkey and man. Anat. Rec., 200:371-399. Gestalt und den Verlauf der Schmelzprismen. Dtsch. Zahn-MundWeber, D.F., and S.H. Ashrafi 1979 Structure of Retzius lines in Kieferheilk., 9:488-514. partly demineralized human enamel. Anat. Rec., 194563-570. Woltgens, J.H.M., P.A. Vingerling, and F. Witjes 1980 Chemical evWeber, D.F., and D.R. Eisenmann 1971 Microscopy of the neonatal idence of two separate apatite phases in human enamel. Arch. line in developing human enamel. Am. J. Anat., 132t375-392. Oral Biol., 25:435-436. Weber, D.F., D.R. Eisenmann, and P.L. Glick 1974 Light and electron