Aventurine feldspars, whose optical effect is due to thin flaky hematite lamellae occurring in crystallographic orientations in respect to the host, have been studied by some mineralogists since the detailed study by Andersen (1915). The origin of hematite in feldspar, however, has not been solved yet. The hematite flakes in labradorite from Labrador, Canada are arranged on many straight lines which are conjectured to be dislocation lines in the host feldspar (Figure 19). Surface microtopographs of these hematite crystals were studied by means of Nomarsky type reflection interference contrast microscopy and ordinary transmission microscopy. The thickness of hematite varies from place to place, thinner portions show greenish interference colour whereas thicker portions reddish. Linear patterns and triangular pits are observed on the (0001) surface of hematite (Figure 21, 22A). When the specimen shown in Figure 22A was heated at 1050℃ for 24 hours, the surface changed to show fine linear patterns (Figure 22B). Some hematite crystals exhibit outlines as if they were melted in the feldspar and decomposed into two or three parts or occassionally into many small parts as seen in Figure 20, 23A. When such specimens (e. g. Figure 23A) are heated at 1100℃ for 24 hours, there appear many black dots in the area originally surrounded by hematite (Figure 21). The origin of the characteristics of surface microtopographs and the outlines of these hematite crystals are considered from this study to be as follows. Some hematites were anhedral when they were included or formed in the host feldspar. After their formation, the hematite reacted with the feldspar forming iron glass around the hematite. The lines and triangular pits are conjectured to be etch patterns on the (0001) surface. Some hematites were decomposed into several parts by prolonged reaction with feldspar. Asterism and chatoyancy phenomena are discussed in some details. Figure 25 shows replicated electron micrograph of tiger's eye etched with HF solution.
