Gem News International Gems & Gemology, Winter 2013, Vol. 49, No. 4

Demantoid Garnet Fossil Pseudomorphs

Emmanuel Fritsch, Julien Raoul, Benjamin Rondeau, and Olivier Segura

IMG - WN13 GNI 157777 636x358
Figure 1. These coral (left) and gastropod (right) fossils are pseudomorphs in demantoid garnet from Ambanja, Madagascar. The coral measures 1.5 × 1 cm. Photo by Pierre-Yves Chatagnier.
In 2010, two extremely rare fossil pseudomorphs in demantoid garnet were acquired from the classic demantoid deposit in Ambanja, Madagascar. One resembled a gastropod, the other a coral (figure 1). The brilliance, lozenge-shaped crystal faces, and green color of these samples were all consistent with the demantoid variety of andradite; this was confirmed by Raman spectroscopy. Both samples clearly showed typical andradite bands at 371 and 516 cm-1, with peaks at 817, 844, and 873 cm-1 forming a triplet. Additionally, an intense quartz signal at 464 cm-1 was observed. The general appearance was that of many small crystals (up to 1 mm in size) grown together, preserving the overall shape and some details of the fossils. Careful observation of neighboring micro-facets in reflected light indicated that both pieces were actually single crystals of andradite (figure 2).

WN13 GNI
Figure 2. Both the coral (left) and the gastropod (right) appear to be single crystals, as revealed by concomitant light reflection (in white) on multiple, parallel rhomboidal crystal faces. Photos by Benjamin Rondeau (left) and Pierre-Yves Chatagnier (right).
It is surprising to observe demantoid replacing a fossil, but some pseudomorphs in gem minerals have been documented, such as pseudomorphs of snail-like gastropods in emerald (P. Vuillet et al., “Les émeraudes de Gachalá, Colombie; historique, genèse et découvertes paléontologiques,” Le Règne Minéral, Vol. 46, 2002, pp. 5–18). Most surprising is that these pseudomorphs were monocrystals. To investigate how the fossils were replaced by demantoid, we submitted the pieces for X-ray tomography analysis to the Laboratoire Français de Gemmologie (LFG) in Paris. They used an RXSolutions DeskTom 130 system; results are given in figure 3. There are no remains of the original fossil. The objects were formed entirely out of andradite, which apparently grew from the solid parts of the animal towards the outside of the shell, and inward toward the hollow parts. Hence, some of the hollow parts were not yet completely filled. Lighter shades of gray were due to less-dense material.

WN13 GNI
Figure 3. X-ray tomography revealed that both fossils are made entirely of demantoid, and that some internal cavities of the organism remained partially empty.
The fact that the present pieces appear polycrystalline but display evidence of being single crystals suggests that the original fossil was replaced by the growth of many small garnet crystals of random orientation. By that theory, they later coalesced and evolved into a single crystal through the process known as Ostwald ripening (I. Sunagawa, Crystals, Cambridge University Press, 2005, p. 258). These pieces are therefore not a pseudomorph stricto senso, as details of the hard parts are lost, but more exactly, recrystallized psedomorphs with relatively well-developed crystal faces.

Emmanuel Fritsch is professor of physics, and Benjamin Rondeau is a researcher, with the University of Nantes (France). Julien Raoul is a gemologist and mineral collector in Madagascar, and Olivier Segura is with the Laboratoire Français de Gemmologie in Paris.