Nanotextures of aragonite in stromatolites from the quasi-marine Satonda crater lake, Indonesia - Tufas and Speleothems: Unravelling the Microbial and Physical Controls - page 217

rapid accretion of corals (Benzerara et al. in rev.),
sponge spicules (Sethmann et al. 2006), and
mollusc nacre (Rousseau et al. 2005) as well as in
abiotic systems deprived of organics (e.g. Nieder-
berger & Colfen 2006). Although it has been pro-
posed that the formation of such mesocrystals may
involve specific and controlled biological processes,
the stromatolite and the abiotic occurrences of ara-
gonite mesocrystals underline the fact that as long
as the molecular processes leading to crystallo-
graphic alignment of these nanodomains are
ignored, little can be concluded about the biospeci-
ficity of such objects.
The evolution of the crystallographic orientation
of the crystals within aragonite laminae is an
additional issue to discuss. Observations of the
Satonda stromatolites by transmitted light micros-
copy suggest that aragonite fibres are globally
oriented perpendicular to the laminae. Interestingly,
such a structure with well-oriented aragonite crys-
tals is also observed in nacre for example, in
which aragonite tablets grow in alternation with
organic matrix sheets. Recently, Coppersmith
et al. (2009) proposed a model to explain how a
greater orientational order develops over a distance
of several aragonite tablets. The model is based on
two assumptions: (1) well-oriented tablets grow
faster than misoriented ones; and (2) the crystal
orientation of a tablet from a given layer is highly
probable to be the same as that of the tablet directly
below its nucleation site. This latter assumption
reflects the presence of pores in the organic matrix
that allow some kind of 'communication' between
the successive aragonite tablet layers. In Satonda
stromatolites, the Mg - Si-rich layers separate the
aragonite layers. However, the map of the orien-
tation of aragonite c-axis (Fig. 6) and TEM images
do not show any 'transmission' of the crystal orien-
tation through the Mg - Si-rich layer. On the FIB foil
observed in the present study, only a few crystals
were found to have their c-axes oriented perpen-
dicular to the laminae, except aragonite fibres at
the top of the underlying laminae. Two issues
remain unresolved: (1) how aragonite crystals
become eventually oriented within an aragonite
laminae? And (2) why aragonite crystals in a
given cluster do not influence the orientation of

aragonite crystals forming on top? More obser-
vations of whole aragonite laminae, also involving
additional techniques that provide a larger scale
view of the crystallographic orientation of arago-
nite, will be needed in order to better understand
how the general orientation of aragonite crystals
eventually evolved towards a preferred crystallo-
graphic orientation of the fibres within a single
lamina.
Finally, the striations observed within the arago-
nite laminae and parallel to them remain unex-
plained. Texturally, they look similar to the
growth bands observed in corals (Fig. 2) and
hence there could be a general mechanism respon-
sible for their formation in stromatolites and
corals. This mechanism would be important to deci-
pher in order to understand whether it is related to
variations in the biological activity and/or
whether it is associated with some temporal period-
icity of the environment. However, we could not
detect in the Satonda stromatolites the significant
Mg and Sr variations observed in corals. The
meaning of such compositional variations is,
however, still discussed in the coral literature (e.g.
Meibom et al. 2004), and further studies detailing
the ultrastructural variations responsible for their
formation are required to better assess the simi-
larities and differences of these features in corals
and stromatolites. One possibility that will require
a more systematic investigation is that these rings
correspond to the 1 - 2 mm thick areas evidenced
by STXM in which the orientation of the aragonite
crystals is roughly the same (Fig. 6). They would
correspond to successive episodes of precipitation
with variations in the growth direction of aragonite
crystals.
In conclusion, the present study provides (1) a
new description of Satonda stromatolites at the
nm-scale, which is, in our opinion, an important
basis for future comparisons; and (2) a new meth-
odological framework offering an unprecedented
description down to the nm-scale of the mineralogy
and texture of carbonate minerals and that might be
of interest to the scientific communities studying
stromatolites, speleothems, travertines and tufas.
Finally, the present study proposes speculative yet
heuristic working hypotheses based on a tentative

Fig. 6. (Continued ) indicated by arrows. Brighter areas absorb more, darker areas absorb less. For a single pixel, the
variation of intensity between the different images is due to linear dichroism. Pixels showing similar variations have a
common orientation of the in-plane projection of the c-axis of aragonite. Dashed curves in (b) delimit areas in which
grains roughly show a consistent crystallographic orientation of the aragonite c-axis with significant deviations toward
the upper right corner. (h) Plot of the variation of absorption with the direction of the polarization vector for different
areas chosen in the FIB foil. The curves can be fit as A cos 2 (u2 f) þ B. The maximum of absorption gives the angular
direction of the c-axis of aragonite (f). Maximum absorption for areas 1, 2, 3, 4 are at 188,548, 1078 and 1558,
respectively. (i) Map of the orientation of the in-plane projection of the c-axis of aragonite in the FIB foil. Areas in black
do not contain aragonite and were not fitted. The spatial resolution (i.e. the size of the pixels that were fitted) of this
map is 50 nm.