Sedimentary Geology 198 (2007) 1 – 27

www.elsevier.com/locate/sedgeo

Grain morphology of heavy minerals from marine and continental

placer deposits, with special reference to Fe–Ti oxides
H.G. Dill
Federal Institute for Geosciences and Natural Resources, P.O. Box 510163, D-30631 Hannover, Germany
Received 9 June 2006; received in revised form 2 November 2006; accepted 10 November 2006

Abstract

The grain morphology of major heavy minerals from various placer-type mineral deposits (aeolian, beach, fluvial, alluvial,
colluvial, residual–eluvial) in Mongolia, USA, Vietnam, Latvia, Germany, Spain, Cyprus, South Africa and Nepal was studied by
means of SEM, conventional petrographic microscopy and visual inspection. Variation of grain morphology is first and foremost
controlled by the physical and chemical processes operative in the environment of deposition and to a lesser extent by the climatic
conditions. The sphericity of grains in the mineral assemblage strongly depends on the crystallographic system of the prevailing
heavy minerals. It improves as minerals such as spinel and garnet, the most common representatives of the cubic crystallographic
system, become more and more prominent among the heavy minerals in the placer material. Even if minerals belong to the same
crystallographic system there may be significant differences as to their roundness. Analysis of the grain morphology of garnet
shows an increase in angularity from the pyrope – through almandine-, grossularite- to the spessartite-enriched garnet solid solution
series in the placers under consideration. Spessartite-enriched garnet solid solution series are widespread in pegmatites and low-
grade metamorphic rocks where they normally form euhedral crystals. Such morphological differences in a solid solution series are
inherited from the source rock and are not modified by sedimentary processes in proximal placer deposits. Thus isometric minerals
such as spinel or garnet are not good marker minerals for depositional environments. Their strong points lie in the field of
provenance rather than environment analysis as far as grain morphology is concerned. For a refinement of the morphology-based
subdivision of depositional environments, particular groups of minerals have to be selected for the various types of placer deposits.
There is no universal mineral ubiquitous to all placers whose morphological changes and surface textures are a direct indicator of a
particular environment of deposition. Redox-sensitive minerals such as Fe–Ti compounds taking an intermediate position on the
Mohs's hardness scale and being fairly resistant to weathering have proven to be a good tool for the distal placer types, spanning
the full range from beach to fluvial placers, and including aeolian types. Ultrastable to stable mineral such as corundum sensu lato,
kyanite or zircon are more appropriate for a morphology-based environmental analysis in proximal types, covering the full
spectrum from fluvial to residual–eluvial and colluvial placer deposits.
© 2006 Elsevier B.V. All rights reserved.

Keywords: Placer types; Continental; Marine; Grain morphology; Climate; Environment of deposition

1. Introduction accessory minerals included in heavy mineral aggre-

The chemical composition of mineral grains, their
intergrowth with other minerals of similar size or with
E-mail address: dill@bgr.de.
0037-0738/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.sedgeo.2006.11.002
gates, may successfully be used to characterize heavy
mineral assemblages with respect to provenance and
supergene alteration (Morton, 1991; Dill, 1998; Morton
and Hallsworth, 1999; Dill et al., 2006, in press). Shape,
surface textures and the size of minerals in clastic
2 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
sediments are often as important as the chemical
composition of the mineral, itself. The typology of
minerals like zircon has successfully been used to draw
a more precise picture of the source rock in which they
have evolved (Pupin and Turco, 1981; Schäfer and Dörr,
1997). The roundness of apatite has been shown to be a
useful environmental indicator by Allen and Mange-
Rajetzky (1992) and Morton et al. (2003). The shape
and surface characteristics of platinum-group elements,
gold, and quartz grains examined with the scanning
electron microscope show transit- and transport-related
changes (Nikolayeva and Yablokova, 1989; Southam
and Beveridge, 1994) and the outward appearance of
particles of precious metals may also provide some
valuable evidence for exploration geologists and guide
them to the primary deposit of placer-type minerals
(Herail et al., 1989; Eyles, 1995; De Oliveira and De
Oliveira, 2000; Duk-Rodkin et al., 2001; Nakagawa
et al., 2005). Scanning electron microscopy (SEM) can
best broaden our knowledge of the morphological and
compositional variations of heavy minerals, supple-
menting conventional petrographic microscopy and
visual inspection under the stereomicroscope during
Fig. 1. Index map to show the position of placer deposits (full dots) referred to in this text and some major towns (full square).
heavy mineral analysis. This technique has proven to be
the most efficient tool for studying grain-surface
textures in a wide range of minerals (Henning and
Störr, 1986; Marshall, 1987; Trewin, 1988; Tucker,
2001). Scanning electron micrographs of heavy miner-
als from different types of placer deposits in continental
and marine environments are discussed in this paper
(Fig. 1). The impacts of the depositional environment,
transport and climate on the morphology of heavy
minerals are the focus of discussion during the present
study of grain morphology. It goes without saying that
morphology and texture as well as the intergrowths
among heavy minerals are also of utmost importance for
the industrial use of placer minerals. In particular, Fe–Ti
compounds such as ilmenite and Ti-bearing magnetite
rank very high among placer minerals of commercial
interest. Mineralogical association and textural relation-
ships, such as exsolution lamellae, grain size, inter-
growth of minerals and the quantity of certain trace
elements, e.g., Cr, V and Nb, are decisive for the final
industrial use of Ti minerals. They find application in a
wide range of branches and have thus been intensively
studied inter alia by Flinter (1959), Wort and Jones
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
(1980), Mücke and Bhadra Chaudhuri (1991) and
Klementová and Rieder (2004). Iron–titanium oxides
are ubiquitous in a great variety of environments of
deposition, and bivalent Fe makes them good redox-
sensitive marker minerals. In view of impact strength,
ilmenite takes an intermediate position on the Mohs's
hardness scale. It is more susceptible to mechanical
indentation and shaping than pure TiO2 modifications
such as rutile or anatase so that its variation in
morphology may be used as a yardstick for environ-
mental-related changes as demonstrated by various
examples in the succeeding paragraphs (Table 1).
The goal of the paper is to provide a morphology-
based scale that may assist in the interpretation of
sedimentary (palaeo)environments and to discriminate
the impact of crystallographic and climatic changes on
grain morphology. To get an idea of the various
depositional environments, the reader is provided with
a set of figures, showing, on one hand, the morphology
of grains, and on the other hand, the geomorphology of
modern placer environments referred to in this text.
2. Methodology
Samples from modern placer deposits were passed
through a sieve and the grain size fraction 63 to 300 μm
was used for follow-up heavy mineral analyses. In some
proximal placer types larger grain size intervals have
also been considered. Neither grinding in a crusher, nor
diluted HCl acid, was used to disintegrate rocks. Doing
so could spoil the result of heavy mineral analyses by
leaching phosphates and introducing man-made
changes to the grain surfaces. During routine analyses,
the heavy minerals (density > 2.9 g/cm3) were extracted
by means of Na polywolframate. After removal of iron
oxide coatings with Na dithionite, translucent heavy
minerals were mounted on glass disks using Canada
balsam and identified under the petrographic micro-
scope, considering between 200 and 300 grains per
sample for mineral analysis and between 100 and
200 grains for morphological studies. In case of gravelly
material and gemstones this number could not be
achieved and an estimate from 20 to 50 grains per
samples was considered. Examination of thin sections
and X-ray diffraction analysis of particles larger than
200 μm supplemented the petrographic studies. The
major tool for shape analysis is scanning electron
microscope (SEM), using an energy-dispersive system
(EDS) to assist in mineral identification and image
analysis for morphological studies. The classification
schemes by Pettijohn et al. (1973) provide adequate
measures to achieve the goal of this study in order to
3
describe the grain morphology for palaeogeographic
purposes (Table 1). In some cases, where SEM runs did
not create unambiguous results, electron microprobe
analyses (EMPA) with a CAMECA SX100 equipped
with five wavelength-dispersive spectrometers and a
Princeton Gamma Tech energy-dispersive system can
overcome these analytical shortcomings encountered at
the SEM (Dill et al., in press). For the presentation of
morphometric data, different ways have been put
forward in the literature, including triangular and pie
chart diagrams (Sneed and Folk, 1958; Benn and
Ballantyne, 1993). A combination of bar diagram and
table provides the best way for the reader to correlate the
mineralogical and sedimentological data obtained
during sedimentological analyses with the data from
literature and the resultant interpretation of the deposi-
tional environment (Table 1). The whole rock chemical
composition of grain size fractions was determined by
X-ray fluorescence in the conventional mode, using
borate beads as flux material. The precision of the
method may be assessed from the lower reporting limits
at 0.01 wt.% SiO2. Monitor samples and more than 50
certified reference materials (CRM) were used for the
correction procedures.
3. Geology, mineralogy and the depositional
environment of heavy mineral accumulations
3.1. Marine placer deposits along the coast of Latvia
3.1.1. Geology and mineralogy
Samples were taken along the western coast of Latvia
near Liepaja, Jurkalne and Ventspils (Fig. 2a,b). Another
sampling site is located at the Veczemji Cliff which
stretches along the eastern coast of the Riga Bight
(Fig. 2c,d).
The Veczemji Cliff near Salaca with its top about 4 m
above sea level, exposes sandstones of the Middle
Devonian Burtnieki Series (Fig. 2c). On a rather narrow
sandy beach, shoreline-parallel accumulations of red
garnet sands may easily be distinguished from the white
beach sand (Fig. 2d and Table 1). The Jurkalne beach
placers along the Kurzeme coast occur in a setting
similar in morphology but different in grain size
distribution. A thick wedge-shaped gravel lag demar-
cates the foot slope of a steep cliff (Fig. 2a,e). These
gravel bars are inclined towards the sea at low angle and
intersected, in places, by small rivulets (Fig. 2e). The
raised Litorina Sea coastal terrace is incised by several
V-shaped creeks with some of these channels carrying
water only episodically (Fig. 2a). The working areas near
Ventspils and Liepaja are characterized by wide sandy
 4
Table 1
Placer mineralization and grain morphology as a function of climate and the environment of deposition

H.G. Dill / Sedimentary Geology 198 (2007) 1–27

H.G. Dill / Sedimentary Geology 198 (2007) 1–27 5

(continued on next page)

 6
Table 1 (continued)

H.G. Dill / Sedimentary Geology 198 (2007) 1–27

The dark area in the bar diagram gives the overall range of roundness, and the shaded areas the roundness of Fe–Ti oxides under study in this paper. For location of each placer locality, see Fig. 1.
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27 7

Fig. 2. Coastal morphology and sites of heavy mineral accumulations. (a) Narrow strandplain with gravel bar (g) and cliff crossed by V-shaped
channel near Jurkalne caused by strong abrasion, Latvia (photo: Ulsts and Bulgakova, 1998). For scale, see the pathway along the channel that is
approximately 3 m in width. (b) Wide sandy strandplain terminated towards land by a belt of partly vegetated dunes (d). The coastal morphological
setting is a case in point for a shore zone in dynamic equilibrium near Liepaja, Latvia (photo: Ulsts and Bulgakova, 1998). For scale see houses in the
background. (c) Narrow sandy strandplain with a cliff carved out of Devonian sandstones at Veczemji Cliff, Latvia. For scale see trees on the
sandstone platform. (d) Close-up view of Veczemji Cliff with “red sand” (rs) accumulation enriched in garnet at MHWL. “WS” denotes clean white
beach sand. “SD” points to the Devonian sandstones. For scale see trees on the sandstone platform. (e) Gravel bars (g) inclined at low angle towards
the sea and intersected by a small creek near Jurkalne, Latvia. For scale see trees in the background. (f) Gravel bars (g) steeply inclined towards the sea
along the south coast of Cyprus. For scale see boats and houses in the background.
8 H.G. Dill / Sedimentary Geology 198 (2007) 1–27

beaches with several runnels, berms and spits, delineat-
ing previous stages of the Baltic Sea, and a conspicuous
belt of dunes bounding the beach on the landward side
(Fig. 2b). This forested abrasion–accumulation plain
originated from the Quaternary Baltic Ice Lake, a
previous stage during the evolution to the modern-day
Baltic Sea (Fig. 2a,b).
Analysis of heavy mineral samples from the sam-
pling sites along the SE coast of the Baltic Sea yielded
a variegated spectrum of minerals with dominantly

Fig. 3. SEM images of placer minerals from Latvia and Vietnam. (White spots in Fig. 3a and 3c are caused by sodium polywolframate/heavy liquid
precipitates). Scales in μm. (a) Titanomagnetite (subrounded to subangular) from a narrow gravelly strandplain at Jurkalne, Latvia. (b) Pseudorutile
(subrounded) from a sandy beach at Liepaja, Latvia. (c) Altered sphalerite (subangular to angular) at Jurkalne beach, Latvia. (d) Ilmenite (rounded to
well rounded) at a sandy beach in front of Veczemji Cliff of Riga Bight, Latvia. (e) Ruby (subrounded) from alluvial–fluvial deposits at Luc Yen,
Vietnam. (f) Ilmenite (rounded) from sand beach along the Gulf of Tonkin, Vietnam.
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
members of the garnet solid solution series (grossularite,
almandine, pyrope–almandine–spessartite). Next in
abundance among heavy minerals are zircon and mona-
zite. Among the opaque heavy minerals ilmenite, titani-
ferous magnetite and their alteration products leucoxene
and pseudorutile are widespread (Fig. 3a,b). As a rather
exotic heavy mineral, corroded grains of sphalerite have
been spotted in the samples from Jurkalne (Fig. 3c).
Geologists from the Geological Institute of the Univer-
sity of Latvia reported ilmenite contents averaging 20 to
30 kg/t, zircon averaging 0.9 to 2.6 kg/t and variable
contents of rutile and monazite in the range 0.1 to 1.2 kg/t
(University of Latvia, 1995). Morphological investi-
gations of mineral grains reveal a high sphericity and
roundness (Fig. 3d). All samples are well sorted.
3.1.2. Environment of deposition and source of heavy
minerals
Ulsts and Bulgakova (1998) have drawn a map of the
coastal zone of the Baltic Sea that can be used as a basis
for a coastal–morphological interpretation. The sam-
pling sites in the Jurkalne area and near Salaca are
interpreted as a microtidal, active abrasion coastal zones
shaped by storms on a periodic basis (Table 1). In the
sheltered Bight of Riga, visible accumulations of heavy
minerals is still going on in a microtidal environment. In
the Ventspils and Liepaja coastal zone a dynamic
equilibrium shore zone exists with accumulation of
heavy minerals by the longshore and transverse marine
drift and aeolian processes in the dune fields. The
longshore drift in the Baltic Sea is directed northward.
The coastal erosional forms and sediment accumulations
of heavy minerals mentioned above are representative of
the most recent stages of the evolution of the Baltic Sea,
that were named Litorina Sea (8 to 4 ka BP) and Limnea
Sea stages (< 4 ka BP) according to their marker fossils
(Kabaliene, 1997; Usaityte, 2000). The Litorina Sea was
a warmer sea of higher salinity as compared with the
present-day Baltic Sea (Yu, 2003). The entire coastal
system bearing the placer deposits under study is
classified as a microtidal wave-dominated littoral system.
The heavy minerals have been washed out of the
glacial and fluvio-glacial Quaternary deposits which acted
as an intermediate repository for the heavy minerals from
the Precambrian crystalline rocks of the Baltic Shield and,
to a lesser degree, from the Late Paleozoic bedrocks
exposed along the coast of the NE Baltic Sea. Ratas and
Kalm (2001) refer apatite, amphibole, pyroxene, zircon,
garnet, monazite, magnetite and ilmenite to the Precam-
brian crystalline basement and hematite, leucoxene,
sulfides and barite to local carbonaceous bedrocks of
early Paleozoic age. The extraordinary high contents of
9
ilmenite and garnet in the Quaternary marine deposits do
not cause very much of a surprise in view of the ilmenite
contents of as much as 11% and garnet of as much as 31%
in some glacial tills which act as a kind of a “prot-ore”
(predecessor of the “real ore”) for the marine placers.
3.2. Continental and marine placer deposits in northern
Vietnam
3.2.1. Geology and mineralogy
In northern Vietnam several placer deposits are
currently worked for ruby and sapphire in the Luc Yen
district, a high relief area underlain by marble and peg-
matites and covered with a thick grus-rich red regolith
(Hofmeister, 2001; Nhung and Truong, 2001) (Figs. 3e
and 4a,b). Favorable conditions for the development of
zircon and ilmenite placer deposits occur in the
northernmost part of the Gulf of Tonkin, in Halong
Bay (Pham, 1994; Bieu, 1998). Current exploration has
led to the delineation of 40 placer deposits measuring as
much as 10 km in length and 3 km in width with zircon,
rutile, monazite and ilmenite present at a grade of 15 to
300 kg/m3 (Fig. 3f).
3.2.2. Environment of deposition and source of heavy
minerals
The primary ruby deposits in the Sino-Vietnam terrane
of northern Vietnam evolved within a metamorphic zone
that formed in the Tertiary during a tectono-metamorphic
event, the large scale left-lateral movements along the
Alialao Shan–Red River shear zone. By contrast,
secondary gemstone deposits, running the gamut from
residual to proximal fluvial placer deposits, formed in a
wide range of Quaternary sediments. Heavy minerals
occur in cavities resulting from the solution of carbonate
minerals in the metacarbonates, as well as in fluvial
sediments that outcrop in valley terraces (Dill, 2005). The
placer system under consideration is transitional from
residual via alluvial into fluvial. The gem-quality
corundum minerals in the regolith strongly contrast with
the silicate and calcareous minerals preserved during
weathering of the bedrock underneath. Due to the
differences in hardness, the “softer” rock-forming miner-
als are rounded to subrounded (Fig. 4b), whereas the
placer gem minerals show a less perfect roundness than
the host silicates and carbonates (Fig. 4c).
The main factors that govern the formation of
coastal and offshore placers may be summarized as
follows (Bieu, 1998). Source rocks along the coastal
zone rich in Ti minerals and zircon were subjected to
strong weathering, as was the case with the gem-
bearing metacarbonates in the Luc Yen area (Fig. 4a,b).
10 H.G. Dill / Sedimentary Geology 198 (2007) 1–27

Fig. 4. Residual placer deposits in Vietnam and Malawi at outcrop. (a) Red weathering loam evolving from metacarbonates intruded by pegmatites
(Pg). Residual-type placer deposits with ruby contained in the upper part of the regolith at Luc Yen, Vietnam. (b) Grus-enriched regolith with
subangular lithoclasts at Luc Yen. (c) Angular ruby concentrated in nests in the regolith at Luc Yen. (d) Regolith developing from meta-ultrabasic
igneous rocks at Chimwadzulu Hill, Malawi. For grain morphology see also Fig. 10a, b.

Between 4000 and 2000 years BP, a tropical and
monsoonal climate sparked strong chemical weathering
in this region in SE Asia and led to a release of heavy
minerals from their source rocks. Titanium minerals
steadily increased through the various levels of chemical
weathering: source rocks (2–3 wt.% ilmenite), bauxitic
saprolite (6–7 wt.% ilmenite), horizons of tropical soils
(8–10 wt.% ilmenite) (Bieu, 1998). Ilmenite placers are
found near Precambrian granitoids and metamorphic
rocks, while zircon and zircon–ilmenite heavy mineral
assemblages are found in lithologies pierced by Triassic
extrusives. Heavy minerals accumulated in beach ridges
and sand bars along the coast by cyclic deposition and
erosion. During the Cenozoic the successive uplift and
denudation in the Red River Graben was accompanied by
a long-term subsidence in the Gulf of Tonkin (Tuyet et al.,
2004). Some ilmenite placers are found in marine sands
from 25 to 30 m depth and interpreted as drowned river
deposits (Bieu, 1998). The modern Ti-bearing beach
placers evolved in a mixed littoral environment (Xinh
et al., 1990). The littoral system in the Gulf of Tonkin
comprises tidal flats, marshes, tidal creeks and channels
(Mathers and Zalasiewicz, 1999). The fluvial-dominated
sector consists of meandering rivers and associated
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27 11

meander belts, floodplain, and fluvial terraces. The wave-
dominated sector in the southern part of the gulf sees
relatively strong summer monsoon winds and is morpho-
logically characterized by sandy spits and alternating
beach ridges and back marshes/swamps (Mathers et al.,
1996; Mathers and Zalasiewicz, 1999).
3.3. Marine placer deposits along the East Coast of the
USA
3.3.1. Geology and mineralogy
Sampling was undertaken along the coastal zone near
Charleston, South Carolina, USA, where placer deposits
with predominantly ilmenite, rutile, magnetite and
grossularite–almandine members of the garnet solid
solution series occur interbedded with arenaceous
sediments on a vast beach (Figs. 5a and 6a; Table 1).
Black sands show low-angle cross bedding and anti-
dunes near the mean high water line (MHW) in runnels
of welded berms (Fig. 5b,c).
served as an example for the so-called “drumstick
model”. The geomorphological setting of these sandy
beach deposits is a regressive, mixed energy barrier
island. Modern beach placer deposits supplied from
distant river sources, as well as local rivers, indicate an
increasing littoral drift during successive marine
transgressions. Placer development and preservation in
this high wave-energy active-margin setting are strongly
controlled by palaeo-shoreline orientation and littoral
cell bounding headlands (Peterson et al., 1987).
3.4. Marine placer deposits along the SW Coast of
Spain
3.4.1. Geology and mineralogy
The sampling sites are located in the Huelva coastal–
estuarine system which opens up towards the Gulf of
Cadiz, Spain (Ruiz et al., 1998). Ilmenite and magnetite
accumulated in black sands within a thick pile of well-
sorted quartz arenites but do not attain ore grade (Fig. 6b).

3.3.2. Environment of deposition and source of heavy 3.4.2. Environment of deposition and source of heavy
minerals minerals
Hayes (1994, 1996) has interpreted this coastal envi- The coastal–estuarine system is a macrotidal system
ronment of deposition as a barrier island system which with a mean tidal range of 2.2 m at the mouth of the

Fig. 5. Coastal morphology and heavy mineral accumulation along the East Coast of the USA. (a) False-color image showing the beach morphology
of Kiawah Island, a regressive, mixed-energy barrier island (photo: Hayes, 1999). (b) Beach placer with ilmenite and magnetite near dune sands. (c)
Antidunes near the mean high water line (MHW) with ilmenite and magnetite in runnels of a welded berm. Scales in cm.
12 H.G. Dill / Sedimentary Geology 198 (2007) 1–27

Fig. 6. SEM images of placer minerals from the USA, Spain, South Africa and Germany. (a) Ilmenite (well rounded) from a sandy beach near
Charleston/Kiawah Island, USA. For coastal morphology and sedimentary structures see Fig. 5. (b) Ilmenite (well rounded) from sandy beach
deposits of the Huelva estuary, Spain. (c) Ilmenite (rounded with pits) from sandy beach deposits at Brand-se-Baai, South Africa. (d) Rutile (rounded
to subrounded with pits) from sandy beach deposits at Brand-se-Baai, South Africa. (e) Ilmenite (subrounded to subangular) from a creek near Lerau-
Leuchtenberg, Germany. (f) Mn-Ilmenite/pyrophanite (subangular to angular) from a creek near Weißenstadt, Germany.

estuary (Davis et al., 2000). The estuarine portion of the most landward part. The tidal channels bearing the
Rio Tinto coastal–estuarine system is characterized by a heavy minerals delivered from the surrounding rocks
net of braided-river channels with gravel lags in the and tidal flats are floored by sand and mud.
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
3.5. Marine placer deposits along the NW coast of
South Africa
3.5.1. Geology and mineralogy
Marine placer deposits enriched mainly in zircon,
ilmenite and rutile are currently operated at Brand-se-
Baai about 250 km NNW of Cape Town, South Africa.
Heavy minerals are concentrated to economic grade
along a sandy beach in front of the rugged west coast
escarpment (De Villiers et al., 1999). Ilmenite and rutile
grains show a high degree of roundness, with some
bearing notches and etch pits (Fig. 6c,d).
3.5.2. Environment of deposition and source of heavy
minerals
Heavy minerals are still being supplemented by
intensive northward longshore drift of the Benguela
Current. They are delivered by raised palaeo-beach
terraces situated some tens of meters above the present-
day sea level which act as an intermediate storage for the
heavy minerals. In addition to the marine longshore drift
as the main driving force for the build-up of the placer
deposits, aeolian transport is also suspected to have been
involved in the concentration process of the heavy
minerals (De Villiers et al., 1999).
3.6. Continental placer deposits at the NW edge of the
Bohemian Massif, Germany
3.6.1. Geology and mineralogy
The NE Bavarian Basement is underlain by calcsi-
licates, amphibolites and gneisses which were intruded
by acidic igneous rocks of Late Carboniferous age.
Together with these acidic intrusions, aplites and
pegmatites of different grain size and mineral compo-
sition, were emplaced in the study area along the
Czech–German border (Dill, 1985). In the southern part
numerous swarms of quartz veins, some of them parallel
to aplitic dykes, occur. They closely resemble in their
outward appearance the “Great Bavarian Quartz Reef”
(“Pfahl”) (Siebel et al., 2005). The landscape of the
uplifted basement was shaped by the Cenozoic drainage
system that gave host to various types of small placers in
gorges and rivulets. Many of them were mined in the
past mainly for tin and gold, but can no longer operated
economically. Poorly shaped grains of ilmenite, cassit-
erite and garnets prevail among the heavy minerals in
these terrigenous placer sediments (Figs. 6e,f and 7a).
Titanium minerals occur mainly in aggregates called
“nigrine”. According to Clark (1993), “nigrine” (also:
Eisenrutil, iron-rutile) is a variety of rutile. Nigrine is
described as black ferruginous rutile that is homoge-
13
neous on a microscopic scale (Ramdohr, 1975). In the
mineral aggregates under study ilmenite is intimately
intergrown with rutile, peppered with inclusions of
ferrocolumbite, wolframite, pseudorutile and leucoxene,
locally replacing ilmenite and rutile from the edge or
along cracks.
3.6.2. Environment of deposition and source of heavy
minerals
The Cenozoic placers at the western edge of the
Bohemian Massif are supposed to have formed on a
low-relief landscape in a tropical wet–dry morphocli-
matic zone, that is characterized by landforms called
peneplains or etchplains by Buedel (1977). The
evolution of such low-relief landscapes with shallow
valleys, low terraces and gravel-topped interfluves was
debated in various textbooks and comprehensive papers
on geomorphology (Thomas, 1994; Twidale, 2002).
Climatic conditions like those postulated for the
deposition of the palaeoplacer in the NE Bavarian
Basement still exist today in tropical Africa (see Section
3.10) and in parts of northernmost Vietnam (see Section
3.2). Patchy heavy mineral accumulations around
Pleystein were shaped by periglacial solifluction
which reworked the Tertiary regolith. Third in the
succession of processes, the small creeks incised into
these solifluction terracettes reworked the Sn- and Ti-
bearing debris once again and provided a final shaping
to the mineral grains. The transport regime along the
slopes is characterized by mass flows that pass upslope
into solifluction/gelifluction sheets and soil creep.
Towards the thalweg, mass flows grade into coarse-
grained gravel deposits of low-sinuosity fluvial stream
channels. The lack of internal structures in the bedsets,
absence of cross-cutting bedforms, channels or scouring
and a coarsening-upward trend in the grain sizes suggest
deposition by a combination of fluvial bedload and
colluvial slope processes (Dill et al., 2006, in press). The
heavy minerals in the continental placers have been
derived from the various granites and their highly
differentiated pegmatitic and aplitic successors located
only a few kilometres away from the present-day fluvial
placers.
3.7. Continental placer deposits in the Asian Continental
Interior, Mongolia
3.7.1. Geology and mineralogy
Two types of placer deposits far away from each
other and very much different as to their mean grain size
and their topographic position are in the center of
interest during this study of the Asian Continental
14 H.G. Dill / Sedimentary Geology 198 (2007) 1–27

Fig. 7. SEM images of placer minerals from the Germany, Mongolia, Nepal and Cyprus. (a) Cassiterite (angular to very angular) from a creek near
Weißenstadt, Germany. (b) Ilmenite (“il” rounded) and sphene (“sp” subrounded) from aeolian sandsheets at Vostochnii, Mongolia. (c) Spessartite-
enriched garnet (angular to very angular/euhedral) from a creek from the north draining into the Katmandu Basin, Nepal. (d) Ilmenite–rutile–
pseudorutile–aggregates (“nigrine“) (subangular) from stream sediments in a creek draining into the Katmandu Basin, Nepal. (e) Titaniferous
magnetite (subrounded) from a mixed sandy to gravelly beach near Mari, Cyprus. (f) Titaniferous magnetite (subangular) from stream sediments in a
creek draining the southern part of the Troodos Ophiolite Complex at Kalavasos, Cyprus.

Interior, Mongolia. In the deposit at Shavaryn Tsaram discrete seams as much as 15 cm thick over- and
trenches have exposed matrix-supported conglomerates underlain by heavy mineral disseminations (Fig. 8).
with angular clasts of garnet up to 3 cm in diameter in a Iron–titanium oxides associated with these garnet
silty matrix. These clasts were concentrated in several aggregates are of smaller size and normally fall in the
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27 15

Fig. 8. Profile typical of a secondary gemstone deposit at Shavarym Tsaram, Mongolia. Various sheet-flood deposits, each of which starts off with heavy
minerals disseminated in a muddy matrix. Angular to subangular pyrope clasts are densely packed in placer seams (seams 1 and 2). Discrete pyrope
seams amenable to open cast mining were emplaced when the flow strength attained its sediment transport peak. Inset “a” shows an angular grain of
pyrope-enriched garnet. Inset “b” an angular grain of fayalite-enriched peridot/olivine from the placer deposit (particulate section – plane-polarized
light) and a chemically equivalent rounded mineral grain from the basanitic host rock some hundred metres away (“c”) (thin section-crossed polars).

range 63 to 300 μm. These beds abundant in heavy
minerals grade towards the trunk river into poorly sorted
arenaceous stream sediments with pyrope and olivine as
major heavy minerals (Table 1). These conglomeratic to
arenaceous heavy mineral deposits occur in the
mountain/ forest steppe in northern Central Mongolia.
They constitute a clastic apron around Upper Tertiary
and Holocene alkali basaltic volcanites in the Hangai
dome (Ufimetsev, 1990).
Another type of placer deposit is largely distributed
in the steppe sensu stricto, near Moltsog els, and
Vostochnii, Mongolia. Towards the south, a similar type
of placer deposit at Biluut uul and Elstein gol,
Mongolia, occurs in what is called the desert steppe.
The arenaceous sediments are medium-grained and
moderately well to well-sorted sands. The particles are
mostly well-rounded. Some grains did not achieve such
a perfect shape and have to be categorized as subangular
16 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
or even angular (Table 1). Among the heavy minerals,
green tourmaline, green amphibole, brown biotite, sphene
and zircon predominate. Ilmenite grains, when present,
show notches and indentations (Fig. 7b). Sporadically,
stibnite and cassiterite were spotted among the heavy
minerals (Table 1).
3.7.2. Environment of deposition and source of heavy
minerals
The source for the gravelly sediments at Shavaryn
Tsaram may easily be identified. It lies in the nearby
basanitic volcanic bedrock. Poor roundness of the
pyrope clasts is convincing evidence for the short
distance of transport (Fig. 8a). On the other hand, the
short distance of transport could not account for the
good sorting value observed in the clast community. A
homogeneous particle size like this was not brought
about by abrasion during the short-distance transport,
but inherited from the source rock where it reflects the
original grain size and shape during crystallization in
the magma. Peridot/olivine behaves in a different way.
It is crushed significantly during transport and its
original roundness worsened, owing to its low stability
against supergene alteration (Nickel, 1973) (Fig. 8b,c).
The lack of internal structures in the bedsets, such as
missing cross-cutting bedforms, channels or scouring
and a coarsening-upward grain size trend, suggest
deposition under conditions of sheet flows (Kor and
Cowell, 1998). These deposits fade out in low-
concentration unconfined flows and in arenaceous
fluvial deposits in a wide mountain valley extending at
high altitude off the primary volcanic-hosted gemstone
deposit.
There is no doubt that the heavy mineral accumula-
tions of the second type in dunes and flat-topped sand
sheets from the desert steppe and steppe sensu stricto
were caused by wind action. All aeolian landforms
described from the steppe sensu stricto and covered to
some extent by grassland belong to the group of
impeded dunes. According to Maninghetti and Carter
(1999), lower to intermediate wind strength may be
deduced from the ripple pattern of the aeolian deposits
under consideration in the Mongolian steppe. Parabolic
dunes evolve in the desert steppe near the desert, proper.
Several slipfaces along the windward side classify these
aeolian placer enrichments as compound barchanoid
ridge dunes (Embabi and Ashour, 1993).
Not all of the detrital material found in these aeolian
deposits achieved the perfect roundness indicative of
mature aeolian sands. The angularity of some grains is
indicative of a short distance of transport by flowing
water, e.g., Vostochnii, Mongolia.
3.8. High-altitude continental placer deposits from the
Himalaya in Nepal
3.8.1. Geology and mineralogy
The drainage pattern of the rivers in the northern half
of the Katmandu Valley and in the neighboring
mountainous region shows a gridiron drainage system
strongly controlled by NW–SE trending swells which
forced the trunk river to flow in a NW–SE direction.
The grain size in the channels is very large, the width of
the valleys, excluding the Katmandu Valley is very
narrow and their morphology mostly V-shaped. To-
wards the flanks of the hills and upstream the channel
deposits imperceptibly merge with poorly sorted gravel
(Dill et al., 2001). The principal heavy minerals found in
the study area in and around the Katmandu basin and the
Ganesh Himal are garnet, zircon and tourmaline of
moderate roundness (Fig. 7c and Table 1). Titanium
oxides are present in the terrigenous sediments,
although of very limited amount, and may be named
as “nigrine” from the compositional point of view (Fig.
7d). They closely resemble the Ti–Fe oxide aggregates
described in more detail from placer deposits in the
Pleystein area, Germany (see Section 3.6).
3.8.2. Environment of deposition and source of heavy
minerals
Ongoing uplift in the Higher Himalaya north of
Katmandu affected the entire Katmandu Valley and
caused a reorganization of the drainage pattern by the
growth of structural topography during the most recent
parts of the Cenozoic. It has triggered linear erosion into
the southern mountain range and the cross-cutting ridges
and as a consequence of this, fluvial megafans have
spread across the Indo-Gangetic plains (Gupta, 1997). In
the study area the narrow fluvial channel systems
became gradually choked with huge amounts of detritus
delivered by alluvial fans, landslides, and a steady talus
creep. As a consequence of these rapid and slow-motion
geomorphological processes, arenaceous stream sedi-
ments are very scarce in these braided- and straight-river
drainage systems. The heavy minerals reported from the
rivers in the preceding chapter may easily be correlated
with the rock-forming minerals in the neighboring
metamorphic and igneous rocks of the Higher Himalaya
(Table 1) – see Gansser (1964) for lithology. In the lower
reaches of the Nepalese drainage system gem-quality
minerals were not spotted throughout this campaign
although several mineralized sites have been recorded
from their catchment areas in the crystalline rocks of
the Higher Himalayas (Basset, 1979, 1985; Harding
and Scarratt, 1986). Gem-quality heavy minerals are
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
scavenged in the alluvial to colluvial aprons developing
downhill of the source rocks and were not delivered to
the lower reaches of the mountain fluvial drainage
system. A case in point is the Hyakule tourmaline deposit
where dislodged blocks from landslides and talus scree
are the sole indications of gemstone mineralization in the
area while the primary deposits in the paragneisses
metacarbonates, gneisses and pegmatites are concealed
and difficult to locate.
3.9. Continental and marine placer deposits along the
southern rim of the Troodos Massif, Cyprus
3.9.1. Geology and mineralogy
The Cretaceous Troodos Complex, Cyprus, is made
up of ultrabasic and basic intrusive rocks, mainly
harzburgite–lherzolite at depth, overlain by the
“Sheeted Dyke Complex”, and an eruptive series of
meta- and pillow-basalts on top (Greensmith, 1994;
Robertson and Xenophontos, 1997). At the eastern and
southern flanks, the Troodos Complex dips underneath
Cretaceous, Paleogene and Neogene marls, chalky
limestones and evaporites. Raised beaches and river
terraces developed during Pleistocene and Holocene
times and constitute the outermost part of the in situ
sedimentary cover of the Troodos ophiolite (Green-
smith, 1994). Heavy minerals samples were collected in
small sandy creeks in the environs of Kalavasos and
along the gravelly beach near Mari. Gravel bars stretch
along the coast and dip steeply towards the sea (Fig. 2f).
Both sampling sites closely resemble each other from
the mineralogical point of view and show a rather
monotonous mineral assemblage with magnetite pre-
vailing over Ti magnetite and different types of
amphibole (Fig. 7e,f and Table 1).
3.9.2. Environment of deposition and source of heavy
minerals
Heavy minerals at Kalavasos came to rest in a narrow
fluvial channel system which laterally grades into debris
flows delivered into the ephemeral drainage system
from the flanks of the Troodos Massif by mass wasting
processes. Less than 20 km from this site, black sands
accumulated on a narrow sandy wave-dominated beach
behind gravel bars. Most of the linear shoreline deposits
in this area are gravel berms with a steep seaward
dipping face in front of vast salt lakes lacking significant
arenaceous interbeds and attesting to a high energy
regime (Fig. 2f). Such gravel sheets have been produced
during transgression, where marine erosion at the base
of a shoreward-migrating gravel beach or spit has cut
into the lagoonal mud behind a barrier. Gravel spits and
17
barrier beaches on an incompetent sedimentary footwall
of coastal salinas do not promote the development of
fixed trapsites and are less favorable for heavy mineral
accumulation than pocket beaches and small gulfs
carved out of the Cretaceous chalky limestones.
The overall coarse grain size of lithoclasts has some
implications for the shape of the heavy minerals.
Constant milling and reworking among these “mill-
stones” in the coastal surge zone gave the grains of
titaniferous magnetite a rather good subrounded shape
when compared with titaniferous magnetite less than
20 km off the coast in the creeks (Table 1 and Fig. 7e).
There is no doubt that the bulk of the titaniferous
magnetite in the heavy mineral assemblages was
introduced to the Mediterranean littoral environments
via ephemeral streams from the basalts of the Troodos
Complex. Octahedral magnetite together with chalco-
pyrite from source basalts in the eruptive series of the
Troodos Complex at Stavrovouniou and magnetite from
the stream sediments are comparable in chemistry and
shape (Figs. 7f and 9a).
3.10. Continental placer deposits in the savanna of
southern Malawi
3.10.1. Geology and mineralogy
The major part of the area under consideration lies in
the tropical semiarid climatic zone (Zahn, 1991).
According to Sutherland (1985) the region of Blantyre
is located today in the morphogenetic region classified
as “humid tropical”. The lithology of Malawi is
characterized by Precambrian metamorphic and igneous
rocks which form part of the various orogens stretching
from Kenya towards Madagascar in eastern Africa. Rift-
related igneous activity from the late Paleozoic (Karroo
System) through the late Mesozoic (Chilwa Province)
has brought about a great variety of volcanic rocks with
a vast spectrum of rare minerals (Bloomfield et al.,
1966; Ministry of Forestry and Natural Resources,
1990; Coakley et al., 1991; Coakley and Mobbs, 2001).
Hypogene and supergene processes form the basis for
the numerous placer deposits with a variegated spectrum
of heavy minerals (Table 1). Titanium minerals and
corundum, in places of gem-quality, are of economic
importance for Malawi. The Mpyupyu Hill heavy
mineral sands with over 11.9 Mt grading 3.8 wt.%
ilmenite and 0.01 wt.% rutile at a depth of 10 m are
among the larger heavy mineral concentrations in
Malawi (Ministry of Mines and Natural Resources and
Environment, 2004). The heavy mineral sands deposited
at Salima, Unga River, Monkey Bay and Linthipe River
along the shoreline of Lake Malawi contain garnet,
18 H.G. Dill / Sedimentary Geology 198 (2007) 1–27

Fig. 9. SEM and BSE (back-scattered electron images) images of placer minerals from Cyprus, Malawi and Germany. (a) Titaniferous magnetite
octahedra (ti) surrounded by a corona of chalcopyrite (cu) in a quartz matrix hosted by metabasalt of the Troodos Complex (source rock) at
Stavrovouniou, Cyprus. (b) Ilmenite (subangular) in fluvio-deltaic deposits prograding into Lake Malawi, Malawi. (c) Almandine–pyrope–
grossularite garnet solid solution (subangular) in fluvio-deltaic deposits prograding into Lake Malawi, Malawi. (d) Titaniferous magnetite octahedra
(angular) at Finishi Village, Malawi. (e) Niobian rutile (dark) intergrown with ilmenite (bright) both of which are being replaced by pseudorutile along
cracks (BSE polished section) from Pleystein, Germany. (f) Spessartite-enriched garnet (subangular) from stream sediments in a creek near Lerau-
Leuchtenberg, Germany.

ilmenite, monazite, and rutile (Ministry of Energy and 700 Mt, grading an average of 5.6% heavy minerals and
Mining, 1997; Coakley and Mobbs, 2001) (Figs. 1 and at Mangochi at 800 Mt, grading an average of 6.0%
9b). The reserves at Salima–Chipoka are placed at heavy minerals. An area in the lower Shire River valley
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
was also subject to an extensive placer exploration
during the late 1950s. The Tengani ilmenite–rutile
deposit is located in the Nsanje District in southern
Malawi where an estimated 2.5 Mt of heavy mineral
sands grading 3% ilmenite and 300,000 t of HM grading
0.3% rutile were delineated (Ministry of Energy and
Mining, 1997).
Corundum is present in gem-quality sapphire and
ruby within the regolith on top of a series of ultramafic
bodies at Chimwadzulu Hill and as opaque crystals in
deeply weathered pegmatites which were intruded into
nepheline gneiss at Thambani, Mwanza (Figs. 4d and
10). The Ministry of Mines and Natural Resources and
Environment (2004) recorded a grade of 75.6 g/m3 for
the Chimwadzulu Hill gemstone deposit, and Yager
(2001) estimated its resources of ruby to be 247 kg.
Fig. 10. Very angular to angular rough stones of different varieties of
corundum from Malawi residual–alluvial placer deposits. Scale grid in
cm. (a) Dark gray and purple corundum (ruby) from Chimwadzulu
Hill. (b) Yellowish, greenish brown and colorless corundum (sapphire)
from Chimwadzulu Hill. (c) Rough opaque corundum from Thambani,
Mwanza deposit.
19
From the Thambani deposit that was worked intermit-
tently between 1923 and 1952, about 1923 short tons of
abrasive-quality corundum were exploited (Carter and
Bennett, 1973).
Placer deposits with kyanite occur at Kapiridimba
and with REE minerals at Kangankunde. Many
tributaries of the Shire River contain heavy mineral
accumulations mainly with monazite, pyroxene and
amphibole, but nowhere do they reach economic
concentrations. This is also true for members of the
spinel solid solution series found in the clastic apron
around the Finishi Village chalcedony deposit (Fig. 9d).
3.10.2. Environment of deposition and source of heavy
minerals
The morphodynamic evolution in the working area in
southern Malawi is controlled by two basic actions.
Both are responsible for the vast low-relief landscape of
the savanna and some residual hills, known as inselbergs
(Dill et al., 2005). Lowering of the basal surface or
planation and incision into the plains at various levels by
mass wasting and fluvial erosion occurred during the
Cenozoic. The Malawian savanna may be taken as a
textbook example to explain the evolution of all types of
continental placer deposits from the proximal residual
through the distal lacustrine placer deposits.
Residual–eluvial placers at Mwanza, Chimwadzulu
Hill, formed as meteoric waters attacked minerals for a
long period of time, reducing quantities of unstable
heavy minerals, such as Fe–Mg-bearing silicates, in the
uppermost part of the regolith and concentrating high-
unit value commodities like gemstones. Planation alone
did not bring this accumulation of heavy minerals into
the reaches of economic operations. Only where the
saprock was stripped of its overlying saprolite, residual
to eluvial placers of the most stable and most valuable
heavy minerals, as the unstable minerals were removed
by hill wash. Most minerals found at Chimwadzulu and
Thambani have sharp crystal facets ruling out any long-
lasting fluvial transport.
Colluvial placers occur near Tengani. These placers
are Quaternary in age (King, 1951) and grade downhill
into fluvial placer deposits near the Shire River.
Unconfined slope wash, such as on the flanks of the
Mulaka Hills, scavenged heavy minerals from heavy
mineral-bearing stringers in the Basement Complex.
The resulting placer is transitional from colluvial into
eluvial.
Fluvial placers across the peneplains in the Central
African savanna mainly contain titanium minerals in
distal stream flow deposits which resulted from
dewatering of the mass flows whereby the heavy
20 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
minerals were drained away. The bedded fluvial
deposits consist essentially of channel-fill coarse-
grained beds, thicker in the lower valley than in the
valley head. Fluvial heavy mineral concentration in the
savanna is mainly of point bar- and river bed-type where
they accumulated in lag gravel, near riffles and in
potholes (Dill et al., 2005).
Fluvio-deltaic (lacustrine) placers in siliciclastics laid
down in the area between Mpyupyu Hill and Lake
Chilwa and along the shorelines of Lake Malawi are the
product of a complex interplay between fluvial–deltaic
and lacustrine processes. Scanning electron micrographs
of placer minerals show a high angularity of mineral
grains. Transport and reworking were not strong enough
to give these grains a better shape and a smoother
surface. The grain analysis of placer deposits around the
lakes in Malawi demonstrates that the minerals were
directly derived from nearby source rocks with little
reworking along the beaches after deposition by fluvial
processes. The placer deposits reflect different stages of
delta progradation under the influence of wave action in
a lacustrine basin.
In Malawi, fluvial as well as the more basinward
fluvio-deltaic and lacustrine placer deposits have been
sourced from different provenance areas. Alkaline
igneous rocks have locally played a bigger role than
the metamorphic rocks of the Precambrian Basement
Complex. Rutile and ilmenite were strongly enriched in
alkali–granites, syenites and foyaites, whereas titano-
magnetite is the most common host of Ti in the
granulite–facies rocks. Potential alkaline igneous
source rocks of rutile and ilmenite, such as nepheline
syenites, not only formed during the Mesozoic, but also
during the Precambrian, and collectively named the
North Nyasa Alkaline Province (Eby et al., 1998). In the
locus typicus at Tengani, as well as around Lake
Malawi, the abundance of almandine and grossularite-
enriched garnets in the heavy mineral assemblages
attests to a metamorphic derivation of the heavy
minerals concentrated in these placer types. Gemolog-
ical studies of grossularite-enriched garnet from Malawi
by Henn (1992) lend support to this derivation. Placer
material such as gemstones, kyanite, zircon and
corundum in residual through alluvial placers was
derived from the mafic bedrock undergoing strong
desilication (Mercier et al., 1999). In corundum from
Chimwadzulu Hill as well as Thambani, diaspore may
be identified. Gibbsite and boehmite are common
constituents in bauxites and laterites converted into
diaspore when undergoing low-grade regional meta-
morphism. Gemstone mineralization took place under
granulite facies conditions at temperature of between
700 and 750 °C and pressure of between 8 and 10.5 kbar
(Malisa and Muhongo, 1990).
4. Discussion
4.1. Grain morphology and the environment of
deposition
4.1.1. The morphology of silicate minerals and Fe–Ti
oxides a tool for environment analysis
For many decades the use of surface textures and
roundness of mineral grains as indicators of depositional
environments has been debated (Marshall, 1987;
Trewin, 1988; Newsome and Ladd, 1999; Tucker,
2001). The basis for the study of roundness and
sphericity of quartz grains was laid by Powers (1953)
and Rittenhouse (1943) who used the visual roundness
method and visual sphericity charts for classification. A
quick look at Table 1 reveals conspicuous differences in
the morphology of heavy mineral grains and an
improving-upward trend in the overall roundness of
heavy mineral grains from residual to aeolian placer
deposits. The bars in the diagram, however, overlap
each other so that any discrimination of related
depositional environments based on roundness alone
would be fraught with some difficulties. The morpho-
logical parameters can, however, be used in connection
with other sedimentary textures and structures to draw a
more detailed picture of the depositional (palaeo)
environment, even on burial (Morton, 1984). In the
present study those heavy minerals strongly susceptible
to alteration during deep burial diagenesis, such as Fe–
Mg silicates or some phosphates, are excluded from
grain morphology analysis. In contrast to these minerals,
the morphology of Ti–Fe minerals discloses their
marine derivation even in early Paleozoic arenites, as
illustrated by the marine sandstones of Bale, Latvia
(Fig. 11). Fresh ilmenite shows a perfect roundness
typical of a nearshore marine environment – for
comparison of modern and palaeoplacers see Figs. 3d,
f, 6a,b and 11a). Even Fe–Ti leucoxene and aluminous
Fe–Ti leucoxene from the same stratigraphic unit have a
high degree of roundness and suggest a marine
derivation of the original mineral, albeit being strongly
corroded (Fig. 11b,c). Morphology-based environment
analysis in palaeoplacers no longer makes sense, when
the boundary into pure TiO2 is crossed. In Fig. 11d
aggregates of tabular anatase mark the final stage of Fe-
release and ilmenite dissolution on burial. Reasonable
results for morphological discrimination may be
achieved among pure TiO2 only for rutile since it is
virtually always detrital.
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27 21

Fig. 11. Alteration of Ti–Fe minerals in a marine palaeoplacer interbedded within early Paleozoic sandstones at Bale, Latvia (heavy minerals are
concentrated by heavy liquid separation, original shape highlighted by white stippled lines). (a) Well-rounded unaltered ilmenite (Fe–Ti). (b)
Rounded altered leucoxene (Ti < Ti). (c) Rounded altered aluminous leucoxene alongside well-rounded ilmenite (Ti ≪ Fe). (d) Aggregates of euhedral
tabular anatase (Ti).

4.1.2. Morphological variation of detrital Fe–Ti oxides
as a function of the depositional environment
Mechanical chipping and reworking creates well
rounded to rounded grains of Fe–Ti compounds
especially in the littoral and aeolian environments of
deposition. Some pits and pock marks in ilmenite grains
of aeolian origin may assist in discriminating aeolian and
littoral ilmenite (Figs. 6c,d and 7f). Even some rutile
grains associated with ilmenite show some pitting of
their surfaces, although less pronounced than ilmenite
(Fig. 6d). Mineralogical work on such sands revealed
that these microtextures of grains resulted from chemical
weathering. Asumadu et al. (1991) suggested that these
features resulted from the intrusion of sesquioxides
along cracks and etching along microfractures and
cleavage planes. By means of SEM analysis during
this study, some unidentified Fe–Al compounds were
detected in some specimens to fill in these notches of
ilmenite without any corrosion or replacement of the
ilmenite itself. It is this kind of aluminous leucoxene that
was also reported from titaniferous sandstones elsewhere
(Fig. 11c). Oxy-hydroxides of Al such as in corundum
minerals and Al silicates such as kyanite may reduce the
transparency of their host minerals. Diaspore was
determined in some of the grains from Malawi by XRD.
Grains deposited on sandy beaches fall into a very
high roundness category (Figs. 3d and 6a,b) that is also
well preserved even in Paleozoic sandstones (Fig. 11a).
Wave action causes grain breakage and abrasion so that
grains obtain their well-rounded shape during long-term
shoreline processes. Pure ilmenite has the highest
roundness score among all heavy minerals under
consideration and may be used as a yardstick when
discriminating depositional environments.
22 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
Some microtidal coastal zones with narrow strand-
plains, erosional cliffs and/ or gravelly beaches, such as
along the southern coast of Cyprus and some sites along
the Latvian coast of the Baltic Sea, have Ti–Fe
compounds of lower roundness than those observed
on sandy beaches (Table 1). The beach types identified
in the study area are either pure gravel beaches or mixed
sand-gravel beaches, using the field criteria published
by Jennings and Shulmeister (2002). Care must be
exercised when interpreting coastal environments based
on roundness values, because the overall grain size
variation of the host sediments plays a more significant
part during shaping of Ti–Fe minerals than the tidal
range in the depocentre.
Heavy minerals from fluvial environments of
deposition have a much poorer roundness than equiv-
alent grains from the aeolian and coastal environments.
Fluvial–lacustrine environments, although being very
much like a fluvial–marine environment with respect to
the hydrodynamic conditions of longshore drift and
wave action, are significantly poorer in their roundness
values. The sites selected in this paper for discussion are
located in the savanna or, in terms of geomorphology, in
a low-relief peneplain. It is in the nature of the savanna
to experience undernourishment of sand, which can be
exploited successfully only at a few places in the fluvial
drainage systems (Sukhoroslov and Yablokova, 2001).
Boulder fields, e.g., basal conglomerates display fitting
breccias and stonelines, are common in savanna or
peneplain drainage systems (Dill et al., 2005). What has
been recorded for the gravelly beaches in the previous
paragraph can be applied to some extent to the fluvial–
deltaic–(lacustrine) deposits in Malawi, where penepla-
nation is still going. It may also be discussed for part of
the fluvial deposits in Germany (Pleystein) where many
of the clasts have been inherited from Miocene deposits
that were undergoing a similar geomorphological
shaping on a peneplained palaeo-landscape.
Among the detrital opaque Fe–Ti oxides, titaniferous
magnetites are less susceptible to abrasion, so that to
achieve a perfect roundness equal as ilmenite takes
much longer than for ilmenite grains ((Fig. 6a, 7e,f and
9a)). These ferrous spinels are second to ilmenite in any
environment analysis and a more reliable tool when it
comes to the determination of provenance areas (Basu
and Molinaroli, 1991).
4.1.3. Morphological variation during formation of
pseudorutile and leucoxene
Pseudorutile, also called arizonite or hydroilmenite,
is still in dispute as to its status as a distinct mineral
phase (Grey and Reid, 1975; Frost et al., 1983; Grey
et al., 1983; Anand and Gilkes, 1985; Mücke and
Bhadra Chaudhuri (1991); Grey et al., 1994). The
impact of post-depositional weathering altering heavy-
mineral assemblages of near-shore marine units has
been discussed by Paine et al. (2005). According to their
results, some primary Fe–Ti minerals appear to have
altered to secondary minerals, such as pseudorutile and
anatase. Pseudorutile in the samples under study shows
a lower roundness than ilmenite and is supposed to have
been delivered into the marine basin. Back-scattered
images (BSE) of heavy minerals taken from the E, C and
B horizons of Baltic soils show a high angularity totally
different from the shape of heavy minerals encountered
along the coast of the Baltic Sea (Vareikiene and
Lehtonen, 2004). Ilmenite and Ti magnetite grains in the
marine sediments are devoid of any remarkable
alteration zones or corrosion pits and thus cannot be
held as a source for these Ti–Fe oxides of higher
angularity. Along with an increase in the Ti/(Ti + Fe)
ratios from Ilmenite (< 0.59) through pseudorutile (0.55
to 0.76), the angularity of grains increases (Dill et al., in
press).
Ilmenite is stable in the beach zone and does not
convert in any of these Fe–Ti compounds under
whatever climatic conditions. The co-existence of Ti–
Fe minerals and sphalerite demonstrates that in this
environment, despite the strong wave action, redox-
sensitive minerals are still well preserved. Pseudorutile
and leucoxene evolved in the provenance area under
more oxidizing supergene or hypogene conditions.
Their grain shape in the beach zone is inherited from
the source rock (Dill et al., in press). Ongoing reworking
of these Fe–Ti minerals, “softer” than ilmenite, does not
enhance the roundness of grains but dissolution of Fe
and Ti. This is corroborated by laboratory trials intended
to improve the recovery of Ti from beach sand
concentrate from Chatrapur, Orissa, India (Sasikumar
et al., 2004). It was observed that mechanical activation
significantly enhances the dissolution of both iron and
titanium in sulfuric acid. The activation energy for
leaching of iron was marginally higher than that of
titanium and both decreased with activation time.
4.1.4. Titanium–iron oxides vs. aluminium oxides and
silicates in proximal placer deposits
The poorest roundness ever observed occurs in
residual to eluvial placer deposits (Fig. 10a,b,c). Alluvial
to colluvial deposits are transitional into these poorly
rounded deposits. They may easily be distinguished from
aeolian and beach deposits, see Fe–Ti compounds in
Figs. 7d and 3b,d, but can no longer be distinguished
from fluvial deposits using Fe–Ti compounds. To draw a
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
boundary between residual–eluvial and alluvial placers
based on grain morphology non-ferrous stable and
ultrastable heavy minerals do much better than Fe–Ti
minerals such as ilmenite. Ruby, sapphire and subpre-
cious corundum from the above placer types are
significantly different with respect to their grain shape
(Figs. 6e and 10). Reworked corundum shows beveled
edges and a smoother surface than the rugged and well-
crystallized in situ corundum from the lower horizons of
the regolith. This is also valid for kyanite and zircon
which can also be held diagnostic to mark this boundary
between different proximal placer and residual deposits.
The most striking feature of the zircons, from, e.g.
Thambani is the absence or only minor development of
(110) prisms and the dominance of (112) pyramids
(Table 1). Some more isometric crystals of zircon show a
“soccer ball” morphology sensu Schaltegger et al.
(1999) with short prisms (110) and (121), (112), and
(011) pyramids. Zircon habits reflect inherited faces,
degree of resorption, and newly grown faces. The
frequent development of several pyramid faces (see
“soccer ball” zircon crystals), together with low
elongation of zircon crystals in Malawi, are indicative
of high-pressure zircon grains. These ultrastable miner-
als are ideal marker minerals not only for the boundary
between autochthonous and allochthonous heavy min-
eral deposits; they also enable source-rock analysis.
Zircon morphology can only be used for proximal
deposits in an area underlain by crystalline basement
rocks, since zircons are readily recycled from pre-
existing sandstones. In the savanna under study, which is
undernourished with respect to sand-sized material, this
does not cause any bother.
4.2. Grain morphology and climatic conditions
Considering the climatic conditions in the study areas,
a clear trend may not be observed among the heavy
minerals with respect to their morphology (Table 1).
Most tropical heavy mineral assemblages score a poorer
roundness value than heavy mineral assemblages from
the temperate morphoclimatic zones, where chemical
weathering is rather moderate relative to the mechanical
impact during fluvial transport and abrasion in the surge
zone of the beaches. The stability of heavy minerals is
crucial as to which degree of roundness may be achieved
in continental environments (Nickel, 1973; Morton,
1984). If Fe–Mg silicates such as pyroxene, amphibole
or phosphates such as apatite or monazite prevail among
the heavy mineral assemblages, under tropical condi-
tions the mineral grains will take a poorer roundness than
under temperate conditions where dissolution is of lesser
23
intensity (Berner et al., 1980). Fayalite-enriched peridot/
olivine in alluvial–colluvial placers, e.g. Shavaryn
Tsaram, Mongolia, is converted into grains of high
angularity and is significantly modified by short-
distance transport and brief residence times in the
sediment, even if the primary mineral in the source
rock is rather well rounded (Fig. 8b,c). This is in
accordance with the order of stability drawn up by Nickel
(1973).
In tropical and subtropical climatic zones supergene
alteration may create skeletal grains and enable SEM
morphology of heavy minerals to act as an indicator of
the rate of weathering (Migon and Thomas, 2002;
Martin-Garcia et al., 2004). Pseudorutile, and to a lesser
degree sphalerite, are representative of such extreme
stages of alteration where only skeletal grains or some
kind of metastasis within ilmenite were left over (Fig.
9e). Replacement textures such as those shown in Fig.
9e do not reflect harsh chemical weathering of the
present-day temperate climate in Europe but have to be
interpreted in terms of palaeoclimatic conditions during
the Late Tertiary (Louis, 1984; Borger et al., 1993).
The climatic influence on the surface texture and
roundness ranks second in relation to the overall
depositional processes in the depocentre. Corrosion
textures such as etch pits, facets, ragged edges and
hacksaw terminations often may have also resulted from
hypogene processes. Such altered minerals are inherited
into the heavy mineral association, as it is the case with
the ilmenite–pseudorutile intergrowth shown in Fig. 7d.
Therefore, care should be taken in interpreting poor
roundness and facets as indicative of dissolution, as
these features can be inherited from the source rock, as it
is the case with its sphericity which is inherent to the
mineral itself.
Source control is the main reason that rock-forming
minerals in some residual placers of tropical morphocli-
matic zones are very much different in roundness from
their placer-type minerals. Chemical weathering
smoothes and bevels edges of rock-forming minerals
in metacarbonate and in meta-ultrabasic rocks (Fig. 4b,
c). Corundum minerals from residual and eluvial
placers, however, are well preserved as angular
fragments even under harsh tropical weathering condi-
tions, irrespective of the bedrock. Corundum occurs in
the regolith by inheritance from the parent rocks. It
significantly alters during long-term exposure under
surface conditions and yields secondary minerals such
as kaolinite, gibbsite and diaspore depending on the
geochemistry of the soil environment. Corundum
minerals may be used to study the effect of weathering
on transported materials given that there is a direct
24 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
correlation with corundum from the source area to see
what the primary minerals morphologically looked like
(Fig. 3e).
4.3. Grain morphology and crystallographic properties
Not surprisingly, heavy mineral assemblages abun-
dant in spinel and garnet, both of which pertain to the
cubic crystallographic system, are characterized by a
high sphericity of grains.
Conventional heavy mineral data obtained during
investigations with the polarizing microscope cannot
answer the question why minerals equal in sphericity
like garnet differ greatly in roundness. The data obtained
from SEM show an increase in angularity from pyrope-,
almandine-, grossularite- to spessartite-enriched garnet
solid solutions. These morphological changes in the
garnet solid solutions originated from evolution of
garnet in the source rock. In this study, pyrope-enriched
garnet evolved from basic mantle-magma (Mongolia) or
during high-grade regional metamorphism, locally,
attaining granulite facies conditions (Scandinavia,
Malawi). Almandine may be correlated with metapelitic
rocks (Scandinavia, Germany) and grossularite is
present in contact metamorphic rocks (Germany,
Nepal) while spessartite is widespread in low-grade
metamorphic rocks as well as pegmatites (Germany,
Nepal). Recycling and recrystallization go through a
minimum in the pegmatitic stage and the preservation
potential of euhedral mineral is high. Spessartite with
well-crystallized faces and a high degree of angularity in
modern detrital sediments may thus be used as a
potential pathfinder to acidic or pegmatitic intrusions
(Figs. 7c and 9f). Among the Fe–Ti compounds under
study, these crystallographic-related differences may
also be seen. Titaniferous magnetite is more angular
than ilmenite under similar depositional conditions, but
because of its cubic lattice, the sphericity of titaniferous
magnetite in heavy mineral assemblages is higher than
that of ilmenite (Table 1).
5. Conclusions
Grain morphology of heavy minerals in modern
placer deposits is predominantly controlled by the
depositional environment and to a lesser extent
influenced by the climatic conditions. To refine
subdivision of the host environments of deposition
based on grain morphology, selection of a particular
group of minerals is recommended for each of the
various categories of placer deposits. There is no
universal mineral such as quartz ubiquitous to most
lithologies and doing well from the residual to the beach
placers. Redox-sensitive minerals, such as Fe–Ti
compounds of intermediate hardness and resistance to
weathering, have proven to be a good tool for the distal
placer types from beach placers to fluvial placers,
including aeolian types, and have proven the viability of
this method also for sandstones as old as early
Paleozoic. Ultrastable to stable mineral, such as
corundum sensu lato, kyanite or zircon, is more
appropriate in proximal placer types covering the full
spectrum from fluvial to residual placers. Isometric
minerals such as spinel or garnet do not do very well as
marker minerals for depositional environments. Their
strong points lie in the field of provenance analysis.
Deep burial and hydrothermal processes may modify the
shape of detrital heavy minerals and cause new heavy
minerals of high angularity to form.
Acknowledgement
I would like to express my thanks to my colleagues
R.-R. Ludwig (Federal Institute for Geosciences and
Natural Resources, Hannover), A. Kathewera and J.
Mwenelupembe (Geological Survey, Zomba, Malawi),
G. Stinkulis (University of Latvia, Riga), S. Khishigsu-
ren, Yo. Majigsuren and J. Bulgamaa (Technical
University of Ulaanbataar, Mongolia), D.R Khadka
and R. Khanal (Geological Survey of Nepal) and several
others colleagues, not named explicitly, who provided
assistance during sampling in Germany, Spain, USA,
Vietnam, Cyprus and South Africa. The sedimentolog-
ical preparation of samples in the laboratory and
quantification of some heavy mineral samples was
performed by I. Bitz. D. Klosa performed numerous
SEM analysis in the laboratories of the Federal Institute
for Geosciences and Natural Resources. S. Kaufhold has
supervised the XRD analyses. One EMP image and its
pertinent data have been provided by F. Melcher and J.
Lodziak who conducted the microprobe analyses. The
paper was reviewed by two anonymous referees for
SEDIMENTARY GEOLOGY. Their comments and the
editorial handling of the paper by K.A.W. Crook are
acknowledged with thanks.
References
Allen, P.A., Mange-Rajetzky, M.A., 1992. Sedimentary evolution of
the Devonian–Carboniferous Clair Field, offshore northwestern
UK: impact of changing provenance. Marine and Petroleum
Geology 9, 29–52.
Anand, R.R., Gilkes, R.J., 1985. Some alumina and silica in weathered
ilmenite grains is present in clay minerals—a response to Frost
et al. (1983). Mineralogical Magazine 49, 141–145.
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
Asumadu, K., Churchward, H.M., Gilkes, R.J., 1991. The origins of
surficial sands within highly weathered terrain in S.W. Australia.
Australian Journal of Earth Sciences 38, 45–54.
Basset, A.M., 1979. Hunting for gemstones in the Himalayas of Nepal.
Lapidary Journal 33, 1492–1520.
Basset, A.M., 1985. The tourmalines of Nepal. Mineralogical Record
16, 413–418.
Basu, A., Molinaroli, E., 1991. Reliability and application of detrital
opaque Fe–Ti oxide minerals in provenance determination. In:
Morton, A.C., Todd, S.P., Haughton, P.D.W. (Eds.), Developments
in Sedimentary Provenance Studies. Geological Society of
London, Special Publication, vol. 57, pp. 55–65.
Benn, D.I., Ballantyne, C.K., 1993. The description and representation
of particle shape. Earth Surface Processes and Landforms 18,
665–672.
Berner, R.A., Sjöberg, E.L., Velbel, M.A., Krom, M.D., 1980.
Dissolution of pyroxenes and amphiboles during weathering.
Science 207, 1205–1206.
Bieu, N., 1998. Factors governing the formation of the coastal and
offshore placers of Vietnam. Proceedings of the thirty-third CCOP
Annual Session in Shanghai, China, 30 October–2 November
1996. Coordinating Committee for Coastal and Offshore Geosci-
ence Programmes in East and Southeast Asia (CCOP): 2. Technical
reports. Bangkok, 259–266.
Bloomfield, K., Gondwe, P.C., Mason, T.P.R., Tchuwa, K.A.,
Habgood, F., Phiri, F.R., 1966. Geological Map of Malawi.–1:
1,000,000, Survey Department, Blantyre.
Borger, H., Burger, D., Kubiniok, J., 1993. Verwitterungsprozesse und
deren Wandel im Zeitraum Tertiär – Quartär. Zeitschrift für
Geomorphologie N.F 37, 129–143.
Buedel, J., 1977. Klima–Geomorphologie. Gebrüder Bornträger
Berlin, Stuttgart. 304 pp.
Carter, G.S., Bennett, J.D., 1973. The geology and mineral resources
of Malawi. Geological Survey Bulletin 6, 1–62.
Clark, H., 1993. Hey's Mineral Index. Mineral Species, Varieties and
Synonyms, 3rd edition. Chapman and Hall.
Coakley, G.J., Mobbs, P.M., 2001. The mineral industry of Malawi.
Mineral Industries of Africa and the Middle East. US Geol. Survey
Minerals Yearbook, vol. 1998-3, pp. 29.1–29.2.
Coakley, G.J., Crockett, R.N., Hammerbeck, E.C.I., 1991. Interna-
tional Strategic Minerals Inventory Summary Report: A Regional
Assessment of Selected Mineral Commodities in Subequatorial
Africa. Bureau of Mines, Washington D.C. 44 pp.
Davis Jr., R.A., Welty, A.T., Borrego, J., Morales, J.A., Pendon, G.,
Ryan, J.G., 2000. Rio Tinto estuary (Spain): 5000 years of
pollution. Environmental Geology 39, 1107–1116.
De Oliveira, S.M.B., De Oliveira, N.M., 2000. The morphology of
gold grains associated with oxidation of sulphide-bearing quartz
veins at São Bartolomeu, central Brazil. Journal of South American
Earth Sciences 13, 217–224.
De Villiers, J., Van Rooyen, M.W., Theron, G.K., Van Rooyen, N.,
1999. Vegetation diversity of the Brand-se-Baai coastal dune area,
west coast, South Africa: a pre-mining benchmark survey for
rehabilitation. Land Degradation and Development 10, 207–224.
Dill, H.G., 1985. Die Vererzung am Westrand der Böhmischen
Masse.–Metallogenese in einer ensialischen Orogenzone. Geolo-
gisches Jahrbuch D 73, 3–461.
Dill, H.G., 1998. A review of heavy minerals in clastic sediments with
case studies from the alluvial-fan through the nearshore-marine
environments. Earth Science Review 45, 103–132.
Dill, H.G., 2005. Lithofacies terrain analysis in search of continental
placer deposits—with special reference to gemstone placers and
25
an outlook on near-surface deposits in Vietnam. In: Hofmeister,
W., Quang, V.X., Nghi, T., Dao, N.Q., Khoi, N.T., Haeger, T.
(Eds.), Gem Materials and Modern Analytical Methods, Hanoi,
pp. 39–54.
Dill, H.G., Kharel, B.D., Singh, V.K., Piya, B., Busch, K., Geyh, M.,
2001. Sedimentology and paleogeographic evolution of the
intermontane Katmandu Basin, Nepal, during the Pliocene and
Quaternary. Implications for formation of deposits of economic
interest. Journal for Asian Earth Sciences 19, 777–804.
Dill, H.G., Ludwig, R.-.R., Kathewera, A., Mwenelupembe, J.A.,
2005. Lithofacies terrain model for the Blantyre Region: implica-
tions for the interpretation of palaeosavanna depositional systems
and for Environmental Geology and economic geology in southern
Malawi. Journal of African Earth Sciences 41, 341–393.
Dill, H.G., Melcher, F., Füssl, M., Weber, B., 2006. Accessory
minerals in cassiterite: a tool for provenance and environmental
analyses of colluvial–fluvial placer deposits (NE Bavaria,
Germany). Sedimentary Geology 191, 171–189.
Dill, H.G., Melcher, F., Füssl, M., Weber, B., in press. The origin of
rutile–ilmenite aggregates (“nigrine”) in alluvial–fluvial placers of
the Hagendorf pegmatite province, NE Bavaria, Germany.
Mineralogy and Petrology.
Duk-Rodkin, A., Barendregt, R.W., White, J.M., Singhroy, V.H., 2001.
Geologic evolution of the Yukon River: implications for placer
gold. Quaternary International 82, 5–31.
Eby, G.N., Woolley, A.R., Din, V., Platt, G., 1998. Geochemistry and
petrogenesis of nepheline syenites: Kasungu-Chipala, Ilomba, and
Ulindi nepheline syenite intrusions, North Nyasa alkaline
province. Journal of Petrology 39, 1405–1424.
Embabi, N.S., Ashour, M.M., 1993. Barchan dunes in Qatar. Journal of
Arid Environments 25, 49–69.
Eyles, N., 1995. Characteristics and origin of coarse gold in Late
Pleistocene sediments of the Cariboo placer mining district, British
Columbia. Canada Sedimentary Geology 95, 69–95.
Flinter, B.H., 1959. The alteration of Malayan ilmenite grains and the
question of “Arizonite”. Economic Geology 54, 720–729.
Frost, M.T., Grey, I.E., Harrowfield, I.R., Mason, K., 1983. The
dependence of alumina and silica contents on the extent of
alteration of weathered ilmenites from Western Australia.
Mineralogical Magazine 47, 201–208.
Gansser, A., 1964. Geology of the Himalayas. Whiley, London.
289 pp.
Greensmith, T., 1994. Southern Cyprus. The Geological Association,
Guide, 50, pp. 1–146.
Grey, I.E., Reid, A.F., 1975. The structure of pseudorutile and its role
in the natural alteration of ilmenite. American Mineralogist 60,
898–906.
Grey, I.E., Li, C., Watts, J.A., 1983. Hydrothermal synthesis of
goethite–rutile intergrowth structures and their relationship to
pseudorutile. American Mineralogist 68, 981–988.
Grey, I.E., Watts, JA., Bayliss, P., 1994. Mineralogical nomenclature:
pseudorutile revalidated and neotype given. Mineralogical Mag-
azine 58, 597–600.
Gupta, S., 1997. Himalayan drainage patterns and the origin of fluvial
megafans in the Ganges foreland basin. Geology 25, 11–14.
Harding, R.R., Scarratt, K., 1986. A description of ruby from Nepal.
Journal of Gemmology 20, 3–10.
Hayes, M.O., 1994. Georgia Bight. In: Davis Jr., R.A. (Ed.) Barrier
Islands, Springer, New York, pp. 281–291, 295–304.
Hayes, M.O., 1996. Barrier Island basics. Geotimes 1996, 17.
Hayes, M.O., 1999. Modern Clastic Depositional environments—
Lecture Notes Research, Inc. Columbia, South Carolina. 10 pp.
26 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
Henn, U., 1992. Granat aus Malawi mit ungewöhnlichen Einschlüssen.
Zeitschrift der deutschen gemmologischen Gesellschaft 41, 17–19.
Henning, K.-.H., Störr, M., 1986. Electron Micrographs (TEM, SEM)
of Clays and Clay Minerals. Akademie-Verlag, Berlin. 350 pp.
Herail, G., Fornari, M., Rouhier, M., 1989. Geomorphological control
of gold distribution and gold particle evolution in glacial and
fluvioglacial placers of the Ancocala–Ananea basin—Southeast-
ern Andes of Peru. Geomorphology 2, 369–383.
Hofmeister, W., 2001. Modelling some mineralizations of typical
Vietnamese gem-deposits. In: Hofmeister, W., Dao, N.Q., Quang,
V.X. (Eds.), Material Characterisation by Solid State Spectrosco-
py: The Minerals of Vietnam, pp. 327–328. Hanoi.
Jennings, R., Shulmeister, J., 2002. A field based classification scheme
for gravel beaches. Marine Geology 186, 211–228.
Kabaliene, M., 1997. Shore line displacement, palaecological condi-
tions and human impact on the southeastern coast of the Baltic Sea.
In: Grigelis, A. (Ed.), The Fifth Marine Geological Conference
“The Baltic” Oct. 6 – 11, 1997 Vilnius, Lithuania, Abstr. Excursion
guide, pp. 114–122. Vilnius.
King, L.C., 1951. South African Scenery, 2nd edition. Oliver and
Boyd, Edinburg. 379 pp.
Klementová, M., Rieder, M., 2004. Exsolution in Niobian rutile from
the pegmatite deposit at Greenbushes, Australia. Canadian
Mineralogist 42, 1859–1870.
Kor, P.S.G., Cowell, D.W., 1998. Evidence for catastrophic subglacial
meltwater sheetflood events on the Bruce Peninsula, Ontario.
Canadian Journal of Earth Sciences 35, 1180–1202.
Louis, H., 1984. Zur Reliefentwicklung der Oberpfalz. Relief, Boden,
Paläoklima 3, 1–66.
Malisa, E., Muhongo, S., 1990. Tectonic setting of gemstone
mineralization in the Proterozoic metamorphic terrane of the
Mozambique belt in Tanzania. Precambrian Research 46, 167–176.
Maninghetti, B., Carter, L., 1999. Across-shelf sediment dispersal,
Hauraki Gulf, New Zealand. Marine Geology 160, 271–300.
Marshall, J.R., 1987. Clastic particles. Scanning Electron Microscopy
and Shape Analysis of Sedimentary and Volcanic Clasts. Van
Nostrand Reinhold, New York. 346 pp.
Martin-Garcia, J.M., Aranda, V., Gamiz, E., Bech, J., Delgado, R.,
2004. Are Mediterranean mountains Entisols weakly developed?
The case of Orthents from Sierra Nevada (Southern Spain).
Geoderma 118, 115–131.
Mathers, S.J., Zalasiewicz, J.A., 1999. Holocene sedimentary
architecture of the Red River delta, Vietnam. Journal of Coastal
Research 15, 314–325.
Mathers, S.J., Davies, J., McDonald, A., Zalasiewicz, J.A., Marsh, S.,
1996. The Red River delta of Vietnam. British Geological Survey
Technical Report WC/96/02. 41 pp.
Mercier, A., Debat, P., Saul, J.M., 1999. Exotic origin of the ruby
deposits of the Mangari area in SE Kenya. Ore Geology Reviews
14, 83–104.
Migon, P., Thomas, M.F., 2002. Grus weathering mantles—problems
of interpretation. Catena 49, 5–24.
Ministry of Energy and Mining, 1997. Mineral Potential of Malawi—
Opportunities for Investment. Ministry of Energy and Mines,
Lilongwe, Malawi. 14 pp.
Ministry of Forestry and Natural Resources, 1990. The Mineral
Resources of Malawi: New Opportunities for Investors. Malamulo
Publishing House, Makwasa, Lilongwe. 29 pp.
Ministry of Mines and Natural Resources and Environment, 2004.
Mineral Potential of Malawi—Opportunities for Mineral Invest-
ment. Ministry of Mines and Natural Resources and Environment,
Lilongwe. 24 pp.
Morton, A.C., 1984. Stability of detrital heavy-minerals in Tertiary
Sandstones from the North Sea Basin. Clay Minerals 19, 287–308.
Morton, A.C., 1991. Geochemical studies of detrital heavy minerals
and their application to provenance research. In: Morton, A.C.,
Todd, S.P., Haughton, P.D.W. (Eds.), Developments in Sedimen-
tary Provenance Studies. Geological Society of London, Special
Publication, vol. 57, pp. 31–46.
Morton, A.C., Hallsworth, C.R., 1999. Processes controlling the
composition of heavy mineral assemblages in sandstones.
Sedimentary Geology 124, 3–29.
Morton, A.C., Spicer, P.J., Ewen, D.F., 2003. Geosteering of high-
angle wells using heavy-mineral analysis: the Clair Field, west
of Shetland, UK. In: Carr, T.R., Mason, E.P., Feazel, C.T.
(Eds.), Horizontal Wells: Focus on the Reservoir. American
Association of Petroleum Geologists, Methods in Exploration,
vol. 14, pp. 249–260.
Mücke, A., Bhadra Chaudhuri, J.N., 1991. The continuous alteration
of ilmenite through pseudorutile to leucoxene. Ore Geology
Reviews 6, 25–44.
Nakagawa, M., Santosh, M., Nambiar, C.G., Matsubara, C., 2005.
Morphology and chemistry of placer gold from Attappadi Valley,
Southern India. Gondwana Research 8, 213–222.
Newsome, D., Ladd, P., 1999. The use of quartz grain microtextures in
the study of the origin of sand terrains in Western Australia. Catena
35, 1–17.
Nhung, N.T., Truong, N.N., 2001. The dependence of gem quality on
the forming conditions in the Yen Bai area. In: Hofmeister, W.,
Dao, N.Q., Quang, V.X. (Eds.), Material Characterisation by Solid
State Spectroscopy: The Minerals of Vietnam, pp. 327–328.
Hanoi.
Nickel, E., 1973. Experimental dissolution of light and heavy-minerals
in comparison with weathering and intrastratal solution. Contribu-
tions to Sedimentology 1, 1–68.
Nikolayeva, L.A., Yablokova, S.V., 1989. Specific features of native
gold from the zone of hypergenesis. Mineralogy, Reports of Soviet
Geologists at the 28th TGC, Washington, July 1989, pp. 178–183.
Paine, M.D., Anand, R.R., Aspandiar, M., Fitzpatrick, R.R., Verrall,
M.R., 2005. Quantitative heavy-mineral analysis of a Pliocene
beach placer deposit in southeastern Australia using the Auto-
GeoSEM. Journal of Sedimentary Research 75, 742–759.
Peterson, C.D., Gleeson, G.W., Wetzel, N., 1987. Stratigraphic
development, mineral sources and preservation of marine placers
from Pleistocene terraces in Southern Oregon, USA. Sedimentary
Geology 53, 203–229.
Pettijohn, F.J., Potter, P.E., Siever, R., 1973. Sand and Sandstone.
Springer, New York. 618 pp.
Pham, V.M., 1994. Titanium and coastal placers of Viet Nam. Joint
Prospecting for Mineral Resources in Asian Offshore Areas,
Annual Session 9.11.93 to 13.11.93, pp. 267–285.
Powers, M.C., 1953. A new roundness scale for sedimentary particles.
Journal of Sedimentary Petrology 23, 117–119.
Pupin, J.P., Turco, G., 1981. Le zircon, minéral commun significatif des
roches endogènes et exogènes. Bulletin e Minéralogie 104, 724–731.
Ramdohr, P., 1975. Die Erzminerale und ihre Verwachsungen.
Akademie-Verlag, Berlin. 1277 pp.
Ratas, M., Kalm, V., 2001. Lithostratigraphy and distribution of tills in
the Saadjärve drumlin field, east-central Estonia. Proceedings of
the Estonian Academy of Sciences. Geology 50, 24–42.
Rittenhouse, G., 1943. A visual method of estimating two-dimensional
sphericity. Journal of Sedimentary Petrology 13, 79–81.
Robertson, A.H.F., Xenophontos, C., 1997. Cyprus. In: Moores
Eldridge, M., Fairbridge Rhodes, W. (Eds.), Encyclopedia of
 H.G. Dill / Sedimentary Geology 198 (2007) 1–27
European and Asian Regional Geology. Chapman & Hall, London,
United Kingdom, pp. 160–171.
Ruiz, F., Gonzalez-Regalado, M.L., Borrego, J., Morales, J.A.,
Pendon, J.G., Munoz, J.M., 1998. Stratigraphic sequence,
elemental concentrations and heavy metal pollution in Holocene
sediments from the Tint–Odiel Estuary, south-western Spain.
Environmental Geology 34, 270–278.
Sasikumar, C., Rao, D.S., Srikanth, S., Ravikumar, B., Mukhopad-
hyay, N.K., Mehrotra, S.P., 2004. Effect of mechanical activation
on the kinetics of sulfuric acid leaching of beach sand ilmenite
from Orissa, India. Hydrometallurgy 75, 189–204.
Schäfer, J., Dörr, W., 1997. Heavy mineral analysis and typology of
detrital zircons: a new approach to provenance study (Saxothur-
ingian Flysch, Germany). Journal of Sedimentary Petrology 64,
451–461.
Schaltegger, U., Fanning, C.M., Günther, D., Maurin, J.C., Schul-
mann, K., Gebauer, D., 1999. Growth, annealing and recrystallisa-
tion of zircon and preservation of monazite in high-grade
metamorphism: conventional and in-situ U–Pb isotope, cathodo-
luminescence and microchemical evidence. Contributions to
Mineralogy and Petrology 134, 186–201.
Siebel, W., Blaha, U., Chen, F., Rohrmüller, J., 2005. Geochronology
and geochemistry of a dyke–host rock association and implications
for the formation of the Bavarian Pfahl shear zone, Bohemian
Massif. International Journal of Earth Sciences 94, 8–23.
Sneed, E.D., Folk, K.L., 1958. Pebbles in the lower Colorado River,
Texas, a study in particle morphogenesis. Journal of Geology 66,
114–150.
Southam, G., Beveridge, T.J., 1994. The in vitro formation of placer
gold by bacteria. Geochimica et Cosmochimica Acta 58,
4527–4530.
Sukhoroslov, V.L., Yablokova, S.V., 2001. Some features of placer
formation under tropical climate conditions. Quaternary Interna-
tional 82, 75–85.
Sutherland, D.G., 1985. Geomorphological control on the distribution
of placer deposits. Journal of the Geological Society (London) 142,
727–737.
Thomas, M.F., 1994. Geomorphology in the Tropics. Wiley,
Chichester. 460 pp.
27
Trewin, N., 1988. Use of the scanning electron microscope in
sedimentology. In: Tucker, M.E. (Ed.) Techniques in Sedimentol-
ogy. Blackwell, Scientific Publications, Oxford, pp. 229–273.
Tucker, M.E., 2001. Sedimentary Petrology. Blackwell, Oxford.
262 pp.
Tuyet, D., Thang, D.V., Van, T.T., Tam, V.T., Dusar, M., Masschelein, J.,
2004. The Cat Ba and Cucu Phuong National Parks and the Ha Long
Bay world natural heritage. Field Excursion Guidebook, Trans-Karst
2004, Hanoi, Vietnam 13 to 18 September 2004, pp. 1–45.
Twidale, C.R., 2002. The two-stage concept of landform and landscape
development involving etching: origin, development and implica-
tions of an idea. Earth-Science Reviews 57, 37–74.
Ufimetsev, G.F., 1990. Morphotectonics of the Mongolian–Siberian
mountain belt. Journal of Geodynamics 11, 309–325.
Ulsts, V., Bulgakova, J., 1998. General Lithological and Geomorpho-
logical Map of Latvian Shore Zone. Baltic Sea and Gulf of Riga.
State Geological Survey of Latvia, Riga.
University of Latvia, 1995. Mineral Deposits of Latvia. Ministry of
Environment and Regional Development, Riga. 21 pp.
Usaityte, D., 2000. The geology of the southeastern Baltic Sea: a
review. Earth-Science Reviews 50, 137–225.
Vareikiene, O., Lehtonen, M., 2004. Heavy minerals in the study of
soil: techniques, their limitations and advantages. Geologija 46,
1–7.
Wort, M.J., Jones, M.P., 1980. X-ray diffraction and magnetic studies
of altered ilmenite and pseudorutile. Mineralogical Magazine 43,
659–663.
Xinh, L.T., Que, T., Hung, T.K., 1990. Bauxite deposit. In: Xinh, L.T.,
Tri, T.V., Dzung, D.H., Dzung, N.T., Chien, L.T., Dzuong, M., Hai,
T.H. (Eds.), Geology and Mineral Resources of Vietnam, vol. 1.
Mineral Development Co, Hanoi, Vietnam, pp. 128–143.
Yager, T.R., 2001. The mineral industries of Burundi, Comores,
Malawi, Mauritius, Reunion, Rwanda and Seychelles. US
Geological Survey Minerals Yearbook 2001, 7.1–7.6.
Yu, S.-Y. 2003. The Litorina transgression in southeastern Sweden and
its relation to mid-Holocene climate variability. LUNDQUA thesis
51. 82 pp.
Zahn, U., 1991. Diercke Weltatlas. Westermann-Verlag, Braunsch-
weig. 275 pp.