Content uploaded by Kenneth Mertens
Author content
All content in this area was uploaded by Kenneth Mertens on Mar 26, 2018
Content may be subject to copyright.
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Determining the absolute abundance of dinoflagellate cysts in recent marine
sediments: The Lycopodium marker-grain method put to the test
Kenneth Neil Mertens
a,
⁎, Koen Verhoeven
a
, Thomas Verleye
a
, Stephen Louwye
a
, Ana Amorim
b
,
Sofia Ribeiro
b,1
, Amr S. Deaf
c
, Ian C. Harding
c
, Stijn De Schepper
d
, Catalina González
d
,
Monika Kodrans-Nsiah
d
, Anne De Vernal
e
, Maryse Henry
e
, Taoufik Radi
e
, Karen Dybkjaer
f
,
Niels E. Poulsen
f
, Susanne Feist-Burkhardt
g
, Jonah Chitolie
g
, Claus Heilmann-Clausen
h
, Laurent Londeix
i
,
Jean-Louis Turon
i
, Fabienne Marret
j
, Jens Matthiessen
k
, Francine M.G. McCarthy
l
, Vandana Prasad
m
,
Vera Pospelova
n
, Jane E. Kyffin Hughes
o
, James B. Riding
o
, André Rochon
p
, Francesca Sangiorgi
q
,
Natasja Welters
q
, Natalie Sinclair
r,w
, Christian Thun
r
, Ali Soliman
s,x
, Nicolas Van Nieuwenhove
t
,
Annemiek Vink
u
, Martin Young
v
a
Research Unit Palaeontology, Krijgslaan 281 s8, 9000 Gent, Belgium
b
Instituto de Oceanografia, Faculdade de Ciências da Universidade de Lisboa, Campo Grande, 1749-016 Lisboa, Portugal
c
School of Ocean & Earth Science, National Oceanography Centre, Southampton, University of Southampton, European Way, Southampton, SO14 3ZH, UK
d
University Bremen, Geosciences Department, Historical Geology/Palaeontology, PO Box 330 440, D-28334 Bremen, Germany
e
GEOTOP, Université du Québec à Montréal, C.P. 8888, succursale "centre ville", Montréal, Qc, Canada H3C 3P8
f
Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen K., Denmark
g
The Natural History Museum, Palaeontology Department, Cromwell Road, London SW7 5BD, UK
h
Geologisk Institut, Aarhus Universitet, Høegh-Guldbergs Gade 2, DK-8000 Århus C, Denmark
i
Université Bordeaux 1, UMR 5805 CNRS OEPOC¹, avenue des Facultés, F-33405, France
j
Department of Geography, University of Liverpool, Roxby Building, Liverpool, L69 7ZT, UK
k
Alfred Wegener Institute for Polar and Marine Research, P.O. Box 120161, D-27515 Bremerhaven, Germany
l
Earth Sciences, Brock University, S. Catharines, Ontario, Canada L2S 3A1
m
Micropaleontology laboratory, Birbal Sahni Institute of Palaeobotany, 53, University Road, Lucknow-226007, India
n
School of Earth and Ocean Sciences, University of Victoria, New Science Building (OEASB) A405, P.O. Box 3065 STN CSC, Victoria, B.C., Canada V8W 3V6
o
British Geological Survey, Kingsley Dunham Centre, Keyworth, Nottingham NG12 5GG, UK
p
Institut des sciences de la mer de Rimouski (ISMER), Université du Québec à Rimouski, 310, allée des Ursulines, Rimouski, QC, Canada G5L 3A1
q
Palaeoecology, Institute of Environmental Biology, Faculty of Science, Utrecht University, Laboratory of Palaeobotany and Palynology, Budapestlaan 4, 3584 CD Utrecht, The Netherlands
r
Geoscience Australia, GPO Box 378, Canberra, ACT, 2601, Australia
s
Karl-Franzens GRAZ University, Institute of Earth Science, Heinrichstrasse 26, A-8010 Graz, Austria
t
IFM-GEOMAR, Leibniz-Institute of Marine Sciences, Wischhofstrasse 1-3, 24148 Kiel, Germany
u
Federal Institute for Geosciences and Natural Resources, Alfred-Bentz-Haus, Stilleweg 2, 30 655 Hannover, Germany
v
CSIRO Petroleum, 11 Julius Ave, Riverside Corporate Park, North Ryde 2113, NSW, Australia
w
Research School of Earth Sciences, Australian National University, Bldg 61 Mills Road, Acton, ACT, 0200, Australia
x
Geology Department, Faculty of Sciences, Tanta University, Tanta 31527, Egypt
abstractarticle info
Article history:
Received 4 September 2008
Received in revised form 28 March 2009
Accepted 8 May 2009
Available online 18 May 2009
Keywords:
dinoflagellate cyst
concentration
Lycopodium clavatum tablets
spike
inter-laboratory calibration
Absolute abundances (concentrations) of dinoflagellate cysts are often determined through the addition of
Lycopodium clavatum marker-grains as a spike to a sample before palynological processing. An inter-
laboratory calibration exercise was set up in order to test the comparability of results obtained in different
laboratories, each using its own preparation method. Each of the 23 laboratories received the same amount of
homogenized splits of four Quaternary sediment samples. The samples originate from different localities and
consisted of a variety of lithologies. Dinoflagellate cysts were extracted and counted, and relative and
absolute abundances were calculated. The relative abundances proved to be fairly reproducible,
notwithstanding a need for taxonomic calibration. By contrast, excessive loss of Lycopodium spores during
sample preparation resulted in non-reproducibility of absolute abundances. Use of oxidation, KOH, warm
acids, acetolysis, mesh sizes larger than 15 µm and long ultrasonication (N1 min) must be avoided to
determine reproducible absolute abundances. The results of this work therefore indicate that the
Review of Palaeobotany and Palynology 157 (2009) 238–252
⁎Corresponding author.
E-mail address: Kenneth.Mertens@ugent.be (K.N. Mertens).
1
Currently at: Section for Aquatic Biology,Faculty of Sciences, University of Copenhagen, Øster Farimagsgade 2D, DK-1353,Copenhagen K, Denmark and Departamento de Geologia
Marinha, LNEG, Estrada da Portela, Zambujal 2721-866, Alfragide, Portugal.
0034-6667/$ –see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.revpalbo.2009.05.004
Contents lists available at ScienceDirect
Review of Palaeobotany and Palynology
journal homepage: www.elsevier.com/locate/revpalbo
Author's personal copy
dinoflagellate cyst worker should make a choice between using the proposed standard method which
circumvents critical steps, adding Lycopodium tablets at the end of the preparation and using an alternative
method.
© 2009 Elsevier B.V. All rights reserved.
1. Introduction
Dinoflagellate cyst concentrations are an important component of
paleoceanographical studies (e.g. Pospelova et al., 2006; González
et al., 2008) and can be determined using the volumetric method (e.g.
Dale et al., 2002; Holzwarth et al., 2007). In general, dinoflagellate cyst
concentrations are calculated by adding a known amount of exotic
markers or a “spike”to every sample according to the method de-
scribed by Stockmarr (1971). The marker commonly used is Lycopo-
dium clavatum Linnaeus (Stag's Horn Clubmoss or Ground Pine).
As noted by Lignum et al. (2008), the so-called ‘standard’
palynological processing methods are still very variable in terms of
initial sample sizes, type and concentration of acids, sieve material
and mesh size, sonication time and strength, number of decanting
cycles and use of heavy liquid separation. This is also apparent in
reviews of the preparation techniques for extraction of dinoflagellate
cysts given by Wood et al. (1996) and more recently by Riding and
Kyffin-Hughes (2004). However, critical evaluation of the effect of
different laboratory procedures on the marker grain technique for
obtaining dinoflagellate cyst concentration has so far never been
attempted. Although it has been reported that several processing
methods such as sonication and chemical treatments can inflict da-
mage on organic-walled microfossils to a certain extent (e.g. Schrank,
1988; Hodgkinson, 1991), the effect on palynomorph concentrations
remain unknown.
This study aims to test the reproducibility of the marker-grain
method, in order to understand the discrepancies in the results fol-
lowing different preparation techniques. Similar efforts to test the
reproducibility of specific laboratory techniques have been done for
other microfossil groups: benthic and planktonic foraminifera (Zachar-
iasse et al., 1978), diatoms (Wolfe, 1997 ), nannofossils (Herrle and
Bollman, 2004)andtheirbiomarkers(Rosell-Melé et al., 2001). It is
therefore timely to carry out a similar exercise with dinoflagellate cysts.
Surface sediment samples from four localities (North Sea, Celtic
Sea, NW Africa and Benguela) were sent to 23 laboratories. The
samples were processed using the palynological techniques routi-
nely used in these laboratories. An equal amount of Lycopodium
tablets, all from the same batch, were added to each sample. The
reproducibility of both absolute and relative abundances for dino-
flagellate cysts is here put to test, and has resulted in a proposal of
recommendations for a standardized method to determine absolute
abundances of Quaternary dinoflagellate cysts with the marker-grain
method. Two laboratories used the volumetric method (Dale, 1976)
for comparison purposes. This study focuses additionally on whether
it is necessary to count 300 or 400 dinoflagellate cysts and on
taxonomy, since notable interlaboratorial differences in nomencla-
ture were recorded.
2. Material and methods
Late Quaternary surface sediment samples from four sites with
different lithologies were used by the 23 different laboratories in-
volved in the project. The North Sea sample consisted of a homo-
genized surface sediment taken using a Reineck boxcorer (51.47°N,
3.48°E, 10 m water depth). The Celtic Sea sample was assembled
through mixing multi-corer samples from Station 8, collected during
several time slots from the Celtic Sea (51.05°N, 5.83°W, 86 m water
depth) (Marret and Scourse, 2002). The sample from Northwest
Africa was a mixture of multicores GeoB9504-4 (15.87°N, 16.67°W,
43 m water depth) and GeoB9503-3 (16.07°N, 16.65°W, 50 m water
depth). The Benguela sample consists of a mixture of sediment
samples collected offshore Walvis Bay, at a water depth of about
200 m during Meteor cruise M63/2. Sample details are given in
Table 1. Each laboratory was given a number, followed by a letter
when the laboratory used more than one processing method.
Laboratory identification and numbers were kept anonymous. A
brief overview of the methods used is described in Sections 2.1–2.5.A
special variation of this method is detailed in Section 2.6 and the
volumetric method is detailed in Section 2.7. Details of the methods
used are given in the Supplementary data.
Homogenization was done using the quartile method. The samples
were oven-dried at a temperature of 58 °C for 24 h. The Lycopodium
spore tablets used are produced and distributed by the Subdepart-
ment of Quaternary Geology, University of Lund, Sweden (http://
www.geol.lu.se/kvg/eng/). Ten Lycopodium clavatum tablets of batch
483216, (X= 18.583 per tablet, s=±1708), were dispatched with the
samples, and a fixed number of tablets was added by each laboratory
to each sample.
2.1. Chemical treatment
Hydrochloric acid (HCl) with a concentration of 6.5–36% was
added for the removal of carbonate. Some 20 to 300 ml was used
depending on the intensity of the reaction. Cold HCl was used in most
of the cases, although some laboratories used hot HCl with a tem-
perature ranging between 42 and 80 °C. Afterwards, the residue was
left to settle (15 min to 42 h). Laboratories that used short settle times
at this step, used centrifugation or sieving to concentrate the sample.
For centrifugation, the rotation speed used varied between 1900 and
3500 rpm, and lasted between 5 s to 10 min.
Demineralised or distilled water was used for rinsing until pH
reached more neutral values of 5 to 7. One to 5 decanting cycles with
intervals of 3 to 24 h were needed depending on HCl concentrations
used. To avoid losing residue during decanting, some laboratories
used centrifugation for concentration of the residue. Extensive
rinsing is necessary for the removal of Ca
2+
, to avoid calcium fluoride
(CaF
2
) precipitation during HF treatment. A few laboratories used
KOH for neutraliz ation (laboratory 2: 1% KOH and l aboratory 18b: 10%
KOH).
The siliciclastic component of the samples was removed by adding
10 to 250 ml of hydrofluoric acid (HF) with a concentration ranging
from 19% to 70%. Commonly a concentration between 40 and 50% was
used. All laboratories used cold HF, except laboratories 12 (42 °C), 2
(50 °C), 6 (60 °C), 10 (70 °C) and 23 (80 °C). Settling times varied
between 12 and 144 h. A few laboratories repeated the HF treatment
up to 3 times before all silicates were removed.
Before neutralisation, about 10 to 300 ml HCl with a concentration
of 6.5 to 36 vol.% was added for the removal of formed fluorosilicates.
Table 1
Description of the samples.
Sample Lithology Dry weight Number of
tablets added
# spores
added
St dev
spores(g)
North Sea Fine-medium sand 10 3 55,749 2959
Celtic Sea Fine silty sand 10 1 18,583 1708
NW Africa Clay 2 2 37,166 2416
Benguela Clay 1 4 74,332 3417
239K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
Mostly cold HCl was used, although some laboratories used hot HCl
with a temperature ranging between 42 and 100 °C. The following
settling time varied between 15 min and 72 h. Again, laboratories that
used short settling times, used centrifugation. The sample was sub-
sequently rinsed with distilled water, until pH reached 5–7. The
rinsing took 1 to 6 decanting cycles with intervals of 3 to 24 h,
depending on the concentrations used. Again, to avoid losing residue
during decanting, some laboratories used centrifuging for the con-
centration of the samples. One laboratory used KOH for the neutra-
lisation (laboratory 2: 1%). A few laboratories skipped the second HCl
treatment and proceeded directly to the rinsing with distilled water
until pH reached values of 5–7. Several of these laboratories used
centrifuging and/or sieving for concentration of the samples. During
rinsing toxic HF was decanted and removed.
One laboratory (laboratory 22b) oxidised three of the samples
(excluding the North-West Africa sample) with Schulze's solution
(70% nitric acid saturated with potassium chlorate).
2.2. Mechanical treatment
Heavy liquid separation for the removal of heavy minerals was
carried out by a few laboratories. Labs 10 and 16 used sodium
polytungstate (SPT) at specific densities to isolate the palynological
fractions.
Between 13 and 1800 s sonication was used to break down organic
matter aggregates by some laboratories. Most laboratories used sonic
baths (Branson™, Sonimasse™, Sonicor™, Eurolab™). Laboratory 8
used a standard oscillating sensor.
2.3. Sieving
Some laboratories pre-sieved before the chemical treatment for
the elimination of the coarse fraction (mesh sizes of 100, 106,120 and
150 µm) and/or fine fraction (mesh sizes of 10, 11 and 15 µm). All the
laboratories added the Lycopodium tablets before pre-sieving, except
laboratory 23.
Sieving after the chemical treatment was used to remove the fine
fraction from the residue. Calgon (sodium hexametaphosphate) was
used to disaggregate the material in a few cases. The sieve mesh sizes
used varied from 6 to 20 µm, and meshes were made of nylon,
polyester, polymer or steel. The devices used were hand, mechanical
and water pressure pumps. Some laboratories sieved without using a
pump.
2.4. Staining and mounting of the slides
Staining with a colouring agent enhances contrast for optical
microscopy and can be used for the detection of pre-Quaternary
specimens (Stanley, 1966). Safranin-O, Fuchsin or Bismark Brown
was used by a few laboratories. Not every laboratory stained the
residue. Finally a few drops of a copper sulphate solution, thymol or
phenol were often added to the residue for the inhibition of fungal
growth.
Slides were mounted on a heated metal plate (65 °C) using a
pipette, by strewing using a spatula or a mix of both methods. The
mounting medium was usually glycerin jelly, but sometimes thymol,
Elvacite, Eukitt, UV adhesive, or Canada balsam was used. Although
sealing is not per se necessary (Poulsen et al., 1990), nail polish or
Plate I. Polykrikos schwartzii extracted from the North Sea sample using different methodologies. Labs are sorted from high (upper lef t corner) to low abundances (lower right
corner).
1. Lab 1a.
2. Lab 20a.
3. Lab 13.
4. Lab 12.
5. Lab 19.
6. Lab 2.
7. Lab 11.
8. Lab 21a.
9. Lab 21b.
10. Lab 22a.
11. Lab 10a.
12. Lab 18b.
13. Lab 1b.
14. Lab 16.
15. Lab17.
16. Lab 10b.
17. Lab 18a.
18. Lab 5.
19. Lab 4.
20. Lab 22b, oxidized. All scale bars are 20 µm.
Table 2
Average percentage of the different taxa in the four samples.
Species name North Sea Celtic Sea NW Africa Benguela
Round brown cysts (RBC) 35.8± 16.0 10.0 ±7.7 3.4± 2.3 62.7 ±17.0
Spiny brown cysts (SBC) 15.5± 12.5 1.7± 3.3 2.3±2.4 8.5 ±8.5
cysts of Alexandrium spp. 0.2 ±0.3 0.5 ±0.9 –0.1± 0.5
cysts of Gymnodinium spp. 0.3 ±0.6 0.3± 0.6 0.0 ±0.1 0.0 ±0.1
Stelladinium spp. 0.3±0.3 0.2± 0.2 0.3±0.3 0.1± 0.4
Lejeunecysta spp. 9.5± 12.0 1.5±1.6 0.4 ±0.5 1.4 ±1.6
Selenopemphix spp. 5.5±1.7 4.8± 2.1 1.0± 0.6 6.5 ±6.3
Tuberculodinium vancampoae 0.0± 0.1 –0.1 ±0.3 0.0 ±0.1
Polykrikos spp. 6.9 ± 3.5 5.7 ± 3.8 1.2± 0.8 1.1 ±0.8
Xandarodinium xanthum 0.2 ±0.3 0.1± 0.1 0.1 ±0.2 0.0 ±0.1
Dalella chathamense –––0.0±0.1
Extremely sensitive cysts (total) 74.3± 7.4 24.8 ±11.2 15.4± 8.2 80.6 ±9.9
Lingulodinium machaerophorum 1.5± 2.5 0.7 ±0.9 86.2± 4.7 0.2 ±0.5
Operculodinium spp. 2.8±1.9 12.3± 3.7 0.5± 0.7 8.4 ± 6.6
Pyxidinopsis reticulata 0.0±0.2 –––
Spiniferites spp. 9.8±3.5 51.8 ±10.7 3.3 ±1.1 5.5±3.2
Quinquecuspis concreta 3.3± 2.1 2.3 ±2.0 0.1± 0.1 1.0±1.5
Trinovantedinium applanatum 0.2± 0.4 1.2 ±1.0 0.2± 0.3 0.3± 0.4
Votadinium spp. 5.8±6.6 0.5± 0.7 0.0±0.1 0.7± 0.7
Moderately sensitive cysts (total) 23.6±7.2 68.9 ±10.7 90.3± 4.2 16.2± 9.7
Nematosphaeropsis labyrinthus 0.0±0.1 0.0± 0.1 0.1± 0.1 2.1 ±2.0
Impagidinium spp. 0.3±0.6 0.15± 0.3 0.0± 0.1 0.0 ± 0.1
Operculodinium israelianum 0.2±0.2 0.0± 0.1 0.4± 0.7 0.4 ±0.7
Pentapharsodinium dalei 0.4± 0.5 2.6 ± 3.5 0.0± 0.1 0.2± 0.5
Polysphaeridium zoharyi 0.4 ±0.6 0.1± 0.3 0.1 ±0.5 0.2 ±0.7
Ataxiodinium choane 0.0±0.1 0.1± 0.1 0.0± 0.0 –
Bitectatodinium spp. 0.6 ±1.1 3.3± 2.0 0.1 ±0.2 0.2 ±0.6
Resistant cysts (total) 0.5± 0.6 6.2± 3.8 0.7± 0.9 3.1±2.5
240 K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
paraffin wax was used to seal the slides to protect the residue from
degradation by dehydration.
2.5. Counting of the palynomorphs and calculation of absolute
abundances
Dinoflagellate specimens were counted only when they comprised
at least half of a cyst. The same criterion was used for other
palynomorphs, also counted by some of the laboratories. Initially
300 dinoflagellate cysts were counted, and subsequently an extra 100
specimens were added. The purpose was to check whether it is
necessary to count 300 or 400 dinoflagellate cysts to obtain rep-
resentative relative and absolute abundances. Indeterminate dino-
flagellate cysts were grouped as Indeterminate spp., and were not
taken into account for the calculation of the relative abundances, since
every observer had a different concept of what counts as an
indeterminate dinoflagellate cyst, and this would introduce observer
bias into the relative abundances. Raw counts together with a sum-
mary of the methodology are available as supplementary data to this
article. Taxonomy follows Fensome and Williams (2004).
Absolute abundances of dinoflagellate cysts were calculated
following the equation by Benninghoff (1962):
c=dc×Lt×t
Lc×w
where
cconcentration=number of dinoflagellate cysts/gram dried
sediment.
241K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
d
c
number of counted dinoflagellate cysts
L
t
number of Lycopodium spores/tablet
tnumber of tablets added to the sample
L
c
number of counted Lycopodium spores
wweight of dried sediment (g)
Maher (1981) devised an algorithm to calculate confidence limits
on microfossil concentrations. A slight correction to this algorithm
was made, since the current study used sediment weight instead of
sediment volume. The confidence limits calculated based on this
algorithm have a 0.95 probability (Z=1.95). It should be noted that
these confidence limits are similar to the total error on concentration
proposed by Stockmarr (1971);(Appendix B). These confidence limits
can then be used in a statistical test to check whether microfossil
concentrations are the same in two different samples (Maher, 1981).
To investigate the reproducibility of results from the different
laboratories, the coefficient of variation (or relative standard devia-
tion) of all counts of a particular sample can be compared. Ideally, the
results should fall within the confidence limits of Maher (1981), and
thus the coefficient of variation calculated from these confidence
limits can be used as a comparison.
2.6. Special methods: the maceration tank method (with HF) and the
washing machine method (without HF)
The maceration tank method (Poulsen et al., 1990; Desezar and
Poulsen, 1994) was used for HF treatment by laboratory 20a. Other
processing steps are similar to those used by the other laboratories
and are detailed in Poulsen et al. (1990) and Desezar and Poulsen
(1994). Each sample is tightly wrapped in filter cloth (25 cm ×25 cm)
with a mesh size of 10 µm, and the filter bags are packed in rubber
foam for protection. The samples are placed inside the maceration
tank and HF is conducted to the tank in PVC tubes. The samples are
treated with cold HF for 7–8 days, after which the HF is drained out
through a bottom-stop cock and led via PVC tubes directly to a waste-
container for used hydrofluoric acid.
With the washing machine method, used by laboratory 20b, no HF
is used. Each sample is tightly wrapped in filter cloth (25 cm× 25 cm)
with a mesh size of 10 µm and the filter bags are packed in rubber
foam for protection. The samples are washed in a standard household
washing machine with a standard household washing powder, after
which carbonates are removed with citric acid at 65 °C. Next the
samples are again given a normal wash with a standard household
washing powder. Finally the remaining minerals are removed by heavy
Plate II. Polykrikos schwartzii extracted from the Celtic Sea sample using different methodologies, sorted from high absolute abundances (upper left corner) to low absolute
abundances (lower right corner).
1. Lab 14.
2. Lab 1a.
3. Lab 13.
4. Lab 3.
5. Lab 19.
6. Lab 12.
7. Lab 1b.
8. Lab 15b.
9. Lab 1c.
10. Lab 21b.
11. Lab 21a.
12. Lab 11.
13. Lab 5.
14. Lab 4.
15. La b 16.
16. Lab 23.
17. Lab 17.
18. Lab 18a.
19. Lab 20a.
20. Lab 2. All scale bars are 20 µm.
Plate III. Lingulodinium machaerophorum extracted from the NW Africa using different methodologies, sorted from high (upper left corner) to low absolute abundances (lower right
corner). (see on page 244)
1. Lab 11.
2. Lab 1a.
3. Lab 14.
4. Lab 13.
5. Lab 19.
6. Lab 10b.
7. Lab 21a.
8. Lab 1b.
9. Lab 12.
10. Lab 17.
11. Lab 21b.
12. Lab 6.
13. Lab18a.
14. Lab 18b.
15. Lab 1c.
16. Lab 15b.
17. Lab 22a.
18. Lab 4.
19. Lab 5.
20. Lab 20b.
21. Lab 16.
22. Lab 8.
23. Lab 23.
24. Lab 3. All scale bars are 20 µm.
242 K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
liquid separation. This method removes the amorphous material very
efficiently. Furthermore, since HF is not used, siliceous constituents
(e.g. diatoms) are not destroyed. Heavy liquid separation with zinc
dibromide (ZnBr
2
) was used at densities of 2.3, 2.0 and 1.8 g/ml to
remove heavy minerals. In order to test the influence of the specific
density of the ZnBr
2
, the NW African sample from laboratory 20b, was
separated using heavy liquid densities of 1.8, 2.0 and 2.3 g/ml.
2.7. Volumetric method
For comparison with the marker-grain method, the volume aliquot
method was performed by laboratories 6 and 8, following Dale (1976).
This method was not used for the North Sea sample because of the
difficulty associated with counting a fixed volume of this sample with
very low abundances.
3. Results
3.1. Relative abundance of dinoflagellate cysts
Quantitative and qualitative disparities between assemblages
recorded by the laboratories may be due to the different processing
methods. It is obvious that aggressive agents could destroy the
more sensitive cysts. To check this dependence of preservation on
Plate II.
243K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
Plate III (see caption on page 242).
244 K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
methodology, it is necessary to group the present species according to
their resistance to degradation. It is assumed that both mechanical
and chemical degradation have similar effects on an assemblage. The
grouping proposed here is similar to the grouping described by
Zonneveld et al. (2001). Cysts not referred to by these authors were
added to a particular group based on the assumption that comparable
morphology (e.g. wall thickness, resistance of structures against
folding) is indicative of similar resistance to decay.
Extremely sensitive cysts: cysts of Alexandrium spp., Dalella
chathamense, cysts of Gymnodinium spp., Lejeunecysta spp., Polykri-
kos spp., round brown cysts (RBC), Selenopemphix spp., spiny brown
cysts (SBC), Stelladinium spp., Tuberculodinium vancampoae and Xan-
darodinium xanthum.
Moderately sensitive cysts:Lingulodinium machaerophorum,
Operculodinium spp., Pyxidinopsis reticulata,Quinquecuspis concreta,
Spiniferites spp., Trinovantedinium applanatum and Votadinium spp.
Resistant cysts: Ataxiodinium choane,Impagidinium spp., Nemato-
sphaeropsis labyrinthus,Operculodinium israelianum,Pentapharsodi-
nium dalei,Polysphaeridium zoharyi and Bitectatodinium spp.
It is evident from the dataset that some species were not recorded
by some observers. One obvious example is Dubridinium spp., which
was often counted by some laboratories as RBC or not counted at all.
To partly reduce this observer bias, we decided to group species into
genera or larger groups (Appendix A). Averages of relative abun-
dances were only calculated when at least 300 dinoflagellate cysts
were counted. The counts from oxidized samples (laboratory 22b)
were also excluded, since all heterotrophic cysts were destroyed. The
average results of the four samples are shown in Table 2. Representa-
tive cysts from the four samples are shown in Plates I–IV.
3.2. Absolute abundances of dinoflagellate cysts
The cyst concentration (absolute abundance) in the North Sea
sample ranges from 570 to 3304 cysts/g, excluding the outliers:
laboratory 1a produced a very high number (8342 cysts/g) and lab-
oratory 22b a very low number (278 cysts/g). The average is
1516 cysts/g with a standard deviation of 698 cysts/g (coefficient of
variation, V=46%). The average coefficient of variation from the
confidence limits of Maher (1981) is 20%. The volumetric method was
not used for the North Sea sample (Table 3).
The cyst concentration (absolute abundance) in the Celtic Sea
sample ranges from 1240 to 5284 cysts/g, excluding the outliers:
laboratories 14 and 1a produced high numbers of 75,633 and
10,961 cysts/g respectively, while laboratory 20a, 2 and 20b give
respectively low values of 1053, 731 and 501 cysts/g. The average is
2583 cysts/g, with a standard deviation of 1342 cysts/g (V= 52%). The
average coefficient of variation from the confidence limits of Maher
(1981) is 25%. Results obtained by the volumetric method give
estimates that are much lower than with the marker grain method. For
the Celtic Sea these values (1160 cysts/g (laboratory 6) and 1167 cysts/
g (laboratory 8)) are even below the lowest value obtained by the
marker grain method (Table 3).
The cyst concentration (absolute abundance) in the NW Africa
sample ranges from 4606 to 38,357 cysts/g, excluding the outliers:
laboratories 11, 1a and 14 produced very high numbers (168,899,
167,651 and 129,236 cysts/g, respectively). The average is
19,441 cysts/g, with a standard deviation of 9148 cysts/g (V= 47%).
The average coefficient of variation from the confidence limits of
Maher (1981) is 23%. As before, the volumetric method gave lower
estimates but within the range of the marker grain method
(11,600 cysts/g (laboratory 6) and 9992 cysts/g (laboratory 8))
(Table 3).
The cyst concentration (absolute abundance) in the Benguela
sample ranges from 30,130 to 298,972 cysts/g, excluding the outliers:
Laboratory 1c produced a high number of 1,455,988 cysts/g, while
laboratories 2 0b and 8 give values as low as 18,472 and 15,910 cysts/g,
respectively. The average is 144,299 cysts/g with a standard devi ation
of 84,159 cysts/g (V= 58%). The average coefficient of variation
from the confidence limits of Maher (1981) is 21%. The volumetric
method used by laboratory 6 yields 53,200 cysts/g (within the
range above) and 8492 cysts/g by laboratory 8. The volumetric
estimate by laboratory 8 is considered to be an underestimation
caused by the destruction of fragile cysts by sonication (see
Discussion); (Table 3).
3.3. Reworked dinoflagellate cysts
About 7% of the recorded dinoflagellate cysts in the North Sea
sample were reworked. The pre-Quaternary cysts recorded in the
North Sea sample were Wetzeliella spp. (dominant), Glaphyrocysta
spp., Cordosphaeridium spp., cf. Oligosphaeridium spp. and cf. Cribro-
peridinium spp. In terms of absolute abundances, reworking shows
the same trends as in situ dinoflagellate cyst absolute abundances.
Very high absolute abundances were recorded in the sample oxidized
by laboratory 22b. This indicates that the robust pre-Quaternary cysts
are more resistant to oxidation. Reworking is very low (less than 1%)
in the samples from the Celtic Sea, NW Africa and Benguela.
3.4. Other palynomorphs
Chlorophycean palynomorphs such as Cymatiosphaera sp. (not
present in Celtic Sea), Pediastrum sp., Pterospermella sp. (not present
in Benguela), Tas man ite s sp., Botryococcus sp. (not present in
Benguela), Mougeotia sp. (only North Sea), Concentricystes circulus
(only NW Africa), Gelasinicysta sp. indet. (only NW Africa) are
recorded in low numbers in all samples, except the North Sea sample.
Faunal remains such as microforaminiferal linings, scolecodonts,
tintinnids, planktonic crustacean eggs and invertebrate mandibles
were encountered in almost every sample. Planktonic crustacean eggs
are very abundant in the North Sea sample.
Pollen and spores are abundant in the North Sea sample. The
assemblage is dominated by pollen (90%). Non-bisaccate pollen
include Quercus,Corylus, Betula, Alnus, pollen of Poaceae, Cyperaceae
and Chenopodiaceae, whereas bisaccate pollen comprise mainly Pinus
and Picea. Some Cedrus pollen is recorded. Reworked pollen and
spores are present in low numbers.
The Celtic Sea sample is dominated by pollen (94%). Non-bisaccate
pollen comprises mainly pollen of Poaceae, Quercus,pollenof
Ericaceae and Chenopodiaceae. Bisaccate pollen is mainly Pinus
pollen. Reworked pollen and spores are very rare.
The sample from NW Africa is also dominated by pollen (95%).
Non-bisaccate pollen comprise mainly pollen of Poaceae, Quercus,
pollen of Ericaceae and pollen of Chenopodiaceae. The bisaccate
pollen are mainly Pinus pollen. Reworked pollen and spores are very
rare.
The Benguela assemblage is dominated by pollen (99%). Non-
bisaccate pollen includes mainly pollen of Poaceae, Asteraceae and
Caryophyllaceae. Bisaccate pollen is mainly Pinus pollen. No reworked
pollen and spores were recorded.
Hyphae and fruiting bodies were counted as fungal remains in
order to check whether the samples were infected by fungi. No
samples showed significant abundances.
The recorded incertae sedis include Cyclopsiella,Halodinium sp.,
Hexasterias problematica (not present in Northwest Africa), Micrhy-
stridium sp. (Celtic Sea and Benguela), Palaeostomocystis subtilitheca
(North Sea and Celtic Sea), Radiosperma corbiferum (Celtic Sea and
Benguela) and Sigmopollis sp. (NW Africa). These were more
abundant in both North Sea and Celtic Sea samples.
Other organisms occurring are the organic linings of calcareous
dinoflagellate cysts, thecamoebians (North Sea, Celtic Sea), chrysomo-
nad cysts (North Sea, Celtic Sea) and diatoms. Diatoms can still be
present when low concentrations of HF are used, possibly combined
245K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
with heavy liquid separation, which enhances the abundance of diatoms
with low densities (laboratories 1c, 9 and 17). Laboratory 20b has good
recovery of diatoms, since the samples are not treated with HF.
4. Discussion
4.1. Is a 300 or 400 dinoflagellate cyst count sufficient to reach reliable
diversities and absolute abundances?
There is no general agreement on the number of cysts which
should be counted to obtain reliable data for diversity and absolute
abundance studies. Most palynologists usually count 300 cysts per
sample, which can provide up to 98% confidence (Germerad et al.,
1968). To check whether it is necessary to count 300 or 400
dinoflagellate cysts, results from counting 300 cysts, plus an
additional 100 cysts are compared using absolute abundances,
species diversity and the Shannon–Wiener Index for all samples
(Table 3). The com parison shows that the disparities in the results are
insignificant: averages of absolute abundances, species richness and
the Shannon–Wiener Index show limited changes compared to the
associated standard deviations. The statistical test of Maher (1981)
indicates that all absolute abundances derived from the 300
dinoflagellate cyst cou nt statistically produc e the same concentration
as from the 400 dinoflagellate cyst count. It can thus be concluded
that a 300 dinoflagellate cyst count is sufficient for generating
reliable diversities and absolute abundance data in Quaternary
studies.
4.2. Reproducibility of relative abundances
The standard deviations of the relative abundances observed in
the grouping based on cyst preservation are always lower than 11.2%.
These relatively small standard deviations suggest that changes in
the relative abundance counts are caused by observer bias rather
than by differences in methodology. Indeed, the highest standard
deviations in the taxonomical groupings are with the taxa RBC, SBC
and Lejeunecysta s.l. and since it can be assumed that the potential
for preservation of these taxa is similar, it is likely that the disparities
in the counts are the result of observer bias. The high standard
deviation for RBC is probably caused by the high numbers of the
morphologically similar Dubridinium spp. and the unfamiliarity of
many observers with Dubridinium spp. Furthermore, an unambig-
uous definition of a round brown cyst is still lacking. The same is true
for the spiny brown cysts, and several poorly defined species fall
within this group. All other standard deviations are lower than 10%,
which we consider an acceptable range for completely independent
dinoflagellate cyst counts. Another possible reason for observer bias
could be related to the use of different illumination techniques
for routine counting of dinoflagellate cysts. Comparison of the use of
phase contrast to interference contrast illumination to count
dinoflagellate cysts on the same slides by laboratory 15 revealed
that phase contrast emphasizes the transparent cysts (Spiniferites s.l.,
Operculodinium s.l., Nematosphaeropsis labyrinthus,etc.),whilst
interference contrast emphasizes the brown heterotrophic cysts
(RBC, SBC, etc.). Despite the observer bias, there is no doubt that
dinoflagellate cyst relative abundance counts by one single observer
are repeatable.
4.3. Explanation of outliers in absolute abundances
The higher numbers can each be explained by examining specific
methodologies employed by particular labs. Labs 1a and 1c lost an
excessive amount of Lycopodium spores due to the use of sieving at
20 µm as shown by Lignum et al. (2008). Labs 11 and 14 experienced
problems with settling after centrifugation and were not confident
that the final residues were suitable for quantitative analysis.
The lower numbers by laboratory 22b are due to the use of
oxidation, which causes preferential destruction of dinoflagellate
cysts. Due to the low amounts of material used in the exercise, the
maceration tank and washing machine method (laboratories 20a
and 20b) did not function optimally and yielded atypical results that
should not be regarded as representative. This might be due to cysts
getting attached to the large filter cloth (25 ×25 cm) used in this
technique (see Discussion,assumption8).Furthermore,oneofthe
samples from NW Africa (laboratory 20b) was separated at specific
gravities of 1.8, 2.0 and 2.3 g/ml. At the specificgravitiesof1.8and
2.3 g/ml, there were almost no dinoflagellate cysts in the slides,
whereas ten times more dinocysts were noted at the specific gravity
of 2.0 g/ml. Further investigation is needed to evaluate the effect of
heavy liquid separation at different specific gravities.
For laboratory 8, the use of a sonic oscillator resulted in destruction
of sensitive cysts, again yielding lower numbers.
4.4. Reproducibility and accuracy of absolute abundances, excluding
the outliers
Total cyst count is less dependent on taxonomical expertise, and
thus probably less influenced by observer bias. The different labo-
ratories participating in the current inter-calibration exercise used
different processing techniques (see Supplementary data). The
Plate IV. Dubridinium spp. extracted from the Benguela sample using differentmethodologies, sorted from high (upper left corner) to lowabsolute abundances (lower right corner).
1. Lab 1c.
2. Lab 3.
3. Lab 19.
4. Lab 11.
5. Lab 13.
6. Lab 1a.
7. Lab 21a.
8. Lab 21b.
9. Lab 6.
10. Lab 16.
11. Lab 18a.
12. Lab 18b.
13. Lab 1b.
14. Lab 23.
15. Lab 10b.
16. Lab 17.
17. Lab 10a.
18. Lab 5.
19. Lab 2.
20. Lab 8. Destructive ultrasonication. All scale bars are 20 µm.
246 K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
reproducibility of estimates of absolute cyst abundances, as
expressed as coefficient of variation in Table 2, shows that there are
differences among the 23 laboratories: the coefficients of variation
are relatively large (46–58%)andnearlytwiceashighasthe
coefficients of variations (20–25%) which are calculated from
Maher (1981). Our results suggest that the determination of absolute
abundances is mainly dependent on processing methodology. In this
light the accuracy also needs to be considered: a better under-
standing of what is causing the variation can only be achieved when
correct absolute abundances of dinoflagellate cysts have been
determined. To estimate whether the absolute abundances give an
accurate picture of the true absolute abundances of the dinoflagellate
cysts, results from the marker-grain method are compared with
independent methods. When compared to the volumetric method,
absolute abundances calculated using the marker-grain method, are
44–63% higher (Table 2). In a similar study, de Vernal et al. (1987),
noted systematically higher concentrations from the marker-grain
method compared to the results from the volumetric method, and
they suggested that significant losses of Lycopodium spores (close to
33% on the average) took place during laboratory procedures. On the
other hand, in a study on Paleogene sediments, Heilmann-Clausen
(1985), found marker-grain estimate s varying between 70% and 129%
of volumetric estimates and on average 2% lower concentration was
calculated from the marker-grain method. Our study confirms the
247K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
observation of de Vernal et al. (1987), and even shows larger
deviations. It should also be noted, that counts from strew slides
made from unprocessed samples show much lower abundances than
the average absolute abundances from the marker grain method.
From these observations, it can be concluded that with most
preparation techniques there are significant losses of Lycopodium
spores, and this is most probably the reason for higher the absolute
abundances using the marker-grain method. Furthermore, there was
no evidence of significant loss of dinoflagellate cysts during the
laboratory preparations, except when oxidation or very long or
destructive sonication was used (see below). Thus, in order to
understand what causes the differences in absolute abundances, one
needs to consider underlying assumptions. Ten assumptions need to
be considered.
(1) Drying samples does not cause decay.
Although drying is often done in palynological preparation, it
should be avoided in organic rich sediments, where drying causes
formation of selenite (gypsum, CaSO
4
·2H
2
O), by reaction of calcium
carbonate with sulphuric acid, usually derived from pyrite decay. The
formation of sulphuric acid significantly affects extremely sensitive
dinoflagellate cysts. In this case, to calculate the weight of the samples,
wet volumes should be used, corrected with dry bulk densities. In our
samples, gypsum crystals were not observed. The homogenized
samples were oven dried before subdivision into smaller batches
and dispatching to individual laboratories. This was done to avoid
differential drying. However, not all laboratories processed the
samples exactly at the same time. Samples were dispatched in
March 2007, and were processed within the following year. The
possibility exists that samples that were processed at a later stage
dried out more. Clustering of amorphous organic matter around the
cysts seems to occur in more dried out samples (most obvious around
Lingulodinium machaerophorum specimens in Plate III), but there were
no clear signs that this process caused changes in the assemblage. This
assumption is thus acceptable.
(2) Samples are homogenous.
It needed testing if samples processed in a similar manner yielded
reproducible results. All samples were processed twice by laboratory
21 (a and b) with the only difference in preparation the addition of
some soap during sieving (Table 4). Following the test by Maher
(1981), for every studied sample, the microfossil concentration in the
quasi-replicas is the same. It can thus be concluded that the samples
are well-mixed and are homogenous. Furthermore, there are few
differences between both samples in terms of relative abundances.
This assumption is thus acceptable.
(3) A single Lycopodium tablet from batch 483216 contains 18,583 ±
1708 spores.
This reference is given by the supplier (Lund University), and
these numbers were calibrated using a Coulter counter. Lignum et al.
(2008) also used a Coulter counter for verification and obtained
16, 971 ± 12 51 Lycopodium spores. We dissolved one tablet in distilled
water and sieving on a 0.25 µm Millipore filter. The filter was cut into
two pieces, mounted on a slide and counted under a transmitted light
microscope. On this filter, 16,993 Lycopodium spores were counted,
which falls within the range proposed by the supplier and Lignum
et al. (2008). A similar exercise has been done for another batch by
Stabell and Henningsmoen (1981) which found similar results. This
assumption is thus acceptable.
(4) There is no degradation of palynomorphs caused by chemical
treatment such as oxidation or acid treatments by HF and
HCl.
Since Lycopodium spores are acetolysed during the manufacturing
process, they can withstand acetolysis. Effects of chemicals on Lyco-
podium show that only colour changes are caused by acetolysis or HCl
treatment (Sengupta, 1975). On the other hand, it has been shown
that acetolysis or oxidation selectively destroys the cysts of the
Polykrikaceae and Protoperidiniaceae (Reid, 1977; Marret, 1993). KOH
treatment causes destruction of the Protoperidiniaceae after 5 min (de
Vernal et al., 1996, and Mertens, pers. observations) and causes
swelling of the palynomorphs. Likewise, methods using H
2
O
2
(Riding
et al., 2007) result in the destruction of protoperidiniacean cysts
(Riding, pers. comm., Hopkins and McCarthy, 2002; Mertens, pers.
obs.). This has also been demonstrated for Late Cretaceous peridinioid
dinoflagellate cysts (Schrank, 1988). Oxidation with Schulze's solution
by laboratory 22b resulted in the near complete destruction of the
RBC, SBC and other heterotrophs in all samples, and led to the relative
enrichment of resistant pollen and reworked non-peridinioid dino-
flagellate cysts. Cold HF and HCl have never been reported to destroy
dinoflagellate cysts. However, hot rinses with HCl after the HF
treatment were particularly harmful to recent peridinoid cysts
(Dale, 1976). Palynomorphs treated with warm HF clearly showed
traces of deterioration: destruction of delicate structures with
fragmentation along sutures and changes in wall texture with a
thickening of the robust structures (Plate I,11,16, Plate III, 6). It can be
concluded that this assumption is acceptable when chemical
degradation is minimized by using only cold hydrochloric and
hydrofluoric acid.
(5) Sonication causes no mechanical degradation of the pollen and
spores or dinoflagellate cysts.
The extensive use of ultrasound will not harm any dinoflagellate
cysts according to Funkhouser and Evitt (1959), however, other
authors report differential damage (e.g. Hodgkinson, 1991). This has
not yet been checked in a quantitative manner for dinoflagellate cysts.
The use of a sonic oscillator, although dependent on frequency
(Marceau, 1969) is extremely damaging: the sonication by laboratory
8 resulted in the destruction of RBC and SBC in the Benguela sample
(Plate IV, 20). Laboratory 18a used an ultrasonic bath for 30 min, and
this resulted in extensive damage to the cysts. Many cysts were
fragmented, often with broken or even lost spines and were often
clustered (Plate I,17,Plate III,13,Plate IV, 11). In addition micro-
foraminiferal linings were often fragmented. This assumption is thus
acceptable when an ultrasonic bath is not used for too long. A limit of
60 s is proposed.
(6) Centrifugation causes no mechanical degradation of the
palynomorphs.
No visible signs were noted that this technique causes degradation
of the cysts. This assumption is thus acceptable.
(7) Sieving causes no loss of palynomorphs.
Lignum et al. (2008) demonstrated that sieving should be done
with a sieve mesh width smaller than 15 µm. Our results confirm this
observation. Laboratories using nylon sieve with widths of 20 µm
(laboratories 1a and 1c) showed extremely high absolute abundances.
This suggests that significant losses of Lycopodium spores occurred
during the sieving process —even larger than the 20% that is proposed
by Lignum et al. (2008). No significant loss of cysts was documented in
this study. It is possible that cysts of Pentapharsodinium dalei pass
through20µmsieves,thisspecieswaspresentinsuchlow
abundances in the studied samples to significantly affect relative or
absolute abundances. This assumption is thus acceptable when mesh
sizes smaller than 15 µm are used.
(8) Decantation causes no loss of palynomorphs.
An experiment was done to determine how many Lycopodium
spores were lost during decanting and sieving. One gram of the NW
248 K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
Africa sample together with one Lycopodium tablet, was processed
with a HCl/HF/HCl cycle, followed by sieving on a nylon mesh of
10 µm. After every decantation, the decanted fluid was filtered
through a 0.25 µm Millipore filter. What remained on the filter was
counted under a transmitted light microscope. Only Lycopodium spores
were left on thefilters, as well as some amorphous organic matter (Table
5). The number of spores will be dependent of the size of the filter used.
Apparently 24% of the Lycopodium spores were lost during decanting.
This is notsurprising, sinceit is well-known that Lycopodium spores float
(e.g. Salter et al., 2002). An extra 1.3% was left on the filter and 1% got
stuck to handling material (e.g. spatula, tube). In the slides only 43.4% of
the Lycopodium spores were found. An additional 30.2% spores were
unaccounted for, and could have been lost during sieving and/or could
have been obscured by other material in the slides to some extent.
Because we did not expect any significant losses to occur during sieving,
we did notcapture sievedmaterial during this experiment. However, we
tested sieving a complete Lycopodium tablet on 10 µm and capture on a
0.25 µm sieve. We found losses to be 0.79% when gently pouring the
dissolved tablet over the sieve and subsequent washing, 0.97% when
using a hand pump to facilitate sieving and 2.01% when using a pipette
tip. Lignum et al. (2006) recorded losses up to 5.8± 1.2% for 15 µm
meshes. It can thus be assumed that only a small part of the missing
spores were pushed through the 10 µm nylon sieve. Presumably, spores
are oftenconcealed by beingobscured by other material, and this plays a
more significant role in explaining the missing amount of spores. Also, it
is possible that due to the texture of the exines of Lycopodium spores, the
spores get more easily caught in the sieves than smoother palyno-
morphs. However, this loss can be easily checked by the observer. This
assumption is thus not acceptable.
(9) Pre-sieving causes no losses.
It is unclear to what extent presieving causes loss of Lycopodium
spores, although it is evident that it should be avoided in samples from
high productivity areas, where high production of amorphous organic
matter forms large clusters in the sediment, which can be discarded
with the large fraction. However; it can be easily checked whether
Lycopodium spores were lost.
(10) Heavy liquid separation causes no loss of Lycopodium spores.
It has been noted that density separation with heavy liquids
can cause incorporation of mineral particles modifying the density
of the heavy liquid (de Vernal et al., 1996). Litwin and Traverse
(1989) recommend pyrite to be removed prior to density separation.
The results of this study do not show any obvious difficulties
with this processing step, although for clarity further study is
suggested.
From these considerations it can be concluded that a significant
amount of Lycopodium spores are lost, mainly during decanting and
sieving. There is little evidence that there is loss of dinoflagellate cysts
during these manipulations (Table 6).
5. Conclusions and recommendations
(1) This study was designed as a comparative one, where the
degree of variability in preparations could be objectively
assessed. The laboratories concerned agreed to take part on
the basis that the results would be presented anonymously, in
order to ensure maximum participation. The point of this work
was to carefully study the techniques used and to encourage
best practice in the future. This initial work presents a firm
basis for more methodological research.
(2) The exercise demonstrated that relative abundances are re-
producible, but underlined the urgent need for taxonomic
intercalibration.
(3) The study also shows that counting 300 dinoflagellate
cysts is sufficient both in terms of diversity and absolute
abundances.
Table 3
Comparison between the marker-grain method and the volumetric method.
Method Variable/sample North
Sea
Celtic
Sea
NW
Africa
Benguela
Marker grain
method
Average (cysts/g) 1516 2583 19,441 144,299
St dev (cysts/g) 698 1342 9148 84,159
Coefficient of variation (%) 46 52 47 58
Coefficient of variation (%)
Maher (1981)
20 25 23 21
Volumetric
method
Average (cysts/g) 1163 10,796 53,200
St dev (cysts/g) 5 1137 0
Coefficient of variation (%) 0 11 0
Difference Cysts/g –1420 8645 91,099
%554463
Table 4
Comparison between the average results after counting 300 dinoflagellate cysts, and counting 400 dinoflagellate cysts.
Variable/sample North Sea
300 cysts
North Sea
400 cysts
Celtic Sea
300 cysts
Celtic Sea
400 cysts
NW Africa
300 cysts
NW Africa
400 cysts
Benguela
300 cysts
Benguela
400 cysts
Average (cysts/g) 1539 1546 2792 2670 33,798 33,684 141,825 142,612
St dev 767 711 1474 1236 43,286 42,193 87,324 88,779
Coefficient of variation (%) 50 46 53 46 128 125 62 62
Species richness 22.00 22.85 24.26 25.26 14.75 16.50 19.13 20.22
St dev 4.67 4.79 5.61 6.02 3.64 4.12 4.94 5.27
Shannon–Wiener index 2.25 2.25 2.29 2.29 0.70 0.72 1.94 1.92
St dev 0.41 0.41 0.30 0.32 0.22 0.23 0.35 0.33
Table 5
The results of the counts of samples processed and counted by Lab 21, processed with
one processing technique. According to the statistical test by Maher (1981), the results
are reproducible.
Lab number Variable/sample North Sea Celtic Sea NW Africa Benguela
21a Dinoflagellate cysts/g 1547 2581 27,851 172,078
95% confidence limits
(Maher, 1981)
1265–
1885
2092–
3327
21,612–
32,060
138,365–
206,955
21b Dinoflagellate cysts/g 1447 2723 24,929 170,888
95% confidence limits
(Maher, 1981)
116 6–
1785
2117–
3354
19,294–
28,216
135,585–
200,884
249K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
(4) Absolute abundance calculations of dinoflagellate cysts are
dependent on processing methodology, since Lycopodium
spores are being lost during different processing steps.
(5) It is possible that some of the laboratories consistently over- or
underestimate concentrations. The addressed problems in
methodology might partly explain these outliers. Future work
should elucidate possible corrections by detailed investigation
of every different processing step.
(6) At the current state of affairs, there are three possible choices
the Quaternary worker can make to calculate reproducible
absolute abundances:
1. Standardize methodology for the extraction of dinoflagellate
cysts.
Since samples can be reproducible when one fixed methodology
is followed (see Section 4.3), a standard methodology is
suggested (Fig. 1). We consider that there are critical steps
that must be avoided in this standard method when preparing
samples for dinoflagellate cyst work: the use of oxidation, KOH,
warm acids, acetolysis, mesh sizes larger than 15 µm, decanting
(substituted by sieving) and sonication longer than 1 min.
During sieving, care should be taken to avoid Lycopodium spores
being forced through the sieve. A certain degree of freedom is
allowed in the number of HCl and HF cycles, length of
ultrasonication (0–60 s), duration of sieving and sieve mesh
size (6–14 µm), Care should be taken to neutralize HF by dilut-
ing at least ten times before sieving. Further studies are required
to fine-tune the method by focusing on designated issues.
2. Adding Lycopodium tablets at the end of processing.
The marker grain method is based on the assumption that
there is no selective loss of fossil and exotic pollen during the
procedures. However, this assumption has never been checked.
Our study suggests that predominantly Lycopodium spores are
lost, and that losses of dinoflagellate cysts are negligible.
Therefore the addition of Lycopodium tablets at the end of the
preparation is suggested, thus limiting the loss of Lycopodium
spores. However, this method is contrary to spiking with an
internal standard before the start of preparation.
3. Alternative methods.
Alternative methods can be used, but may not yield better
results. The use of microbeads was introduced by Ogden
(1986), but often results in much higher abundance estimates,
apparently because of difficulty in sustaining an even suspen-
sion of the particles in the stock solution: the higher specific
gravity of microspheres causes them to settle three to four
times more rapidly than pollen grains (McCarthy,1992). Other
marker-grain methods, such as the Eucalyptus globulus
marker-grain method (Matthews, 1969), has also been used
(e.g. de Vernal et al., 1987). However, it is not knownwhether
these methods give more reliable results. The aliquot method
gives more accurate results than the Lycopodium method in
our study, but unfortunately not much is known about the
precision of this method.
Acknowledgements
André Catrijsse (VLIZ), Karin Zonneveld and James Scourse are
thanked for providing samples. John Lignum (Kingston University)
and Richard Telford (Bjerknes Centre for Climate Research) are
thanked for fruitful discussions. Jane E. KyffinHughes and James B.
Riding publish with the permission of the Executive Director, British
Geological Survey (NERC). Ana Amorim refers to project MICRODYN-
POCTI/CTA/45185/2002. Three anonymous reviewers are thanked for
their constructive comments.
Table 6
Results of an experiment to look into the effects of manipulations on loss of Lycopodium
spores. Shown is the number of Lycopodium spores lost during each manipulation. It is
supposed that one tablet contains 18,583 spores, so the % is calculated by dividing the
number of counted spores by 18,583 spores.
Counted Lycopodium spores %
HCl treatment
First decantation 916 4.9
Second decantation 267 1.4
Third decantation 2485 13.4
HF/HCl treatment
First decantation 6 0.0
Second decantation 143 0.8
Third decantation 650 3.5
Left on filter (not washed off) 242 1.3
Left in tube +stuck on spatula 187 1.0
Found on slides 8067 43.4
Total 12963 69.8
Missing spores 5620 30.2
Fig. 1. Flow-chart of the proposed standardized method. AOM stands for amorphous
organic matter.
250 K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
Appendix A. Species list.
Species name Grouped under North Sea Celtic Sea NW Africa Benguela
Achomosphaera andalousiensis Jan du Chêne 1977 Spiniferites s.l. x x x
Cysts of Alexandrium affine (Ioue and Fukuyo 1985) Balech 1985 Cyst of Alexandrium spp. x x
Cysts of Alexandrium tamarense (Lebour 1925) Balech 1985 Cyst of Alexandrium spp. x x
Ataxiodinium choane Reid 1974 Ataxiodinium choane xxx
Bitectatodinium spongium Zonneveld 1997 Bitectatodinium spp. x x x
Bitectatodinium tepikiense Wilson 1973 Bitectatodinium spp. x x x x
Tectatodinium pellitum Wall, 1967 emend . Head 1994 Tectatodinium spp. x
cf. Tectatodinium pellitum Wall, 1967 emend. Head 1994 Tectatodinium spp. x
Brigantedinium cariacoense (Wall 1967) Lentin and Williams 1993 Round Brown Cyst x x x x
Brigantedinium majusculum Reid 1977 ex Lentin and Williams 1993 Round Brown Cyst x x
Brigantedinium simplex Wall 1965 ex Lentin and Williams 1993 Round Brown Cyst x x x x
Cyst of Protoperidinium americanum (Gran and Braarud 1935) Balech 1974 Round Brown Cyst x x x x
Dalella chathamense McMinn and Sun 1994 Dalella chathamense x
Diplopelta? symmetrica Pavillard 1993 (Dale et al., 1993) Spiny Brown Cysts x
Dubridinium ulsterum Reid 1977 Round Brown Cyst x x x
Dubridinium caperatum Reid 1977 Round Brown Cyst x x x x
Echinidinium aculeatum Zonneveld 1997 Spiny Brown Cysts x x x x
Echinidinium bispiniformum Zonneveld 1997 Spiny Brown Cysts x x
Echinidinium delicatum Zonneveld 1997 Spiny Brown Cysts x x x x
Echinidinium granulatum Zonneveld 1997 Spiny Brown Cysts x x x x
Echinidinium transparantum Zonneveld 1997 Spiny Brown Cysts x x x
Echinidinium cf. transparantum Zonneveld 1997 Spiny Brown Cysts x x x
Cyst of Gymnodinium catenatum Graham 1943 Cyst of Gymnodinium spp. x x x x
Cyst of Gymnodinium microreticulatum Bolch et al., 1999 Cyst of Gymnodinium spp. x x
Cyst of Gymnodinium nolleri Ellegaard and Moestrup 1999 Cyst of Gymnodinium spp. x x x x
Impagidinium aculeatum (Wall 1967) Lentin and Williams 1981 Impagidinium spp. x
Impagidinium pallidum Bujak 1984 Impagidinium spp. x
Impagidinium paradoxum (Wall 1967) Stover and Evitt 1978 Impagidinium spp. x x x
Impagidinium patulum (Wall 1967) Stover and Evitt 1978 Impagidinium spp. x x x
Impagidinium sphaericum (Wall 1967) Lentin and Williams 1981 Impagidinium spp. x x x
Impagidinium strialatum (Wall 1967) Stover and Evitt 1978 Impagidinium spp. x
Impagidinium velorum Bujak 1984 Impagidinium spp. x x
Islandinium? cezare de Vernal et al., 1989 ex de Vernal in Rochon et al., 1999 Spiny Brown Cysts x
Islandinium minutum Harland and Reid in Harland et al., 1980 Spiny Brown Cysts x x x x
Leipokatium invisitatum Bradford 1975 Lejeunecysta s.l. x
Lejeunecysta diversiforma (Bradford 1977) Artzner and Dörhöfer 1978 Lejeunecysta s.l. x
Lejeunecysta marieae Harland in Harland et al., 1991 ex Lentin and Williams 1993 Lejeunecysta s.l. x
Lejeunecysta oliva (Reid 1977) Turon and Londeix 1988 Lejeunecysta s.l. x x x x
Lejeunecysta paratenella (Benedek 1972) Zonneveld and Marret xxx Lejeunecysta s.l. x x x
Lejeunecysta sabrina (Reid 1977) Bujak 1984 Lejeunecysta s.l. x x x x
Lingulodinium machaerophorum (Deflandre and Cookson 1955) Wall 1967 Lingulodinium machaerophorum xxxx
Nematosphaeropsis labyrinthus (Ostenfeld 1903) Reid 1974 Nematosphaeropsis labyrinthus xxxx
Operculodinium centrocarpum sensu Wall and Dale (1966) Operculodinium s.l. x x x x
Operculodinium israelianum (Rossignol 1962) Wall 1967 Operculodinium israelianum xxx x
Operculodinium janduchenei Head et al., 1989 Operculodinium s.l. x x x x
Operculodinium sp. II? Marret, 1994 Operculodinium s.l. x
Operculodinium sp. A of Vink (2000) Operculodinium s.l. x
Cyst of Pentapharsodinium dalei Indelicato and Loeblich III 1986 Cyst of Pentapharsodinium dalei xxxx
Polykrikos kofoidii Chatton 1914 Polykrikos spp. x x x x
Polykrikos schwartzii Bütschli 1873 Polykrikos spp. x x x x
Polysphaeridium zoharyi (Rossignol 1962) Bujak et al., 1980 Polysphaeridium zoharyi xxxx
Pyxidinopsis reticulata (McMinn & Sun 1994) Marret and de Vernal 1997 Pyxidinopsis reticulata x
Quinquecuspis concreta (Reid 1977) Harland, 1977 Quinquecuspis concreta xxxx
Selenopemphix crenata Matsuoka and Bujak, 1988 Selenopemphix s.l. x
Selenopemphix nephroides Benedek 1972; emend. Bujak in Bujak et al.,1980; emend.
Benedek and Sarjeant 1981
Selenopemphix s.l. x x x x
Cyst of Protoperidinium nudum (Meunier 1919) Balech 1974 Selenopemphix s.l. x x x x
Selenopemphix quanta (Bradford 1975) Matsuoka 1985 Selenopemphix s.l. x x x
Spiniferites belerius Reid 1974 Spiniferites s.l. x x x x
Spiniferites bentorii (Rossignol 1964) Wall and Dale 1970 Spiniferites s.l. x x x x
Spiniferites bulloideus (Deflandre & Cookson 1955) Sarjeant 1970 Spiniferites s.l. x x x
Spiniferites delicatus Reid 1974 Spiniferites s.l. x x x x
Spiniferites elongatus Reid 1974 Spiniferites s.l. x x x
Spiniferites hyperacanthus (Deflandre and Cookson 1955) Cookson and Eisenack 1974 Spiniferites s.l. x x x x
Spiniferites lazus Reid 1974 Spiniferites s.l. x x x
Spiniferites membranaceus (Rossignol 1964) Sarjeant 1970 Spiniferites s.l. x x x x
Spiniferites mirabilis (Rossignol 1964) Sarjeant 1970 Spiniferites s.l. x x x x
Spiniferites pachydermus Rossignol 1964 Spiniferites s.l. x x x
Spiniferites ramosus (Ehrenberg 1838) Loeblich and Loeblich 1966; emend.
Davey and Williams 1966
Spiniferites s.l. x x x x
Stelladinium reidii Bradford 1975 Stelladinium spp. x x x
Stelladinium stellatum (Wall and Dale 1968) Reid 1977 Stelladinium spp. x x x x
Trinovantedinium applanatum (Bradford 1977) Bujak and Davies 1983 Trinovantedinium applanatum xxx x
(continued on next page)
(continued on next page)
251K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
Author's personal copy
Appendix B. Error calculation according to Stockmarr (1971)
According to Stockmarr (1971) total error is e=ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
e2
1+e2
2+e2
3
q
where
e
1
= error on number of spores in marker tablets
e2=ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
cysts counted
p
cysts counted = error on dinoflagellate cysts counted
e3=ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
spores counted
p
spores counted = error on the number of spores counted
.Appendix C. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.revpalbo.2009.05.004.
References
Benninghoff, W.S., 1962. Calculation of pollen and spores density in sediments by
addition of exotic pollen in known quantities. Pollen et Spores 6, 332–333.
Dale, B., 1976. Cyst formation, sedimentation, and preservation; factors affecting
dinoflagellate assemblages in recent sediments from Trondheimsfjord, Norway.
Review of Palaeobotany and Palynology 22, 39–60.
Dale, B., Dale, A.L., Fred Jansen, J.H., 2002. Dinoflagellate cysts as environmental
indicators in surface sediments from the Congo deep-sea fan and adjacent regions.
Palaeogeography, Palaeoclimatology, Palaeoecology 185, 309–338.
Desezar, Y.B., Poulsen, N.E., 1994. On palynological preparation technique. American
Association of Stratigraphic Palynologists, Newsletter 27 (3), 12–13.
de Vernal, A., Larouche, A., Richard, P.J.H., 1987. Evaluation of palynomorph concentra-
tions: do the aliquot and the marker-grain methods yield comparable results?
Pollen et Spores 19 (2–3), 291–304.
de Vernal, A., Henry, M., & Bilodeau, G., 1996. Techniques de préparation et d'analyse en
micropaléontologie. Les cahiers de GEOTOP, 3, Université de Québec a Montréal,
unpublished report, 28 pp.
Fensome, R.A., Williams, G.L., 2004. The Lentin and Williams index of fossil
dinoflagellates, 2004 edition. AASP Contributions Series Number 42, 909 pp.
Funkhouser, J.W., Evitt, W.R., 1959. Preparation techniques for acid-insoluble micro-
fossils. Micropalaeontology 5, 369–375.
Germerad, J.H., Hopping, C.A., Muller, J., 1968. Palynology of Tertiary sediments from
tropical areas. Review of Palaeobotany and Palynology 6, 189–348.
González, C., Dupont, L.M., Mertens, K., Wefer, G., 2008. Reconstructing marine
productivity of the Cariaco Basin during marine isotope stages 3 and 4 using
organic-walled dinoflagellate cysts. Paleoceanography 23 (PA3215). doi:10.1029/
2008PA001602.
Heilmann-Clausen, C., 1985. Dinoflagellate stratigraphy of the Uppermost Danian to
Ypresian in the Viborg 1 borehole, Central Jylland, Denmark. Serie A/Danmarks
Geologiske Undersøgelse 7, 1–69.
Herrle, J.O., Bollman, J., 2004. Accuracy and reproducibility of absolute nannoplankton
abundances using the filtration technique in combination with a rotary splitter.
Marine Micropaleontology 53, 389–404.
Hodgkinson, R.I., 1991. Microfossil processing: a damage report. Micropalaeontology 37,
320–326.
Holzwarth, U., Esper, O., Zonneveld, K., 2007. Distribution of organic-walled dino-
flagellate cysts in shelf surface sediments of the Benguela upwelling system in
relationship to environmental conditions. Marine Micropaleontology 64, 91–119.
Hopkins, J.A., McCarthy, F.M.G., 2002. Postdepositional palynomorph degradation in
Quaternary shelf sediments: a laboratory experiment studying the effects of
progressive oxidation. Palynology 26, 167–184.
Lignum, J., Jarvis, I., Pearce, M., 2008. A critical assessment of standard processing
methods for the preparation of palynological samples. Review of Palaeobotany and
Palynology 149, 133–149.
Litwin,R.J., Traverse,A., 1989.Basic guidelinesfor palynomorphextractionand preparation
from sedimentary rocks. In:Feldman, R.M., Chapman, R.E., Hannibal, J.T. (Eds.),Paleo-
techniques: Paleontological Society, Special Publication, vol. 4, pp. 87–98.
Maher Jr., L.J., 1981. Statistics for microfossil concentration measurements employing
samples spiked with marker grains. Review of Palaeobotany and Palynology 32,
153 –191.
Marceau, L., 1969. Effets, sur le pollen, des ultrasons de basse frequence. Pollen et Spores
11, 14 7–16 4.
Marret, F., 1993. Les effets de l'acétolyse sur les assemblages de kystes de dinoflagellés.
Palynosciences 2, 267–272.
Marret, F., Scourse, J., 2002. Control of modern dinoflagellate cyst distribution in the
Irish and Celtic seas by seasonal stratification dynamics. Marine Micropaleontology
47, 101–116.
Matthews, J.,1969. The assessment of a method for the determination of absolute pollen
frequencies. New Phytologist 68, 161–166.
McCarthy, F.M.G., 1992. Quaternary climate change and the evolution of the mid-
latitude western North Atlantic Ocean: palynological, foraminiferal, sedimentolo-
gical, and stable isotope evidence from DSDP sites 604, 607 and 612, unpublished
PhD dissertation, Department of Geology, Dalhousie University, Halifax, 270 pp.
Ogden III, J.G., 1986. An alternative to exotic spore or pollen addition in quantitative
microfossil studies. Canadian Journal of Earth Sciences 23, 102–106.
Pospelova, V., Pedersen, T.F., de Vernal, A., 2006. Dinoflagellate cysts as indicators of
climaticand oceanographic changes during the past 40 kyr in theSanta Barbara Basin,
southern California. Paleoceanography 21 (PA2010). doi:10.1029/2005PA001251.
Poulsen, N.E., Gudmundsson, L., H ansen, J.M., Husfeldt, Y., 1990. Palynologica l
preparation techniques, a new maceration tank-method and other modifications.
Geological Survey of Denmark, Series C 10. 24 pp.
Reid, P.C., 1977. Peridiniacean and glenodinicacean dinoflagellate cysts from the British
Isles. Nova Hedwigia 29, 429–463.
Riding, J.B., Kyffin-Hughes, J.E., 2004. A review of the laboratory preparation of
palynomorphs with a description of an effective non-acid technique. Revista
Brasileira de Paleontologia 7 (1), 13–44.
Riding, J.B., Kyffin-Hughes, J.E., Owens, B., 2007. An effective palynological preparation
procedure using hydrogen peroxide. Palynology 31, 19–36.
Rosell-Melé, A., Bard, E., Emeis, K.C., Grimalt, J., Muller, P., Schneider, R., Bouloubassi, I.,
Epstein, B., Fahl, K., Fluegge, A., Freeman, K., Goñi, M., Guntner, U., Hartz, D.,
Hellebust, S., Herbert, T., Ikehara, M., Ishiwatari, R., Kawamura, K., Kenig, F., de
Leeuw, J., Lehman, S., Ohkouchi, N., Pancost, R.D., Prahl, F., Quinn, J., Rontani, J.F.,
Rostek, F., Rullkotter, J., Sachs, J., Sanders, D., Sawada, K., Schultz-Bull, D., Sikes, E.,
Ternois, Y., Versteegh, G., Volkman, J., Wakeham, S., 2001. Precision of the current
methods to measure alkenone proxy UK'37 and absolute alkenone abundance in
sediments: results of an inter-laboratory comparison study. Geochemistry,
Geophysics, Geosystems 2, 1–28 20 00GC00141.
Salter, J., Murray,B.G., Braggins, J.E., 20 02. Wettableand uns inkable: the hydrodynamics
of saccate pollen grains in relation to the pollination mechanism in the two New
Zealand species of Prumnopitys Phil. (Podocarpaceae). Annals of Botany 89,
133–144.
Schrank, P., 1988. Effects of chemical processing on the preservation of peridinoid
dinoflagellates: a case from the Late Cretaceous of NE Africa. Review of
Palaeobotany and Palynology 56, 123–140.
Sengupta, S., 1975. Experimental alterations of the spores of Lycopodium clavatum as
related to diagenesis. Review of Palaeobotany and Palynology 19, 173–192.
Stabell, B., Henningsmoen, K.E., 1981. Capsules with Lycopodium spores for absolute
diatom and pollen analysis. Nordic Journal of Botany 1 (5), 701–702.
Stanley, E.A., 1966. The problem of reworked pollen and spores in marine sediments.
Marine Geology 4, 397–408.
Stockmarr, J.,1971. Tablets with spores used in absolute pollen analysis. Pollen et Spores
13, 615 –621.
Wolfe, A.P., 1997. On diatom concentrations in lake sediments: results from an inter-
laboratory comparison and other tests performed on a uniform sample. Journal of
Paleolimnology 18, 261–268.
Wood, G.D., Gabriel, A.M., Lawson, J.C.,1996. Palynological techniques —processing and
microscopy. In: Jansonius, J., McGregor, D.C. (Eds.), Palynology: Principles and
Applications, vol. 1. American Association of Stratigraphic Palynologists Foundation,
Dallas, TX, pp. 29–50.
Zachariasse, W.J., Riedel, W.R., Sanfilippo, A., Schmidt, R.R., Brolsma, M.J., Schrader, H.J.,
Gersonde, R., Drooger, M.M., Broekman, J.A., 1978. Micropaleontological counting
methods and techniques; an exercise on an eight metres section of the lower
Pliocene of Capo Rossello, Sicily. Utrecht Micropaleontological Bulletins 265 pp.
Zonneveld, K.A.F., Versteegh, G.J.M., de Lange, G.J., 2001. Palaeoproductivity and post-
depositional aerobic organic matter decay reflected by dinoflagellate cyst
assemblages of the Eastern Mediterranean S1 sapropel. Marine Geology 172,
181 –195.
Appendix A (continued)
Species name Grouped under North Sea Celtic Sea NW Africa Benguela
Tuberculodinium vancampoae (Rossignol 1962) Wall 1967 Tuberculodinium vancampoae xxx
Votadinium calvum Reid 1977 Votadinium spp. x x x x
Votadinium spinosum Reid 1977 Votadinium spp. x x x
Xandarodinium xanthum Reid 1977 Xandarodinium xanthum xxxx
Appendix A (continued)
252 K.N. Mertens et al. / Review of Palaeobotany and Palynology 157 (2009) 238–252
