Microgaming Casinos Usa Players

Knapp
et al.—Morphology
of Discostella
stelligera
and D. elentarii
New
Zealand
Journal of Marine
and Freshwater
Research,
2006, Vol. 40: 429–438
0028–8330/06/4003–0429 © The Royal Society of New Zealand 2006
429
A comparison of the morphology and ultrastructure of the diatoms
(Bacillariophyceae) Discostella stelligera and D. elentarii from two
lakes in Fiordland, New Zealand
J.M. Knapp1,2
P.C. Furey1
R.L. Lowe1,3
1Department of Biological Sciences
Bowling Green State University
Bowling Green
OH 43403–0218, United States
email: [email protected]
2School
of Natural Resources and Environment
University of Michigan
440 Church St, Ann Arbor
MI 48109, United States
3University
of Michigan Biological Station
Pellston
MI 49769, United States
Abstract Morphological and ultrastructural
variability of the diatoms, Discostella stelligera
and Discostella elentarii, were examined from lakes
Te Anau and Manapouri, located in Fiordland on
the South Island of New Zealand. The objectives
of this study were to examine and compare the size
variability, frustule morphology, and ultrastructural
detail of these taxa, which were recently collected
from locations of a different origin than the type
localities. To support the current descriptions of
the newly erected genus of Discostella and recently
described species, D. elentarii, information on
morphological characteristics across valve patterns
was compared. A principal components analysis
of the external valve features of D. stelligera and
D. elentarii measured by light microscopy did
not separate the two taxa. This close similarity in
external morphology highlights the need to optically
dissect the valves underneath the light microscope
to clearly differentiate between the convex/concave
M05069; Online publication date 14 July 2006
Received 1 November 2005; accepted 9 June 2006
nature of the D. stelligera valves compared with
the flat valve of D. elentarii. New external and
internal morphological details of D. elentarii were
documented with a field emission scanning electron
microscope.
Keywords Cyclotella; Discostella stelligera;
D. elentarii; diatoms; field emission microscope;
Lake Te Anau; Lake Manapouri; New Zealand;
morphology; ultrastructure
INTRODUCTION
The diatom genus Cyclotella (Kützing) Brébisson
has been a recent focus of systematists, and clades of
related species have been transferred to newly erected
genera (Skabichevskii 1975; Håkansson 2002). The
stelligeroid taxa of Cyclotella have been transferred
to a new genus, Discostella (Houk & Klee 2004).
Cyclotella sensu lato has been characterised by the
presence of two distinct patterns on the valve face
consisting of marginal, radially arranged striae, and a
distinctly different pattern in the central area (Lowe
1975). Cyclotella sensu lato has fultoportulae located
in a ring near the margin and, in some species, also
in the central area (Lowe 1975). The rimoportulae
are few and located on the costae or at the edge of
the central area (Round 1990). Within the genus
Cyclotella, there are c. 15 species of stelligeroid
taxa, which are distinguished from other Cyclotella
by the presence of a stellate pattern in the central
area (Houk & Klee 2004). These taxa have marginal
fultoportulae and one rimoportula located at the valve
edge between two costae. The external expression
of the rimoportula is more centripetal to the valve
margin than that of the fultoportulae (Lowe 1975;
Houk & Klee 2004). The genus Discostella was
erected to accommodate a clade of species whose
placement of the fultoportulae and rimoportulae
were inconsistent with the current description of
Cyclotella (Houk & Klee 2004).
Discostella stelligera (Cleve et Grunow) Houk
et Klee, the type species from Lake Rotoaira, New
430New Zealand Journal of Marine and Freshwater Research, 2006, Vol. 40
Zealand, maintains the two morphologically distinct
patterns of the valve face that consist of marginal
radially arranged striae and a distinctly different
central area (Houk & Klee 2004). These morphological
patterns of D. stelligera are further characterised by a
concentrically undulate valve face with a stellate pattern
composed of alveoli or external ridges. Stellate patterns
composed of alveolae that are open to the inside are
always present on valves that are convex, whereas
concave central areas lack alveolae and have an illdefined or “ghost” stellate pattern. The frequency and
mechanisms of regulation for the development of these
morphological patterns (i.e., degree of development or
pattern of the alveolate stellate pattern) are not known.
Inside the valve, the collared fultoportulae are located
between every one to four marginal costae with satellite
pores situated laterally on either side of the process. The
single rimoportula is located between two costae and
is externally expressed with a simple opening (Houk
& Klee 2004).
Discostella elentarii (Alfinito et Tagliaventi) Houk
et Klee, a recently described stelligeroid species, was
first collected from Lake Monowai on the South
Island of New Zealand (Alfinito & Tagliaventi
2002). Discostella elentarii has flat valves with a
large, colliculate central area with scattered punctae.
A pore at the end of a shortened stria marks the
external expression of the fultoportulae located
every one to two striae. The interior of the valve is
smooth in the centre with costae and alveoli in the
marginal area. The fultoportulae are located between
the costae and are surrounded by two satellite pores.
There is one sessile rimoportula per cell. Some cells
have a faint stellate pattern in the central region of
the valve interior that does not penetrate the siliceous
layer (Alfinito & Tagliaventi 2002).
Using light microscopy (LM), it can be difficult
to differentiate between D. stelligera and D.
elentarii given the ill-defined stellate pattern on
the external concave valve face of D. stelligera and
the faint stellate pattern on the internal valve face
of D. elentarii. The purpose of the present work
was to describe and compare the size variability,
frustule morphology, and ultrastructural detail of
D. stelligera and D. elentarii to better differentiate
between them. This study was centred on LM and
scanning electron microscopy (SEM) observations
of recent collections from two fiord lakes, and differs
from previous studies of these taxa that have focused
on collections from the type localities. Observations
included the use of a field emission SEM, which
enabled close examination and documentation of fine
ultrastructural detail of D. elentarii that previously
Fig. 1 Map of New Zealand showing the location of
Discostella stelligera in Lake Rotoaira (star) as reported
by Houk & Klee (2004), D. elentarii in Lake Monowai
(circle) as reported by Alfinito & Tagliaventi (2002) and
collection sites of both taxa in Lake Te Anau (triangle) and
Lake Manapouri (square) from this study.
had not been described for this genus. This work
supports recent name changes for D. stelligera and
D. elentarii, and provides a modern record of two
additional New Zealand locations for these taxa.
The descriptions of the two fiord lakes are provided
to contribute knowledge of the habitat of these two
newly transferred taxa.
MATERIALS AND METHODS
Study sites
Diatom samples were collected from lakes Te Anau
(45°12′S, 167°47′E) and Manapouri (45°31′S,
167°27′E) on the South Island of New Zealand
(Fig. 1). Previous studies of D. stelligera focused
on collections made from Lake Rotoaira (39°01′S,
175°42′E), a caldera on the North Island, and previous
studies of D. elentarii focused on collections from
Knapp et al.—Morphology of Discostella stelligera and D. elentarii
Lake Monowai (45°87′S, 167°45′E) of the South
Island in the same Fiordland region as lakes Te Anau
and Manapouri (Fig. 1).
Lake Te Anau originated from the union of four
glacial valleys resulting in highly complex basin
morphologies (Carter & Lane 1996). It is the largest
lake on New Zealand’s South Island, and it occurs
at 203 m a.s.l. (Jolly 1968; Glasby et al. 1991). The
lake has a surface area of 347 km2, with a long axis of
60 km. Lake Te Anau has three fiord arms, the North
Fiord, Middle Fiord and South Fiord. There are 26
islands within the lake and eight named basins, six
of which are over 250 m deep (Jolly 1968; Glasby
et al. 1991). Lake Te Anau can be classified as an
oligotrophic, warm monomictic lake (Jolly 1968;
Glasby et al. 1991). The lake is relatively clear with
an average Secchi disc reading of 10 m (Wells et al.
1998). The main inflows are located at the head of
the lake and in the three fiord arms, and the main
outflow is into the Waiau River, which flows to Lake
Manapouri. The catchment surrounding Lake Te
Anau covers an area of 2998 km2 and is composed
of 43% native forest, 37% tussock, 12% lake, and
4% lowland scrub (Jolly 1968; Glasby et al. 1991).
The western part of the catchment contains primarily
metamorphic rock with some granite and sandstones,
whereas the eastern part of the catchment is primarily
gravels (Jolly 1968; Glasby et al. 1991).
Lake Manapouri is smaller than Lake Te Anau
(surface area 143 km2) with the greatest breadth and
length being 9.6 and 19.3 km, respectively (Ellwood
et al. 2001). It is a deep (max. 444 m), glacial lake
fed primarily by the Waiau River, which drains Lake
Te Anau (Ellwood et al. 2001). Lake Manapouri has
an average Secchi disc reading of 6.5 m (Wells et al.
1998) and can be classified as a warm, monomictic
lake. The catchment surrounding Lake Manapouri
is c. 3000 km2 and is composed of tussock-covered
hills or native forestland (Ellwood et al. 2001).
Sample collection and analysis
Periphyton was collected from littoral areas from various
arms of Lake Te Anau on 18 February 2002, and Lake
Manapouri on 21 and 22 February 2002. Macrophytes
were shaken in a bag with water to remove epiphyton.
Epilithon was scraped off rocks using a spoon, and
epipsammon and epipelon were carefully removed
using a pipette. Samples were air dried onto aluminum
foil and processed in the laboratory.
Diatoms were cleaned in preparation for LM
(Round et al. 1990) and SEM (Postek et al. 1980).
Twenty-five ml of sample were removed and boiled
in nitric acid in a 1:1 sample to acid ratio until the
431
volume returned to the original volume of 25 ml. The
samples were rinsed with distilled water and allowed
to settle for 8 h, followed by decanting. This rinsing
process was repeated until the pH of the suspension
was neutral (c. 10 rinses). The cleaned material was
then concentrated, air dried onto coverslips and
processed for either LM or SEM analysis.
For LM, coverslips were permanently mounted
onto glass slides with Naphrax® mounting medium
and analysed under oil immersion at 1000× using
an Olympus BX51 photomicroscope with high
resolution Nomarski DIC optics. Digital images
were captured with a monochromatic camera
(Spot®) attached to the microscope and recorded
on computer. Size variability and external valve
features were measured digitally using SPOT ®
software (v. 4.01, Diagnostic Instuments, Inc.)
calibrated against a stage micrometer and recorded
for 53 D. stelligera valves and 211 D. elentarii
valves. The density of D. stelligera valves was
much lower than that for D. elentarii valves, thus
an unequal number of valve measurements were
obtained for the two taxa. Measurements included
the valve diameter, the diameter of the central area
(from valve centre to the inner start of the striae),
striae length, number, and density, the ratio of striae
length to radius of the central area, and the ratio of
the centre radius to the valve radius. Striae density
was calculated in two ways. First, the conventional
striae/10 µm was calculated. Next, the number of
striae/90° was calculated. This second calculation
provides a standardised summary of the striae
density to ensure that the size of the cell does not
alter the proportion of the valve from which the
striae are counted. Summaries of the morphological
characteristics were calculated using JMP (v. 5.1).
A principal components analysis (PCA; Primer 5 v.
5.2.9) (Legendre & Legendre 1998) was performed
on standardised morphological measurements based
on a correlation matrix to test whether the external
morphological variables other than the flex of the
central area (flat, concave, convex) could distinguish
between D. stelligera and D. elentarii or if different
size dependent trends were present for each taxa. A
PCA was performed on measurements that applied to
all valves (no inner stellate measurements included),
first for both taxa considered together and then for
each taxon separately. A PCA was also applied to D.
stelligera data including measurements of the inner
stellate pattern.
For SEM, coverslips were mounted on specimen
stubs and either sputter-coated with a gold/palladium
alloy (10 nm) and analysed using a Hitachi S2700
432New Zealand Journal of Marine and Freshwater Research, 2006, Vol. 40
SEM, or coated with 1.5 nm of iridium and analysed
on a Hitachi S4800 field emission variable pressure
scanning electron microscope (FEVP).
RESULTS
External valve morphology
Valves of D. elentarii were flat, either with a
colliculate valve face (Fig. 2–7) or with a light
ghost striae pattern in the centre area (Fig. 8–13).
In contrast, valves of D. stelligera were concave or
convex in the centre, either with ghost striae or a
more developed stellate pattern in the centre (Fig.
14–19). Under the light microscope, valves with
less distinct flexing or stellate characteristics were
observed. For example, Fig. 20 shows a valve with
a developed stellate pattern in the central area, but a
flat valve face. Fig. 21 depicts a valve with a shadow
stellate pattern in the central area, but the shadowed
central area suggests the presence of a convex valve
face. Similarly, the slightly degraded valve in Fig.
22 shows a valve with shading in the central area,
suggesting flex, and the absence of a clear stellate
pattern. Optical dissection of these valves under
the light microscope is necessary to determine the
convex flex of the valve.
Valves of D. elentarii had diameters that ranged
from 4.0 to 25.9 µm (Fig. 2–13), whereas D. stelligera
valve diameters ranged from 5.0 to 17.1 µm (Fig.
14–19; Table 1). Discostella elentarii had a broader
range in centre diameter than D. stelligera (Table 1).
The minimum striae length was the same for both
taxa, but the maximum striae length was longer for
D. elentarii than for D. stelligera (Table 1). The
range in striae/10 µm was similar across both taxa;
however, the range in striae per 90° was greater for
D. elentarii than for D. stelligera (Table 1). The inner
stellate pattern of D. stelligera ranged from 1.7 to
8.3 µm in diameter with 5–16 alveoli and was with
or without a central alveolus (Table 1).
PCA of the external morphological variables other
than the flex of the central area (flat, concave, convex)
did not clearly separate D. stelligera and D. elentarii
(Fig. 23). When examined together or separately by
taxon, the first two PCA axes explained over 90%
of the variance (PCA-A, B, C), with over 69% of
variance explained by the first PCA axis. Component
loadings of the first PCA axis represented changes in
morphological structure with size (Table 2). The first
axis had positive correlations with most variables,
and negative correlations with striae length (µm)/
centre radius (µm), and striae/10 µm (Table 2).
Although different size-dependent trends were not
apparent for D. elentarii and D. stelligera, variance
of the second PCA axis was explained by slightly
different morphological characteristics. Part of the
variance of the second PCA axis for both taxa was
explained by striae length (µm), and striae length
(µm)/centre radius (µm) (Table 2). Additional
variance of the second PCA axis for D. elentarii
was explained by cell diameter and total number
of striae, the latter having a higher loading value
and a negative relationship (Table 2). In contrast,
additional variance of the second PCA axis for D.
stelligera was explained by centre radius (µm)/total
radius (µm), and striae density (Table 2). When
the measurements of the inner stellate pattern of
D. stelligera were included in the PCA (PCA-D),
the variance was spread over three axes with the
presence or absence of the centre alveolus explaining
most of the variance of the third PCA axis (Table
2).
Spines were documented in one or more rows on
the valve face of D. elentarii, and their arrangement
varied from no spines to spines encircling the entire
valve face (Fig. 24–27). In contrast, spines were not
observed on valves of D. stelligera (Fig. 28–30).
Every valve of D. elentarii examined using FEVP
contained a fimbriate margin (Fig. 38–40). The
fultoportulae were expressed externally as a round
opening on the margin of the striae (Fig. 41, 42);
however, the external expression of the rimoportula
was not observed. External expression of the punctae
located on the margin of each costa, including those
around the fultoportulae, was observed using both
SEM and FEVP (Fig. 40, 41). The central area of D.
elentarii was flat with punctae scattered throughout
(Fig. 24, 25). These punctae have not been previously
documented, and it is believed that they are now
visible owing to the high resolution achieved using
FEVP.
Internal valve morphology
Collared fultoportulae, surrounded by two satellite
pores, occurred between every two to four costae
on D. elentarii (Fig. 43–45). Punctae were found
fairly consistently on the margin of every costa
of D. elentarii valves, and were only occasionally
absent (Fig. 44–47). The punctae of D. elentarii were
covered with a cribrum that did not appear clearly in
previous internal valve micrographs of Discostella. It
is not known if this cribrum is present in D. stelligera
as valves were not observed under FEVP because
of limited access to the FEVP. No more than one
nearly sessile rimoportula was observed on valves
Fig. 2–22 A series of light micrographs showing the size range and external
morphology of Discostella elentarii and D. stelligera. Fig. 2–7 D. elentarii
with a smooth, flat valve face. Fig. 8–13 D. elentarii with ghost striae in the
centre area. Fig. 14–19 D. stelligera with a convex valve face with a distinct
stellate pattern in the centre. Fig. 20–22 Valves where the characteristics
of these two taxa are less clear. Fig 20 A flat valve with a stellate pattern.
Fig. 21 A valve with ghost striae that needs to be optically dissected to determine the flex of the valve. Fig. 22 A D. stelligera valve where the stellate
pattern is less distinct and optical dissection is required. Scale bar: 10 µm.
Knapp et al.—Morphology of Discostella stelligera and D. elentarii
433
16.0
5.0
8.3
1.7
0.69
0.68
0.36
0.38
21
13.5
4.0
4.5
14.5
14.7
8.0
8.6
5.1
3.4
1.0
1.0
17.5
11.1
210
53
D. elentarii
D. stelligera
25.9
17.1
4.0
5.0
n
1.9
2.5
StriaeRatio
length
Striae per
Striae per
centre radius/
(µm)
10 µm
90°
valve radius
min. max. min. max. min. max. min. max.
Centre
diameter
(µm)
min. max.
Cell
diameter
(µm)
min. max.
Taxa
Table 1 Summary of cell measurements for Discostella stelligera and D. elentarii.
Centre
stellate
diam. (µm)
min. max.
Centre
stellate
alveoli (no.)
min. max.
434New Zealand Journal of Marine and Freshwater Research, 2006, Vol. 40
Fig. 23 2-dimensional PCA ordination of the first two
principal components axes for the morphological measurements (not including inner stellate measurements) of
both taxa—Discostella stelligera (closed circles) and D.
elentarii (open circles).
Fig. 24–37 External and internal views of Discostella ➤
elentarii and D. stelligera. Fig. 24–27 External valve
view of D. elentarii. Fig. 28–30 External valve view
of D. stelligera. Fig. 31–34 Internal valve view of D.
elentarii showing both flat and slightly stellate centre area.
Fig. 35–37 Internal valve view of D. stelligera showing
distinct stellate pattern. Scale bars: 2 µm.
of both D. stelligera and D. elentarii (Fig. 48–50).
Discostella elentarii had an inward projection whose
position was uncertain as to whether it occurs on the
girdle band or the inside edge of the valve. Closer
examination of both the valve margin and single
copula of the girdle is required to better understand
the location of this structure (Fig. 51, 52). Also
on this ambiguous part of the cell, a row of pores
was noted (Fig. 53–55). There was no previous
record of these pores, and the taxonomic importance
is unknown. These pores were only visible using
the FEVP, and as no valves of D. stelligera were
observed under FEVP, it is unknown if they are
present on D. stelligera.
DISCUSSION
Owing to the range of variability in morphology
and heterovalvar nature of stelligeroid taxa, it
can be difficult to distinguish between taxa with
Knapp et al.—Morphology of Discostella stelligera and D. elentarii
435
Table 2 Principal components analyses (PCA) of the correlation matrix for the morphological characteristics of
Discostella stelligera and D. elentarii. PCA component loadings of the morphological characteristics: A, both taxa; B, D.
elentarii; C, D. stelligera (no inner stellate measurements); D, D. stelligera (including inner stellate measurements).
A
B
C
D
Axis 1Axis 2Axis 1Axis 2Axis 1Axis 2Axis 1Axis 2Axis 3
Eigen values (λ)
6.14
1.29
6.41
1.09
5.53
1.81
Cell diam. (µm)
0.397 0.133 0.390 0.366 0.414 0.160
Striae length (µm)
0.352 0.388 0.355 0.455 0.305 0.492
Striae length (µm)/centre radius (µm)
–0.303 0.498 –0.315 –0.419 –0.299 –0.481
Centre radius (µm)/total radius (µm)
0.330 –0.463 0.343 –0.001 0.316 –0.465
Centre diam. (µm)
0.400 0.005 0.391 –0.061 0.422 –0.023
Total no. of striae
0.394 –0.025 0.386 –0.680 0.413 –0.004
Striae/10 µm
–0.218 –0.607 –0.231 –0.061 –0.165 –0.533
Striae/90°
0.394 –0.025 0.386 –0.061 0.413 –0.004
No. of inner stellate alveoli
Inner stellate diam. (µm)
Presence of centre alveolus
Total variance explained (%)
76.8
16.1
80.1
13.6
69.2
22.6
Cumulative variance explained (%)
76.8
92.9
80.1
93.7
69.2
91.8
7.08
0.365
0.267
–0.261
0.278
0.373
–0.364
–0.143
0.364
0.303
0.356
–0.085
64.3
64.3
1.84
–0.160
–0.491
–0.477
0.462
0.022
0.000
0.524
0.000
0.104
–0.076
0.011
16.7
81.0
1.05
–0.014
–0.077
–0.089
0.089
0.021
–0.098
–0.290
–0.098
0.203
0.044
0.912
9.6
98.5
436New Zealand Journal of Marine and Freshwater Research, 2006, Vol. 40
Fig. 38–42 External view of Discostella elentarii captured by field emission variable pressure scanning electron
microscope. Fig. 38–40 Fimbriate margin on external valve face (a arrows). Fig. 41, 42 External expression of
the fultoportulae (b arrows). Fig. 40, 41 External expression of the punctae located on the margin of each costa
(c arrows). Scale bars: 1 µm.
similar or overlapping morphological characteristics
such as those of D. stelligera and D. elentarii. It
is probable that D. stelligera and D. elentarii are
sibling species. Discostella stelligera was originally
described from New Zealand material (Houk & Klee
2004) but has been reported from many localities in
both hemispheres (Houk & Klee 2004). Discostella
elentarii is thus far endemic to New Zealand and
may have recently descended from D. stelligera,
making separation of the two species difficult. Our
initial examination of the stelligeroid valves from
lakes Te Anau and Manapouri did not result in a clear
differentiation between the two taxa owing to the
morphological variability in the central area. Upon
further examination, it was clear that the concave/
convex nature of the valves is a critical factor in
distinguishing between these taxa, especially when
ghost striae are present. The PCA analysis of the
morphological features demonstrated the overlap
in external valve features between D. stelligera and
D. elentarii. Our study emphasises the importance
of optically dissecting valves of D. stelligera and
D. elentarii to determine the flat or concave/convex
nature of the valve when the stellate characteristics
of the centre area appear similar; i.e., on concave
Fig. 43–50 Internal view of Discostella elentarii cap- ➤
tured by field emission variable pressure scanning electron
microscope. Fig. 43–45 Fultoportulae between every
second to third costa. Fig. 46, 47 Collared fultoportulae
with satellite pores on either side. Fig. 48–50 Rimoportula between two costae with satellite pores on either side.
Scale bars: 1 µm (Fig. 43–45, 48–50), 500 nm (Fig. 46),
300 nm (Fig. 47).
Fig. 51–55 Internal morphological characteristics of ➤
Discostella elentarii captured with field emission variable
pressure scanning electron microscope. Fig. 51, 52 Open
girdle with anti ligula. Fig. 53–55 Pores on girdle band
with unknown taxonomic importance. Scale bars: 1 µm
(Fig. 51–53, 55), 500 nm (Fig. 54).
D. stelligera valves with more of a shadow stellate
pattern than a well developed stellate pattern of the
convex valve, where the centre alveoli perforate the
basal siliceous layer.
The external morphological features of D.
stelligera and D. elentarii were generally similar
to those reported in previous studies. The largest
valve diameter of D. stelligera was lower than that
reported by Houk & Klee (2004; 5–40 µm) and
Knapp et al.—Morphology of Discostella stelligera and D. elentarii
437
438New Zealand Journal of Marine and Freshwater Research, 2006, Vol. 40
similar to that reported by Lowe (1975; 5–25 µm).
The range in striae density for D. stelligera (8.6–14.7
striae/10 µm) was similar to that reported in other
studies (Houk & Klee 2004, 8.9–11.5 striae/10 µm;
Lowe 1975, 10–14 striae/10 µm). The range in valve
diameters of D. elentarii (4.0–25.9 µm) was similar
to those reported by Alfinito & Tagliaventi (2002,
6–25 µm). The range in striae density (8–14.5 striae/
10 µm) extended the upper and lower densities from
those previously reported (Alfinito & Tagliaventi
2002, 9–10 striae/10 µm).
The external and internal valve characteristics
of D. stelligera and D. elentarii reported in this
study were similar to those previously documented.
This study provides further documentation of the
arrangement and variability in the spines of D.
elentarii. In addition to providing high quality
images of the fine ultrastructural details, this study
introduced a morphological feature not documented
before on diatoms. The row of pores on the inside
margin of the valves of D. elentarii was visible on
every cell examined. Increased development and
access to technological advances in microscopy
will further our understanding of fine ultrastructural
details of diatom morphology such as those observed
in this study.
Our results support the findings previously
recorded for these closely related species. With the
recent separation of Discostella, our study contributes
evidence of different and new morphological details
of D. elentarii in addition to describing D. stelligera
and D. elentarii from recent collections from two
new locations of different origin than the type
localities. A PCA of the external valve features
measured by LM did not distinguish between the
two taxa and highlights the need to optically dissect
the valves underneath the light microscope to clearly
differentiate between the convex/concave valve face
of D. stelligera compared with the flat valve face of
D. elentarii. The pores on the inside margin of the
valves of D. elentarii have not been documented
before, and the fimbriate margin and the presence of
the cribrum have not been recorded for this species
until now.
ACKNOWLEDGMENTS
We thank Barry Biggs and the National Institute for Water
and Atmospheric Research Limited. This research was
funded in part by a National Science Foundation grant to
R. Lowe (INT–9417225).
REFERENCES
Alfinito S, Tagliaventi N 2002. Morphology and taxonomy
of Cyclotella elentarii spec. nova, a newly
described centric diatom form a New Zealand lake.
Algological Studies 105. Archiv für Hydrobiologie
Supplement 143: 29–38.
Carter GS, Lane MR 1996. Modelling surface oscillations
in New Zealand lakes. New Zealand Journal of
Marine and Freshwater Research 30: 341–353.
Ellwood MJ, Hunter KA, Kim JP 2001. Zinc speciation
in Lakes Manapouri and Hayes, New Zealand.
Marine and Freshwater Research 52: 217–222.
Glasby GP, Stoffers P, Renner RM, Fenner J 1991.
Mineralogy and geochemistry of sediments from
Lake Te Anau, New Zealand. New Zealand Journal
of Marine and Freshwater Research 25: 43–56.
Håkansson H 2002. A compilation and evaluation of species
in the general Stephanodiscus, Cyclostephanos
and Cyclotella with a new genus in the family
Stephanodiscaceae. Diatom Research 17: 1–139.
Houk V, Klee R 2004. The stelligeroid taxa of the genus
Cyclotella (Kützing) Brébisson (Bacillario­phyceae)
and their transfer into the new genus Discostella
gen. nov. Diatom Research 19: 203–228.
Jolly VH 1968. The comparative limnology of some
New Zealand lakes 1. Physical and chemical.
New Zealand Journal of Marine and Freshwater
Research 2: 214–259.
Legendre P, Legendre L 1998. Numerical science. 2nd
ed. Amsterdam, Elsevier Science. 853 p.
Lowe RLL 1975. Comparative ultrastructure of the valve of
some Cyclotella species (Bacillariophyta). Journal
of Phycology 11: 415–424.
Postek MT, Howard KS, Johnson AH, McMichael
KL 1980. Scanning Electron Microscopy: A
Student’s Handbook. Williston, VT, Ladd Research
Industries. 305 p.
Round FE, Crawford RM, Mann DG 1990. The Diatoms:
Biology and Morphology of the Genera. New York,
Cambridge University Press. 747 p.
Skabichevskii AP 1975. A new species of Diatomea
Stephanocyclus skabitsch gen. nov. Ukrayinskyi
Botanichnyi Zhurnal 32: 205–209.
Wells RDS, Clayton JS, De Winton MD 1998. Submerged
vegetation of Lakes Te Anau, Manapouri,
Monowai, Hauroko, and Poteriteri, Fiordland,
New Zealand. New Zealand Journal of Marine and
Freshwater Research 32: 621–638.