REPRODUCTIVE SEASONALITY AND SIMULTANEOUS
HERMAPHRODITISM IN TWO SPECIES OF SIPHONARIA
(GASTROPODA: PULMONATA) FROM THE SOUTHEAST COAST
OF SOUTH AFRICA
PURBA PAL AND ALAN N. HODGSON
Department of Zoology and Entomology, Rhodes University, Grahamstown, 6140, South Africa
(Received 5 February 2004; accepted 18 June 2004)
ABSTRACT
The reproductive cycles of two species of siphonariid limpet (Siphonaria capensis, which has planktonic
larvae, and S. serrata, which has intracapsular development) inhabiting a rocky shore in the Eastern Cape
of South Africa were investigated. A histological study of the gonads established that both species are
simultaneous hermaphrodites. In addition, the gametogenic cycle was similar in both species.
Spermatogenesis was continuous throughout the year whereas oogenesis was interrupted briefly in the
winter months, with mature oocytes more abundant during the late spring and early summer months.
Spawning in S. capensis, as determined by gonad index and egg mass counts in the field, commenced in
late spring (October) and peaked in summer, with about one-third of the animals laying eggs at any one
time. The presence of egg masses on the rocks throughout spring/summer, and the gradual (over 3
months) summer decline in the gonad index of S. capensis suggests that this species may be a partial
spawner.
INTRODUCTION
Siphonariid limpets (Pulmonata: Basommatophora) are abundant on warm temperate to tropical intertidal rocky shores
(Hodgson, 1999). They are hermaphrodites and most species lay
benthic gelatinous egg masses on rocks. Two types of development have been described for siphonariids: planktonic, in which
veliger larvae hatch after about 1 week from egg capsules
embedded in the egg mass; and intracapsular, where crawling
juveniles emerge after 3 –4 weeks (Chambers & McQuaid, 1994a,
b; Hodgson, 1999). Although a number of aspects of the biology
of siphonariids have been investigated (for review see Hodgson,
1999), information on reproductive cycles including gametogenesis and seasonality of spawning is generally lacking. Hodgson
(1999) suggested that siphonariids with seasonal reproduction
would produce gametes seasonally, although some authors
(Marcus & Marcus, 1960; Berry, 1977; Hodgson, Bernard &
Lindley, 1991) have noted that both eggs and sperm were always
present in the gonad. Based on information from two species only,
Hubendick (1978) stated that siphonariids were protandric
hermaphrodites. Thus it remains to be established whether
species of Siphonaria have a distinct gametogenic cycle involving
either simultaneous or sequential hermaphroditism.
Two of the commonest and geographically widespread species
of Siphonaria on South African rocky shores are S. capensis Quoy &
Gaimard, 1833 and S. serrata Fischer, 1807 (Kilburn & Rippey,
1982; Chambers & McQuaid, 1994a). Despite the abundance of
these pulmonate limpets, it is only recently that reproductive
studies have been forthcoming. Hodgson et al. (1991) and Pal &
Hodgson (2002) described spermatogenesis and oogenesis in these
species at an ultrastructural level, Chambers (1994) undertook a
brief study of the annual pattern of egg laying for S. serrata, and
Pal & Hodgson (2003) described the structure of the egg masses
of S. capensis and S. serrata. Although sympatric in their
distribution, these species have different modes of larval
Correspondence: A. N. Hodgson; e-mail: [email protected]
J. Moll. Stud. (2005) 71: 33–40
doi:10.1093/mollus/eyi003
development (Chambers & McQuaid, 1994a, b); Siphonaria
capensis has planktonic development and S. serrata intracapsular
development. It is not known whether they have similar
reproductive patterns. If the life histories and population
dynamics of these important grazing gastropods is to be fully
understood, more complete information on gametogenesis and
spawning is required.
The aims of this study, therefore, were to investigate and
compare the pattern of gametogenesis in S. capensis and S. serrata,
and to determine whether they are sequential or simultaneous
hermaphrodites. An additional aim was to examine the pattern of
spawning (egg laying) in S. capensis. As Chambers (1994) had
already described the seasonal variation in gonad index and egg
laying of S. serrata, this was not repeated in the current study.
MATERIAL AND METHODS
All animals were collected from an aeolian sandstone platform at
Kenton-on-Sea (338420 S, 268410 E) in the Eastern Cape, South
Africa.
Gonad index of Siphonaria capensis
Mean monthly gonad index was estimated from 30 animals (shell
lengths 15– 24 mm) collected each month from September 1999
to December 2000. Animals were brought back to the laboratory,
dissected and the blotted wet weight of both the total body weight
(excluding shell) and gonadal tissue (which can be easily
separated from the body) was recorded to the nearest 0.01 g.
Gonad index (GI) was calculated using the formula: GI ¼ (wet
gonad weight/total wet weight) £ 100.
Gametogenesis in Siphonaria capensis and S. serrata
To determine the gametogenic condition of S. capensis and
S. serrata the gonads from five animals of each species per month
(August 1999 to November 2000) were fixed in 10% aqueous
Journal of Molluscan Studies Vol. 71, No. 1 q The Malacological Society of London 2005, all rights reserved.
P. PAL AND A. N. HODGSON
The gonad of siphonariids contains a number of acini in which
the gametes develop (Hodgson, 1999). Because oogenesis is
asynchronous between acini (Pal & Hodgson, 2002), and
different stages of oogenesis can be found within an acinus, it
was decided to determine the number of oocytes at different
stages of maturity in a number of acini. Five histological sections
were selected from different regions of the gonad of each
individual of each species. The number of oocytes of each
developmental stage, from five acini in each section, were then
counted. For each species the mean (^ SE) number of oocytes of
each stage was calculated. Stages of oogenesis quantified were:
(1) previtellogenic oocytes (which included both early oocytes
with a relatively small cytoplasmic area compared with the
nuclear area, and previtellogenic oocytes containing more than
one nucleolus); (2) early vitellogenic oocytes (characterized by a
large germinal vesicle); (3) late vitellogenic oocytes with marked
eosinophilia; and (4) mature oocytes, when they have detached
themselves from the acinar wall and moved closer to the lumen
(Fig. 1A, B).
For spermatogenesis, acini were classified as: stage A (early),
containing mainly spermatocytes; stage B (mixed), containing
both spermatocytes as well as Sertoli cells with spermatids; stage C
Table 1. A three-factor nested ANOVA comparing number of egg masses
at two different sites A and B (with two subsites each) from September
1999 to December 2000.
Sources
d.f.
MS
F
P
Time
15
17.890
95.466
< 0.0005
Site
1
44.689
9.950
. 0.10
Subsite (Site)
2
4.491
23.967
< 0.0005
Time* Site
15
4.251
22.686
< 0.0005
Residual
1566
0.187
Data log (x þ 1) transformed. Cochran’s test P . 0.05. Significant P values
(, 0.05) are in bold.
Bouin’s fluid for at least 7 days. Following fixation, the tissues
were dehydrated in a graded ethanol series (50 – 100%) and
embedded in Paraplast (via xylene). Serial sections, 5 mm thick,
were cut on a Leica RM 2035 microtome and stained with
haematoxylin and eosin (Humason, 1981).
For a quantitative assessment of gametogenesis, both oogenesis
and spermatogenesis were classified into various stages of
development.
Figure 1. A, B. Acini of Siphonaria serrata (A) and S. capensis (B) showing different stages of oogenesis. C–F. Acini of S. capensis showing different stages of
spermatogenesis. C. Stage A, spermatocytes only. D. Stage B, spermatocytes (sc) and spermatids (st); C. E. Stage C, late spermatids (note also that one
acinus contains spermatids only and a second oocytes). F. Stage D, spent. Abbreviations: ev, early vitellogenic oocyte; mo, mature oocytes; po,
previtellogenic oocyte; sc, spermatocytes; st, spermatids; vo, late vitellogenic oocyte. Scale bars: 0.1 mm.
34
REPRODUCTIVE SEASONALITY AND HERMAPHRODITISM IN SIPHONARIA
variable. Cochran’s test was used to check homogeneity of
variances, and transformations [log (x þ 1)] were done when
needed (Underwood, 1997).
A Student-Newman-Keuls test was used as a post-hoc test in all
the analyses (Zar, 1984).
RESULTS
Gonad index and gametogenesis in Siphonaria capensis
The gonad index (GI) of S. capensis increased in spring (October
1999) and was greatest in summer (January– February 2000)
(Fig. 2). During early spring the gonads contained a large
number of previtellogenic and early vitellogenic oocytes (Fig.
3A). By mid-spring the number of late vitellogenic and mature
oocytes had begun to increase reaching a peak in summer
(December – February) (Fig. 3B). From February to May 2000
there was a decline in the GI (Fig. 2) suggesting that spawning
was occurring during this time. The decline in GI was
accompanied by a decrease in the number of late vitellogenic
and mature oocytes in the gonad acini (Fig. 3B). The GI
remained very low throughout the autumn and winter months
(Fig. 2). During this time, except for a few previtellogenic
oocytes, the gonad of S. capensis was devoid of any oogenic activity
(Fig. 3A, B).
Sperm production was continuous throughout the sampling
period although the animals were spermatogenically most active
in the autumn and winter months (Fig. 4). A very low frequency
(around 1%) of spent acini was found in August and September
2000 only (Fig. 4).
Figure 2. Mean (^ SE) monthly gonad index of Siphonaria capensis from
September 1999 to December 2000.
(late or mature) where spermatids were seen with or without
Sertoli cells as well as spermatozoa; and stage D (spent) (Fig.
1C– F). For spermatogenesis the sampling method was similar to
that for oogenesis (i.e. five animals; five sections per gonad and five
acini per section) except that the occurrence of different stages
were expressed as a percentage for five animals in each month.
Seasonality of spawning in Siphonaria capensis
To determine whether spawning was seasonal in S. capensis, egg
masses were counted once a month (at spring low tide, either new
or full moon) between September 1999 and December 2000 at
Kenton-on-Sea. Siphonaria capensis lays egg ribbons in rock pools,
on vertical walls and wave-cut flat platforms. As it was not known
whether microhabitat would influence the timing of spawning,
this study was restricted to horizontal platforms only. Sampling
was undertaken at two sites (hereafter referred to as sites A and
B), about 65 m apart. The number of egg masses was counted in
25 random quadrats (each 0.0625 m2) at each of four subsites,
two nested in site A (I and II) and two nested in site B (III and
IV). Subsites I and II covered an area of approximately 25 m2
each and were about 5 m apart. Subsites III and IV were
approximately 15 m apart and covered areas of 20 and 24 m2,
respectively. A three-factor nested ANOVA (time and site as
fixed factors, subsite as random factor nested in site) was used to
analyse the data. To meet the assumptions of normality and
homogeneity of variances, data were transformed [log (x þ 1)]
and Cochran’s test was used to check for homogeneity of
variances (Underwood, 1997). Data were analysed using
Statistica Statsoft (version 6).
Spawning cycle of Siphonaria capensis
Siphonaria capensis mainly spawned in the summer months (Fig.
5). In 1999, spawning commenced in spring with the greatest
number of egg masses laid between October 1999 and January
2000 (Fig. 5). Very few egg ribbons were found for the rest of the
year although at site B, egg masses were relatively abundant in
March 2000 (Fig. 5). An analysis of variance revealed that the
number of egg masses differed significantly over time (Table 1),
with the mean number of egg masses being lowest in February,
June and August 2000 and highest in November and December
1999 (Table 2; Fig. 5). The number of spawn differed
significantly within sites, i.e. between subsites (Table 1).
There was a significant interaction between time and site
(Table 1). While the temporal pattern (timing) of spawning by
S. capensis was similar at both sites, the period of spawning
was longer, and the number of egg masses laid greater, at site B
(Fig. 5).
Number of egg masses in relation to density of Siphonaria
capensis
Number of egg masses in relation to density of Siphonaria
capensis
The follow-up 6-month (September 2001 to February 2002)
study of spawning in S. capensis confirmed that egg laying began
in spring (September – October), with the greatest density of egg
masses in summer (November 2001 and February 2002; Fig. 6A).
The mean number of egg masses differed significantly within sites
but not between sites (Tables 3 and 4) although the density of egg
masses was greatest at site B during the peak of spawning
(November) (Fig. 6A).
The density of sexually mature S. capensis ($10 mm shell
length) differed significantly between sites with a greater number
of animals at site B (six to 10 limpets per 0.0625 m2) as well as
within sites over time (Tables 5 and 6; Fig. 6B). A comparison of
egg mass density per limpet at sites A and B revealed that there
was no significant difference between sites (Table 7). The highest
number of egg masses per individual (Table 8) was laid in
As reproductive output (number of egg masses) of S. capensis
varied between sites A and B during the 16 months of sampling,
the study was repeated to test the hypothesis that the variation
between the sites was influenced by the density of adult limpets
(shell length $10 mm). The number of adults and egg masses was
counted in 25 quadrats (each 0.0625 m2) each month at sites A
and B between September 2001 and February 2002. Data were
analysed using two three-factor nested ANOVAs (time and site as
fixed factors, subsite as the random factor nested in site) with the
density of individuals and number of egg masses as the dependent
variables. To investigate the temporal pattern of egg laying with
respect to density of animals, a three-factor nested ANOVA was
conducted (time and site as fixed and subsites as random factors)
with mean number of egg masses per individual as the dependent
35
P. PAL AND A. N. HODGSON
Figure 4. Percentage of different spermatogenic stages in Siphonaria
capensis from August 1999 to November 2000. Stage A, mainly
spermatocytes; Stage B, spermatocytes and spermatids; Stage C, late
spermatids; Stage D, spent.
Figure 3. Mean number (^ SE) of oocytes of Siphonaria capensis (from five
animals; 25 gonad acini per animal) at different stages of development (A,
previtellogenic and early vitellogenic; B, late vitellogenic and mature)
from August 1999 to November 2000.
Table 2. Results of the Student-Newman-Keuls test (post-hoc test) to
determine the differences in the mean number of egg masses (per
0.0625 m2) between September 1999 and December 2000.
Months
Mean
1
2
Feb 2000
0.007
X
Jun 2000
0.007
X
Aug 2000
0.007
X
Sep 1999
0.014
X
Jul 2000
0.025
X
Dec 2000
0.139
X
X
Oct 2000
0.142
X
X
May 2000
0.158
X
X
Nov 2000
0.246
X
Apr 2000
0.253
X
Sep 2000
0.421
Jan 2000
0.621
3
4
5
6
Figure 5. Mean (^ SE) number of Siphonaria capensis egg masses at two
sites from September 1999 to February 2000.
X
November (mean 0.34 per 0.0625 m2), i.e. about one-third of the
population had spawned.
X
Mar 2000
0.843
X
Oct 1999
0.858
X
Dec 1999
1.185
X
Nov 1999
1.190
X
Gametogenesis in Siphonaria serrata
Siphonaria serrata had a similar gametogenic cycle to that of
S. capensis. Previtellogenic oocytes were more abundant in spring
Crosses in different columns indicate a significant difference.
36
REPRODUCTIVE SEASONALITY AND HERMAPHRODITISM IN SIPHONARIA
Table 4. Results of the Student-Newman-Keuls test (post-hoc test) to
determine the differences in the mean number of egg masses (per
0.0625 m2) between September 2001 and February 2002.
Months
Mean
1
2
3
Sep 2001
0.05
X
Dec 2001
0.12
X
Oct 2001
0.13
X
Jan 2002
0.15
X
Feb 2002
0.30
X
Nov 2001
0.31
X
Crosses in different columns indicate a significant difference.
Table 5. A three-factor nested analysis comparing density of animals
(.10 mm) at two different sites from September 2001 to February 2002.
Sources
d.f.
MS
F
P
Time
5
1.757
3.803
< 0.005
Site
1
113.595
20.538
< 0.05
Subsite (Site)
2
5.531
11.972
< 0.0005
Time* Site
5
1.588
3.437
< 0.01
Residual
586
0.462
Data log (x þ 1) transformed, Cochran’s test P . 0.05. Significant P values
(, 0.05) are in bold.
Table 6. Results of the Student-Newman-Keuls test (post-hoc test) to
determine the differences in the density of animals (per 0.0625 m2)
between September 2001 and February 2002.
Figure 6. A. Mean (^ SE) number of egg masses of Siphonaria capensis at
two sites from September 2001 to February 2002. B. Density of S. capensis
(mean ^ SE) at two sites from September 2001 to February 2002.
MS
F
P
Time
5
1.082
26.186
< 0.0005
Site
1
1.754
5.883
. 0.2
Subsite (Site)
2
0.298
7.22
< 0.001
Time* Site
5
0.193
4.671
< 0.0005
Residual
586
0.041
1
Oct 2001
1.54
X
Nov 2001
1.58
X
Sep 2001
1.66
X
Jan 2002
1.70
X
Dec 2001
1.71
X
Feb 2002
1.91
2
X
X
Table 7. A three-factor analysis comparing number of egg masses/individual at two sites from September 2001 to February 2002.
Sources
Table 3. A three-factor nested analysis comparing number of egg masses
at two different sites from September 2001 to February 2002.
d.f.
Mean
Crosses in different columns indicate a significant difference.
(August– October), with greater numbers of late vitellogenic and
mature oocytes throughout the summer (e.g. December 1999)
(Fig. 7A, B). In the winter months few late vitellogenic and
mature oocytes were present in the gonad acini (Fig. 7B).
Spermatogenesis occurred throughout the year although the
acini showed greater activity during the late winter and early
spring (Fig. 8). A few acini were found spent in March,
September and November 2000.
Sources
Months
d.f.
MS
F
P
Time
5
1.126
9.105
< 0.0005
Site
1
0.167
0.291
. 0.5
Subsite (Site)
2
0.574
4.637
< 0.02
Time* Site
5
0.112
0.908
. 0.5
Residual
586
0.124
Significant P values (, 0.05) are in bold.
DISCUSSION
Generally molluscs from the south and southeast coast of South
Africa reproduce in spring/summer (McGwynne & van der
Horst, 1985; Lasiak, 1986, 1987; Chambers, 1994; Gray, 1996;
Henninger & Hodgson, 2001), whereas those on the west coast
are autumn and/or winter breeders (Branch, 1974; Griffiths,
1977; Gray, 1996). The present study of southeast coast Siphonaria
capensis, together with the study of Chambers (1994) on S. serrata,
Data log (x þ 1) transformed, Cochran’s test P . 0.05. Significant P values
(, 0.05) are in bold.
37
P. PAL AND A. N. HODGSON
Table 8. Results of the Student-Newman-Keuls test (post-hoc test) to
determine the differences in the mean number of egg masses per
individual between September 2001 and February 2002.
Months
Mean
1
2
Sep 2001
0.05
X
Dec 2001
0.07
X
Oct 2001
0.14
X
X
Jan 2002
0.17
X
X
Feb 2002
0.22
Nov 2001
0.34
3
X
X
Crosses in different columns indicate a significant difference.
Figure 8. Percentage of different spermatogenic stages in Siphonaria serrata
from August 1999 to November 2000. Stage A, mainly spermatocytes;
Stage B, spermatocytes and spermatids; Stage C, spermatids and
spermatozoa; Stage D, spent.
1994a) and in the shade (Hodgson, personal observation). The
egg ribbons of S. serrata (which spend 3 – 4 weeks on the shore) are
able to resist desiccation because of structural modifications to the
mucous matrix (Pal & Hodgson, 2003) and a low surface area to
volume ratio (Chambers & McQuaid, 1994a). The reproductive
cycle of S. capensis and S. serrata is possibly linked to food
availability, as is the case for S. diemenensis (Quinn, 1988) and
S. japonica (Liu, 1994).
Although S. capensis and S. serrata have a distinct breeding
season, sperm were present in the gonad of both species all year
round and very few spent acini were observed. This indicates that
spermatogenesis is a continuous process. Continuous spermatogenesis has also been recorded in S. hispida, although sperm
production was greatest after egg laying (Marcus & Marcus,
1960). Although spermatogenesis in S. capensis and S. serrata
occurred throughout the year, activity was lower in summer
(December 1999 to February 2000) when compared with late
winter/early spring (August and September 1999; July and
August 2000). The conclusion that sperm production is
continuous is supported by observations of the hermaphrodite
ducts of these species. In both, sperm were present in the seminal
vesicle region of the duct throughout the year (Pal, unpublished
data).
In S. pectinata, mature oocytes were observed in the gonad
throughout the breeding season and spent gonads were found
after spawning (Ocaña & Emson, 1999). By contrast, the gonads
Figure 7. Mean number (^ SE) of oocytes of Siphonaria serrata (from five
animals; 25 gonad acini per animal) at different stages of development (A,
previtellogenic and early vitellogenic; B, late vitellogenic and mature)
from August 1999 to November 2000.
has shown that spawning in these species also commences in late
spring and continues in summer. This means that S. capensis and
S. serrata spawn during the hottest months. For a species that lays
benthic egg masses that are exposed during diurnal low tides, it
would seem more advantageous to lay eggs when desiccation is
less severe (e.g. autumn or winter). Egg ribbon desiccation,
however, may not be critical in S. capensis because the veligers
emerge within 3 – 4 days of egg laying. Furthermore, many egg
masses are laid in shallow rock pools (Chambers & McQuaid,
38
REPRODUCTIVE SEASONALITY AND HERMAPHRODITISM IN SIPHONARIA
BRANCH, G.M. 1981. The biology of limpets: physical factors, energy
flow, and ecological interactions. Oceanography and Marine Biology: An
Annual Review, 19: 235 –380.
CHAMBERS, R.J. 1994. The conflict between adaptation and constraint: The
case of the siphonariid limpets. Ph.D. thesis, Rhodes University.
CHAMBERS, R.J. & MC QUAID, C.D. 1994a. Notes on taxonomy,
spawn and larval development of South African species of the
intertidal limpet Siphonaria (Gastropoda: Pulmonata). Journal of
Molluscan Studies, 60: 263–275.
CHAMBERS, R.J. & MC QUAID, C.D. 1994b. A review of larval
development in the intertidal limpet genus Siphonaria (Gastropoda:
Pulmonata). Journal of Molluscan Studies, 60: 415 –423.
CREESE, R.G. 1980. Reproductive cycles and fecundities of two
species of Siphonaria (Mollusca: Pulmonata) in south-eastern
Australia. Australian Journal of Marine and Freshwater Research, 31:
37–47.
DUNCAN, C.J. 1975. Reproduction. In: Pulmonates (V. Fretter & J.
Peake, eds), 1. 309 –365. Academic Press, London.
GERAERTS, W.P.M. & JOOSSE, J. 1984. Freshwater snails
(Basommatophora). In: The Mollusca (A. S. Tompa, N. H. Verdonk
& J. A. M. van den Biggelaar, eds), 7. 141–207. Academic Press,
London.
GRAY, D.R. 1996. Studies of the biology and ecology of the high shore South
African limpet, Helcion pectunculus (Mollusca: Patellogastropoda). Ph.D.
thesis, Rhodes University.
GRIFFITHS, R.J. 1977. Reproductive cycles in littoral populations of
Choromytilus meridionalis (Kr.) and Aulacomya ater (Molina) with a
quantitative assessment of gamete production in the former. Journal of
Experimental Marine Biology and Ecology, 30: 53–71.
HELLER, J. 1993. Hermaphroditism in molluscs. Biological Journal of the
Linnaean Society, 48: 19– 42.
HENNINGER, T.O. & HODGSON, A.N. 2001. The reproductive cycle
of Helcion pruinosus (Patellogastropoda) on two South African boulder
shores. Journal of Molluscan Studies, 67: 385–394.
HIRANO, Y. & INABA, A. 1980. Siphonaria (pulmonate limpet) survey
of Japan. I. Observations on the behaviour of Siphonaria japonica
during breeding season. Publications of the Seto Marine Biological
Laboratory, 25: 323–334.
HODGSON, A.N. 1999. The biology of siphonariid limpets (Gastropoda:
Pulmonata). Oceanography and Marine Biology: An Annual Review, 37:
245–314.
HODGSON, A.N., BERNARD, R.T.F. & LINDLEY, D.S. 1991.
Comparative spermatology of four sympatric species of Siphonaria
(Pulmonata: Basommatophora). Journal of Molluscan Studies, 57:
309–322.
HUBENDICK, B. 1978. Systematics and comparative morphology of the
Basommatophora. In: Pulmonates, 2A (V. Fretter & J. Peake, eds),
1–47. Academic Press, London.
HUMASON, G.L. 1981. Animal tissue technique. W. H. Freeman, New
York.
IWASAKI, K. 1995. Foraging and spawning rhythms of the pulmonate
limpet Siphonaria sirius (Pilsbry): switching of activity period by a
diurnal forager. Journal of Molluscan Studies, 61: 275 –288.
JARNE, P., VIANEY-LIAUD, M. & DELAY, B. 1993. Selfing and
outcrossing in hermaphrodite freshwater gastropods (Basommatophora): where, when and why. Biological Journal of the Linnaean Society,
49: 99– 125.
JOSKA, M.A.P. & BRANCH, G.M. 1983. The reproductive cycle of the
trochid gastropod Oxystele variegata (Anton, 1839). Veliger, 26: 47–51.
KILBURN, R. & RIPPEY, E. 1982. Sea shells of southern Africa.
Macmillan, Johannesburg.
LASIAK, T. 1986. The reproductive cycle of the intertidal gastropod
Turbo coronatus Gmelin 1791, on the Transkei coast. South African
Journal of Zoology, 21: 153 –155.
LASIAK, T. 1987. Observations on the reproductive cycles of Cellana
capensis (Gmelin, 1791) and Patella concolor (Krauss, 1848) (Gastropoda: Patellidae). South African Journal of Zoology, 22: 195–199.
LEVINGS, S.C. & GARRITY, S.D. 1986. Notes on reproduction of a
tropical pulmonate limpet. Siphonaria gigas (Sowerby). Veliger, 29:
86–90.
of S. capensis and S. serrata had a few spent acini only, although
oogenesis was interrupted in the winter months.
As it was not possible to monitor spawning in individual
S. capensis, whether an individual underwent partial or complete
spawning could not be determined. The gonad index (GI) results
revealed that the decline in GI took place over 3 months
and during this time the number of mature oocytes in the gonads
also decreased gradually. This suggests that the limpets might lay
eggs more than once during the breeding season.
Chambers (1994) noted that some egg masses of S. concinna and
S. serrata were present on shores all year round, although in
S. serrata the number of egg masses per individual was low in the
winter months. Similarly, in this study a small number of egg
masses of S. capensis were found in winter. Joska & Branch (1983)
in a study of the South African prosobranch Oxystele variegata,
commented that there will always be some individuals in a
population that spawn out of the cycle, but they cannot be taken
as representative. One advantage of haphazard spawning by
some individuals is that the new recruits, if they survive, may
contribute to the existing population at any time of the year
(Williamson & Steinberg, 2002). At peak spawning the mean
number of egg masses per individual was 0.34, indicating that not
all limpets spawn at any particular spawning time. Similar
findings were obtained by Quinn (1988) for S. diemenensis and
Ocaña & Emson (1999) for S. pectinata.
A number of studies have shown that siphonariids spawn
during a particular phase of the moon or time of the day (Zischke,
1974; Creese, 1980; Hirano & Inaba, 1980; Branch, 1981;
Levings & Garrity, 1986; Chambers, 1994; Iwasaki, 1995).
Although the present study did not examine the relationship
between the timing of spawning in S. capensis and the lunar cycle
or day/night cycle, it was found that egg masses were present on
the shore after both a new and full moon. A more detailed study is
needed to determine whether there is relationship between
spawning and the lunar cycle in S. capensis and S. serrata.
Although Heller (1993) concluded that the majority of
pulmonates are simultaneous hermaphrodites, Hubendick
(1978) suggested that siphonariids are protandrous hermaphrodites, a conclusion based on the observations of Marcus &
Marcus (1960) and Zischke (1974) on S. hispida and S. pectinata,
respectively. Duncan (1975) and Geraerts & Joosse (1984) also
noted that some basommatophoran limpets could be protandric
before becoming simultaneous hermaphrodites. Jarne, VianeyLiaud & Delay (1993), however, observed that this protandric
phase could be very short in terms of the life cycle of the animal.
Siphonaria capensis fits this pattern with a short male phase
(individuals about 9 mm long) before becoming a simultaneous
hermaphrodite at a shell length of $ 10 mm.
ACKNOWLEDGEMENTS
The authors wish to thank L. Vat, F. Porri, J. Gush and
J. Erlandsson for help in the field. J. Erlandsson is also thanked
for statistical advice. The National Research Foundation, South
Africa and Andrew Mellon Foundation, Rhodes University
provided financial support. This study forms part of the doctoral
thesis by p.p. submitted to Rhodes University, South Africa.
REFERENCES
BERRY, A.J. 1977. Gastropoda: Pulmonata. In: Reproduction of Marine
Invertebrates (A. C. Giese & J. S. Pearse, eds), 4. 181–226. Academic
Press, New York.
BRANCH, G.M. 1974. The ecology of Patella Linnaeus from the Cape
Peninsula, South Africa. 2. Reproductive cycles. Transactions of Royal
Society of South Africa, 41: 111–160.
39
P. PAL AND A. N. HODGSON
LIU, J.H. 1994. Distribution and population dynamics of three
populations of Siphonaria on rocky intertidal shores in Hong Kong.
Journal of Molluscan Studies, 60: 431 –443.
MC GWYNNE, L.E. & VAN DER HORST, G. 1985. Patterns of
reproduction in three sandy beach whelks of the genus Bullia Griffith.
Journal of Molluscan Studies, 51: 190 –197.
MARCUS, E. & MARCUS, E. 1960. On Siphonaria hispida. Universidade
Sao Paulo. Faculdade de Filosofia, Ciencias e Letras. Boletim de Zoologia, 23:
107–139.
OCAÑA, T.M.J. & EMSON, R.H. 1999. Maturation, spawning and
development in Siphonaria pectinata Linnaeus (Gastropoda: Pulmonata) at Gibraltar. Journal of Molluscan Studies, 65: 185 –193.
PAL, P. & HODGSON, A.N. 2002. An ultrastuctural study of oogenesis
in a planktonic and a direct developing species of Siphonaria
(Gastropoda: Pulmonata). Journal of Molluscan Studies, 68: 337–344.
PAL, P. & HODGSON, A.N. 2003. The structure of the egg ribbons of a
planktonic and intracapsular developing siphonariid limpet (Gastro-
poda: Pulmonata). Invertebrate Reproduction and Development, 43:
243 –253.
QUINN, G.P. 1988. Ecology of the intertidal pulmonate limpet Siphonaria
diemenensis Quoy et Gaimard. II. Reproductive patterns and
energetics. Journal of Experimental Marine Biology and Ecology, 117:
137 –156.
UNDERWOOD, A.J. 1997. Experiments in ecology: their logical design and
interpretation using analysis of variance. Cambridge University Press,
London.
WILLIAMSON, J.E. & STEINBERG, P.D. 2002. Reproductive cycle of
the sea urchin Holopneustes purpurascens (Temnopleuridae: Echinodermata). Marine Biology, 140: 519–532.
ZAR, J.H. 1984. Biostatistical analysis. Prentice Hall, New Jersey.
ZISCHKE, J.A. 1882. Spawning, development and growth in the
pulmonate limpets Siphonaria pectinata Linnaeus, 1758 and Siphonaria
alternata Say, 1882. Veliger, 16: 399 –404.
40