11 551-564 Arkhipkin
4/10/0 4:37 pm
Page 551
J. Moll. Stud. (2000), 66, 551–564
© The Malacological Society of London 2000
ADAPTATIONS FOR COLD WATER SPAWNING IN
LOLIGINID SQUID: LOLIGO GAHI IN FALKLAND WATERS
A.I. ARKHIPKIN, V.V. LAPTIKHOVSKY and D.A.J. MIDDLETON
Fisheries Department, Falkland Islands Government, P.O. Box 598, Stanley, Falkland Islands.
Email: [email protected]
(Received 30 March 2000; accepted 5 June 2000)
ABSTRACT
Inshore spawning sites of the cold water squid Loligo
gahi were found in the waters of East Falkland
(Southwest Atlantic), where there is a major fishery
based on this species. Egg masses occurred in algal
beds, often at the outer (seaward) edge, with ambient
water temperatures of 6.5–9°C and salinity 33.75–
33.85‰. They were attached to the stipes of the kelp
algae Lessonia spp. and Macrocystis pyrifera from
0.5 m to 2.5 m off the bottom at 8–20 m depths. The
overall density of egg masses was low. The egg mass is
a bundle of elongated gelatinous translucent capsules
with each capsule firmly attached to the kelp stipe at
its basal end. The capsules are mainly 50–60 mm in
length and contain an average of 70 fertilized eggs.
Sampled egg masses consisted of 4–161 capsules and
from 138 to 11,487 eggs. Large egg masses ( 50 capsules) were apparently formed by several females at
different times, as embryos in different capsules were
at various stages of development. Eggs laid in winter
are significantly larger than those laid in summer. In
comparison with tropical and temperate Loligo spp.
L. gahi have short egg capsules containing a small
number of eggs, but the eggs (2.2–2.5 mm diameter)
and hatchlings (3.1–3.4 mm mantle length) are large.
These are probable adaptations for cold water spawning and development.
INTRODUCTION
The Patagonian long-finned squid Loligo gahi
Orbigny, 1835 is a common inhabitant of the
shelf waters of South America from southern
Peru and Chile in the Pacific to the colder
waters of southern Argentina and the Falkland
Islands in the Atlantic (Roper, Sweeney &
Nauen, 1984). This small squid (normal adult
size 120–160 mm mantle length, ML) is an
important fishery resource within the Falkland
Islands Interim Conservation and Management
Zone (FICZ). In the 1990s, the annual catch of
this species varied from 26,000 to 98,000 tonnes
(FIG, 1998). It is thought that L. gahi under-
takes ontogenetic vertical migrations (Hatfield
& Des Clers, 1998). They move from the inner
shelf to the shelf edge and continental slope
(mainly down to 200–300 m depth) as juveniles,
feed and grow in the deep water as immature
and maturing adults and, upon maturation,
return to shallow waters to spawn (Hatfield,
Rodhouse & Porebski, 1990; Hatfield & Rodhouse, 1994). Mating occurs both in deep water,
with sperm storage within the special seminal
receptacle on the buccal membrane of a female,
and in shallow waters, with sperm storage near
the gills within the mantle cavity of a female
(George & Hatfield, 1995). It is assumed that
the first deepwater mating is a ‘trigger’ for final
sexual maturation in females, while the shallow
water mating allows fertilization of eggs during
spawning (Rasero & Portela, 1998).
Loliginid squids are generally shallow water
spawners (Hanlon, 1998), although their egg
masses have sometimes been found on the bottom as deep as 507 m in Loligo forbesi (Lordan
& Casey, 1999) and 720 m in Loligo opalescens
(Butler, Fuller & Yaremko, 1999). Loliginids
usually form dense spawning aggregations and,
in the case of Loligo vulgaris reynaudii, their
spawning sites have a patchy distribution along
the coast (Sauer, McCarthy, Smale & Koorts,
1992). Females attach gelatinous capsules containing eggs to different objects on the bottom,
sometimes making large concentrations of egg
capsules (‘egg beds’) greater than 3 m in diameter (Sauer et al., 1992). If egg capsules are
already present at a spawning site, females tend
to attach their capsules to those that have been
laid already, or to nearby objects. Thus, egg
capsules in one particular egg mass or egg bed
may be at different stages of embryonic development (Drew, 1911).
Although small numbers of fully mature
females have been found in inshore trawl
catches, prompting the assumption of shallow
water spawning (Hatfield et al., 1990), neither
11 551-564 Arkhipkin
552
4/10/0 4:37 pm
Page 552
A.I. ARKHIPKIN, V.V. LAPTIKHOVSKY & D.A.J. MIDDLETON
the spawning sites nor egg masses of L. gahi
have previously been described. Local SCUBA
divers have previously indicated the rare presence of egg masses, strongly resembling those
of loliginids, in shallow bays around East Falkland, but these were never examined (George
& Hatfield, 1995). A shallow water marine
survey around the Falkland Islands recorded
the presence of loliginid egg masses at several
sites around the Islands (Tingley, Saunders,
Harries & King, 1996) but again these were not
studied. The specific aims of this study were to
locate inshore spawning sites of L. gahi around
East Falkland (near main fishery grounds) by
diving, and to investigate the morphology of
egg masses and egg capsules at various stages of
embryonic development in different seasons.
These contribute to the broader objective of
understanding the adaptations for cold water
reproduction in this high latitude loliginid
species.
MATERIALS AND METHODS
Twenty egg masses resembling those of loliginid
squids were collected (or sub-sampled) throughout
1999 in the region of Port William (at the eastern tip
of East Falkland, see Fig. 1) either by SCUBA divers,
dredging, or found stranded on the beach. Forty six
egg masses were sampled during a diving survey conducted from the r/v Dorada in November 1999. A
further two egg masses, collected in April 1992 and
April 1995 and preserved in alcohol, were examined
to establish their stage of development.
Previous observations during dives near Stanley,
and at various sites around the Islands (Tingley et al.,
1996), suggested that egg masses resembling those of
a Loligo sp. were generally found attached to algae
(‘kelp’) stipes. Initial choice of dive sites during the
research cruise was therefore based on the charted
occurrence of extensive kelp beds. Actual dive positions were determined both by weather conditions
and an on-site evaluation of kelp bed extents. Divers
entered the water at the seaward margin of kelp, as
visible from the surface, and the area surveyed was
generally restricted to this outer edge. However, at
many sites kelp was found to extend beyond the area
visible from the surface, while some sites allowed
inspection of the seabed deeper than the kelp bed.
Four divers, diving in pairs, participated in the
research cruise. Dives generally involved 30 minutes
survey time. The divers searched along a linear route
either in a single direction or around the perimeter of
a rectangular area, depending on the topography of
the site. During the dive the number of egg masses
observed was recorded, along with the approximate
number of capsules in each mass. On dives where egg
masses were found, the first egg mass encountered
was sampled whole. Capsules were sampled from up
to four other egg masses per dive. Between site varia-
bility in factors such as visibility, kelp density and egg
mass density (and hence time spent sampling), combined with the aim of covering a long coastline in a
fairly short period, prevented simple quantification of
the area surveyed. Instead egg mass density was
expressed as egg masses observed per minute dive
time.
At each diving station, temperature and salinity
were measured using a CTD profiler SBE-25 (SeaBird Electronics Inc.). Calibration of the profiler had
been made by the manufacturer.
After sampling, egg masses were kept in seawater
in a refrigerator for a maximum of one day after
which they were analysed in the laboratory either
ashore or on board the ship. Every egg capsule was
examined in 29 egg masses. In the 38 remaining egg
masses a sample of 2 to 5 capsules was analysed. For
each capsule, the total length was measured to the
nearest 1 mm under. The total number of eggs within
the capsule was counted by squeezing the capsule
between two glass slides. Stages of embryonic development were assigned using the scale described for loliginid squid by Arnold (1965). This scale was found to
be appropriate for L. gahi with the single exception
that its embryos had larger yolk sacs at stages 25–27.
When a capsule contained embryos at slightly different stages the earlier stage was used in analyses. The
maximum egg diameter was measured in 20–40 eggs
for each embryonic stage observed.
The frequency distributions of egg diameter at different stages of embryonic development in the austral
summer (October to March) and winter (April to
September) were compared using the approximate
Kolmogorov-Smirnov two-sample test, computed
using the method described by Sokal & Rohlf (1981).
Confirmation that the egg masses were those of
Loligo spp. was based on a number of morphological
characteristics including the chromatophore arrangement on the head of embryos in late stages of development (Sweeney, Roper, Mangold, Clarke & von
Boletzky, 1992).
RESULTS
Location of spawning sites and egg mass
occurrence
The egg mass density encountered around the
coast of East Falkland during the period of the
research cruise varied considerably (Fig. 1).
Squid eggs were found around almost the entire
coastline of East Falkland with the notable
exception of the central part of Falkland Sound.
The highest density was found on the north-east
coast.
There was considerable local variation in egg
mass density. Eggs could be entirely absent
from a site within a few kilometres of a site with
relatively high densities. Within a site there was
also noticeable patchiness in egg mass density.
11 551-564 Arkhipkin
4/10/0 4:37 pm
Page 553
COLD WATER SPAWNING ADAPTATIONS IN LOLIGO GAHI
At some sites egg masses would be encountered
throughout the dive, whilst at other sites most
of the dive could pass without observing any egg
masses. All egg masses found during the course
of the research cruise were attached to kelp,
both Lessonia spp. and Macrocystis pyrifera
(Fig. 2). Areas of rock and sand adjacent to
kelp beds were checked at several sites, without
discovering any egg masses, but sites further
offshore were not investigated. Egg masses
tended to occur on short, solitary kelp strands.
This was particularly noticeable in the case of
M. pyrifera where the majority of the kelp tends
to be rooted in dense, ‘communal’ holdfasts (i.e.
supporting a mass of fronds). No egg masses
were found in such clusters of kelp stipes.
Typically the M. pyrifera stipes supporting egg
masses had no large frond and extended only
two or three metres off the seabed. Clusters on
553
Lessonia spp. were generally found on small,
frondless stipes, although a few clusters were
found on healthier looking plants (although the
maximum stipe diameter on which clusters were
attached was about 2cm). Egg masses were
attached approximately 0.5m to 2.5m off the
seabed.
Environmental conditions
It was only possible to study the environmental
characteristics of the spawning sites of L. gahi
during the research cruise in November 1999
(the transition period from austral spring to
summer). Egg masses occurred in the whole
spectrum of hydrological parameters encountered during the study (Figure 3). However,
maximum egg mass abundance was found at a
water temperature of 7.5–7.8°C, salinity of 33.8‰
51˚ 30'S
d
n
a
l
k
l
t
Fa
s
Ea
52˚ 00'S
Egg masses observed
per minute dive time
1
0.5
0.01
52˚ 30'S
60˚W
59˚W
58˚W
Figure 1. Position of dives around the coast of East Falkland during the Falkland Islands Fisheries Department
research cruise ZDLH1-11-1999, November 1999 (crosses), and density of L. gahi egg masses (expressed as egg
masses observed per unit dive time). Circle diameter is proportional to the square root of egg mass density. The
rectangle encloses the Stanley/Port William area where additional samples were obtained.
11 551-564 Arkhipkin
554
4/10/0 4:37 pm
Page 554
A.I. ARKHIPKIN, V.V. LAPTIKHOVSKY & D.A.J. MIDDLETON
Figure 2. Egg mass attached to stalk of Lessonia spp. Photograph by David Eynon, Falkland Images.
oxygen concentration of 9–9.3 ml/l, and water
density of 26.4 kg/m3.
Egg mass morphology and egg numbers
The egg masses look like bundles of elongated
balloons (capsules) with each capsule attached
strongly to the algal stipe at its basal end. The
number of capsules in the sampled egg masses
varied from 4 to 161. As in other loliginid squid
(see Drew, 1911), the egg capsule is spindleshaped and consists of two gelatinous layers.
The outer layer is more resilient and surrounds
a more viscous inner layer that contains the
eggs. In egg masses collected in winter the
length of capsules varied from 33 to 77 mm
(mean 54.93; SD 8.87) and egg number ranged
from 13 to 155 (73.32; SD 22.85). In egg
masses collected in spring-summer the length of
capsules varied from 18 to 86 mm (mean 53.64;
SD 10.86) and egg number ranged from 0 to
144 (mean 71.67; SD 22.37). There was no
significant difference in mean capsule length or
number of eggs per capsule between winter and
summer laid egg masses (two sample t-tests
using Welch’s modification for unequal variances: t –1.5673, p 0.1188 for capsule
length; t –0.8068, p 0.4209 for egg number).
Larger capsules usually contained more eggs
(Fig. 4). The size distribution of the capsules
(Fig. 5A,B) and number of eggs per capsule
(Fig. 5C,D) are symmetrical but show significant deviations from normality in the tails of
the distributions. The total number of eggs in
the sampled egg masses varied from 138 to
11,487 eggs. Egg masses containing up to 2500
eggs were most frequent (Fig. 6A).
Stages of embryonic development
Almost all stages of the embryonic development were observed in sampled egg masses,
from fertilized eggs (stage 1–2) to embryos
which were ready to hatch (stages 29–30) (Fig.
6B). Each egg capsule contained eggs at similar
stages of embryonic development.
11 551-564 Arkhipkin
4/10/0 4:38 pm
Page 555
COLD WATER SPAWNING ADAPTATIONS IN LOLIGO GAHI
555
Figure 3. The relationship of egg mass density (in numbers observed per unit dive time) to temperature, salinity,
oxygen concentration, and water density at diving stations around the coast of East Falkland in November 1999.
Figure 4. The relationship between the number of eggs in a capsule and capsule length. Kendall’s rank correlation, τ 0.39, p 2.2e–16.
11 551-564 Arkhipkin
556
4/10/0 4:38 pm
Page 556
A.I. ARKHIPKIN, V.V. LAPTIKHOVSKY & D.A.J. MIDDLETON
Figure 5. Frequency distributions and quantile-quantile plots of capsule length (A, B), and number of eggs per
capsule (C, D) for all sampled egg capsules. The Shapiro-Wilk normality test for the distribution of capsule
length yields W 0.9974, p 0. 02404, and for eggs per capsule W 0.9929, p 2.753e–06.
As in other loliginids (Boletzky, 1989), the
perivitelline space appeared around the yolk
just after fertilisation due to the chorion swelling. The maximum egg diameter of fertilised
eggs (Stages 1–3) was 2.0–2.5 mm (Fig. 7a).
During the cleavage, blastodisc formation and
cellulation (Stages 4–12) the egg diameter
increased slightly up to 2.1–3.2 mm. The sharp
increase in egg size was observed at the start of
organogenesis of the embryos (Stage 17). At
hatching, the egg diameter achieved 4.7–6.2 mm
(Fig. 7g). The distribution of egg diameter
within sample capsules at different stages of
embryonic development was compared between
summer (October to March) and winter (April
to September) (Fig. 7). Kolmogorov-Smirnov
two-sample tests suggest that the distributions
are significantly different, with larger eggs in
winter than summer, in four of the six stage
groupings for which comparison was possible
(Table 1).
It is notable that despite a considerable
11 551-564 Arkhipkin
4/10/0 4:38 pm
Page 557
COLD WATER SPAWNING ADAPTATIONS IN LOLIGO GAHI
557
Figure 6. (A) Total number of eggs in an egg mass, for all egg masses where each egg capsule was analysed, and
(B) monthly occurrence of different stages of embryonic development in egg masses. The latter includes both
the 1999 samples and the samples from April 1992 and 1995.
Table 1. Comparison of egg diameter frequency distributions (Fig.
7) at different stages of embryonic development. D is the Kolmogorov-Smirnov two-sample test statistic, and p the associated
(two-tailed) probability that the two samples drawn from the same
distributions, calculated by the approximate method for large
samples described by Sokal & Rohlf (1981, p.443).
Embryonic
stage grouping
1–3
4–12
13–16
17–20
25–26
27–29
D
0.2870
0.2309
0.4287
0.1793
0.4333
0.2631
p
7.74e–05
5.59e–05
2.89e–14
0.187
2.44e–04
0.084
Median egg diameter
Apr–Sep
Oct–Mar
2.5
2.5
2.7
3.0
3.5
4.0
2.4
2.4
2.5
3.0
3.3
4.2
11 551-564 Arkhipkin
558
4/10/0 4:39 pm
Page 558
A.I. ARKHIPKIN, V.V. LAPTIKHOVSKY & D.A.J. MIDDLETON
Figure 7. Comparison of egg diameter frequency distributions in (A) April to September and (B) October to
March, at different embryonic stages, and using boxplots (C). The width of the box is proportional to sample
size. If the notches of two plots do not overlap then the medians are significantly different at the 5 percent level
(McGill et al. 1978). Embryonic stages: (a) 1–3, (b) 4–12, (c) 13–16, (d) 17–20, (e) 25–26, (f) 27–29, (g) 30.
11 551-564 Arkhipkin
4/10/0 4:39 pm
Page 559
COLD WATER SPAWNING ADAPTATIONS IN LOLIGO GAHI
increase in egg diameter, both length and width
of the egg capsule did not change during embryonic development. To demonstrate this, egg
capsules containing 75–85 eggs with embryos at
different stages of development were taken as
an example (see Fig. 8). It is probable that the
inner jelly layer of the egg capsule had dissolved so allowing the eggs to occupy more
space inside the capsule.
559
summer in stages 1 to 20. On a few occasions
capsules were encountered where all eggs were
dead and a developmental stage could not be
assigned. These are not included in Table 2 and
the numbers involved are too small to detect
any seasonal difference (2 out of 1271 capsules
examined in summer, 1 of 137 capsules examined in winter).
Multiple maternity of eggs within egg masses
Egg mortality within the capsules
The percentage of dead eggs within the sampled capsules was very low and did not exceed
4.96% (Table 2). The greatest incidence of dead
eggs was observed for stages from 17 to 20. In
winter, the proportion of dead eggs was more
than four times greater than that in spring and
Among 29 egg masses that were examined
completely, 15 contained eggs that were all at
the same stage of development (i.e. all eggs
within a maximum range of one point on the
Arnold scale). The remainder had capsules with
eggs at different (up to 7) stages of embryonic
development. Large egg masses usually con-
Figure 8. The relationship between capsule length and the stage (297 capsules from 37 different egg masses).
Kendall’s rank correlation, τ 0.06, p 0.11.
Table 2. Percentage of dead eggs (of all eggs examined) in sampled egg capsules.
Month sampled
Apr–Sep
Oct–Mar
Stages 1–12
(cleavage and formation)
of the blastodisc
Stages 13–16
(cellulation)
Stages 17–20
(early
organogenesis)
Stages 25–30
(late
organogenesis)
0.34 (3802)
0.07 (33426)
0.86 (2431)
0.20 (36297)
4.96 (2679)
1.19 (13577)
1.36 (1105)
0.91 (7588)
11 551-564 Arkhipkin
560
4/10/0 4:39 pm
Page 560
A.I. ARKHIPKIN, V.V. LAPTIKHOVSKY & D.A.J. MIDDLETON
tained capsules of very variable size with eggs at
different stages of development (Fig. 9). This
probably reflects multiple maternity of the egg
clusters within a mass, which is common for
loliginid squids (Drew, 1911), rather than different rates of egg development in different egg
capsules. The simultaneous occurrence of capsules at very different stages of embryonic
development (stages 14 and 28, for example) in
the same egg mass supports this conclusion.
Within large egg masses with eggs at multiple
stages of development, it is possible that those
egg capsules containing eggs at the same stage
of development were laid by a single female.
The total number of eggs in these ‘single female
egg masses’ ranged usually from 400 to 1300,
which corresponds to the total egg number
observed in the oviducts of mature females. Our
preliminary studies have shown that egg number in the oviducts varies from 500 (110–120
mm ML) to 1300 (190–200 mm ML), sometimes
reaching 4300 in large, rare females of 280–300
mm ML. Egg masses containing 3000–9500 eggs
at the same developmental stage are therefore
assumed to have been laid simultaneously by
several females.
Hatching and paralarvae
One egg mass studied contained embryos at
development stage 30 (ready to hatch). During
its analysis in the laboratory (an hour after its
collection), paralarvae started to hatch from the
egg mass, probably due to a sudden increase in
water temperature in the jar. Each paralarva
emerged from the egg and then penetrated
easily through the outer gelatinous layer of the
egg capsule. After paralarva escapement the
hole in the outer layer of the capsule was
quickly filled by the surrounding jelly and dis-
Figure 9. Numbers of egg capsules with embryos at different stages of development in three separate egg
masses (A, B and C).
11 551-564 Arkhipkin
4/10/0 4:39 pm
Page 561
COLD WATER SPAWNING ADAPTATIONS IN LOLIGO GAHI
561
Figure 10. General morphology and head chromatophore arrangement of embryos at stage 28 (A) and newly
hatched paralarvae (B) of Loligo gahi. Ventral (a) and dorsal (b) views. Orange chromatophores are shown as
empty circles, and brown chromatophores as black circles. The scale bar is 1mm.
appeared leaving the capsule appearing intact,
but with an empty egg inside.
The chromatophore arrangement of the
paralarva head (Fig. 10) was typical for Loligo
spp. hatchlings (Segawa, Yang, Marthy & Hanlon, 1988; Vecchione & Lipinski 1995), confirming that the egg masses studied belonged to
Loligo sp. There is only one species of this genus
in Falkland’s waters: Loligo gahi. The mantle
length of newly hatched paralarvae varied from
3.1 to 3.4 mm (mean 3.2 mm ML).
DISCUSSION
Squid of the family Loliginidae are demersal
spawners laying their egg masses on different
substrates on the bottom (Hanlon, 1998). Some
species tend to spawn inshore at depths 15–20 m
(L. vulgaris reynaudii, Sauer et al., 1992; Lolliguncula brevis, Hall, 1970), other species prefer
to spawn deeper on the shelf and continental
slope (L. forbesi, Lordan & Casey, 1998). The
inshore spawning L. v. reynaudii is rather flexible in choosing spawning sites, from relatively
protected bays to exposed parts of the coast,
but always below the energetic turbulent zone
(15 m depth, Sauer et al. 1992). In contrast to
the latter species, L. gahi almost always chooses
rather exposed sites, often on the outer edge of
the kelp forest, with significant surge and tidal
movements. The spawning substrate of L. gahi
seems to be unique among loliginid squids, with
eggs laid on stipes of algae at shallow depths.
Most loliginids studied lay their egg capsules
one by one on a sandy bottom, anchoring them
in the sand by the sticky basal tip of the capsule
(L. opalescens, McGowan, 1954; L. plei, Waller
& Wicklung, 1968; L.v. reynaudii, Sauer &
Smale, 1993). They usually form large ‘egg beds’
several metres in diameter containing as many
as 6700 egg capsules (Augustyn, Lipinski, Sauer,
Roberts & Mitchell-Innes, 1994). Females occasionally attach a few egg capsules to different
objects on the bottom such as rope moorings,
twigs and crab pots (L. forbesi, Holme, 1974).
The reason for the preference of L. gahi to
spawn onto algal stipes is unclear. In Falkland
waters there are no potential fish predators
capable of eating the squid egg capsules such
as, for example, the sparid Spondyliosoma
emarginatum in South Africa which attacked
the egg capsules of L. v. reynaudii (Sauer &
Smale, 1993). Only benthic invertebrates could,
therefore, potentially feed on squid eggs. When
offered egg capsules in the laboratory aquarium, spider crabs easily opened the egg capsules
using their claws, and removed and ate the eggs
11 551-564 Arkhipkin
562
4/10/0 4:39 pm
Page 562
A.I. ARKHIPKIN, V.V. LAPTIKHOVSKY & D.A.J. MIDDLETON
(pers. obs.). Thus, laying the egg capsules on
smooth algae stipes at some distance above the
bottom should protect them from foraging by
invertebrates. Continuous movement of the egg
masses together with the algae may prevent
significant coverage by organic debris and
enhance egg respiration. Finally, locating the
egg masses on algal stipes may afford protection in the high-energy near-shore environment.
Despite surveying near the end of the spawning period of the abundant second cohort of
L. gahi (Hatfield & Des Clers, 1998), the total
density of egg masses found at the edge of the
kelp forest was low. It is possible, therefore,
that the main spawning grounds of L. gahi are
located in deeper waters (50 m depth) where
fully mature and spawning females have been
frequently caught by trawling (Rasero &
Guerra, 1998; our data). Two facts, however, do
not favour deepwater spawning. Firstly, we
have never observed egg masses of L.gahi
attached to any substrate other than algal stipes,
and suitable algae do not occur deeper than
20 m with any great frequency. Secondly, egg
masses have not been reported from the bottom
trawl fishery for L. gahi despite extensive monitoring of the fishery by scientific observers
(George & Hatfield, 1995). Further investigations are needed to clarify the status of the
inshore spawning sites of L. gahi.
L. gahi is the coldest water spawning species
among the loliginids, the majority of which
spawn at ambient temperatures greater than
11–14°C (Boletzky, 1987). Development of other
loliginid embryos is greatly slowed or even
arrested at temperatures less than 11°C (Segawa
et al., 1988; Sauer et al., 1992). However, live
and developing embryos of L. gahi were found
at temperatures well below this 11°C limit
(6.5–8°C).
In loliginids the length of egg capsules seems
to be dependent on the maximum size attained
by fully mature females. Relatively small loliginids (100mm ML 200 mm, L. opalescens)
lay short egg capsules (60–90 mm in length,
Yang, 1986), whereas the large L. forbesi from
Azores (300–400 mm ML) produces long egg
capsules (180 mm, Segawa et al., 1988). L. gahi
is characterised by small mature females
(mainly 140–170 mm ML) and fits well into the
start of this sequence, with egg capsules of
mean length 55 mm. The total number of eggs
within each egg capsule does not, however,
depend on female size across different species.
It is approximately the same, for example, in
the small L. gahi (70–80 eggs) and the large
L. forbesi (85–100 eggs, Segawa et al., 1988).
Despite the small size of mature females in L.
plei (95–126 mm ML) they have 200–300 eggs
in their capsules (Waller & Wicklung, 1968). In
L. opalescens, the total number of eggs ranged
from 107 to 199 (mean 156 eggs) (Yang, Hixon,
Turk, Krejci, Hulet & Hanlon, 1986). Thus, the
cold water spawning loliginids tend to have
fewer eggs in each egg capsule than warm water
spawning loliginids. In contrast, both egg diameter and hatchling mantle length are greater
in cold water spawning loliginids than in those
which spawn in warm water. Egg diameter
ranges from 1.0–1.6 mm in the warm water L.
pealei (Summers, 1983) to 3.0–3.3 mm in the
cold water L. forbesi (Segawa et al., 1988), while
hatchling ML varies from 1.8 mm in L. pealei
(McConathy, Hanlon & Hixon, 1980) to 4.3–4.9
mm in the cold water L. forbesi (Segawa et al.,
1988). Thus, the egg diameter (2.2–2.5 mm) and
hatchling size (3.1–3.4 mm ML) of L. gahi correspond well with its cold water spawning habit.
The occurrence of capsules without eggs,
and capsules containing only a few eggs, was
observed in the L. gahi egg masses and is a wellknown feature in other loliginids. Females of L.
opalescens usually lay empty egg capsules in the
early portion of the spawning period (Yang et
al., 1986). The collective formation of egg
masses by several females, and the addition of
egg capsules to existing egg masses, found in L.
gahi, are also characteristic for loliginids. These
have been exploited to attract mature females
to spawn in aquaria by the attachment of artificial silicon egg capsules to the bottom (Yang
et al., 1986).
Despite the fact that we have not observed
any spawning specimens of L. gahi during our
diving study it is possible to reconstruct their
spawning strategy taking into account the diversity and morphological structure of their egg
masses in nearshore waters. At least some L.
gahi arrive in shallow waters to spawn, presumably in small schools containing several animals.
They choose exposed, shallow shelf areas to
spawn and lay their egg masses on algal stipes
to protect them from predation by benthic
invertebrates. L. gahi possess a ‘dispersed’ type
of spawning, with rather short egg capsules containing a number of eggs that is near the minimum for loliginid squids. This is probably an
adaptation to cold water spawning when a prolonged egg development makes egg masses
more vulnerable to predation and displacement
from the algae by surge during stormy weather.
On the other hand, the relatively large eggs
of L. gahi produce large hatchlings that are
capable of attacking large plankton prey, and of
11 551-564 Arkhipkin
4/10/0 4:39 pm
Page 563
COLD WATER SPAWNING ADAPTATIONS IN LOLIGO GAHI
escaping from potential predators, both of which
are advantageous for starting life in a coldwater environment. The same adaptive features
are observed in the North Atlantic cold water
loliginid L. forbesi (Boyle, Pierce & Hastie,
1995). Acquisition of these adaptive features
was doubtless of great importance to a loliginid
squid that undertakes only limited migration,
such as L. gahi, in the occupation and utilisation
of the resources of the productive, cold-water
shelf and slope around the Falkland Islands.
ACKNOWLEDGEMENTS
We thank Emma Jones and Steve Waugh of the Falkland Islands Fisheries Department, and David Eynon
of South Atlantic Marine Services, for taking part in
the dives described. Paul Ellis of Sulivan Shipping
Services provided egg masses taken during dredging
and Lian Butcher assisted in the laboratory analysis
of samples. The crew of the Dorada provided invaluable assistance and local knowledge. We thank the
Director of Fisheries, John Barton, for supporting this
work.
REFERENCES
AUGUSTYN, C.J., LIPINSKI, M.R., SAUER, W.H.H.,
ROBERTS, M.J. & MITCHELL-INNES, B.A. 1994.
Chokka squid on the Agulhas Bank: life history
and ecology. South African Journal of Marine
Science, 90: 143-154.
ARNOLD, J.M. 1965. Normal embryonic stages of the
squid, Loligo pealii (LeSueur). Biological Bulletin,
128: 24-32.
BOLETZKY, S. VON. 1987. Embryonic phase. In:
Cephalopod life cycles, vol. II (P.R. Boyle, ed.),
5-31. Academic Press, London.
BOLETZKY, S.VON. 1989. Recent studies on spawning,
embryonic development, and hatching in the
Cephalopoda. Advances in Marine Biology, 25:
85-115.
BOYLE, P.R., PIERCE, G.J. & HASTIE, L.C. 1995. Flexible reproductive strategies in the squid Loligo
forbesi. Marine Biology, 121: 501-508.
BUTLER, J., FULLER, D. & YAREMKO, M. 1999. Age
and growth of market squid (Loligo opalescens)
off California during 1998. CalCOFI Reports, 40:
191-196.
DREW, D.G. 1911. Sexual activities of the squid
Loligo pealei (Les.). I. Copulation, egg laying and
fertilization. Journal of Morphology, 22: 327-359.
FIG, 1998. Fisheries Department fisheries statistics,
Vol. 3. Falkland Islands Government Fisheries
Department, Stanley.
GEORGE, M.J.A. & HATFIELD, E.M.C. 1995. First
records of mated female Loligo gahi (Cephalopoda: Loliginidae) in the Falkland Islands. Journal
of the Marine Biological Association of the U.K., 75:
743-745.
563
HALL, J.R. 1970. Description of egg capsules and
embryos of the squid, Lolliguncula brevis, from the
Tampa Bay, Florida. Bulletin of Marine Science, 20:
762-768.
HANLON, R.T. 1998. Mating systems and sexual selection in the squid Loligo: how might commercial
fishing on spawning grounds affect them? CalCOFI
Reports, 39: 92-100.
HATFIELD, E. & DES CLERS, S. 1998. Fisheries management and research for Loligo gahi in the Falkland
Islands. CalCOFI Reports, 39: 81-91.
HATFIELD, E.M.C. & RODHOUSE, P. 1994. Migration
as a source of bias in the measurement of cephalopod growth. Antarctic Science, 6: 179-184.
HATFIELD, E.M.C., RODHOUSE, P.G. & POREBSKI, J.
1990. Demography and distribution of the Patagonian squid Loligo gahi d’Orbigny during the
austral winter. Journal de Conseil, 46: 306-312.
HOLME, N.A. 1974. The biology of L. forbesi Steenstrup (Mollusca: Cephalopoda) in the Plymouth
area. Journal of the Marine Biological Association
of the U.K., 54: 481-503.
LORDAN, C. & CASEY, J. 1999. The first evidence of
offshore spawning in the squid species Loligo
forbesi. Journal of the Marine Biological Association of the U. K., 79: 379-381.
MCCONATHY, D.A., HANLON, R.T. & HIXON, R.F.
1980. Chromatophore arrangements of hatchling
loliginid squid (Cephalopoda, Myopsida). Malacologia, 19: 279-288.
MCGOWAN, J.A. 1954. Observation on the sexual
behaviour and spawning of the squid Loligo opalescens at La Jolla, California. Californian Fish and
Game, 40: 47-54.
MCGILL, R., TUKEY, J.W. & LARSEN, W.A. 1978. Variations of box plots. American Statistician, 32: 12-16.
RASERO, M., PORTELA, J.M. 1998. Relationships
between mating and sexual maturation of Loligo
gahi females in Falkland waters. Journal of the
Marine Biological Association of the U. K., 78: 673676.
ROPER, C.F.E., SWEENEY, M.J., NAUEN, C.E. 1984.
FAO species catalogue. Vol. 3. Cephalopods of the
world. An annotated and illustrated catalogue of
species of interest to fisheries. FAO Fisheries Synopsis, 125: 1-277.
SAUER, W.W.H., MCCARTHY, C., SMALE, M.J. &
KOORTS, A.S. 1993. An investigation of the egg
distribution of the chokka squid, Loligo vulgaris
reynaudi, in Krom Bay. South African Bulletin of
Marine Science, 53: 1066-1067.
SAUER, W.W.H. & SMALE, M.J. 1993. Spawning
behavior of Loligo vulgaris reynaudii in shallow
coastal waters of the south-eastern Cape, South
Africa. In: Recent advances in cephalopod fisheries
biology (T. Okutani, R. K. O’Dor & T. Kubodera,
eds) 489-498. Tokai University Press, Tokyo.
SAUER, W.W.H., SMALE, M.J. & LIPINSKI, M.R. 1992.
The location of spawning grounds, spawning and
schooling behaviour of the squid Loligo vulgaris
reynaudii (Cephalopoda: Myopsida) off the Eastern Cape Coast, South Africa. Marine Biology, 114:
97-107.
11 551-564 Arkhipkin
564
4/10/0 4:39 pm
Page 564
A.I. ARKHIPKIN, V.V. LAPTIKHOVSKY & D.A.J. MIDDLETON
SEGAWA, S., YANG, W.T., MARTHY, H.J, & HANLON,
R.T. 1988. Illustrated embryonic stages of the
eastern Atlantic squid Loligo forbesi. The Veliger,
30 (3): 230-243.
SOKAL, R.R. & ROHLF, F.J. 1981. Biometry, 2nd edn.
W.H. Freeman & Co., New York
SUMMERS, W.C. 1983. Loligo pealei. In: Cephalopod
life cycles, vol. I (P. R. Boyle, ed.) , 115-142. Academic Press, London.
SWEENEY, M.J., ROPER, C.F.E., MANGOLD, K.M.,
CLARKE, M.R. & VON BOLETZKY, S. 1992. ‘Larval’
and juvenile cephalopods: a manual for their identification. Smithsonian Contributions to Zoology,
513, 1-282.
TINGLEY, G., SAUNDERS, G., HARRIES, D. & KING, J.
1996. The first shallow marine survey around the
Falkland Islands. A report prepared for the Falkland
Islands Government. IC Consultants Ltd (ICON)
and Brown & Root Environmental, London.
VECCHIONE, M. & LIPINSKI, M. 1995. Descriptions of
the paralarvae of two loliginid squids in southern
African waters. South African Journal of Marine
Science, 15: 1-7.
WALLER, R.A. & WICKLUNG, R.I. 1968. Observations
from a research submersible - mating and spawning
of the squid, Doriteuthis plei. BioScience, (February
1968): 110-111.
YANG, W.T., HIXON, R.F., TURK, P.E., KREJCI, M.E.,
HULET, W.H. & HANLON, R.T. 1986. Growth,
behavior, and sexual maturation of the market
squid, Loligo opalescens, cultured through the life
cycle. Fishery Bulletin, 84: 771-798.