-
Deep-Sea Research Vol 39 No 7/8, pp 1177-1200 1992 0198-0149/92
$5 00 + 0 00 Printed m Great Britain © 1992 Pergamon Press Ltd
Models of the early life history of Euphausia superba--Part I.
Time and temperature dependence during the descent-ascent
cycle
EILEEN E. HOFMANN,* JORGE E CAPELLA,'~ ROBIN M Ross$ a n d
LANGDON B . QUETIN:~
(Recetved 31 December 1990, m revtsed form 25 November 1991,
accepted 4 December 1991)
Abs t r ac t - -A time- and temperature-dependent model was
developed to simulate the descent - ascent behavior of the embryos
and early larval stages of the Antarctic knll, Euphausza superba
This model combines laboratory measurements of temperature effects
on developmental times, density and physiology of krlll embryos and
larvae and the observed water temperature structure in the
Bransfield Strai t-South Shetland Islands region Simulations with
observed vertical tempera- ture profiles from this region show that
embryos that develop at temperatures less than 0°C hatch relatively
deep (~1000 m) or hit the bot tom before hatching The presence of
warm (1-2°C) Circumpolar Deep Water (CDW), between 200 and 700 m,
results in hatching depths of about 700 m The sinking rate pattern
characteristic of the embryos of Euphausta superba retains the
embryos in the CDW, where development is accelerated Larval ascent
rate through the C D W is rapid, so larvae reach the surface before
metamorphosing into the first feeding stage, and have sufficient
carbon reserves to drift at the surface for several weeks before
needing to find food These results suggest that the sinking rate
pat tern characteristic of embryos of Antarctic krlll may be part
of a reproductwe strategy that evolved in response to the thermal
structure of its environment The complementary component of this
reproductive strategy is the observed correlation between the
distribution of krlll schools containing reproducing individuals
and the presence of C D W With this reproductive strategy, the
spawning regions of Antarctic krlll are m areas where oceanxc
conditions enhance the probability of survival of its embryos and
non-feeding larvae
I N T R O D U C T I O N
THE majority of the 85 known species of euphausuds live In
oceanic eplpelagic and mesopelagic environments. Embryos of some of
these euphausild species are heavier than water and sink from the
moment of release After hatching, larvae ascend through the water
column in order to reach the surface waters where they feed. For
embryos and early non-feeding larval stages of Antarctic krill,
Euphausm superba, this descent-ascent behavior spans a wide depth
range.
The descent-ascent cycle of embryos and larvae of E superba was
first suggested by MARR (1962) on the basis of net tows taken
during the H M S Dtscovery cruises (1925-
*Depar tment of Oceanography, Old Dominion University, Norfolk,
VA 23529, U S A tMesosca le Air -Sea Interaction Group, Florida
State Umversxty, Tallahassee, FL 32306, U S A :~Marlne Science
Instxtute, University of Callforma, Santa Barbara, CA 93106, U S
A
1177
-
1178 E E HOFMANN et al
Table 1 Imttal knl l embryo s m k m g rates obtamed f rom
laboratory measurements
Flme Stoking rate (h) (rn day 1) Reference
12 140-320 MARSCHALL (1983)
12 175 QUETIN and Ross (1984a) 12 201 Ross and QUETlN (1985)
1939 and 1950-1951) m the Antarctic, and later confirmed by
other investigators (HEMPEL and HEMPEL, 1986) As the krill embryo
sinks, it passes through six distinct developmental stages:
multiple cell, blastula, gastrula, limb bud, pulsating heart or
twitching, and hatching (QuETIN and Ross, 1984a) As the larva
ascends, it passes through two non- feeding (Nauplius,
Metanauplius) and several feeding (Calyptopls 1-3) developmental
stages
The descent-ascent cycle described by MARR (1962) was refined by
QUETIN and Ross (1984a) who showed that embryos of E. superba do
not sink at the initial high rates (Table 1) throughout development
The sinking rate pattern described by QUETIN and Ross (1984a) is
characterized by a decrease in sinking speed to a minimum during
the gastrula stage, between 48 and 72 h after release, that ranges
from 51-73 m day -1 (QuETIN and Ross. 1984a; Ross and QUETIN, 1985)
to 80-90 m day -1 (GEORGE and STROMBERG, 1985) Sinking rates then
accelerate before hatching to about the initial sinking rate This
pattern is independent of temperatures and salinity (QuETIN and
Ross, 1984a) Whether the sinking rate pattern of the embryos of E
superba confers advantages to these early stages is unknown.
Another pattern relevant to reproduction of E superba IS the
observed spatial segregation between sexually mature krill and
subadult krlll during the spawning season. This spatial segregation
has been observed in several regions: the Scotia and northern
Weddell Seas (MAKAROV, 1970), the waters west of the Antarctic
Peninsula (KocK and STEIN, 1978, JAZDZEWSKI el a l . , 1978,
MAKAROV, 1979, WITEK et a l . 1981', QUETIN and Ross, 1984b;
SIEGEL, 1988, and refs cited within), the Bransfield Strait region
(QuETIN and Ross, 1984b; KALINOWSKI, 1982; KITrELL and JAZDZEWSKI,
1982, WOLNOM~EJSKI et al, 1982) West of the Antarctic Peninsula in
the Bransfield Strait region (Fig 1) schools of reproducing krill
are generally found to the north of the South Shetland Islands,
with schools composed of juvemle or non-reproducing krlll inside
the Strait (QuETIN and Ross, 1984b) The spatial separation of
schools of reproducing and non-reproducing krill can be the result
of either physical or behavioral processes
The first objective of this research was to investigate what
processes underlie the pattern of sinking rate during embryonic
development. The second objective was to simulate the effects of
actual temperature regimes on the rate and timing of this pattern
To achieve these two research objectives a model was developed to
simulate the effect of ambient water temperature on developmental
and physiological processes during the descent and ascent of
embryos and early larval stages of krfll
The region of interest for this modeling study is the Bransfield
Strait-South Shetland Islands (cf Fig 1). Circulation in this
region can be complex (CAPELLA, 1989) and has an
-
The early life history of Euphausia superba--Part I 1179
60 ° S
61° l
62 °
I I I I I I I I I I
2O
/ t5
I I I
I I
I 2o
. I /
i " ( • / i
Elephant Island
6ao L II I I I - - Z I
I I i / I - - i I
/ 70 -05" 0 15 /
I *
I 64~
6s
°°-67 = w 6 6 ° 6 s ° 6 4 ° 6 ~ o ~2o 6To 6 o o s 9 o s s o 5 'z
o s'6 o s'5o 54 ° 5~ °
Fig l Composite temperature distribution at 500 m The
distribution was constructed using historical temperature data
collected during the last 10 years from XBT and CTD observations
(CAPELLA et al, 1992b) Dots indicate the location of stations from
which the temperature measurements were obtained Contour interval
is 0 5°C Dashed lines indicate regions of sparse temperature
observations Also shown are observed distributions of krlll schools
containing gravld ( 0 ) and nongravid ( A ) females observed during
the austral summer of 1982 by QUETIN and Ross
(1984b) Gerlache Strait is identified as Ger Str
important effect on the distribution of early life stages of
krill. Therefore, understanding developmental (i e biological)
effects independent of circulation effects is necessary to
accurately interpret the space and time-dependent embryo-larva
particle trajectories calculated with the Lagrangian model, which
combines the developmental and circulation effects (CAPELLA et al,
1992a)
The time- and temperature-dependent model is described in the
following section This is followed by a discussion of simulated
descent-ascent profiles of krlll embryos and larvae Finally, these
modeling results are discussed in relation to the more general
understanding of whether factors such as the pattern of sinking
rate of the embryo and the observed spatial distribution of schools
or reproducing Antarctic krill E superba contribute to recruitment
success in this species
-
1180 E E HOFMANN et al
METHODS
Time- and temperature-dependent model
Simulated descent and ascent patterns from the time- and
temperature-dependent model are influenced greatly by temperature
effects on developmental time of the krlll embryo and larva Because
early development in krill IS equlproportional (Ross et al , 1988)
the biological processes in the model are formulated in terms of
fraction of total development time. Equlproportlonal development
means that the duration of any developmental stage is the same
proportion of total development at all temperatures_ The proportion
IS not the same for all stages, however In the descent portion of
the model, stoking rate of an embryo stage is calculated from the
difference between embryo and water density In the model, embryo
wet weight (mass) and diameter are dependent on the fraction of
total embryo developmental time (from release to hatching), which
is controlled by temperature. Stated mathematically
d(dlameter) _ diameter change(development) (1) dt
and
d(wet weight) = wet weight change(development) dt
(2)
From the embryo diameter and wet weight, embryo density can be
calculated as
6 embryo wet weight flembryo ---- 7/. (embryo diameter) 3
(3)
From embryo density, sinking rate can then be calculated from
Stokes' Law as
2 r 2 _/Pembryo __ 1t sinking rate = ~ v g ~ Pw (4) )
where r is the embryo radius, v is the kinematic viscosity of
water, g is the gravitational a c c e l e r a t i o n , Pembryo lS
the embryo density and Pw is the ambient water density For the
calculations presented in the following sections the value of v was
1.787 x 10 -6 m 2 s -1, which is the kinematic viscosity for
seawater at 0°C With a known sinking rate, a time- dependent depth
profile for the embryo can be determined Ambient pressure does not
have a direct effect on embryo stoking rates (Ross and QUETIN,
1985).
The ascent portion of the model simulates the behavior and
physiology of the non- feeding larval stages Larval ascent rate is
dependent on ambient water t empera tu re
d(ascent) dt - ascent change(temperature). (5)
Larval ascent rate can be used to construct a depth profile, as
for the embryo. Additionally, carbon loss from respiration in the
embryo and the larva is estimated as
d(carbon) _ respiration(development) (6) dt
-
T h e ear ly life h i s tory of Euphausm superba--Part I
1181
where carbon use at any time is dependent on the fraction of the
total developmental time. Descriptions of the krill embryo and
larva measurements used to formulate the terms on the right side of
equations (1)-(6) follow.
Embryo and larva developmental ttme
Embryo developmental times were measured at temperatures of -
1,0, 1 and 2°C, which encompass the range of temperatures
experienced by Antarctic krlll (Ross et a l , 1988) Total
developmental time of embryos is faster at temperatures greater
than 0°C and ranges from 5-8 days (MARSCHALL, 1984; IKEDA, 1984,
ROSS et a l , 1988). As part of the experiments reported in Ross et
al (1988), developmental times of five embryo stages-- single cell
to early gastrula (SC-eG), early gastrula to gastrula (eG-G),
gastrula to early limb bud (G-eLB), early limb bud to late limb bud
(eLB-1LB) and late limb bud to hatch (1LB-N1)--were also measured
Developmental time for a given embryo stage follows the same
pattern as found for total developmental time Relattonships between
developmen- tal time and temperature for the first, third and fifth
embryo stages were best described by exponentially decreasing
functions (Table 2) Development of the second and fourth stages was
best described by linear functions (Table 2)_ These developmental
relationships are used in the descent portion of the model
Larval developmental time is sensitive to temperature, despite
the relatively narrow temperature range in regions where krlll are
found (Ross et a l , 1988) The relationships between larval
developmental time and temperature for the first three larval
stages-- Naupllus (N1) to Metanaupllus (MN), N1 to Calyptopls 1
(C1), and N1 to Calyptopls 2 (C2)--used in the ascent portion of
the model are exponentially decreasing functions (Ross et a l ,
1988, Table 2) One additional finding of importance is that larvae
reared at
Toble 2 Equattons used to calculate embryo and larval
developmental nines, embryo dtameter, and larval ascent rate
Developmental ume m hours and larval ascent rate m m day 1 are
predzcted f rom temperature,
T Embryo dtameter m ktm ts predzctedfrom developmental ttme,
x
Var i ab l e T e m p e r a t u r e E q u a t i o n
D e v e l o p m e n t a l t imes
S C - e G - 1 ° C - 2 ° C y = 1 225e -2 a s l r + 24 147
e G - G T < 0°C y = 33 3 - 10T
T ->/l°C y = 33 3
G - e L B - 1 ° C - 2 ° C y = 11 851e 1 123Tq_ 62 404
e L B - 1 L B - 1 ° C - 2 ° C y = 110 15 + 14 8 T
1Lb-N1 - 1 ° C - 2 ° C y = 37 258e 0 90v:r + 108 644
N 1 - M N - 1 ° C - 2 ° C y = 38 36e l 4L7 + 225 29
N 1 - C 1 - 1 ° C - 2 ° C y = 78 26e -~ ~ 3 r + 417 93
N 1 - C 2 0°C-2°C y = 3 2 0 58e i 10r + 752 22
E m b r y o d i a m e t e r - I °C y = 626 605 + 104 936x - 247
258x 2 + 192 003x 3
0°C y = 621 557 + 153 305x - 355 408x z + 260 204x 3
l °C y = 620 460 + 150 220x - 334 003x 2 + 238 811x 3
2°C y = 621 505 + 138 389x - 325 546x 2 + 260 204x 3
M e a n y = 621 557 + 153 305x - 355 408x 2 + 260 204x 3
L a r v a l ascent ra te - I ° C -< T - < 0°C y = ( - 0
208 - 0 0 1 1 T ) P s
0°C -< T -< I °C V = ( - 0 208 - 0 043T)P ,
-
1182 E E HOFMANN et al
- 1°C generally fail to develop beyond the C1 stage, suggesting
that the larvae undergo a period of strong cold sensitivity (Ross
et a l , 1988)
Embryo dtameter and wet weight
Throughout experiments on developmental times of embryos and
larvae, diameter and wet weight of embryos from each of four broods
reared at four temperatures ( - 1,0, 1 and 2°C) were monitored (see
Ross et a l , 1988, for rearing techniques) Diameters of five
embryos from each of the 16 brood- temperature combinations were
measured under a dissecting microscope at about 12 h intervals. At
36 h intervals, 10 Individual embryos from each brood- tempera ture
combination were weighed with a Cahn electrobalance, diameters of
five of these same embryos were measured prior to weighing Embryos
were discarded after any handling. Diameter measurements have an
individual precision of +10 j~m, and are within the brood standard
deviation of 9-15 ~m (1-2%) for any one measurement period
(methodology in QVETIN and Ross, 1984a).
Evidence suggests embryos are fragile and deform easily during
the first 3 ~ h, and measurements during this time period may be
overestimates Estimates of initial diameter (
-
The early hfe history of Euphausta superba--Part I 1183
680
E
660
E tB r~ g 640
. .Q
E t.u
620
680
~" 670
• - 1 ° C
~ , 0 ~ C
1 o c
a 2 ~ C
• x • • ) / • = o •
tn • ~,
• ~ ~ta • a, t3 • ~ • • o & i •
• !
o
1 I I I
02 04 06 08 Fract ion Development T ime
%
• A = t3
a i
10
• 660
650
O 640 1 ~ ~ ~ ~ C ..O E LU 63O
b 6200 I t 016 I I
02 04 08 10 Fract ion Development T)me
Fig 2 Relationship between krlll embryo diameter and development
(a) Average diameter ot knll embryos as a function of fraction of
embryo development time, for four temperatures Each point is the
mean (n = 20) for embryos from all four broods combined at any one
temperature (b) Curve fits for the relationships between embryo
diameter and fraction of development time for the four
temperatures, coefficients m Table 2 Data from
Ross and Quetm
Larval ascent rate
The krill embryo hatches at depth into the first of several
non-feedang larval stages, N1 Once hatched, the larva must swim
back to the lighted surface to began feeding because at is a
herbavore (MARSCHALL, 1985) Newly hatched larvae are either
neutrally buoyant or have a shght negative buoyancy (MARsCHALL,
1984; Ross et a l , 1985) and sink unless swimming actively. Thas
negatave buoyancy and records of early larvae in very deep tows
(>2000 m) have led to suggestions that the Nauplius I and II are
weak swimmers that continue to sink for several days (MARSCHALL,
1984; HEMPEL and HEMPEL, 1986). However, because the larva is
capable of swammang upward at the tame of hatching, it as assumed
in this model that the larva starts ats upward ascent immediately
after hatching
Ross et al (1985) observed that larvae swam in a spiral and
measured upward vertical displacement rates of nauplii of the order
of 170 m day -1. In general, the average rate of vertical
displacement increases wath increasing temperature. For the model,
larval ascent rate was parameterized on the basas of data in Ross
et al. (1985) as two hnear functaons (Table 2).
The value used for Ps, the fraction of tame the larva spends
swlmmang, was chosen by comparing the depth of the various larval
stages in the model wath observed depths from
-
1184 E E HOFMANN et al
f=
,$
g ..Q E LU
142
140
138
136
134
132 620
Y= -2s° ~ a
S t I t s c - e G 630 640
144 y ~
143
142
141 eG-eLB
140 t I I I l I
150
148
146
144
~'~ 142 J~ E uJ
y= -3400911 + 0 7539x R = 098 C
[ I I I
180
170
160
150
140 640
y= -311 4229 + 0 7099X R- 1 0 O ~
ILB-NI I I I J I I I I I
650 660 670 680 690
Embryo Diameter (,urn) 638 644 C~6 648
E m b r y o D t a m e t e r (#rn)
Fig 3 Embryo wet weight as a function of diameter for embryo
stages (a) SC-eG, (b) eG--eLB (stages II and III combined), (c)
eLB-1LB, and (d) ILB-N1 Open squares indicate measured wet weight
and diameter values Fdled squares in&care wet weight and embryo
diameter values obtained by inverting the embryo stoking rate
profiles gwen m
QUETIN and Ross (1984a)
trawl data For example, in simulations with a Ps of 10%, larvae
did not reach the surface before running out of carbon. With a Ps
of 50%, metanauplli occurred at the surface, contrary to
observations. A Ps of 30% gave larval depth distributions that
match observations (HEMPEL and HEMPEL, 1986) and is the value used
in the reported simu- lations.
Embryo and larva respiratzon
Carbon budgets for embryos and larvae also are calculated, which
allows estimation of the carbon reserves available to larvae as
they ascend through the water column_ Respiration rates of
known-age krill embryos at 0°C and larvae at three temperatures, -
1, 0 and 2°C, increase with increasing age, even wtthln a stage,
and are affected by temperature (QtJE~N and Ross, 1989)
The relationships between embryo and larva oxygen consumption
rates and the fraction of total developmental time (Fig 4) were
used in the model to estimate carbon used by the embryo or larva
with a standard conversion constant of 0.385/~g carbon used per tzl
of oxygen consumed (QuETIN and Ross, 1989) Embryo or larva carbon
use gives the respiration loss on the right side of equation
(6).
Initial carbon content of the embryo is 15 pg (IKEDA, 1984; Ross
and QUETIN, 1989) This carbon content must sustain the embryo
throughout ItS descent and the larva throughout its ascent until It
is able to begin feeding. Ross and QUETIN (1989) found that the
point-of-no-return (PNR) for the C1 was 10-14 days or 7.5/~g
carbon. At the PNR, the animal wall not survive even if food
becomes available. Thus, 7.5/zg carbon was taken to be
-
The early hfe history of Euphausm superba--Part I 1185
Fig 4
O
X
t r , r ,
gg ~.tu
r r
o
n l
? O 100
X • T 80
nr' '¢1 = p 60
, 0
~ ~ 20
¢= 0 836fl" 1~ (0 7195x) FI= 0 88 a
ta tla q5
tl o ta tl o t~
= 0
0 0 ' 0~2 L 0'4 ' 0'6 ' 0'8 ' 0
F r a c t i o n D e v e l o p m e n t T i m e
y=11 103"10A(0 8489x) R= 089 b
O ¢g
@,..,~.o o o l a r v a t t i i i
0J2 0 4 0J6 0L8 1 0
F r a c h o n D e v e l o p m e n t T~me
Knll (a) embryo and (b) larval resplratxon rate as a function of
fractxon of development time Data from QUETIN and Ross (1989)
the threshold value in the model, once the carbon content
decreased below this value the simulation was ended because the
larva was past the PNR and would not survive.
R E S U L T S
In the Bransfield Strait-South Shetland Islands region, the
krill embryos and larvae are typically exposed to temperatures that
range between - 1 and 2°C. In the following section, we present
simulated depth profiles for krill embryos and larvae exposed to
constant temperatures of - 1 and 2°C, which bracket the range of
temperatures of interest for this study Next we describe
simulations in which the knll embryos and larvae descend and ascend
through the vertical temperature structure observed at various
locations around the South Shetland Islands and in the Bransfield
Strait
Constant temperature simulations
The model at - I°C (Fig 5a) correctly simulates the sinking rate
pattern of krill embryos (Fig 5b) measured by QUETIN and Ross
(1984a) and Ross and QUEXIN (1985) Stoking rates are high (~180 m
day -I) initially, then decrease to about 70 m day -1 during the
gastrula to early limb bud stage. As the embryo develops into the
limb bud stages, sinking rates again increase, reaching a second
maximum just prior to embryo hatching. However, initial sinking
rates from the model tend to underestimate observed stoking rates.
This discrepancy is related to the difficulty in accurately
estimating mmal embryo diameter. Small differences in embryo
diameter have a large effect on the calculation of embryo density
[cf equation (3)] and hence stoking rate. As previously mentioned,
imtial embryo
-
1186 E E HOFMANN et al
-3 0
200
400 g
600
800
1000
1200
-2 i
T ° C
0 1 i i
I I
3
160
120
z_ 8 O
0 >-
m 40
n 0 0
i i i J i i i i i i
- ~ O O b
II
II, '," ,l'V,I , v, , I , 2 4 6 8
T I M E (DAYS) 10
100
-D 8 0
< 6 O
LU 4o
20
I I I I I I
C
N 1 C I
I , , r , I I I 1 5 2 5 3 5
T I M E (DAYS)
Fig 5 (a) Temperature dlstnbuUons used for the constant
temperature slmulaUons (b) Simulated embryo stoking rate profile at
a constant temperature of -I°C The vertical lines along the bottom
correspond to the embryo developmental stages SC-eG (I), eG-G (II),
G-eLB (III), eLB-ILB (IV), and ILB-N1 (V) Embryo sinking rates from
laboratory measurements are also shown O, embryos that developed at
0°C (QuETI~ and Ross, 1984a), A embryos that developed at I°C (Ross
and QUETIr~, 1985) The develop- mental txme recorded for these
stoking rate experiments was scaled to -I°C using the relatlonshnps
for each embryo developmental stage given m Table 2 (c) Simulated
larval ascent rate at a constant temperature of -I°C The vertical
hnes along the bottom axis
correspond to the larval developmental stages N1 and C1
d iameter m e a s u r e m e n t s tend to be overes t imates and
result in lower sinking rates m the m o d e l . H o w e v e r , the
range of sinking veloci t tes can be quite large (cf Figs 2 and 3
in QUEX]N and R o s s , 1984a) and those obta ined from the m o d e
l are wel l within measured values
E m b r y o s at - 1°C sink to approximate ly 1100 m before
hatching (Fig 6a) whereas at 2°C embryos hatch at about 600 m (Fig
6b). A t the t ime of hatching, the carbon content of the e m b r y
o at - 1 ° C 1s 14 8/~g C, only a 1% loss from the low respiration
rate The carbon content of embryos at 2°C is similar
Af ter hatching larvae at - 1°C require approx]mate ly 25 days
to c o m p l e t e the ascent to the surface (Fig. 6a) Larval
ascent rate at - 1 ° C ts low, only about 40 m day -1 (Fig 5c).
During the ascent the larvae pass through two deve lopmenta l
stages, the M N , at about 10
-
The early hfe history of Euphausta superba--Part I 1187
0 15
200
400
~ 600 w a
8 O O
1000
1200
\ h a t c h /
[ i i i i i i f i ~ i i i i i i 2 4 6 6 10 15 25 35
TIME (DAYS)
13
11=8 < O
200
4O0 g -r I.- 600
o 800
V.2~ °,
15 b
13
o
9
1000 C I
I
1200 I I , i - [ I I 7 0 2 4 10 20
TIME (DAYS)
Fig 6 Simulated depth profile (sohd hne) of a krlll embryo-larva
parncle exposed to constant temperatures of (a) - 1°C and (b) 2°C
Dashed hnes are carbon use by the embryo-
larva particle
days, and C1, the first feeding stage, prior to reaching the
surface When the larva reaches the surface its carbon content has
decreased to approximately 11 Hg C (Fig 6a).
The ascent time for the larvae at 2°C (Fig. 6b) IS much shorter
than that at - 1 ° C The shallower hatching depth of approximately
600 m and the increased ascent rate at 2°C mean that the larva only
takes about 10 days to reach the surface. About half way through
its ascent, at a depth of about 300 m, the larva develops into the
C1 stage. Upon reaching the surface the carbon content of the larva
is about 13.5 Hg C (Fig 6b)
Observed temperature stmulattons
The vertical temperature profiles used for the following
s~mulations are from an XBT and CTD data set from the region of
Bransfield Strait and the South Shetland Islands obtained during
the past several years (CAPELLA et a l , 1992b). The temperature
profiles selected for the simulations were chosen to illustrate the
effects of the different water masses found in the model region.
These temperature profiles correspond to areas where krlll schools
have been observed and are consistent with the regions where krlll
embryos
-
1188 E E HOFMANN et al
400
600
D 800
T °(C)
-3 -2 -1 0 0
200
1000
1200
1 2 i i
a
200
400 g
600
a 800
1 O00
" ' - 13
ryo
h a t c h
1 5
13
z 11 O m
r r < O
CI I
12000 2 4 6 10 210
T I M E (DAYS)
Fig 7 (a) Temperature profile from the southern Gerlache Strait,
near Anvers Island (64 9°S, 64 3°W) (b) Simulated depth profile
(sohd hne) and carbon usage (dashed line) for
a krlll embryo-larva pamcle exposed to the temperature profile
shown m (a)
were released for the Lagranglan tracing experiments described
in the following paper (CAPELLA et al., 1992a)
Southern Gerlache Strait nearAnvers Island. At the southern end
of the Gerlache Strait near Anvers Island temperatures in the upper
100 m are cold, reaching - 2 ° C (Fig 7a) Be low 100 m temperatures
increase to almost 0°C near the bottom at 400 m. The cold water in
the upper water column is winter water formed by cooling during the
previous winter The warmer water at depth is CDW, which is found
throughout the eastern part of the Bell ingshausen Sea (GORDON and
BAKER, 1982)
The cold water in the upper 100 m extends the developmental
time, and hence the period of initially high sinking rates, so the
embryo hits the bottom at 400 m in about 3 5 days (Fig_ 7b). The
embryo has passed through only two of its five developmental stages
and must complete the remainder of its development on the bottom
Carbon loss (Fig 7b) results in about a 1% depletion of the initial
carbon content of the embryo
After hatching, the larva ascends (Fig 7b) reaching the surface
in approximately 8 days The initial larval ascent rate is high (50
m day i), decreases to approximately 20 m day -1 as the larva
encounters the cold winter water, and again increases just prior to
reaching the surface. During ascent the larva develops into the MN
and C1 stages and remains as a C1
-
The early hfe history of Euphausla superba--Part I 1189
200
4 0 0
600
121 800
1000
120(
-3 -2 1 2 3 0 i i i
8
T °(C)
-1 0 \,
hatch o o 800
9
F 0 2 4 6 10 20
TIME (DAYS)
Fig 8 (a) Temperature profile from the western Bransfield Strata
between Decepuon and Low Islands (63 l°S, 61 5°W) (b) Simulated
depth profile (sohd hne) and carbon usage (dashed hne) for a krdl
embryo-larva particle exposed to the temperature profile shown
m (a)
until reaching the surface Carbon content of the larva on
reaching the surface is 13.7#g C (Fig 7b), about a 9% depletion of
the imtial carbon content of the embryo
Instde Bransfield Strait. In the western Bransfield Strait at a
location approximately half way between Decept ion and Low Islands,
tempera ture in the upper 200 m IS nearly isothermal at - I ° C
(Fig. 8a). The warmer water below 200 m coincides with a band of C
D W that extends into the western portion of the Bransfield Strait
from Drake Passage through the gap between Smith and Snow Islands
(CAPELLA et al., 1992b) This tempera- ture profile ~s from the edge
of the CDW and therefore the warmest temperatures observed are only
0°C
The cold temperatures at this location extend embryomc
developmental time to about 7 days Consequently, the embryo hits
the bot tom about 1 day prior to hatching (Fig. 8b) and the carbon
usage is small. After hatching the larva reaches the surface in
about 13 days, passing through two developmental stages (MN and C1)
as it ascends (Fig 8b). About 13% of the initial carbon (Fig 8b)
content of the embryo is used throughout the descent-ascent
cycle.
-
1190 E E HOFMANN et al
-3 0
200
400
600
a 800
1000
1200
T °(C) -2 -1 0 1 2 ' 'a
0
200
40O g ~ 600
a 800
1000
1200
~X hatch
13
11§ rr < O
9
CI I 2 4 10 2[0
TIME (DAYS)
Fig 9 (a) Temperature profile from north of Smith Island (62
3°S, 62 8°W) (b) Simu- lated depth profile (sohd line) and carbon
usage (dashed line) for a krlll embryo-larva
particle exposed to the temperature profile shown m (a)
North of the South Shetland Islands Temperature profiles from
north of the South Shetland Islands (Figs 9a, 10a and l l a ) all
show increasing temperature with depth, which results from the
presence of C D W Below 200 m, temperature Increases to about l°C
and remains constant at this temperature to the bottom
The simulated depth profiles at the three locations show
essentially the same behawor High initial sinking rates place the
embryo near the top of the 1°C water in about 1 5-2 days (Figs 9b,
10b and 11b)_ The sinking rate then decreases and the embryo
hatches in about 5 - 6 days at depths of 680-790 m. Carbon usage by
the embryo is small at all locations After hatching the larva swims
upwards and reaches the surface in about 12-15 days (Figs 9b, 10b
and l i b )
Western Bransfield Stratt near the Antarcttc Penmsula. Waters in
the region around the AntarcUc Peninsula are cold, less than - I °
C at all depths (Fig. 12a), which increases embryo hatching time to
almost 10 days However , before hatching the embryo hits the bottom
at 1000 m, about 2 days prior to hatching. Because high initial
sinking rates persist for almost 3 days, the embryo reaches depths
of about 500 m early in development and early in the descent cycle
Carbon use by the embryo (Fig. 12b) is about 1/xg C, which is a 6%
decrease In the initial carbon value. This is the largest depletion
due to embryo resptration in all the scenartos tested
-
The early hfe history of Euphausta superba--Part I 1191
-3 0
200
400 g
600 a
BO0
1000
1200
T °(C) -2 -1 0 1 2
0
200
400
600
o 800
1000
1200
r-T-r-- I=4--4, i i ,
U ' ' " "
" ~ h a t c h
CI I
[ I I I I I . ii , , i 2 4 6 10 20
TIME (DAYS)
15 i
b
13
z 11
9
Fag 10 (a) Temperature profile from north of Snow Island (62
2°S, 61 5°W) (b) Simulated depth profile (sohd hne) and carbon
usage (dashed hne) for a krlll embrycr-larva
parucle exposed to the temperature profile ~hown in (a)
Krlll embryos and larvae reared at - 1°C do not continue to
develop past C1 (Ross et a l , 1988) Whether the embryos or larvae
are critically sensitive to low temperatures during a specific
period sometime during development is not known_ For the
comparative purposes of this study, consideration of the larval
ascent profile and vertical ascent rate when exposed to cold
temperatures (Fig. 12a) is appropriate because larvae exposed to -
1 ° C temperatures early in development surwve until late C1 At the
cold temperatures found near the Antarctic Peninsula, the larva
required about 25 days to complete its ascent to the surface. At
approximately 750 m the larva developed into a C1 Larva ascent rate
was low due to the low temperatures and, consequently, the carbon
use (Fig. 12b) was high Carbon use throughout the descent-ascent
cycle represents a 26% depletion of the imtlal carbon content of
the embryo
D I S C U S S I O N
Temperature effects and Implications for the descent-ascent
pattern
The constant and observed temperature slmulaUons illustrate the
importance of temperature in determining the hatching depth of
krill embryos, the total descent-ascent
-
1192 E E HOFMANN et al
200
400
600
Lu a
8OO
1000
1200
-3 -2 0
T° (C)
-1 0 1 2
0 - - 1 - - - I - - - - 4 - I I I
200
4OO g
600
° I \1 1000 CI 1200 i i = = i . . 0 2 4 10 20
TIME (DAYS)
15
13
z 11o 07 < O
9
Fig 11 (a) Temperature profile from north of King George Island
(60 8°S, 58 5°W), (b) Simulated depth profile (sohd hne) and carbon
usage (dashed hne) for a knll embryo-
larva particle exposed to the temperature profile shown in
(a)
time, and the carbon reserves of the larvae on reaching the
surface (Table 3). In regions where depths are shallow (
-
The early life history of Euphaus,a superba--Part I 1193
T °(C)
-3 0 1 2 0 i i I
a
2 0 0
4 0 0
6 0 0
a BOO 1000
1200
-2 -1
I 0 15
"E 400200 ~ombr,/---~'''''--'-''''--I'--~'o " ' ' ' " [ / I i i
i b 13 ~,~,
6 0 0 11
< tO
8OO
9
1000 I
1 2 0 0 I I i , i i i I _._. i I I I I 7 2 4 6 8 10 20 30 40
TiME ( D A Y S )
Fig_ 12 (a) Temperature profile from near the Antarctic
Peninsula (63 5°S, 61 4°W) (b) Simulated depth profile (solid line)
and carbon usage (dashed line) for a krlll embryo-
larva particle exposed to the temperature profile shown in
(a)
temperatures greater than 0°C. Thus, development 1s accelerated,
and the embryo hatches in 5-6 days The high sinking rate prior to
hatching coincides with the embryo shedding its outer gelatinous
sheath in preparation for hatching The increased sinking rate does
not persist for very long and as a result does not cause the embryo
to sink much deeper in this brief time. Thus, these simulations
lead to the suggestion that the sinking rate pattern characterisUc
of krill embryos may be the result of an evolutionary adaptation by
this animal to the thermal structure of its enwronment .
Larval ascent rate in warm water is high so the larva usually
reaches the surface before developing into the first feeding stage
The larva then has sufficient carbon reserves to drift at the
surface for several weeks before food must be available
Krill school distmbutions and relatton to water mass
dtstrtbuttons
During six austral summers, schools of Antarctic krlll in the
Bransfield Strait-South Shetland Islands region have been
identified and sampled as described in QUETIN and Ross (1984b)
These schools can be categorized as reproducing or non-reproducing
and by being located over C D W or not. Reproducing schools have
greater than 20% gravid females, l.e with red thelyca, swollen
carapaces and fully developed ovaries The
-
Fab
le 3
Su
mm
arv
of
re~u
lt~ fr
om t
he e
mbr
yo-l
arva
stm
ulat
lons
Dep
th o
f T
ime
to
Per
cen
tag
e d
evel
op
men
t T
~me
for
To
tal
Per
cen
tag
e C
h
atch
(m
) ha
tch
(day
s)
on
bo
tto
m
asce
nt (
days
) tt
me
(day
s)
rem
ain
ing
-1°C
10
90
8 4
--
24 6
33
75
2°
C
600
4 8
--
10 2
15
90
G
erla
che
Str
ait
425
7 5
56
10 5
18
91
W
este
rn B
rans
fiel
d S
trai
t 66
0 6
9 19
14
1
21
87
No
rth
Sm
ith
Isl
and
700
5 5
0 13
5
19
85
Sn
ow
Isl
and
79
0 6
1 0
15 4
21
5
84
No
rth
Kin
g G
eorg
e Is
lan
d
680
5 2
0 13
8
19
87
Ant
arcU
c P
enin
sula
-Wes
tern
Bra
nsfi
eld
Str
mt
1025
9
8 17
25
3
35 1
74
> z z
-
The early life history of Euphausta superba--Part I 1195
6 0 ° S I I I I I I I I I I I
Drake Passage
61 = - E~,pham ~ f ~ : C ' ~ m
Klnge • / - 4 - # c%?. , , • /
Is. •
b," " Smith Is , , , , . o ~ \ E _ ~ ' ~ ~ ' " ~ ' "
• "o
e[ _ f stern Weddell Sea 6 5 ° 1 0 ~ " ~ )1 t r I ( I I I I I
I
; 4 o W 6 2 ° 6 0 ° 5 8 = 5 6 ° 5 4 °
Fig 13 Distribution of schools of Euphausla superba in the
Bransfield Strait-South Shetland Islands region with gravid females
greater than 20% of the population (n = 56) Observations are from
seven austral seasons (1981-1982 through 1987-1988) The solid
circle (0) represents all years except 1983-1984, which are
represented by O For five of these years seasonal coverage was
good, from early December through February, with some observations
m March, for 2 years, observations were limited to December-early
January Schools were identified with a simrad echo-sounder as the R
V Polar Duke transected areas where krIll are historically found,
and a 1-m Isaacs Kidd Mtdwater Trawl was used to collect animals
Detads of the materials and methods used to collect the krill are
given in Ross and QUET]N (1983) The solid line is the 0°C isotherm
at 500 m, indicating the approximate southern boundary of CDW
obtained from the temperature data presented in CAPELLA et al
(1992b) Geographic names are abbreviated as Eastern Belhngshausen
Sea (E B S ),
Brabrant Island (B I ) and Snow Island (S I )
geographica l locanon of each school was re la ted to the
presence or absence of C D W by the
0°C i so therm at 500 m (Fig. 13). In all years except 1983-1984
the major i ty of the krtll schools were usually found to the
nor th of the South Shet land Islands, outs ide Smith Island, in
the eas tern Be lhngshausen
Sea, and in the sou thern Bransfield Strait. SIEGEL'S (1988)
analysis of the dis t r ibut ion of the
d i f ferent life stages of krdl m this region f rom several
years confirms that gravid and
spawning adults occur along the cont inenta l s lope, a l though
a small n u m b e r of adults
spawn in the deep basra of Bransf ield Strait Grav]d females
were rarely found in the
central or eas tern Bransfield Strait or m Ger l ache Strmt,
except in austral s u m m e r 1983- 1984. CROXALL et al. (1988)
show that p reda tors on krill did poor ly in the seasons 1977-
1978 and 1983-1984, bo th seasons fo l lowed years when strong
E1 Ni f io /Southern Oscil-
la t ion events occur red The 1983-1984 season ~s cons idered
unusual by several o ther
invest igators as well because of poor spawning success (WITEK
and KITrEL, 1985) or a
reverse m the usual seasonal t rends in mean month ly abundance
(SIEGEL, 1988). W h e n only categor ical r e fo rmat ion such as
the above is avai lable , a n o n p a r a m e t n c
-
1196 E E HOFMANN et al
Table 4 Observed frequen O' o f reproducmg and non- reproducmg
schools o f Euphausla superba found over C D W (>0°C at 500 m)
or over colder water m the Bransfield Stratt, Gerlache Stratt and
outside the South Shetland Islands Data are for schools sampled
durmg five summer seasons (early December through early March),
1981-1982 to 1982-1983, and 1984-1985 to 1986-1987 The season fo l
lowmg the strong El Nt~o (1983-1984) was omttted f rom the analysts
Z 2 = 5 048, contingency coeffictent= 0 21, stgntficant (P < 0
05) Numbers m
parentheses are eapected values
CDW Cold deep Total
Reproducing 23 15 38 (17 4) (20 6)
Non-reproducing 27 44 71 (32 6) (38 4)
Total 50 59 109
contingency coefficient is a useful statistic for determining
whether the two sets of attributes are not related or if the
observed association represents a genuine relationship in the
population (SIEGEL, 1956)_ The contingency coefficient calculated
for data combined for all years except for 1983-1984 (Table 4) was
statistically significant (P < 0 05)_ This evidence that the
observed association between schools of reproducing krill and CDW
is not a result of chance alone provides support for the hypothesis
that the distribution of reproducing schools may be part of a
reproductive strategy that locates spawning populations in water
masses where development and survival of the embryos and larvae is
enhanced.
Schools of reproducing krlll also have been observed in the area
around the tip of the Antarctic Peninsula where warm water is not
present at depth. At colder temperatures, the krill embryos sink
deeper, and often these embryos hit the bottom (cf Table 3),
especially when released in water that is less than 500 m deep.
Once on the bottom the fragile embryos may be destroyed
mechanically or lost by predation The occurrence of juvenile krlll
forms in the Antarctic Peninsula region suggests either that some
portion of the embryos survive or that the krill larvae observed in
this region are advected there by the prevailing circulation This
point is addressed with the Lagranglan calculations presented in
the following paper (CAPELLA et al . 1992a) The final answer,
however, remains to be determined by additional observations,
experiments and modeling studies
It has been suggested that the descent-ascent behavior of the
embryos and larvae of E superba IS a mechanism that serves to
rapidly transport the embryos away from predation by adults. While
this may be a benefit of the descent portion of the cycle, it does
not explain the strong correlation between krill spawning regions
and water mass distributions VORONINA (1974) attempted to explain
the distribution of E superba in the Southern Ocean by considering
the factors that affect the survival of the larvae Her hypothesis
was that the krill are restricted to the regions of the Weddell Sea
and East Wind Drift circulations because the presence of dense
water at depth would prevent the krill embryos from sinking to
depths greater than the larvae would be able to swim back to the
surface
-
The early hfe history of Euphausm superba--Part I 1197
/
Temperature (°C)
\ \
i t I I I
\ \ / . . / /
/
Fig 14 Distribution of embryos of Euphausta superba, denoted by
*, overlaid on the tempera- ture distribution at 500 m obtained
from GORDON and BAKER (1982) Contour interval is 0 5°C Embryo
locations were obtained from FRASER (1936), MARR (1962), MACKINTOSH
(1972), HEMVEL
(1979), HEMPEL et al (1979), HEMPEL and HEMVEL (1986) and RONG
(1989)
and surwve. This hypothesis has since been rejected on the bas~s
of laboratory measure- ments that show that krill embryos are
always denser than sea water and would continue to sink through
dense water at depth, even under pressure (MARSCHALL, 1983; QUETIN
and Ross, 1984a; Ross and QUETIN, 1985). Ad&tionally, for all
scenarios tested with the model, the krill embryos sank through all
parts of the water column
Throughout the southern ocean, warm water is found between 300
and 600 m (GORDON and BAKER, 1982) in response to the northward
transport of surface water induced by the zonal c~rcumpolar wind
stress. The hmlted number of observations of krill embryos outside
the Antarctic Peninsula region suggest that the correlation between
reproducing krill schools and CDW may be circumpolar m nature (Fig
14) However, quantifying the extent of th~s relationship must await
additional observations.
Evoluttonary imphcatmns
The circumpolar circulation in the Southern Ocean developed when
Drake Passage opened approximately 20 milhon years ago (KENNETr,
1982) in the early Miocene The circulation and water mass structure
similar to that of the present day developed m the late
-
1198 E E HOFMANN et al
M i o c e n e (KENNETI" and BARKER, 1990), a b o u t 10 m i l h
o n yea r s ago . T h e s t e m f o r m of the
E u p h a u s i a c e a m a y h a v e d e v e l o p e d a p p r
o x i m a t e l y 300 mi l l ion yea r s ago (BURKENROAD,
1963), p r e d a t i n g d e v e l o p m e n t o f t he A n t a
r c t i c C i r c u m p o l a r C u r r e n t . E v e n if E
superba did n o t a p p e a r unt i l p r e s e n t day c i r c u l
a t i o n p a t t e r n s d e v e l o p e d , 10 mi l l i on yea r
s w o u l d be
suf f ic ien t for t he r e p r o d u c t i v e s t r a t egy p
r o p o s e d h e r e to e v o l v e , e spec i a l l y g iven the
s t r o n g n a t u r e o f the t ra i t
A n e f f e c t i v e e v o l u t i o n a r y s e l e c t i o n
s t r a t egy is o n e tha t is app l i ed to the r e p r o d u c t
i v e
b e h a v i o r o f an o r g a n i s m T h e r e a re m a n y e
x a m p l e s o f m a r i n e spec ies w i th s p a w n i n g
s t r a t eg i e s t ha t t a k e a d v a n t a g e o f f e a t
u r e s in the e n v i r o n m e n t (e .g PARRISH et al., 1981',
SHERMAN el a l , 1984). F o r t h e s e spec ies the v a r i o u s
r e p r o d u c t i v e s t r a t eg ie s a re d e s i g n e d
to
e n h a n c e the su rv iva l o f the y o u n g T h e r e p r o
d u c t i v e s t r a t egy of E superba, which Inc ludes b o t h
the p a t t e r n o f s ink ing ra tes d u r i n g e m b r y o n i
c d e v e l o p m e n t and w h e r e and w h e n
s p a w n i n g occurs , m a y h a v e e v o l v e d in r e s p
o n s e to the e f f ec t o f t he t h e r m a l s t r u c t u r e
o f ItS
e n v i r o n m e n t on the p r o b a b i l i t y o f su rv iva
l of its y o u n g I f such sugges t ions a re t rue , t h e n
an i n t r i gu ing a r e a for f u r t h e r r e s e a r c h is
t ha t o f the p a l e o c l r c u l a t i o n o f the S o u t h e
r n O c e a n and the c o n c u r r e n t e v o l u t i o n a r y a
d a p t a t i o n o f the krill
A~knowledgements--Thls research was supported by the National
Soence Foundation, Dtviston of Polar Programs, under grant number
DPP-85-15486 to E E H , DPP-86-06440 and DPP-85-18872 to R M R and
L B Q We thank David Schnetder, Andrew Clarke and one anonymous
reviewer for constructive comments on this paper
R E F E R E N C E S
BURKENROAD M D (1963) The evolution of the Eucarlda (Crustacea,
Eumalacostraca), in relation to the fossil record Tulane Studtes m
Geology, 2(1), 3-16
CAVELLA J E (1989) Circulation and temperature effects on the
development and distribution of the eggs and larvae of the
Antarctic krlll, Euphausta superba A modeling study Ph D
Dissertation, Texas A&M University, College Station, TX, 162
pp
CAPELLA J E , L B QUETIN, E E HOFMANN and R M Ross (1992a)
Models of the early life htstory of Euphaustasuperba--Part II
Lagranglan calculations Deep-Sea Research, 39, 1201-1220
CAPELLA J E , R M ROSS, L B QUETIN and E E HOFMANN (1992b) A
note on the thermal structure of the upper ocean m the Bransfield
Strait-South Shetland Islands region Deep-Sea Research, 39,
1221-1229
CROXALL J P , T S MCCANN, P A PRINCE and R ROTHERY (!.988)
Reproductive performance of seablrds and seals at South Georgia and
Slgny Island, South Orkney Islands, 1976-1987 Imphcatlons for
Southern Ocean monitoring studies In- Antarcttc Ocean and resources
vartabdtty, D SAHRHAGE, edttor, Springer- Verlag, Heidelberg, pp
261-285
GEORGE R Y and J STROMaERG (1985) Development ol eggs of
Antarctic krlll Euphausta superba Polar Biology, 4, 125-133
GORDON A L and T N BAKER (1982) Oblectlve contouring and grid
point data set, and the atlas data set In Southern Ocean atlas,
Columbta University Press, New York, 11 pp , 233 plates
FRASER F C (1936) On the development and dtstrlbutlon of the
young stages of krtll (Euphausta superba) Discovery Reports, 14,
3-192
HEMPEL I (1979) Vertical dlStrlbutton of eggs and naupln of krdl
(Euphausta superba) south of Elephant Island Meeresforschung, 29,
53-59
H~MPEE I and G HEMPEE (1986) Field observattons on the
developmental ascent of larval Euphausta superba (Criastacea) Polar
Btology, 6, 121-126
HEMPEL I , G HEMPEL and A DE C BAKER (1979) Early life history
stages of krlll (Euphausta superba) to Bransfield Strait and
Weddell Sea Meeresforschung, 27,267-281
IKEDA T (1984) Development of the larvae of the Antarcttc knll
(Euphausta superba Dana) observed in the laboratory Journal of
Experimental Marzne Btology and Ecology, 75, 107-117
-
The early life history of Euphausta superba--Part l 1199
JAZDZEWSKI K , J DZIK, J POREBSKI, S RAKUSA-SUsZCZEWSKI, Z WITEK
and M WOLNOMIEJSKI (1978) Biological and populatlonal studies on
knll near the South Shetland Islands, Scotia Sea and South Georgia
in the summer of 1976 Polskw Archlwum Hydroblologu, 25,607~531
KALINOWSKI J (1982) Distribution and stocks of krlll in the
Drake Passage and the Bransfield Strait, during the BIOMASS-FIBEX
expedition Polish Polar Research. 3,243-251
KENNETT J (1982) Marine Geology Prentice Hall, Inc , New Jersey,
813 pp KENNEYr J P and P F BARKER (1990) Latest Cretaceous to
Cenozoic climate and oceanographic developments
in the Weddell Sea, Antarctica an ocean-drilling perspective In
Proceedings of the Ocean Drdhng Program Scientific Results. P F
B~.RKER and J P KENNETr. editors, U S Government Printing Office.
Washington. DC, Vol 113,937-960
K1TIEL N and K JAZDZEWSKI (1982) Studies on the larval stages of
Euphausta superba Dana (Crustacea. Euphauslacea) in the Southern
Drake Passage and in the Bransfield Strait In February and March
1981 during the BIOMASS-FIBEX expedition Pohsh Polar Reaearch. 3.
273-280
KOCK K -H and M STEIN (1978) Krlll and hydrographic conditions
off the Antarctic Peninsula Meeres- [orschung, 26 .7%95
MACKINTOSH N A (1972) Life cycle of Antarctic krdl in relation
to ~ce and water conditions Discovery Report~. 36. 1-94
MAKAROV R R (1970) Separate existence of various age groups of
Antarctic krlll Information Bulletin of the Soviet Antarctic
Expedition, 7,547-550
MAKAROV R R (1979) Size composition and conditions of existence
of Euphausta superba Dana (Crustacea Euphausiacea) in the eastern
part of the Pacific sector of the Southern Ocean Oceanology, 19,
582-585
MARR J (1962) The natural history and geography of the Antarctic
knll (Euphausta superba Dana) Discove O' Reports. 32, 37-465
M~,RSCHALL H P (1983) Sinking speed, density and size of
Euphausl ld eggs Meerefforschung. 30 .1 -9 MARSCHALL H P (1984)
Development . swimming, and feeding of early stages of krdl
Antarctw Journal o t the
United States, 19, 137-138 MARSCHALL H P (1985) Structural and
functional analyses of the feeding appendages of krlll larvae
In
Antarctic nutrient cycles and food webs. W R SIEGFmED, P R CONDY
and R M LAWS. editors, Spnnger- Verlag. Berlin, pp 346-354
PARRISH R H . C S NELSON and A BAKUN (1981) Transport mechanisms
and reproductive success of fishes in the California Current
Bzologtcal Oceanography, 1(2). 175-203
QUET1N L B and R M Ross (1984a) Depth distribution of Euphausta
superba embryos, predicted from stoking rates Manne Biology. 79,
47-53
QUETIN L B and R M Ross (1984b) School composition of the
Antarctic knll, Euphausla superba, in waters west of the Antarcuc
Peninsula. austral summer 1982 Journal of Crustacean Biology, 4.
Special Number 1, 96-1(16
OUET1N L B and R M Ross (1989) Effects of oxygen, temperature
and age on the metabolic rate o1 the embryos and early larval
stages of the Antarctic krlll Euphausla ~uperba Dana Journal of
E~pertmental Marine Biology and Ecology. 125, 4 3 4 5
RONG W (1989) Distribution and abundance of eggs and larvae of
Antarctic knll (Euphausm superba Dana) in Antarctic Peninsula area
Chinese Journal of Oceanology and Ltmnology, 7(2). 97-104
Ross R M and L B QUETIN (1983) Spawning frequency and fecundity
of the Antarctic krlll, Euphausm superba Marine Biology, 77(3).
201-205
Ross R M and L B QUETIN (1985) The effect of pressure on the
sinking rates of the embryos of the Antarctic krIll, Euphausta
superba Deep-Sea Research. 32, 79%807
Ross R M and L B QUETIN (1989) Energetic cost to develop to the
first teedlng stage of Euphau~ta superba Dana. and the effect of
delays in food avaflabdlty Journal o] E~perlmental Marme Biology
and Ecology. 133,103-127
Ross R M . L B QUETIN and M O AMSLER (1985) Euphausia superba A
prehmmary report of three areas ot investigation Antarctic Journal
of the United States, 19(5), 153-155
Ross R M , L B QUETIN and E KIRSH (1988) Effect of temperature
on developmental rimes and survival of early larval stages of
Euphausta superba Dana Journal of Experimental Marine Biology and
Ecology. 121, 55-71
SIEGEL S (1956) Nonparametrtc statistics for the behavioral
sciences. McGraw-Hill Book Co , Nev, York, xvla 312 pp
SIEGEL V (1988) A concept of seasonal variation of krlll
(Euphausmsuperba) dlstr lbunon and abundance west ot the Antarctic
Peninsula In Antatctw ocean and resources variabthty. D SAHRHAGE.
editor, pp 219-230
-
1200 E E HOFMANN et al
SHERMAN K , W SMITH, W MORSE, M BERMAN, J GREEN and L EJSYMONT
(1984) Spawning strategies of fishes m relation to circulation,
phytoplankton production, and pulses m zooplankton off the
northeastern Umted States Marine Ecology Progress Series, 18,
1-19
VORONINA N M (1974) An at tempt at a functional analysis of the
distributional range of Euphausta superba Marine Biology.
24,347-352
WITEK Z , J KALINOWSKI, A GRELOWSKI and N WOLNOMIEJSKI (1981)
Studies of aggregations of krdl Euphausm superba Meere3forschung,
28. 228-243
W1TEK Z and W KIt-fEE (1985) Larvae of the species of the genus
Euphausta (Euphauslacea, Crustacea) m southern part of the Drake
Passage and Bransfield Strait during the BIOMASS-SIBEX (December
1983- January 1984) Poh~h Polar Research. 6(1/2), 117-132
WOLNOMIEJSKI N , H CZYKIETA, R STEPNIK and J J~CKOWSKA (1982)
Biological characteristics of Euphausta ~uperba Dana m the Southern
Drake Passage and the Bransfield Strait in February-March 1981
(BIOMASS-FIBEX) Pohsh Polar Research. 3,259-271
