ICES STATUTORY MEETING 1993
CM 19931L: 57
HISTOLQGICAL STUDY ON LARVAL NUTRITIONAL CONDITION OF THE
SOUTIIWEST ArLANTIC ANCHOVY, ENGRAULIS ANCHOITA (HUBBS AND
MARJNI 1935), CAUGHT IN THREE HYDROGRAPBICALLY DIFFERING
FRONTAL SYSTEMS OF THE SOUTHWEST-ATLANTIC
by
Andreas Sieg
University ofHamburg
- Elbelabor Große Elbstraße 268
22767 Hamburg
GERMANY
ABSTRACT
During the 11 th cruise of R. V. METEOR in autumn 1989, larvae of the Southwest Atlantic
anchovy, Engraulis anchoita (HUBBS & MARINI 1935), were caught in three areas ofhigh
oceanographic diversity in the Southwest Atlantic. One objective was to compare the amount
of weil or insufficiently fed fish larvae, when they have grown up in hydrographically differing
environments. The sampies were taken in subtropical waters on the Brazilian shelf
characterised by upwelling events, in a region off Uruguay which is intluenced by the
freshwater outflow of the Rio de la Plata, and in a tidal mixing front on the Argentine shelf
around the Valdez peninsula which is intluenced by the cold Malvinas current. Hauls were
carried out with a modified MOCNESS equipment, and about 1000 larvae were selected and
prepared using standard histological techniques. The histological appearance of gut, liver,
pancreas and of musculature were selected as -condition criteria, and the height of the midgut
mucosa was measured. A simultaneous classification of organ developmental stages allowed a
stage related comparison of the individual larval nutritional condition. The histological analysis
c1assified more larvae in insufficient nutritional conditions caught off Brazil, than off Uruguay
and Argentina.
'
.
. INTRODUCTION·
.
. '
One task of the 11th cruise of R.V. METEOR in 1989 was to investigate the' biologicaloceanographic conditions~ which are assumed to ;urect variability in recruitment. of the
!
' . .
Southwest-Atlantic Anchovy, Ellgraulis anchoita (HUBDS & MARIN! 1935). Presuming fish
stock recruitment ~o depend on the survival ofthe early !ife stages (reViewed by t\1AY: 1974),
which is control!ed by development and
density of larval
food and on the frequency of
.
,
predators (BLAXTER 1969, HOUDE 1987, HUNTER 1984, NELLEN 1986), main attention
was directed towards those environmental conditions which should affect mortality during the
.
•
I
.
'
A
early life stages. As' part of the Oceml Sciellce In Relation To Livillg ResOllrces
I
•
(OSLRlSARP) program, the investigation had to be c~rried out in a marine area of high.
oceanographic diversity in order to dleck its effect o~ quality and density of phyto-' and
zooplankton, the nutritional condition ofthe fish larvae, a,nd thc small-scale distribution of fish
eggs ~d ~arvae and their predators. Tbe Southwest-Atla~tic anchovy was particularly suitable
for this kind of study because of its long spawning season and its extended spawning ~ea,
.
- .
.
i
.
covering the shelf from South Brazil down to Patagonia. The research area was subdivided
.
into . sections off Brazil, Uruguay and Argentina regarding the different hydrographical
situations (Fig. 1). As depicted in Figure 2, the Brazilian area is characterised by upwelling
events in the subtropical waters on the shelf off Sa:nt~ Marta (Area B, Fig. 1), while the
Uruguayan section (Area U, Fig. 1) is affected by the freshwater outflow from the Rio dc
la .
.
~
.
Plata, by cotd Subantarctic coastat waters, and by the tropical Brazit current, which result in
. water bodies with different salinities. Thc area of the Ärgentinean shelf Mound thc Valde~
peninsula (Area A, Fig. 1) is influenced by tidat mixing fr~nts and the cold Malvinas Current. .
~
~
or insufficiently
fed fish tarvae .emptoying
Among 'others, it was intended to etassitY well .
,
biochemical (RNAIDNA fa!ios (CLEMMESEN 1989) an~ trypsine activities (UEBERSCHÄR
1988» and histological standard techniques, aod to compare their frequency according to thc
three hydrographically differing environm~nts. For the his~otogical approach, the appearanc,e of
gut, liver. pancreas and of musculature should. be taken as condition criteria, and the .midgut
cell height should be measured simultaneousty..
.
~
Several authors discussed devetoped fish larvae to jbe. less susceptible' to short-time
i~sufficiency of food suppty (BLAXTER & EHRLIC~ 1974, RICE et al. 1987) due to
enh~ced energy reserves (POWELL &. 'CHESTER ~1985, YlN & BLAXTER 1~87).
Consequently, sarnples of developed larvae should be composed of less individuals in starving
conditions than those with many younger ones;. which have becn exposed to the same
~',
.
:
insufficient food suppty. Storage ability shoutd be more linked to organ than. to sorriatic
development or agc. Several.studics have pointed out th~t different temperature can result in
different organ developmental stages at the same length (GORBUNOVA 1981, JOHNSON
i
- 2 -
.
e
1983, JOHNSON & FELTES 1984, SIEG 1992a, 1992b). For that reason, measurements of
length d~ not seem to be reliable enough to estimate potential energy reserves. Both might be
essential for a surplus of energy, enough food to digest and enough organie ability. The larval
organ system develops ability and quality of food digestion during ontogenesis, at least with a
functional stoma~h. Hence, a histological classification of organ development should allow to
estimate on the potency of a fish larva to store energy, although this method does not give
clear infonnation about the fat reserves. Following these ideas, organ developmental stages of
larval Engraulis anchoita were classified prior to the definition ofthe nutritional condition.
Moreover, it is essential to consider organ devetopment in order to· compare larval nutritional
condition classified by measurements of the midgut cell height. In this study, the attempt was
made to employ the method described by THEILACKER & WATANABE in 1989 also for
larvae which had their midgut epithelium already folded. However, during the development
•
from a stretched to a folded structure, gut epithelium height is markedly d~creased, and later
increases again. Consequently, midgut cell height as criterion of nutritional condition can only
be compared within a defined developmental stage.
This paper concentrat~s on the· presentation of prelimin~ results of the histological
investigation of larval f,'ngralllis allchoita, since the analyses of zooplankton densities arid
biochemical conditions arc still in progress
MATERIAL & METHODS
Three sections/boxes were studied along the east coast of South America off Brazil, Uruguay
and Argentina (Fig. 1). The area off Uruguay (Box U) was visited twice beca:Use of
unfavourable weather conditions during the first visit. All sections were subdivided into
transects running from near coast to off coast. Positions of the transects were selected for
biological sampling according to the abundance of anchovy eggs and/or larvae and to different
water bodies. Biological sampling was repeated twice to four times on each transect in order to
compare sampIes from different times of the day. The fish larvae were caught using a modified
MOCNESS equipment (WIEBE el al. 1976, 1985, NELLEN el al.. 1988). Since larval
Engralliis anchoita were supposed to be concentrated in the upper 50 m of the water column
(CIECHOMSKI et al. 1986), the gear was lowered to a depth of 50 m at maximum, and the
overall time ofthe tows was 12 to 15 minutes. The larvae were immediately fixed in a buffered
6% formaldehyde seawater solution. All specimen were individually checked for sampling
damage under a microscope, and up to 45 specimen per sampIe were transferred and stored in
new fixative at 4°C until further preparation. Larvae were· coded to exclude arbitrary
- 3 -
,------------------
- - - - - - - - - -
..
evaluations by knowing the individual's identity. Standard length of the fixed larvae was
detennined to the next millimetre below. About 1000 larvae were dehydrated in alcohol series, .
.
immersed in benzyl benzoate. and finally embedded in paraffin. Sagittal seetions of 5 J.1m were
mounted and stained using a mixed PAS (Periodie Acid Schiffs) - HE (HematoxyIin~Eosin) .
i
techniqtie.
,
The histological grading covered the following organs/tissues:
Table 1:
- Liver/hepatocytes: Glycoge~
shrinkage/separations.
content,
cytoplas~,
nucIei,
cellular
•
- Pancreas: Cellular ratio zymogen/cytoplasm'
- Foregut: Cell height
-l\1idgut:
I~tracellularvacuoles,
mikrovilIi, cytoplasm, cell separations
- lIindgut: Pinocytotic vacuoles
,
- Gut content: Distinguished between. food in, the fore- and in the mid-.
/hindgut
-l\1usculature': Fibre separation, fibrils visible, prese~ce/absence of the .
connective tissue
- Cartilage of the jaws and the branchial elements: Shrunk chondrocytes
Tissues/cells were graded according to the histological cIassification by O'CONNELL (1976 &
. 1980) for larval Ellgraulis mordax, but in the present
~tudy, the histological indices' given
above were graded into five cIasses ranging from 1 (best) to 5 (worst condition). In addition to
the histological interpretations, midgut cell height was :measured following the method of
THEILACKER & \VATANABE iJresented in 1989. It was attempt~d to employ the meth?d
also ~or larvae which had the midgut epithelia already fo!ded. These midgut folds are usually
composed of cells of different height: the cells on the) top of the rolds are stret~hed and .
columnar, those ~t thc bottom between the rolds are Io~~r and cubic. Therefore, midgut cells
were chosen for ~alysis which were located in the middle of the height in each rold. Se~eral
measu~ements were take~ of the ventral row in the post~rior region of the rnidgut, the range
wa's recorded, and the mean value was calculated to be connected with the histological score.
!"
- 4 -
.....
RESULTS& DISCUSSION
Leng!h Freqllencies & Developmen!al Stages
Standard length frequency of selected and fixed larval Engralilis anchoita ranged from 3to
12 mm, detennined to thc next mm below (Figs. 3-6). Selection of larvae for histological
analysis concentrated on the smaUer individuals because of their· higher susceptibility to
starvation. Consequently, the length frequencies are biased by this selection and by the fact that
not all caught larvae were prepared for histology. Nevertheless, since no smaller larvae could
be sampled in the Brazilian section (Box B), a comparison between Box Band Box A'
(Argentina) indicates that the population off Brazil was further deveIoped than the individuals
•
offthe Valdez peninsula.
Higher temperatures influence larval metabolism, velocity of growth and organ development,
as weil as the relation betw~en organ developmental status and length. Measuring larvallength
provides onty restrictive information about the deve10pmental stage, which is essential ~o. know
in order. to estimate the development-dependent susceptibiIity to' starvation, and to allow
comparisons of midgut cell height, in general. Six deveIopmental stages of larval Ellgralilis
anchoila were defined (Table 2).
Data were calculated for each arealbox separately in order to check how different water
temperature influences developmentat stage (OS) at standard length (SL) (Figs. 7-10).,
Individuals offBrazil (Fig: 7~ T (CO) at the surface = 23.5-21.9) measu'red 4 mm below ranged
from DS III to IV (Arabian numbering in the graphs) with a median at OS 111, while at the
same standard length, larvae of the colder waters off Patagonia (Fig. 1O~ T caC) at the
surface = 16.4-12.7) fluctuated between DS II and III with a median at OS 11. OS VI was first
c1assified at a SL of 11 mm off Argentina. Some larvae caught olTBrazil reached this stage at a
SL of 6 mm already. At these orders of magnitude, bias due to the length measurement to the
next miIlimetre below can be regarded to be neglectable. Since thc slope of the regression
calculated for the medians decreased during growth, it Carl be assumed that this phenomenon is
mainly restricterl to the very early larval development: either the Brazilian larvae had a smaller
their early lifetime.
hatching length, or they had a higher velocity of organ development during
.
,
Figures 11 to 14 depict the frequencies of developmental stages determined from the prepared
material of each arealbox. According to the larger standard length measured olT Brazil (Fig. 3),
organ development was enhanced, too (Fig. 11). Both length and DS f!equencies in the
Uruguayan area (Figs. 4 & 5, 12 & 13~ Box Ul & U2) showed a trend towards higher values
, - 5 -
Table 2:
DS I: Jolk Sac Stage - jolk mass clearly detectable or remnants anterior to the
developing liver
DS 11: No jolk detectable - first small swim bladder bulge - straight digestive
tract • ventral jaw with a single layer of chondrocytes
DS III: Swim bladder anla~e with multiplied cell layers. pancreas starts cranial
extension - esophagus epithelium height increased ~ gut epithelium still unfolded ventral jaw possesses severallayers of chondrocytes
DS IV: Swim bladder further increased, shows small cavity and firstly' a cranial
tubular branch - ventral midgut mucosa folded, dorsal row starts folding - hindgut
still unfolded
DS V: Swim bladder with different types of cell layers - ventral and dorsal midgut
mucosa folded - hindgut starts folding
DS VI: Swim bladder either witl) a very thickened wall or strongly extended
because of first filling - esophagus cell height furt her increased, now with several
celllayers - pancreas extends cranial beneath thc glomerulum and caudal above the
anterior part ofthe midgut - hindgut folding completed
during the second biological sampling (U2), which ean be attributed to the time interval of 10
days between both visits. Anyway, when determining numbers of weIl fed and starving larvae
according to the sampled areas/boxes, these data have to be considered beeause of the bigger
and older larvae caught off Drazil, which have had more time to store energy reserves.
Otherwise, insuffieient feeding eonditions off Brazil eould be underestimated.
lvlidgut Ce/I Height In Relation 1'0 Histological Grading
Different cell indices were cited in order to check the relation between histological scoring and
midgut cell height. Dest histological appearance should co-occur with high midgut epithelium,
whereas a shmnken mucosa is to be expected when eellular breakdown ean be observed.· DS I
was not eonsidered, since jolk sac larvae do not starve. All individuals were taken for the first
comparison, which hepatocytes were filled with glycogen (PAS-positive. reacting substances in
the liver were defined as glycogen according to O'CONNELL & PALOMA (1981). These
larvae were in optimal
condition (Class 1), since glycogen
is regarded to be an energy
reserve
.
.
,
(LOVE 1974, O'CONNELL & PALOMA 1981, POWELL & CHESTER 1985), which ean be
completely exhausted within a single day during food deprivation (WATANABE 1985). This
was tme for the next eomparison, when liver glycogen was c1assified as "5": cell inc1usions
- 6 -
•
·were compIeteIy absent, but the individuals of this group still possessed hepatocytes as weil as
mid- and hindgut cells in good condition (refer to TabIe 1). The next group included those
larvae, which showed midgut cell separations and dark cytopIasm, and for the last check all .
larvae were cited, which scores given for hepatocyte cytoplasm and elose cell contact summed
up to 8 or higher, 'which midgut cells showed the conditions of the previous comparison, and
which hindgut cells lacked any cell incIusions (refer to Table 1). The results of these
comparisons are summarised in Figures 15 to 19. Other organs/tissues, as there were cartilage,
foregut, musculature, and pancreas, did not show such as clear histological alterations as
reported by O'CONNELL 1976 & 1980. Individuals ~ithout glycogen can not be c1assified as .
having been in emaciated condition. According to O'CONNELL & PALOMA (1981), this can
be due to normal daily periods of feeding and nonfeeding. In contrast, the histological pictures
cited for the last comparison (very right columns in the Figs. 15-19) are charaeteristic for
starving larvae with irreversible cell damages (comp.
•
O'CONNELL
1976,
1980,
THEILACKER 1978). No larvae of OS II and III of all transects were found in such bad
condition (refer to Figs. 15 & 16, bath right columns). Consequently, rio information is
available about the respective decrease of ~idgut cell height, when Engralilis anchoita larvae
pass into starvation.
There are many overlappings in the frequencies of midgut cell height and the different
histological classification when comparing OS IV to VI (Figs. 17-19). For example, in Figure
)7 (OS IV), it is not possible to reIate a larva with amidgut cell height of 12,5 J.lm to a
histological condition. Only OS V (Fig. 18) showed an obvious deviation in midgut cell height
betwecn weIl fed individuals, and the starving ones ofthe very right column.
There are two explanations for this finding: either larvae attached to bad histological co~dition
had not starved long enough to give elearer deviations from "normal" midgut cell heights, or
they had starved until cellular damage, but the midgut cell heights did not vary more than those
of fed individuals of the same developmentaI stage. Both could be true: THEILACKER &
WATANABE (1989) presentcd measurements of midgut cell height in larval E'ng,.alllis '
mordax, ranging from about 18.7
do~
to 7.3 J.lffi (average=:-:I1.8 J.lm) for individuaIs of
SL<4 mm, which had starved for three days at 15.5 °C, Atllarvae had a straight, unfolded gut
epithelium. Referring to ~he material of this study (Figs. 15 & 16), when citing liver glycogen
classified as "1" (comp. tempo range, same DS, same SL), the larvae between the 10wer and
upper quartiles (indicated by the frame) were out ofthe range ofthe average value reported by
the authors. Consequently, even starved Ellgralilis anchoita larvae of these stages can be
expected to have a lower mucosa cell height.· On th~ other hand, the histological features taken ..
for both right comparisons in the graphs of Figurcs 15 to 19 are based on O'CONNELL's in
Si/li study on Ellgraulis mordax in 1980, in which he attributed these findings to emaciated
larvae. From this referencc, thc respective larvae of this study should have starved, but
~
7 -
measurements of their midgut cell height did not yield, in frequency distributions, ,,,hich are
sufficiently divergent from those of fed individuals. In conclusion, further calibration work is
, needed in order to obtain more reliable information on the definite relation between inidgut cell
height and larval histologY at different temperatures, levels of starvation and developmental
I
stage.
'
Nlitritiollal COllditiOl1S 011 1'/1(: Tral1sects Sampfet!
Because of the findirigs given above,' the measurements of midgut cell height were not
considered for the classification of different nutritionalconditions on the transects sampled.
Figures 20 to 28 depict the frequencies oflarval.E.l1graulis allchoita on each transect according
to their different nutritional condition. Larvae of Grade 1 had food in the mid- or hindgut.
Some individuals had food in the esophagus, however, this was ignored in the evaluation,
because it could result from both normal feeding as weil as stress-induced food intakc during
net time. Grade 2 in these graphs ineludes alllarvae who had no food in thc midgut, but which
hepatocytes werc full ofglycogen (deflned as Class 1, refer to Chap. Material & A1cthods) or
which had glycogen remnants (defined as Classes 2-4). Individuals wcre classified to the worst
Grade 4; when their organs/tissues showed alterations as described for a starving condition in
the graphs of the Figures 15-19: the scores of hepatocyte
cytoplasm and e10se cell contact
,
,
summed up to 8 or higher, the midgut cells showed
dar~
cytoplasm and cell separations, and
the hindgut cells lacked any cell inclusions (refer to 1.'able 1). All larvae who had better
histological conditions than the individuals of Grade 4 but did not have cell inclusions in the
.
'
liver were attached to Grade 3.
;
. 1
In general, mostofthe E;,groulis clIIchoitalarvae seemed 'to prefer fecding during daylight, as
indicated by the empty rectangles above the columns ineluding
larvae graded "1". This is in
,
.
accordance with data of SANCHEZ el al (1991), who described a daily feeding pattern over a
14 h period started after sunrise. Glycogen was present in the liver during day and night.
~
According to WATANABE (1985), who found at 15-20°C a decrease of liver glycogen in
goby larvae already 5h, and a total reduce one day after food deprivation, larval Ellgralilis
Clnchoila should' show a decreased glycogen level at dawn duc to non-feeding during night.
This would be in line with O'CONNELL & PALOMA 1981, who found daily circles in the
hepatocyte's glycogen quality. Grade 2 in the' Figures 20 to 28 ineludcs larval livers which
glycogen contents scored from Class 1 to 4, and, therefore, it includes not only the best
glycogen dass" 1". Nevertheless, many larvae caught in the early morning had empty guts, but
best glycogen' level. In conclusion, in spite of daily fluctuations in the hepatocyte glycogen
content, a complete consumption should not occur when exposed to normal food conditions.
- 8 -
•
, First measurements ofRNNDNA ratios of Ellgraulis allcllOita larvae caught at station 995 off
Brazil revealed about 80 % of thc sampled individuals with SL 5-10 rnm to be in bad
nutritional condition (CLEMMESEN-BOCKELMANN, pers. comment): This seems to match:
the histological findings, which classified 81% of the respective larvae (SL=4-12 mm) i~ a'
suboptimal condition (refer to Fig. 21). This percentage consists ofindividuals with reversible
suboptimal cell conditions (Grade 3) and .of those without having had the opportunity to
survive due to cellulär collapse (Grade 4). From these findings it cari be concluded that
determining RNNDNA ratios is not reliable enough without further calibration work to detect
fish larvae with irreversible celldamages. The latter have passed the "point of no return", and
contribute to larval mortality rates due to starvation.
Figures 20 to 28 were based on individuals of all developmental stages (11-V~) to give a first
•
general survey ofthe feeding situation on the transects sampled. In the future, it is intended to
evaluate the histological data in relation to developmental stages to allow comparisons of the
frequencies of different nutritio,nal condition at each station, on each transect, and finally in
each area investigated. Unfortunately, no larvae in early stages were caught off Brazil, where
the food supply seems to have been suboptimal. They would have made possible, a comparison
of the nutritional condition of early post jolk sac stages, which are assumed to be most
susceptible to starvation (cornp. RICE el ai. 1987). However, THORISSON (1992) discussed
, metamorphosis as another critical stage to afTect the energy budget in the Iife history of marine
fish larvae. This idea and the histological 'appearance of the developed larvae of the Brazilian
section stress the necd of studies on nutritional condition also for later stages than those of the
eluly post jolk sac period..
HUBOLD (1980) measured tower ~alues for chlorophyll-a concentrations and zooplankton
densities in' t~ose areas off south Brazil and the Rio de la Plata estuary, which were under
dominant intluence of the' tropicat w~ters. ,In contrast, higher concentrations of both
paramet'ers we~e ,foundi,n the Subantarctic water. This may indicate a lower primary
production off Brazil as compared to the Argentine section. ALHEIT el ai. (1991), however,
'presented ~ata ofthe ch,lor~phyII-ameasurementsdeteimined on this cruise ofR.V. METEOR,
which showed quite similar values off BraziJ, Uruguay and Argentina, tluctuating between 1.5
to 5 mg/m3 • Since the zooplankton analysis and the measurement ofRNNDNA ratios are stilI
in progress, a final valuation of the ecological situation at time of the cruise can be given not
before the end ofthis year.
- 9 -
REFERENCES :
ALHEIT, 1., 1. OE CIECHOMSKI, L. DJURFELDT, C. EBEL, M. EHRLICH, 1.C. ELGUE,
. G. MANTERO, Y. MATSUURA., H. MIANZAN, W. NELLEN, C. ODEBRECHT.
F. RAMlREZ, R. SANCHEZ, G. SHAFFER & M.D. VINAS. 1991. SARP studies
. on Southwest Atlantic anehovy, Ellgralilis allchaita. off Argentina, Uruguay and
, Brazil. ICES C.M. 1991/ L: 48: 32 pp.
BLAXTER, lH.S. 1969. Development: eggs and
larvae~
111: Hoar, W. S. and D. l Randall
(eds.): Fish Physiology 3. Academic Press. New York, San Franciseo, London: 177-
252.
BLAXTER, lH.S. & K.F. EHRLICH. 1974. Changes in behaviour during starvation of
herring and plaice larvae. 111: Blaxter, -1. H. S.. (ed.): The early life history of fish.
Springer Verlag, Berlin, Heidelberg, New York: 575-588.
CLEM.MESEN; C. 1989. RNAIDNA ratios of laboratory-rearedand wild herring larvae
determined with a highly sensitive fluorescence method. l Fish. Biol. 35 (Supplement
A): 331-333.
CIECHOMSKI, 1.D. OE, RP. SANCHEZ &
c.A.
Lasta. 1986. Evaluacion de la biomasa de
adultos desovantes, distribucion vertieal y variacion euantitativa de la intensidad de
los desves de la anchoita (l.. .l1graulis anchoita) durante la primavera de 1982. Rev.
luvest. Des. Pesq. 5: 30-48.
GORBUNOV A. N.N. 1981. Larvae ofthe genus Villciguerria (Gonostomatidae) with keys. J.
Ichthyol. 21 (4): 138-141.
HOUDE, E.D. 1987. Fish early lire dynamics and recruitmentvariability. Ameriean Fisheries
Society Symposium 2: 17-29
HUBOLD, G. 1980. Hydrography and plankton off southern Brazil and Rio de La Plata,
. August· November 1977. Atlantica, Rio Grande 4: 1-22.
HUNTER, lR. 1984. Interferences regarding predation on the early life stages of eod and
other fishes.ll1: Daht, E., D. S. Danielssen, E. Moksness and P. Solemdal (eds.): The
propagation of eod, Gadus morhua, Part 1. Institute of Marine Research. F10devigen
Biologieal Station: 533-562.
- 10 -
•
.
,
.
, JOHNSON, RK. 1983. Variation in Vinciguerria nimbaria (Pisces, Photiehthyidae) with
, eomments on thestatus ofthe Red Sea population. Bull. Oeean. Fish.?: 243-255.
JOHNSON, RK., & R.M. FELTES. 1984. A new species of Vinciguerria (Salmoniformes:
Photichthyidae) from' the Red Sea' and Gulf Of Aqaba, with comnients on the
depauperaey ofthe Red Sea mesopelagic fish fauna. Fieldiana: ZooI., n. S.: 22: 35.
LOVE, R.M.1974. The chemical biology of fishes. Academic Press, London, New York:
547 pp.
MAY, RC. 1974. Larval mortality in marine fishes and the eritieal period coneept. jll: Blaxter,
1. H. S. (ed.): The early life history of fish. Springer Verlag, Berlin, Heidelberg, New
•
York: 3-19.
NELLEN, W. 1986. A hypothesis on the feeundity ofbony fish. Meeresforsch. 31 (1): 75-89.
.
,
NELLEN, W.,' D. SCHNACK & B. ZEITZSCHEL. 1988. Expeditionsbericht über die
METEOR-Reise 5, Abschnitt ,3. Berichte aus' dem Zentrum rur Meeres- und
Klimaforschung der Universität Hamburg 0: 165 pp.
O'CONNELL, C.P. 1976. Histological eriteria for diagnosing the starving condition in early
post yolk sac larvae of the northem anchovy, El1graulis mordax Girard. J. Exp. Mar.
Bio1. Ecot. 25: 285-312.
O'CONNELL, C.P. 1980. Pereentage ofstarving northern anehovy, Ellgralilis mordax, larvae
in the sea as estimated by histologieal methods. Fish. Bull. 78 (2): 475-489.
O'CONNELL, C.P.,& P.A PALOMA 1981.
Histoch~miea~
indications of liver glycogen in '
sampIes of emaciated and robust larvae of the northern anehovy, Ellgraulis mordax.
Fish. Bull. 79 (4): 806-812.
POWELL, AB., & 1\.1. CHESTER. 1985. Morphometric indices ofnutritional condition and
Trans. Anl. Fish. Soe. //4: 338-347. ' ,
sensitivity to starvation ofspot larvae.
,
,
RICE, J.A, L.B. CROWDER & F.R.BINKOWSKI. 1987. Evaluating potential sources of
mortality for larval bloater (Coregolllls hoyi): Starvation, and vulnerability to
predation. Can. 1. Fish. Aquat. Sei. 44: 467-472.
- 11 -
SANCHEZ, R.P., J.D. OE CIECHOMSKI, C.A. LASTA & R.A. Guerrero. 1991. A drift
study of vertical distribution and mortality of E'ngraulis anchoita eggs and larvae.
ICES C.M. 1991/ L: 22: 24 pp.
SIEG, A. 1992a. Histological study of organogenesis in young stages of the mesopelagic fish
Vincigllerria (Photichthyidae). Bull. Mar. Sei. 50 (1): 97-107.
SIEG, A.
1992b. A histological study on the nutritional condition of ,larval and
metamorphosing fishes of the genus Vincigllerria (Photichthyidae) sampled in two
contrasting environments. J. Appl. Ichthyol. 8: 154-163.
THEILACKER, G.H. 1978. Effect of starvation on, the histological and morphological
characteristics of jack mackerei, Trachllrlls symmetriclIs, larvae. Fish. Bull. 76 (2):
403-414.
THEILACKER, G.H. 1986. Starvation-induced mortality of young sea-caught jack mackerei,
Trachllrlls symmelriclls, determined with histologieal arid morphological methods.
Fish. Bull. fi./(I): 1-17.
THEILACKER, G.H., & Y. WATANABE. 1989. Midgut cell height defines nutritional status
of laboratory raised larval northem anehovy, Engralliis 11lordax. Fish. Bull. fi7: 457,469.
THORISSON, K. 1992. Is metamorphosis a critieal stage in the early life of marine fishes?
ICES C.M. 1992/ L: 37: 16 pp.
UERERSCHÄR, B.F.R. 1988. Determination ofthe nutritional condition ofindividual marine
fish larvae by analysing their proteolytic enzyme aetivities with a highly sensitive
fluorescence technique. Meeresforseh. 32: 144-154..
WATANABE, Y. 1985. Histological changes in the liver and intestine of freshwater goby
larvae during short-term starvation. Bull. Jap. Soc. Sei. Fish. 51 (5): 707-709.
WIEBE, P.H., K.H. BURT, S.H. BOYD &
A.W. MORTON.
1976. A multiple
openingicIosing net and environplcntal sensing system for sampling zooplankton. J.
Mar. Res. 34 (3): 313-326.
- 12 -
•
r - - - - - - - - ------
WIEBE, P.H., A.W. MORTON, A.M. BRADLEY, R.H. BACKUS, J.E. CRADDOCK, V.
BARBER, TJ. COWLES & G.R. FLIERL. 1985. New development in the
MOCNESS, an apparatus for sampling zooplankton and micronekton. Mar. Bio!. 87:
313-323.
YIN, M.C., & lH.S. BLAXTER. 1987. Feeding ability and survival during starvation of
marine fish larvae reared in the laboratory. 1. Exp. Mar. Biot. Ecol. 105: 77-83 .
•
- 13 -
0,
FI ori .nopons,"
30·s
8
•
Rio de I. Plata
BOX U
Peninsula ..
Valde: .,..;;
,,
~
.
~,
BOX A
200m
.•.•'
50·w
60·w
Fig. 1: Location of sampling areas/boxes and transects during I\1ETEOR cruise 1113. Box B =
Box offBrazil, Box U = Box offUruguay, Box A = Box of Argentina.
-1-
{al
VERVWEAK
WIND
O
WEAK
STAOt..lG
WIND
WIND
~~""",,-,,'""'-J"~~_ _SSU~RiF1FA~CEfJE~KMA;.;ANN'"";;;;;;';~
,
lAYEA
A
S
T
{bI
•
(cl
ZONEOF
STABILIT'Y
-
Fig. 2: Sehematie diagrammes on environmental situations in the three different areas sampled
during METEOR eruise 11/3: a) The wind-driven eoastal upwelling regime off Cape Santa
Marta Grande, Brazil (Box B); b) The estuarine and shelf-break fronts off Uruguay (Box U);
c) The tidal mixing front off Peninsula Valdez, Argentina (Box A). Reproduced from BAKUN
& Parrish. 1991. lCES J. mar. Sei. 48.
-ll-
Absolute Frequenc
Absolute Frequency
140.------.:-..:-----------T.,----.
N ~ 367
DBox D
140,-----.:-..:-----------,--------,
J :lO
l~O
100
100
N::. 11:2.
- - -
so
so
60
60
~
....-
40
40
n~-
:0
0
6
S
9
JO
11
20
12
o '---'--,-L.L-4-,-,-.w.-6...L.J:=-8--Q--'-0-J-I--J:~--JJ--.J
13
Standard Lc:ngul (Olm below)
Standurd l.-rngth (mm betaw)
Fig. 4: Length frequency of prepared EngraulJi...
anchoita larvae caught in the Uruguayan a r .
(1. run).
Fig. 3: Length frcquency of prepared EngrauJis
anchoita larvae caught in the Brazilian area.
l20
Absolute F~que.nc)'
HO , - - - - - - - - - - - - - - - - . - - - - - - - - ,
OBox A
·47J
120
100
100
Ahsolu.tt".
I'Tt"qUCl1C)"
1~O . - - - - - - -
I
DBox 112
N • 60
r-
...-
e
SO
60
c--
c--
40
20
20 .
o L-_ _-=::LJ.-..l...l-.l...L-..l...l-.l...L-..l..==-9
JO
11
...J
12
.13
0
13
4
6
7
n~~
9
10
11
12
13
Standard Lenglh (mOl below)
Standard Length (mm below)
Fig. 6: Length frequency of prepared Engraulis
anchoita larvae caught in the Argentine area.
Fig. 5: Length frequency of prepared EngrauJis
anchoita larvae caught in thc Uruguayan area
(2. run).
- Ul -
Sampling Area Off Brazil (Surface Temp. Range: 23.5 - 21.9 Oe)
DevelopmentaI Stage
1Or--------------..;..--------...,
+
,,>
••..••••••
Y = a b log x
a " -0.4
b '="6.4 - ,",
r2:' 0,999 '
9
8
7
N = 367
- - - . - - • - ,- ' • -'- '- - - - -, " - -
6
5
4 -3
2
n
...
=3
n ;; 46
n I; 80
D t
5
6
4
3
•
.. . ,. ..
.. - .
"..
84
7
-
-
n • 80
8
.' -
...
n
;11
41
'D.
9
n
24
10
=6
n*3
11
12
13
Standard Length (mm below)
Fig. 7: Sampling area off Brazi1: Frcqueney distributions of different devclopmental stages within
the standard length c1asses. Tbe horizontal line is the median, the boxes indicate the lower and
upper quartiles, and the vertieal Iines cxtend out to the extremes. Tbe dots give values occuring
more than 1.5 times away from the intcrquartilc range. The respective "n" are given above the
x-axis. The logarithmic regression was calculatcd for the mcdians.
1. Sampling Area Off Uruguay (Surface Temp. Range: 19.1 - 18.0 0c)
Developmental Stage
•
10,---------------------9
Y
a
= a + b log x
= "1,636
N
8 -, \j =- 7~4 - - ,
7
r2 = 0,9986
= 112
'0 ----. -... --.. :- .
6 -- - ' ,
5
4
3
2 .,- ,
- - - - - - . - - -
-
-'
-
...
..'
~
. . n~27 n-43 n~29 'n*10 Da3
oL..._I.----.JI-----l---l----l.---l..---l..---'---J---L...--J
2
3
4
5
6
7
8
9
10
11
12
13
Standard Length (mm below)
Fig. 8: Sampling area off Uruguay - 1. run: Frequency distributions of different developmental
stages within the standard length classes. The horizontal line is the median, the boxes indicate the
lower and upper quartiles, and tbe vertical lines extend out to the extremes. Tbe dots give values
occuring more than 1.5 times away from the interquartilc range. Thc respectivc "n" are given
ahove the x-axis. Tbe logarithmic regression was calculated for the mcdians.
- IV-
2. Sampling Area Off Uruguay (Surface Temp. Range: 20.6 - 18.5 0c)
Devc1oproental Stage
10.....-----------------------,
. .
.
:.
Y =at b log x
a = -3,077
8 -b"= "9,09-·
7
r2 = 0,999
9
6
N
= 60
- - - - - -
5
3
2
I .
...
D-3
0
3
2
12
.Q •
4
5
.
D.
n " 16
6
13.
n. 9
n. 5
: n-3
8
9
10
7
11
12
13
Standard Length (rom below)
Fig. 9: Sampling area off Uruguay • 2. run: Frequency distributions of different developmental
stages within the standard length classes. The horizontal line is the median, the boxes indicate the
lower and upper quartiles, and the vertical lines cxtend out to the extremes. The dots give values
occuring more than 1.5 times away from the interquartile range. Thc respective "n" are given
ahovc the x-axis. Tbc logarithmic regression was calculated for the mcdians.
Sampling Area Off Argentina (Surface Temp. Range: 16.4 - 12.7 Oe)
Developmental Stage
10.....---------------------...,
Y = a + b log x
a = -1,68 .
8 - b' =' 6;87 - 7 .. r2 =0,998
9·
N
= 471
6 _.
5
4
3
2 --
~r.,r·- -
-. -
.
.
-
-
-
-
-
-
•
1
n _ 50 n = 101 n' 129 n = 101 n' 50 n' 23 n = 10
O'---'------'---'-----"----'----'----'-----'---'--.....L--'----'
6
9
4
5
10
2
J
7
8
11
12
13
Standard Lcngth (mm below)
Fig. 10: Sampling area off Argentina: Frequcncy distributions of different developmental stages
within the standard length c1asses, The horizontal line is the median, thc hoxes indicate the lower
and upper quartiles, and the vertical lines cxtend out to the extremes. The dots give values
occuring more than 1.5 times away from the interquartile range. The respective "n" are given
above the x-axis. Tbc logarithmic regression was calculated for the medians.
- V·
•
Absolute Frequeney
Absolute Fre.queoC)'
250..---------------,.---__,
250 . . - - - - - - - - - - - - - - - . , - - - - - ,
DBox 1Jl
= 112
OBoxB
N = 361
200
200
--
]50
100
150
100
,
50
50
3
4
o ~_:__........._=_-.u~:___"W--...J...c===:=5=---6
5
Devc10pmental Stage
Development.ll
_ . 11: Frequency of larval developmental stages
of prepared EngrauJis anchoita larvae caught in
the Brazj] ian area.
Ab$ointt~
:
r
Fig. 12: Frequency of larval developmental tages
of prepared Engraulis anchoita larvae caught in
the Uruguayan area (1. run).
Absolute Frequency
Fn-quency
~'vr----------------r--__,
N
--7---'
Sta~e
250r--------------,...----,
OBoJl[U2
60
OBoxA
= 471
200
)
150
100
100
50
50
Developmental
Developmenlal
Sta~e
Fig. 13: Frequency of larval developmental tages
of prepared Engraulis anchoita larvae caught in
thc Uruguyan arca (2. run).
- VI-
Sta~
Fig. 14: Frequency of larval developmental stages
of prcpared Engraulis anchoita larvae caught in
thc Argentine arca.
Relation: Histological Classes - Midgut Cell Height
Developmental Stage 11
Midgut Cell Height (Jtm)
,
- -. -. - i
,..
_ t
25
. J
.&
. . . . .•
.
20
15
N" 64
~.....
;.
..
.;.
t
'"
,.,'
..
'
I ...
10
.
• I.
I
-' I .
5
na 41'
Liver
Glyc:ogea
I
I
1
.. n" 2.'l
Liver
Glycopea
M.O Cell Sep.
&: CP Coodit
I
Fig. 15: Relation between thc histological nutritional cJasses and the rcspcctive midgut cell
hcights for all larvae of thc dcvelopmental stage II. Tbc horizontal line is the median, thc boxes
indicatc the Jower and upper quartilcs, and the vcrtical lines extcnd out to the extremes. The dots
give unusual valucs occuring more than 1.5 times away from the interquartile range. Thc
rcspectivc "n" are givcn above the x-axis.
Relation: Histo]ogica] Classes - Midgut Cell Height
Developmental Stage III
Midgut Cell Height (p.m)
-
.1 •
30 -:-
)
I.
1
1
25
-I -
-l-
I.
I
........--if-
-I -
-I -
.1
I
I
1
I
I
. I
,
10
,
)
I
,
I
-
1
1
1
5
0
n= 116
n =40
LiVCf
Liver
Cla.u.:
-I"
. I· .
-I I
I
4~'&
Glycos en
Olycogea
= 156 ..
I
I
20
N
!-s
1 MO Ccli Sep.
I &: ~ Coadit
I Hep. eo.d. & MG C.II Sep. & MG
I Cp C'oacl. a: HO Cell InclullioDI
Fig. 16: Relation between the histological nutritional c1asses and the respective midgut cell
heights for all larvae of the developmentaJ stage BI. The horizontal line is the median, the hoxes
indicate the Jower and upper quartiles, and the verticaJ lines extend out to the extremes. Tbe dots
give unusual valucs occuring more than 1.5 times away from the intcrquartile range. The
respective "n" are given above the x-axis.
- VIl -
•
Relation: HistoIogicaI Classes - Midgut Cell Height
Developmental Stage IV
Midgut Cell Height (Jlm)
30
. ,-:-
- -' - -
.',',.·
:
_
-
t
I'
I
/"
t .
~
J _
.. I.
N'" 116,
,
·
_.~...:. f~r+,·i·d'-;b.'
20 --,
15
···.
·
-
25
-l-
-1 -
-
10 : - -
-,V.
-
- '- .•
•
I
I
.' --=r- . ~ . ---c:=:r
:~-
..
~
-
-
-, '-t ..
-, ..
- 1-', -
~
5 --
,
..
-
.:
':
n* 79
OL-.o----'----'-.l.____'.....;..____'-.L._-l-'-____'.l......____'+---J____'_ _~____'____'___I
elast.:
e,Fig.
14-5.3-$
Liver
Uver
Olycopea
Glycagea
.
I
MG CeU Sep.
1
&: CF Ooadit.
SUt."4-$A:3.5."-5
I Hep. CODd. &: MO Cell Sep. cl MG
I ('P Coad. a: HO C.e-II laduaioas
17: Relation betweell the histological nutritional c1asses and the rcspective miogut ceH
heights for a11 larvae of the developmental stage IV. Tbc horizontal linc is thc median, thc boxcs
indicate the lower and upper quartilcs, and thc vcrtical lines extend out to the extremes. Thc oots
givc unusual values occuring more than 1.5 times away from the interquartile range. The
respective "n" are givcn above thc x-axis.
Relation: HistologicaI Classes - Midgut CeII Height
DeveIopmentaI Stage V
Midgut ('-eU lIeight (I/rn)
•
;
JO
- f
25 , •
•
20
·:
.
.
-
t
. No: 172
·
-' .
•
15
10
_
5
,0
Ga:
81
t
~.
_
..
··-
-~
n
-= 61
. 1 ..
L..-;~-'--.l.-,-~_...L;..~4---_J,;",.;.-'-F-;.,..".J.-'--'--'-;;.J....";";"-'-:...;,.,;,.J
I
4-5&3.'
Liver:
MG c.lI Sep.
CP Coadit.
Clan:
Liver
OIYCOSell
I
GlycOSeD
t"
; SUt,a4-'&.3-'&4-$
• Hep. Coad. ci MG Cell Sep. &: MG
: C'P Coad. &; HG Ceoll Induaioaa
Fig. 18: Relation between the histological nutritional c1asses and the respective midgut cell
heights for all larvae of the developmental stage V. The horizontal line is the median, the boxes
indicate the lower and upper quartiles, and the vertical lines extend out to the extremes. Tbe dots
give unusual values occuring more than 1.5 times away from the interquartile range. The
respective "n" are given above the x-axis.
- VIlI -
,---------
-
--
----
Relation: Histological Classes - Midgut Cell Height
Developmental Stage VI
Midgut Cell lIeight (lLm)
,. ..
,..
-, ... ...
, ..
...
"
....
~
...
',,' :
.~ ~.~._.~
*.. ,.#
... , .
~~ ~~: ~D·~
','
25
".
I
t
20
.
:
IS --
:
.
---
.
-
-
I
..
10 --
'
I'
..
•
:
•
_:.
,.
..:,,'
I
•
•.
;.
I
I
.. \
n=
75
'
I
_ _....J
OL-.~_.....L..~-+-..;,.."....I....-~~----L~...-..~J....~~...........J~
CIaSI:
Livf'f
Liver
OJyc:ogea
OIYCOIeG
I
.. - 5 &: 3 - .$
I SU I 8 .t: .. - S &. ,
I
MO <'eil Sep.
.t. CP Coadit
I Hep. Coad,
C'P Omd.
*
5& 4-S
MG ('.eil Sep. &. MG
HO C1e1l Iacluiou
cl;
a:
Fig. 19 Relation between the histological nutritional classes and the respective midgut ceH heights
for a11 larvae of the developmental stage VI. The horizontal Hne is the median, the boxes indicate
the lower and upper quartiles, and thc vcrtical Hnes extend out to thc extremes. The dots give
unusual values occuring more than 1.5 times away from the intcrquartile range. The respective "n"
are given above the x-axis.
Nutritional Conditions Of Larval EngrauJis anchoita
Transect Brazil t
]00%
75%
.---------------.iiiiiioiioo,-----..,
_. - .' - - . -
~
.'. - - - - . • .
50% .:., •• - ' . - •
25%
-. - - • - •••
Station/Numher
, _ Grade 4
0
oe Larvae/Day
Grade 3
0
Grade 2
Time (UTC)
111 Grade
1
I
Fig. 20: Nutritional conditions of larval Engraulis ,1nchoit:lcaught on transect Brazil 1. Histological
grades are defined as foHows: Grade t = Food in the mid- or hindgut; Grade 2 = No food in the
gut. hut glycogcn in the hcpatocytcs; Grade ~ = No food & no glycogcn, hut hetter histology than of
Grade 4; Grade 4
= Sum
of scoring of hepatocyte cytoplasm and cell contact
~
8, midgut cells dark
& separated, hindgut cells without inclusions. X-axis givcs the station sampled, number of analysed
larvae. and time of the haul. Stations were grouped according to same position and daytimc. Day &
night is indicatcd by the respective symboles above thc columns. NC
- IX-
= near coast. OC = off coast.
e
Nutritional Conditions Of Larval Engraulis ancboita
Transect Brazil 2
75%
- - - - - -
50%
- - - - - -
25%
~
t...::=.J
'-:.:.!:-,----::!o,,,----:±"';----L..:~~_±_!-L-L_:!:__~~,_w-
90S
t\..
l~
eH)
'7.
..~,'-;,..: -'-,-'-"'::!::--:::
....,..--:!
...:-'
",,1l n-)4
11.25 11:U
I oe 1
Station/Numher Of Larvae/Day Time (UTC)
I_Grade 4 DOrade 3 DOrade 2 _Grade 1
I
-Pig. 21: Nutritional conditions of larval Engraulis ancboita caught on transect Brazil 2. Histological
grades refer to Pig. 20. X-axis gives the station sampled, number of analysed larvae, and time of
the haul. Stations were grouped according to same position and daytime. Day & night is indicated
by the respective symboles above the columns. NC = near coast, OC = off coast.
Nutritional Conditions Of Larval EngrauJis ancboita
Transect Brazil 3
75%
--
50% -- - -
25%
Q
~
L."O'::-:'O-:,+.:o"c--:::!:"••,.----J---""~,,-:,~o..-:,O=:
.. ,---'----,"~"-:l::-;,.";-I::::.,,,.----J"""""-:::IO\-:-II
~l.~U-'-:,=",:-'-'-1.:::I.\-:-1I~"~"-'-:I*.-'O~
n-"Z3
,,_13
".,2"
n.'"
",,21
n_'1'
02~9
10;20
H~j
~:<B
12:36
1,7:02
Ioe I
Slalion/Number Of Larvae/Day Time (UTC')
\_ Grade 4 D ('.rade 3 D ('.rade 2 • Grade 1
Fig. 22:
I
utritional conditions of larval Engraulis anchoita caught on transect Brazil 3.
grades refer to Fig. 20. X-axis gives the station sampled, number of analysed larvae,
the haul. Stations were grouped according to same position and daytime. Day & night
by the respective symboles above the columns. NC = near coast, OC =
- X-
Histological
and time of
is indicated
off coast.
Fig. 23: Nutritional conditions of larval Engraulis anchoita caught on transect Uruguay 1.
Histological grades refer to Fig. 20. X-axis gives the station sampled, number of analysed larvae,
and time of the haul. Stations were grouped according to same position and daytime. Day & night
is indicated by the respective symboles above the columns. NC = near coast, OC = off coast.
Nutritional Conditions Of Larval Engraulis anchoita
Transect Uruguay 3
100% , - - - -
o
000.
•
-------------,--,
75%
50%
25%
~1u.:'1l1l1l·9!J~.
".
I_
I~ I : " 11M 1m.
~l • ..,
1):»1,...1020:1)22.
.yu.
a." •.
Station
Grade 4
I
lQt.s IlfiCl1lU
I~
1061 1066 H91 UM
HOl Uf'2
-"
OIJl
umber Of Larvae/Day Time (UTC')
0
Grade J
0
Grade 2 _ Grade 1
I
Fig. 24: Nutritional conditions of larval Engraulis anchoita caught on transect Uruguay 3 (both
run ). Histological grades refer to Fig. 20. X-axis givcs the station saropled, nurober of analysed
larvae, and time of the hau!. Stations were grouped according to same position and daytime. Day
& night is indicated by the respective symboles above the columns. NC = near coast, oe = off
coast.
- XI-
Nutritional Conditions Of Larval EngrauJis anchoita
Transect Argentina 1
100% 0 0
00
r------------......,,;;.-.,:.=.,
75%
50%
25%
~
1111'
111411\0
U14
1120
... u
1l2.1
1lI's
• •)<1
Cla)5
Clal2
12:.12
lS:n
lLI2
02~n
•• )5
;1)
I_
e
t111
.~
01:54
UI)
litt
loe I
Statioll/Number Of LarvaelDay Time (UTC)
Grade 4
0
Grade 3
0
Grade 2 • Grade 1
I
Fig. 25:
utritional conditions of larval Engraulis anchoita caught on transect Argentina l.
Histological grades refer to Fig. 20. X-axis gives thc station sampled, number of analysed larvae,
and time of the hau!. Stations were grouped according to same position and daytime. Day & night
is indicated by the respective symboles above tbe columns. Ne = near coast, oe = off coast.
Nutritional Conditions Of Larval EngrauJis anchoita
Transect Argentina 2
o
75%
50%
25%
o
~"""""~"""""'--::'lln::---::'Il:""----'----'"":":,,:,.,,...i--:':1l'::-"...1...-J...-L"'7.,,'::-,•.........",,'::-,,-"--~---:~
a.U
11:UI
• .t
Ioe I
:ll.
Station/Numbcr Of Larvae/Day Time (UTC)
\_ Grade 4
0
Grade 3
0
Grade 2 • Grade 1
I
utritional conditions of larval Engraulis anchoita caught on transect Argentina 2.
grades refer to Fig. 20. X-axis gives the station sampled, number of analysed larvae,
and time of thc haul. Stations were grouped according to same position and daytime. Day & night
is indicated by the respective symboles above tbe columns. Ne = near coast, oe = off coast.
Fig. 26:
His~o'ogical
- XII-
Nutritional Conditions Of Larval Engraulis anchoita
Transect Argentina 3
75%
50%
- - - - - -
25%
1I.9
!lS4
1I~)1
1l~22
"':49
17;)1
Station/Numbn Of LarvaelDay Time (UTC)
/_ Grade 4
0
Grade 3
0
Grade 2 • Grade 1
I
Fig. 27: Nutritional conditions of larval Engrcwlis anchoita caught on transect Argentina 3.
Histological grades refer to Fig. 20. X-axis gives the station sampled, number of analysed larvac,
and time of tbe hau!. Stations were grouped according to same position and daytime. Day & night
is indicated by the respective symboles abovc tbe columns. Ne = near coast, OC = off coast.
Nutritional Conditions Of Larval EngrauJis anchoita
Transcct Argentina 4
100%
,.----,...,-=- 0
---------"oiiiiiiiio
75%
50%
-- - - -
25%
I I
Ne
1Ie3
1170
1171
1166
111)
,,·)1
1l:21
1I~)li
1168
,,_,,7
':)1:53
10:46
06~
16:(11
Station/Number Of Larvae/Day Time (UTC)
I_
Grade 4
0
Grade 3
0
Grade 2 _ Grade 1
I
Fig. 28: Nutritional conditions of larval Engraulis anchoita caught on transect Argentina 4.
Histological grades refer to Fig. 20. X-axis gives tbe station sampled, number of analysed larvae,
and time of the bau!. Stations were grouped according to same position and daytime. Day & night
is indicated by the respective symboles above the columns. NC = near coast, OC = off coast.
- XIII -