.~

ll)

iu,..,"

•.'-J' 1I,. .1~ .~II.

! ..

c. N. E. X. O.

C. O. B.

rI

•

A GROWTH MODEL FOR NEPHROPS NORVEGICUS

FROM BISCAY BAY IN FUNCTION OF

PERIODICITY OF M(O~~~~

by Blbllop ~

. .\..",.~;/" .

Gerard CONM{ -~~;J..~.CM 1975/K : 10

'.

•

SUMMARY.

Size frequency distributions, proportion of soft individuals and

proportion of berried females were sampled in Nephrops commercial landings

from Biscay Bay betweenApril 1970 and April 1971. Age classes 11 to V yield

most of captures. During fishable life span male and female Norway lobsters

would mault in Spring ; in Autumn all males and age 11 females would mault again.

Females become mature at end of age 11 and spawn there on every year, skipping

Autumn mault while incubating. Von Bertalanffy curves were fitted to data of

modal or mean carapace sizes referred to age at Spring and Autumn moult. . 116 91(1 -0.077(t+0.438» f I dregress10n equat10ns are L = • -e or ma es an

6 4(1 -0.1785(t+O.435» f f 1 h L' 1 hL = 5.0 -e or ena es, w ere lS carapace engt 1n mn

andtage in years. Regression equations for weight in grams versus carapace

size in mm are W = .00039 L3 • 18 for males and W =.00081 L2

•97

for females.

A computer model simulates variation of mean size and mean weight growth rates

in function of relative increase in size at moult and periodicity in rate of

moulting.

•

This paper not to be cited without prior reference to the author

Contribution C.M. ]975/K 10

... / ...

•

A growth model for Nephrops norvegicus from Biscay Bay

in function of periodicity of moult+ ,

by

Gerard CONAN

Centre Oceanologique de Bretagne (CNEXO) , B.P. 337, 29273.BREST-Cedex (France)

Usual methods for estimating growth are unsuccessfull when applied

to Norway lobster ; individuals are not directly ageable and progression

through time of modes in the size frequency distributions is not readily inter­

pretable. Present work deals with a growth model based on interpretation of

life cycle of Nephrops.

HATERIAL AND 11ETHODS.

From April ]970 to April ]97], size frequency distributions were

sampled twice a month from nothern Biscay Bay boat landings, randomly selected

at harbour "Le Guilvinec". Number of soft individuals and berried females

were recorded for each .5 mm size class.

A Fortran program was written for computing, printing and plotting

relative size frequency distributions resulting from stratified sampies

regrouped over any selected time period. Component age groups were separatcd

from total size frequency distributions by Hasselblad' s technique (1966) as

programmed by Tomlinson (]970). Von Bertalanffy's equations were ajusted by

least square regression without linear transformation (Tomlinson and AbraQson,

196] ) •

RESULTS.

Regression of Weigl1t versus carapace size yielded the equations

Log W =-3.407 + 3.184 L, for the males, and Log W= -3.093 + 2.973 L fore e

the females (weicht is grams, length in millimeters). Differences significant

at 95% level were found between sexes by F test both for slope and position

of regression lines.

+ Contribution n° 407 du Departement Scientifique du Centre Oceanologiquede Bretagne.

• :~ .

Scasonal variations of sex ratio and proportion of berried females

are presented in Fig.I •. During July, August and Septenber a weIl oarked peak

period in the proportion of berried feroales is reaehed (up to 75%) ; it is

likely that reproduetion period is sharply defined und thus distinet age

groups should exist within the population. Proportion of fenales, as sho~m

by sex ratio, deereased in eateh along with disappearanee of berried individuals

and stabilized around 35% from Oetober to December.

size frequency distributions were regrouped in wide time sequenees

eorresponding to described seasonal variations (Fig.2). During July and August

a very high, nearly constant, proportion of berried females is present in all

size elasses. From September to November srnaller size elasses are recruited ;

in smaller size classes proportion of berried fe~~les is small or null. During

Winter months distributions are unimodal. They eorrespond to the reeently reerui­

ted size elasses ; very few berried fenales are eaught. From February to April

1arge size elasses of fenales reappear ; these females are no longer berried.

Seasonal variations of the proportion of "soft" individuals

are presented in FiG. 1. ClassifyinG in eategory "soft" per:::lits a qu;ek

diagnosis for individuals in intermolt stages D4 , E, Al' AZ' BI' BZ (Draeh,

1939 ; Drach and Tchernigovtzeff, 1967).3 maxima were reaehed by the propor­

tions of soft individuals anong males and among fema1es ; these peaks are like­

ly to correspond to periods of high moulting activity. During April, Uay and

June 1970 (Fig.2), und later during February, ~farch and April 1971, all size

classes of both sexes seem to be uniforcly affected by maulting; while during

September, Oetober and November 1970 all si~e elasses of males, but only smal­

ler, reeently reeruited, size elasses of fenales, eontain individuals in the

state of moulting.

For analysing the size frequency distributions, it was taken into

aeeount that during mault periods, each age graup yields two subdistributions

one is produced by individuals in premoult stages C, D, E, whieh have not

yet grown, the othe! by individuals in postmolt stages A and B whieh have

already grown in slze. Number of eompanents in sampled distributions is thus

variable ; the nodes nay double in number during molt periods, or even be

fully eoneealed when the distance between means of subdistributions is slight

compared with their standard deviations.

Equations for relative inerease in size at moult (Thomas, 1965)

were used in order to estimate thc number of moults which separated successivc

modal or menn sizes during internoult periods. Starting from smallest detecta­

ble modal size, hypotheses using I, 2, 3 und 4 moult inerements were tested.. .. / ...

i3.

for reaching next Denn sizes of conponent subgroups in the S.F. distributions

of Fig.3. Expected size frequency distributions were adjusted to the observed

ones by Hasselblad's technique. Best fit solutions suggest that the first

two subgroup nean sizes are separated by two noults in the size frequency

distributions of the males, and by one in the size frequency distributions

of the females. Results of size frequency distribution analysis are presented

in table ).

SYNTHESIS

- Life cycle.

!Wo periods of mault per year are likely to exist, one in Spring,

the other in Autumn. In this case 2 Spring moults would be represented in

Fig. I ; a cycle of 3 ma1ts per year is unlikely 'since distances between mean

sizes of age groups would not be expiainabie.

Best interpretation wou1d be that 4 age groups yie1d most of the

catch of Nephrops. Young individuals are recruited in Autumn at a mean size

of 21 mm ; 4 years Iater males reach 45 mm by means of 2 mou1ts a year, and

would possibly there on mou1 t 1ess frequent1y, at the same age females·.·reach

39 mm after mouiting twice the first year, and from then on only once every

Spring.

About 9 months after recruitment, most of fernales would spawn ; they

wou1d remain in berry until the next Spring(I), thus skipping Autumn moult,

but would become progressively unavailable to trawl fishery at end of sumner(2);

next they would reapper in catch in ear1y spring, after hatching of brood and

would·moult and spawn again before end of summer(3, 4)

- Growth interpretation.

In order to evaluate parameters of von Bertalanffy· growth equations,

each estimated modal or mean size in a cohort was referred to the age at

start of the moult period which had originated it. For non recruited age

groups size frequency data from 1Iillis (1972) were used ; snailest mode in these

distributions is at 4 mm carapace length corresponding to 12 mm total length

.... / ...I) Figueiredo and Thomas (1967) indicate that period of berry is of six to

seven nonths ; in Portuguese waters spawning is in August, hatching inFebruary.

2) Rice and Chapman (1971) found that berried fena1es show tendency for burro­wing, thus being 1ess easi1y caught by trawl.

3) Thomas (1964) shows, on basis of ovary naturity that 90% of fernales spawna~nually in scottish watcrs.

4) Figueiredo and Barraca (1963) state that shortly after spawning a new setof eggs begins to deve10p in ovaries of berried fernales so that spawningwill again take place in ensuing summer.

I.~.

(O'Riordan, 1964) ; compared with mean size of first post larval stage this

is 3.5 mn less than indicated by Figueiredo and Thomas (1967) after Santucci

(1926) and 1 mrn more than indicated by Jorgensen (1925) ; Poulsen (1946)

estimated that pelagic larval stages last 2 to 3 weeks, thus the 4 mm mode was

attributed to age group 0 born during preceeding Spring. The set of data taken

from llillis and the one collected from Biscay Bay, were fitted together by trial

and error, using least square analysis ; in the trial time 1ags of 0 and one

year were nscribo d betwecn thc two sets, the no tit:le lag hypothesis was

finally rctained for both sexes. Nephrops would be recruited at agc 11, and

their growth equations wou1d be L = 116.91 (1 - e-0 •077 (t + 0.438» for the

1 56 4 (1 -0.1785 (t + 0.435» f h f 1 I .on es or L = .0 - e or t e ema es. n str~ct

acceptance these curvcs should only be used for finding out mean individual

sizes reached at each moult period ; in between these time points growtll is

either nullduring intermoult periods, or is a function of instantaneous rate

of moulting and mean relative increase in size at moult during moult periods •.

- Computer simulation of growth :

A model was designed forretracing age group growth in mean size

and mean weight. Size frcquency distribution is first calculated, and then

converted to a weight distribution by applying weight-lenght relation­

ship to median sizes of size classes ; ultimately weightcd averages of

size and weight distributions yield the estir.J.ates of mean size and mean

weight. The model is applied independently to each sex.

Initial size frequencies at recruitment, f(L), are assumed to be

normally distributed. Distribution mean is calculated by applying von Bertalanffy

equation to the age at moult preceeding recruitment standard deviation 0

is obtained by Hassclblad's techniquc from youngest fully recruited age group

in the catch. f(L) is divided into 14 size intervals equal to 0/2 ; a

discreet distribution F(L) is substituted by integrating f(L) over each intervnL

F (L )m

L. +0/4= JfI(1.)dL

L .0'4m

T. • h d' . f th. . 1Im ~s t e me ~an s~ze 0 m ~nterva,

1 < m < 14

Mean relative increase in size r is calculated for each moult

period as ratio of mean individual sizes (L., L. '1) reached at end ofn n-present and preceeding noult periods. }lean individual sizes are calculated

fromvon Dertalanffy equation.

r = L /L l'n n-

Proportion P of individuals having Mou1ted at time T of a mou1t'l'

period is estimated by integrating instantaneous relative rate of moulting

p (t) from t start of moult period, up to T (fig. 4).o ... / ...

5.

'\..

P =0 f (t) .at't' 0

pet) is derived from Set), proportion of "soft" individuals in

age group. If pet) represents the proportion of individuals in an age group

reaching stage D4 (stnrting to bn soft) anu p'(t) thc proportion of individ~als

reaching B2 (finishing to b~ soft), ~~hile 0 is the average time during which

an individual remnins soft, then ()' (t) :.: p(t - 8). At time T :

1"

G(t) :: ! p(t)ct'(-0

pet) is aS3uraed to fit a r.ormd C~j:~lC~ thus 3(t) is defined by 3

parameters T (time at which graatest pro~crr.ion of individuals reach D4), a'

the standard deviaticn of p(t) ~nd G. An algo~ithc based on a polynomial

approximation of the 110I1:.al cquation is ~scd fot" modeling numerically Set)

T,a' and 0 are estimated by leas~ r,~U;lre fitting of Set) to time sequence of

proportions of soft indi.~liduals oboerycd durine a moult period. Thp. ~m.~ ~f -::~.~g, •.

res i8 minimized by simple iter~tion p~cceS3.

Lets deHne ircqu\:!ncy of non t'lou1ted individuals remaining in size

F(L ).0 -p )m ~

and frequency of individuals h~"'ing moulte:l <:n·.1 grown from Im to (Lm.r)

F(L ) .P.m '(

then mean size at tio~ l ~s

_ m=14 Cl"'14L • (t F(L ).(I-P ).L ) + (~ Y(L ).P .L .r)'[ In" m -1 J:l -= mm=l • m-.

Hean weight is calculatf.d by applying -;;eight length relationship

in each size class

== m=14 h ~"'1l~ bW .. (t F (L ).(I-P ).aL ) -} (E F(II ).P .a.(1. .r»

T m") m 't m :::..= I . rl '1;' In

Mean size gLowth curves obtnincd by this method simulate quite well

observed modal sizes at aße (Hg. 6). Hcwever goodness of fit decreases towards

older age groups which oight obey to a different rythmof moulting. Mean

weight growth curves can ea~ily bc intcgrated piece wise in a yield model by

numerical methods(Ricker» ~958). Fer ~~ny crustacean species proportion of

soft individuals in popu1;;tim.... !.~;: 1.-·no~,;:: to follov regular time variations ; the

... / ...

6.

present model could possibly be adapted to other species either for modeling

growth or for predicting proportion of individuals in soft, premoult or post­

mault conditions in the captures.

LITERATURE CITED

DRACH, P., 1939. ~1ue et cyc1e d'intermue chez 1es Crustaces Decapodes. Ann.Inst. Oceanogr. 19 : 103-391.

DRACH, P. and C. TCHERNIGOVTZEFF, 1967. Sur la methode de dete~ination desstades d'internue et son application generale aux Crustaces.Vie Milieu, 8, 3 : 595-610.

FIGUEIREDO, M.J. and I.F. B:JUL\CA, 1963. Contribucao para 0 conhecimento dapesca e da bio10gia d Langostim (Nephrops norvegicus L.)na costa portußesa. Notas Estud. Inst. Bio1. marit. (Lisb.),28, 44 p.

FIGUEIREDO, B.J. and lI.J. THOHAS, 1967. Nephrops norvegicus (Linnaeus 1758)Leach-a review. Oceanogr. Mar. BioZ. Ann. Rev.~ 5 : 371-407.

I~SSELBLAD, V., 1966. Estimation of parameters for a mixture of normal distri­butions. Technometrics~ 8 (3) : 432-444.

l11LLIS, J.P., 1972. Juvenile Nephrops caught in the Irish Sea. Nature (Lond.)237 : 280-281.

JORGENSEN, O.M., 1925. The early stages of Nephrops norvegicus~ from theNorthumber1and plankton, together with a note on the post­larval deve10pment of Homarus vuZgaris. J. Mar. Biol. Assoe.U.K. 13 : 870-879. .

0' RIORDAN , 1964. Nephrops norvegicus the Dutlin Bay prawn, in lrish waters.Scient. Proc. R. Dubl. Soc., sero B, 1 : 131-157.

POULSEN, E.M., 1946. Investigations on the Danish fishery for and bioloßYof the NOrYlay lobster and deep-sea prawn. Rep. Dan. Biol.Sta~ 48 : 27-49.

RICE, A.L. and C.J. CHAP~UU~, 1971. Observations on the burrows and burrowingbehaviour of two nud dwel1ing decapod crustaceans, Nephropsnorvegicua and GonepZa::c rhomboides. Mar. Biol. (Berl.) 10 :330-342.

RICKER, W.E., 1958. IIandbook of computations for bio10gical statistics offish populations. Fish. Res. Board Can. Bull. 119 : 300 p.

THOHAS, H.J., 1965. The growth of NOrl-my lohster in aquaria. Rapp. P.-V. Rcmn.Cons. Int. Explor. Mer, 156 : 209-217.

TOtaINSON, P.K., 1970. Program for separating mixture of normal distribu­tions. Unpub1ished manuscript, 2 p. Ca1ifornia Departmentof Fish and Game, Operations Research Branch, Lang Beach,CA 90802.

... / ...

l

TOMLINSON, P.K. and N.J. ABRAllS0N, 1961. Fitting a von Bertalanffy growthcurve by least squares. CaZif. Dep. Fish Game~ Fish. BuZZ.,116 : 69 p.

Table 1.

Interpretation of progression of modalsizes during, sampling period.

Males

April 1970 (1) 27 33 39 47 53Hay ( 1) 27 ,33(5) 37(4) 41 (4) 45(5) , 53June ( 1) 27 31 (4) 33(3) Lf5 49July 27 33 39 45August 27 35 39 45September ( 1) 21 27 35 39 45 49October (1) 23 45(3) 51November ( 1) 23 31 37 45' 51December 23 3 1 55February 71 ( 1) 25 31 45March (1) 23 29 39 45April ( 1) 23 27 31 35 39 45

49

49

27(4) 29 (3) 33 37 1+127(4) '29(3) 35(4,3)27 (4) 29(3) 35 3925(2) 29(2) 33 39(4) 41 (3) 4527(2) 31 (2) 35(2) 39(2)25(2) '33(2) 37(2)25(2)

Females

April 1970 ( 1 )

May ( 1)June (1)

JulyAugust

( 1)September 21October ( 1)November ( 1)DecemberFebruary 71 ( 1 )Harch ( 1)April ( 1 ),

2323 31 (2) Berried females are notreadily available to fishery23 25(2) at end of Autumn'and in Winter.23 25(2) 31(4,3)23 27(4,3)31

25(4) 31 33(3) 37(3) 4127(3) 35(4)

(1) Months during which "soft" individuals were found in the catch.(2) Modal sizes in size frequency distributions, of "soft" individuals.(3) Modal sizes in size frcquency distributions of berried fecales.(4) Modal sizes of individuals in premoult stages.(5) Modal sizcs of individuals in postmoult stages.

"---~

80~------------------------------------------------,

70

40

30

20

10

...

~i", ...."1 \. .I \.. \I 6. \

I "'1 \. 0...-·°\·./. . \

0_ . I·0..... f'·.....ql

"\

/5

10

5

0 0A M J J A S 0 N" D J F M A

• 0 Percentage of berried females. o--._.~ Percentage of "soft" females.

'" .. &-'-'-6Percentage of females in catch. Percentage of "soft" males.

Fig. Seasonal variations of parameters in catch during sampling period.

e'"

KlO.O 100.0

4

10.0

/.0

10.0

1.0

\.0 1.0 Y\~X

0.10111 15 19 " 23 27 31 35 39 43 47 51 55 59 I1 15 19 23 27 31 35 39 43 47 51 55 59

100.0

3

0.1

11 15 19 23

Fig. 2a Polygons for size frequencies regrou~2d over moultand intermoult periods. Males.

I) Moult. May 8 - June 18 1970.2) Interrlloult, egg bearing. July 2 - August 27 1970.3) Moult. September 7 - November 11 1970.4) Interrnoult. November 30 1970 - February 1 1971.5) Moult. February 15 - Ap~il 19 1971.

I>

47 51 55 59

Ordlnat. t % frequencle. Abselsle: coropOC8 lenQth In mm. X l total ... : soft

100. 100.0

/\4

10.0 10.0

1.0 1.0

0./ ~ 0.1 • i , i i ' i i 1T"""f i·i i • li i' ., •• i I. i i' i.11 15 19 23 27 31 35 39 43 47 51 55 5911 15 19 23 27 31 35 39 43 47 51 55 59

e 100.0

3

Fig. 2L Polygons for size frequencies regrouped ovcr moultand intermC'ult periods. Females.

1) Moalt. H3.Y 8 - June 18 1970.2) Intermoult, egg bearing. July 2 - August 27 1970.3) Hoult. September 7 - November 11 1970.4) Intermoult. November 30 1970 - February 1 1971.5) Mault. February 15 - April 19 1971 •

$

V <l0./

I1 15 19 23 27 31 35 39 43 47 51 55 59

Ordinate : % frequencies Absci.so , coropoce lenglh in mm. V I tolol <I' .oft Alberried

A

1.0 1.0 "

1\\ "0.1 0.111 15 19 23 27 31 35 39 43 47 51 55 59 11 15 19 23 27 31 35 39 43 47 51 55 59

e 100.0 100.0

25

100 10.0

100.0 1000

APRIL 70 JULY 70

10

1.0

10.0

10

l>

0151 55 5911 15 19 23 zr 31 35 39 43 47

e .100.0

MAV 70

100.0

AUGUST 70

51

100Kl.O

10 10

\ l> l>-+

01 0111 15 19 23 27 31 35 . 39 43 47 51 55 59 11 15 19 23 27 31 35 39 43 47 51 55 59

e 100.0 100.0JUNE 70 SEPTEMBER 70

100

1.0

Ordinate I % frequencie, Ab,c/no' coropoce l,n9th In mm.

1.0

\ )jf-I(

0.1

11 15 19 23 27 3\ :35 39 43 47 51 55 59

X I 10101 ... I soft

Fi~ Polygons for monthly regrouped size frequencies.3a HaIes.

1000 100.0

OCTOBER 70 . FEBRUARY 70

10.0

1.0

0.1

IS

elOO

100

1.0

~

0.1+II.....,.TTI5...,..,....,..,.19...,.... '2'3' , '27' , '31' "35 39 43" '47' , '51' , 's's' , 'S'9' •

1000

NOVEMBER 70 MARCH 70

x

1.0

10.0

1.0

10.0

~ ~

0.1 0.1 •• i " 5'9' •11 15 19 23 27 31 35 39 43 47 51 55 59 11 IS 19 23 27 31 35 ~ 43 47 51 55

_100.0 DECEMBER 70 100.0 APRI L 70

100

1.0

100

1.0

O.l-h-.,-.-rT"'..........,...,.~rT"'.-.-.,...,.:r-r.,..,...-.-.......-~.,......, ..........":':r..-rTT"',..,..,.::rr-r::'":,...,..11 15 19 23 27 31 35 39 43 47 51 55 59

O.l+O-....,...,...,...,...,...,...,..~~~~.-r-.-r-~..,..,,....,.....,..,~~~~~.,....,.,....,.,....,.,....,~

I1 15 19 23 27 31 35 39 43· 47 51 55 S9

Ordinate: % frequencies A bscisso I coropoce len~t in mm. X I totol t> • lOft

Fig. 3Polygons for monthly regrouped size frequencies.3b Males •

....

1000 100.0

APRIL 70 JULY 70

100

v

10

0.1-lIl............1,...5......."T""T19"T"'O......2~3........r2...7"T""T..3..1......n-35...............3"T9.....,'..,4'...,3'~'......i4""'7"""""T'5l...".,'...,'M5~5'........' r,5...·9"T'.,..,

100

10

0.1+II..........1~5.............19"T""T"T"'02..,3..........2,...7......."T3"T1......·'..,·-3~5....' ,...,r·3T•9"T'.,.'.."43..,'M·......'4'='7r' ...·.,.'5="1..'..,'..,''=5'=5....' ,...,T·5:':·97•...•

100.0 100.0

MAY "70 AUGUST 70

10.0

J X1.0 10

A

0.1 0.111 15 19 23 27 31 35 39 43 47 51 55 59 11 15 19 23 27 31 35 39 43 47 51 55 59

10.0

100.0JUNE 70

100.0SEPTEMBER 70

10.0

1.0 vv

x

0.1"+.,...,..,...,15.,...,.......I9~........2....3........,...2r7.......r3l.......T"""3"T's""""""3"9"""'''i'43'''''i''''''''47'''''r1i''''~5'''''1''-''''''''5'''''s''''''''''r'5r9r, ....,15 19 23 27 31 35 39 43 47 51 55 59

Ordinate 1 % frequencies A bsciSIQ r coropoce ""9th in mm V 1 total <l 'soft A tberrled

Fig. 3 Polygons far monthly regrouped size frequencies.3c Females.

1000 100.0

ocrOBER 70

10.0

1.0

10.0

1.0

FEBRUARY 71

v

0.1 +11.,....."T'"T15"....~197"....-::'2-.3 ,-,..,':'2'="7.......r:3'!:"1,...,..":3~5TTT3::r9:r-r"T4:r3:"'""T"T7477'"""::'5,...,1 '-""':'55'7"'""':5'=9"

e 1000. ,

1000

NOVEMBER 70 MARCH 71

10.0

1.0

10.0

1.0

0.1 0.111 15 19 23 27 31 35 39 43 47 51 55 59 11 15 19 23 27 31 35 39 43 47 51 55 59

_100.0DECEMBER 70 100.0

APRIL 71

10.0 1\ 10.0

A

1.0 1.0

O.I+.-..............T'T"rr..,..,,....,..,...,...,...,.,.......,.,............,..,,....,..,...,...,...,.....,.,............,..,,..,...,..,...,...,...,....,11 15 19 23 27 31 35 39 43 47' 51 55 595955

o Ih.,...rM...........,...,..,...rM~..,...,..,....,..,...,rT"T..,...,..,...rM~..,...,..,...rMrT"T..,...,..,.....,.11 15 19 23 27 31 35 39 43 47 51

Ordlnat. I % frequencl., Abacisso I corapaC8 l.n9th In V I total -4 t soft Alberried

F~ Polygons for monthly regrouped size frequencies.3d Females •

•

";!

•

p (t)

ort'(t)

~ ,

e

ct lime t=t

S (I): ~

Pt' ~

0'

T

Fig. 4 Graphical representadon of Set), proportion of "soft" individuals,and of PT , proportion of individuals having moulted at time T.

P (t): instantaneous relative rate of initiation of moulting.p' (t): ins'tantaneous relative rate of termination of moulting.,

•• •

time

•

,-

j

x

x

Females•

_...-lB

-- A---,..,...." ...-x

Males

o•

xx

...1 B..../.: A

""".../_--/

//

/'./-_....

>Y" "X xx xx X

"..../

/ .---/

./

""

•30

35

,....50

~'-'

Q)N

• .-1III

Q)(JCIl 45Cl<CIlI-lCIl(J

t:::CIlQ)

:L:

40

25

20

Fig. 5a Mean·size growth curves. A) Computer simulated,B) obtained from von Bertalanffy equation fitted

to modal or mean carapace sizes referred to age at startof moult period. Dots and crosses represent observed modalor mean sizes at age.

A

1o11

A

1om

A...

le·onr

.A

I...0 .:sr

•

A

Io

e nA

I.0

:EI.·AI Aga

•

'--140

120

_100I/l

Eo...Cl.....

o::::J

:2> 60'0.5

c:o(1)

:2:40

Males

/'./

.---::: ./"/

./,/

./

,..

",",

20 :-------~

A 0 A 0 A 0

I II I III I IV

Fig. Sb Male mean weight growth curve.

• • •

Age

•

140

120

_100(/)

EEöl

Ci:J

"0

:~ 60"0.scoQ)

~

40

20

Females

----- -.­.--.--

.­- ­.-.--

....---A.............

--.--

... . ..

A 0 A 0 A 0 A 0 i 0 A 0 A

I II I III I IV I V VI I VII I Age

Fig" Sc Female mean weight grow_urve. e·11 • • •, •LL

--_.-