Effect of feeding attractants on the behaviour and performance of juvenile penaeus monodon fabricius r hartati, m r p briggs

Behavioural trials monitoring the feeding response of the shrimp were used to gauge the attractant qualities of the substances, A growth trial recording the feed intake, feed assimilatio

Aquaculture and Fisheries Management 1993, 24, 613-624

Effect of feeding attractants on the behaviour and

performance of juvenile Penaeus monodon Fabricius

R HARTATI & M R P BRIGGS Institute of Aquaculture, University of Stirling,

Stirling, Scotland

Abstract A study was conducted to obtain a measure of the potency of some potential and

commercially used feeding attractants for Penaeus monodon Fabricius Behavioural trials

monitoring the feeding response of the shrimp were used to gauge the attractant qualities of the substances, A growth trial recording the feed intake, feed assimilation, growth, food conversion and survival rates of the shrimp was used to assess further the feeding stimulant properties of the substances.

Replicate groups of juvenile shrimp were fed semi-purified diets containing 1-5% by weight

of a range of potential feeding attractants.

In the behavioural trial, diets containing taurine and a yeast extract were found to be significantly preferred to the control and all other diets,'However, none of the substances appeared to act as potent feeding stimulants, producing statistically similar feed intake and assimilation rates to the control diet However, taurine and an amino acid mixture designed to mimic a clam extract promoted the best performance of the attractants tested in terms of growth rate and feeding efficiency.

Overall, the behavioural response of the shrimp to the feeding attractants was found to show similarities to the effects of attractant supplementation of feed on subsequent ongrowing performance, but not significantly so.

Introduction

Feeding attractants can play an important role in enhancing food detection and feeding stimulation Their role can be especially important in slow-feeding crustaceans, reducing leaching of essential water-soluble nutrients present in the diet (Deshimaru & Yone 1978; Heinen 1980) by increasing the rate of food consumption

Many studies have been made of the factors involved in stimulating feeding in crustaceans, particularly with lobsters, crabs and freshwater prawns However, little is currently known about the effect of attractants on the feeding behaviour and ingestion rate of

Penaeus monodon Fabricius, one of the most widely cultivated crustacean species.

Most of the studies on feeding attractants to crustaceans have analysed the effects of a range of concentrations of various potential attractants on behaviour, particularly with regard to their effects on the crustacean chemosensory apparatus (Carr 1978; Carr & Thompson 1983; Carr, Netherton & Mistead 1984; Carr & Derby 1986; Harpaz, Kahan, Galun & Moore 1987; Harpaz & Steiner 1987, 1990; Nakamura 1987) In addition, growth trials have been conducted with crustacean species using both commercial (Farmanfarmaian, Lauterio, Tejada, Manger, Moore & Park 1979; Murai, Sumalangkay & Pascual 1981; El Hag 1984; Costa-Pierce & Laws 1985) and semi-purified diets (Deshimaru & Yone 1978; Pascual 1980) containing various attractants

Correspondence: Mr M, R, P, Briggs, Institute of Aquaculture, University of Stirling, Stirling FK9 4LA, Scotland,

613

Olfactory stimulation by a variety of chemicals has been demonstrated by many authors in decapod crustaceans (Heinen 1980) Substances (usually of low molecular weight, <1000) which were found to be strong stimuli when tested singly on decapods were amino acids, nucleotides and quaternary ammonium compounds, while sugars, alcohols, starches, fatty acids and several other compounds have usually given lesser responses or none at all (Heinen

1980; Carr & Derby 1986).

The aim of the present work was to obtain a measure of the potency of various potential

and commercial feeding attractants to P monodon by way of both behavioural and

ongrowing studies This two-phase approach also allowed an assessment as to whether the behavioural responses to feeding attractants can predict the effects of attractant supplemen-tation on subsequent ongrowing performance This measure is important as the addition of feeding attractants might elicit an increase in appetite and subsequently food intake and growth (Lindstedt 1971) and also improve shrimp survival and food conversion (Heinen 1980)

Materials and methods

Stage 1: Diet choice behavioural trial

The experimental shrimp were obtained from the University of Stirling Institute of Aquaculture hatchery and reared in a holding tank until they reached the desired weight Prior to the trials the shrimp were held in a common holding tank equipped with water recirculation and were fed twice daily with chopped mussel and green beans at the rate of 20% wet body weight/day Following each behavioural trial, each group of shrimp was kept in individual cages made of mesh-covered PVC pipe measuring 16cm in length and 11cm in diameter All cages were placed in the holding tank so that they could be used for subsequent trials after allowing at least 2 days for shrimp recovery

A semi-purified diet was used as a medium for testing the potential feeding attractants as

has been used in earlier studies with Macrobrachium rosenbergii De Man (De Proenca 1990),

Penaeus japonicus Bate (Deshimaru & Yone 1978) and P monodon (Pascual 1980).

The composition and proximate analysis of the basal diet is presented in Table 1 Crude protein was determined by Kjeldahl analysis, expressed as total N x 6-25, crude lipid by the Soxhlet method, ash by combustion at 450°C for 12h and carbohydrate as the remainder of the dry weight Moisture was determined by drying at 105°C for 24h and gross energy by combustion in a micro-bomb calorimeter All results were based on the mean of triplicate analyses

The attractants, included at a rate of l-57o in each diet (except the control) were as follows:

Diet 1 (T) : Taurine

Diet 2 (B/G) : A mixture of 50% betaine and 50% glycine

Diet 3 (AA) : A mixture of L-amino acids based on the composition of an extract

of short-necked clam {Tapes japonica: Deshayes) as described by

Deshimaru & Kuroki (1974) (Table 2) Diet 4 (AMP) : Adenosine 5'-monophosphate

Diet 5 (TMAH) : Trimethylamine hydrochloride

Ejfect of feeding attractants on P monodon 615

piet 6 (F) : Finnstim, a commercial chemoattractant developed by the Finnish

Sugar Co Ltd The product is based on betaine and a mixture of stimulatory amino acids

Diet 7 (YE) : Yeast extract, a natural flavour enhancer

Diet 8 (C) : Control diet, using 1-5% a-cellulose in place of an attractant All shrimp were starved for 48h prior to the trials Each of the seven test diets and the control were tested against each of the other diets in paired comparisons (Table 3) Each set

of two diets was tested against one another for the relative attractiveness of the diets Every combination of diets was replicated five times for statistical comparison A total of 28 combinations of diets was tested Tests consisted of a single group of eight shrimp being introduced into an aquarium and being offered a choice of two diets

Five replicate groups of eight juvenile shrimp, mean weight 35 ± 5mg, were tested in separate perspex tanks each measuring 36 x 19-5 x 28cm One end of each tank was divided into two freely available, equal compartments (X and Y) by a piece of black acrylic plate measuring 12 cm^ A section of black PVC pipe was fixed to the bottom of the tanks at the opposite end as a shelter The tanks were half-filled with 91 of sea water at 25%o salinity The

Table 1 Basal composition of the diets

Ingredient % dry weight

Casein (vitamin free) 31-0

Gelatin 10-2

Dextrin 35-7

Astaxanthin 0-1

Vitamin premix* 2-0

Mineral premix*' 4-0

Cod liver oil 6-0

Lecithin 3-0

Cholesterol 0-5

a-cellulose 6-0

Attractant/a-cellulose*** 1-5

S o u r c e " "

Proximate analysis

Protein 40-8

Lipid 9-1 •

Carbohydrate 40-2

Ash 3-8

Fibre 6-1

Moisture 8-0

Total energy (Kcal/g) 4-9

* Vitamin premix (in lOOg premix): thiamine-HCl 150mg, riboflavin 500mg, niacinamide 2000mg, pantothenic acid

750 mg, inositol 10000 mg, biotin 15 mg, folic acid 37-5 mg, pyridoxine-HCl 150 mg, P-aminobenzoic acid 1000 mg, choline chloride 20000mg, ascorbic acid 25000mg, a-tocopherol acetate lOOOmg, menadione (sodium bisulphide) lOOmg, p-carotene 100mg, cholecalciferol 15mg, cyanocobalamine lmg, a-cellulose powder 39181-5mg.

** Mineral premix (in lOOg premix): CaHPO4.2H2O 75-0276g, MgSO4.7H2O 12-75g, NaCI 6 Og, KCl 5-Og, FeSO4.7H2O 0-25g, ZnSO4.7H2O 0-55g, CUSO4.5H2O 0 0785g, MnSO4.4H2O 0 2538g, COSO4.7H2O 0-0478g, CaIO2.6H2O 0 0295g, CrCl3.6H2O 0 0128g.

*** a-cellulose used in control diet instead of attractant.

**** All chemicals and other dietary ingredients obtained from Sigma Chemicals Ltd except lecithin from Lucas Meyer Ltd.

Table 2 Composition of the amino acid mixture

Amino acid % dry weight Taurine 43-6 Aspartic acid I.4 Threonine 0-8 Serine 1.6 Glutamic acid 6-8 Proiine 1-0 Glycine 21-6 Alanine 8-6 Cystine 0-4 Valine 1-0 Methionine 0-6 Leucine 1'2 Isoleucine 0-6 Tyrosine 10 Phenylalanine 1-4 Histidine-HCl 0-6 Lysine 1-6 Arginine 6-2

temperature of the sea water in the tanks was kept constant at 29 ± 1°C All tanks were placed

in dim lighting conditions on a 12:12h light:dark cycle

The diets were offered at the rate of 20% wet body weight at the beginning of each trial either in compartment X or Y, while the same amount of control diet or another test diet was placed in the other Aeration was turned down during the trial The trial was started and the number of shrimp located in each compartment counted every minute for a total of 30 min The q^ttractivene^s of the diets is expressed in terms of feeding activity as shrimp time units, i.e the number of shrimp feeding on each diet, each minute for 30 min These data were converted to give a measure of the percentage of the maximum possible response to each diet in order to gauge the overall attractiveness of the diets The results were ranked and

Table 3 Test group comparisons used in the behavioural trials

Test diet T B/G AA AMP TMAH F YE C

B/G , .

AA • " *

AMP

TMAH

F

YE

T — Test diet 1 (Taurine).

B/G — Test diet 2 (Betaine-Glycine mixture).

A A — Test diet 3 (Amino acid mixture).

AMP — Test diet 4 (AMP).

TMAH — Test diet 5 (TMAH).

F — Test diet 6 (Finnstim).

YE — Test diet 7 (Yeast extract).

C — Control diet (Without attractant).

Effect of feeding attractants on P monodon 617

analysed for significant differences between paired samples by the Wilcoxon-Mann-Whitney two-samples test (Steel & Torrie 1980),

Stage 2: Feeding trial

The diets analysed in the behavioural trial were retested in a feeding trial Results of the behavioural trials were compared with these results to determine whether the subsequent ongrowing performance of shrimp could be predicted by behavioural studies

Two replicate groups of five shrimp were used in this stage of the study Experimental animals, rearing conditions and test diets were the same as those used in stage 1

All shrimp were starved for 48 h prior to the trial The feeding regime was 10% of the wet body weight/day given in two equal portions at 0900 and 1600 h for 6 days/week with no feeding on the weekly sampling day

After l-5h (because it was found that shrimp started producing faeces 2h after first feeding) all food remaining uneaten was siphoned, filtered through Whatman filter paper no,

54 and dried in the oven for 24 h to calculate the actual food intake based on the following equation:

Daily feed intake (g feed/g shrimp/day)

Fl

[{wt, + Wt2)/2]d

One and a half hours later, the faeces produced were siphoned, filtered and dried in the oven to determine feed assimilation efficiency based on the following equation:

Food assimilation efficiency (%)

Where

Fl : Food ingested (g/wk)

= food given — food wasted (g)

FP : Faeces produced (g/wk)

vv^i : Mean wet body weight at day 1 (g)

Wt2 : Mean wet body weight at day n (g)

d : Time (days)

All uneaten food and faeces were collected before both the morning and evening feedings and the water lost during each siphoning was replaced This resulted in a water exchange rate

of approximately 30%/day,

The feeding experiment was conducted for 28 days The shrimp were weighed individually and counted every 7 days and the feeding regime adjusted after each weighing The performance of each test diet was evaluated as specific growth rate (SGR), food conversion ratio and survival rate

One-way analysis of variance (arc-sine transformed for the percentage values) and Duncan's multiple range tests (P < 0-05) (Steel & Torrie 1980) were employed to establish the effect of feeding attractants on shrimp feed intake and assimilation, growth, survival and feeding efficiency

Stage 1: Diet choice behavioural trial

The results of the paired sample comparisons were looked at in terms of feeding activity (shrimp-time units), i.e the number of shrimp feeding on each diet for each of the 30 min of each trial Results showed that most of the substances, particularly taurine, the yeast extract and the amino acid mixture, were more attractive (P < 0-05) than the control while the betaine/glycine mixture and adenosine 5'-monophosphate were not

Results from the response of shrimp to each diet during the paired comparisons as a

percentage of the maximum possible response showed that there were significant {P < 0-05)

differences between diets (Table 4 and Fig 1) The rank attractiveness of the test diets in decreasing order were as follows (test diets with a common underline were not significantly different):

T Y E F A A T M A H A M P C B / G

Overall, the shrimp response data indicated that taurine and the yeast extract were

significantly [P < 0-05) preferred over all other attractants and the control In addition, the

amino acid mixture and Finnstim were preferred (P < 0 05) over the control, but that TMAH, AMP and betaine/glycine were not

Stage 2: Feeding trial

The daily feed intake results showed that there were no significant differences between treatments and that the shrimp fed on the control diet had a similar daily feed intake (0-35 g/g shrimp) to those fed the diets containing attractants (Table 5) Among the test diets which contained attractants, the shrimp fed diet 4 (containing AMP) had the highest daily feed intake followed by diet 1 (Taurine) Both of these diets also resulted in a fairly constant feed ingestion rate, while that of shrimp fed any of the other diets tended to decline during the trial period

Table 4 Response of shrimp fed the test diets as a percentage of the maximum possible response

Replicate

1

2

3

4

5

6

7

X

+SEM

T 310

27-0

33-4

36-0

33-2

30-8

31-0

318

1-06

B/G 120 10-1 5-6 13-6 13-0 10-4 5-0 9-9 1-30

AA 14-6 16-6 23-2 15-8 14-5 25-6 28-8 19-9 2-22

AMP 14-6 9-0 12-2 18-6 15-6 7-2 19-6 13-2 1-82

Test diet

T M A H 102 24-0 10-2 15-4 17-2 12-4 24-6 17-7 2-08

F 19-8 31:0 23-2 17-2 23-0 16-4 20-8 19-6 1-31

YE 27-8 31-0 25:6

31 6 34-6 33-4 37-4 31-6 1-52

C

9 8 7-2 13-0 14-6

9 8 9-0 14-6

I M 110

Effect of feeding attractants on P monodon 619

E 3

O

c

o.

at

40 -I

3 0

20

10

-c

4 Test

5 Diet

Figure 1 Mean response (+SEM) of shrimp to test diets as a percentage of the maximum response.

The differences in feed assimilation efficiency between the test diets were insignificant (P

< 0-05) However, the highest feed assimilation was achieved by shrimp fed on diet 4 (AMP) and the lowest by shrimp fed on diet 7 (YE) (Table 5),

Statistical analysis showed that weight gain of shrimp fed on diets 1 (T) and 3 (AA) was significantly (P < 0-05) higher than that of shrimp fed on diet 2 (B/G), but not than those fed the control or any other test diet The SGR of shrimp fed on diets 1 (T) and 3 (AA) was also significantly (P < 0-05) better than shrimp fed on diet 2 (B/G) (Table 5 and Fig, 2),

Table 5 Effect of the feeding attractants on the daily feed intake, feed assimilation efficiency, weight gain, specific

growth rate, food conversion ratio and survival of the shrimp (+SEM) over 28 days

Test

diet

1

2

3

4

5

6

7

8

Attractant

Taurine

Betaine/

Glycine

Amino acid

mixture

AMP

TMAH

Finnstim

Yeast

extract

Control

DFI 0-33"

(0-02) 0-28"

(0-05) 0-29"

(0-06) 0-35"

(0-05) 0-25"

(0-05) 0-28"

(005) 0-26'' (0-05) 0-35"

(0-07)

FAE 92-44'"' (0-41) 92-75""

(1-22) 93-81"

(0-80) 93-90"

(0-74) 92-69""

(0-87) 92-27""

(0-15) 89-732"

(0-29) 93-45"

(0-14)

WG 1-32"

(0-20) 0-45"

(0-10) 1-21"

(0-26) 0-79""

(0-10) 0-85"?

(0-03) 0-81""

(0-22) 0-89""

(0-13) 0-79""

(0-12)

SGR 3-45"

(0-46) 1-52"

(0-31) 3-29"

(0-33) 2-45""

(0-17) 2-53""

(0-06) 2-46""

(0-53) 2-55""

(0-31) 2-25""

(0-28)

FCR 0-91"

(0-01) 2-75°

(0-50) 1-08"

(0-05) 1-60""

(0-08) l-23b (0-02) 1-49""

(0-54) 1-39""

(0-03) 2-05""

(0-52)

SR 80" (0) 60" (0) 60" (0) 60" (0) 50° (0) 60" (20) 50" (10) 70" (10) DFI — Daily feed intake (g),

FAE — Feed assimilation efficiency (%),

WG —Weight gain (g),

SGR — Specific growth rate (% body weight/day),

FCR — Food conversion ratio,

SR — Survival rate (%),

"h — For single criterion, mean values in the same column bearing the same superscript are not significantly different

{P < 0-05),

•a 3 T3 O

E

4 •

31

0

-T

0)"'

T

m

T

;<

T

0 'S MA

1

-T

LJ.

T

^'

T _

4 5 6 Test Diet

Figure 2 Mean SGR (+SEM) of shrimp fed the test diets.

Diet 2 (B/G) also produced the poorest FCR of 2-75, which was significantly (P < 0-05) worse than the FCR for the diets containing taurine (diet 1) at 0-91, the amino acid mixture (diet 3) and TMAH (diet 5) (Table 5) Although the survival rate of shrimp during the trial was not significantly different between treatments or from the control, the diet containing taurine as an attractant gave the best survival (80%) and the yeast extract and TMAH the worst (50%) (Table 5)

Discussion

Diet choice

The diet choice comparison technique adopted in this study was able to reveal significant differences between the behaviour of shrimp towards diets containing different feeding

attractants Shrimp were shown to have a significant {P < 0-05) preference for diets

containing taurine and the yeast extract over all other diets, particularly the control, demonstrating the attractant qualities of these substances Additionally, the diets containing these two attractants were seen to elicit the highest (P < 0-05) percentage response figures (about 32% of the maximum possible) of the diets

This evidence suggests that P monodon might have a taurine-specific antennular chemoreceptor similar to that found in Panulirus argus Travis (Fuzessery, Carr & Ache

1978) This may be related to the fact that taurine is the most abundant p-amino acid in the

prey of shrimp and in their own bodies (Watanabe & Konusu 1972; Fuzessery et al 1978;

Smith, Miller & Mead 1987) The diet containing the amino acid mixture (43-6% taurine) as

an attractant produced a lesser response than taurine alone, but was still preferred (P < 0-05) over the control diet Similar results to these were found by Deshimaru & Yone (1978) with

P japonicus although they also found glycine to be particularly effective.

In this study the response of shrimp fed a diet containing the betaine/glycine mixture was not significantly different to those fed the control diet, whereas both glycine and betaine

alone have previously been shown to improve diet attractiveness for P monodon (Murai et

al 1981) In contrast, Finnstim, composed largely of betaine, proved more attractive (P <

0-05) to the shrimp than the betaine/glycine mixture and the control in this trial It may

Effect of feeding attractants on P monodon 621

therefore be that betaine and glycine are antagonistic when used together as attractants, as has been found for proline, alanine, arginine and taurine in lobsters due to competitive interaction for receptor sites (McLeese 1970; Johnson & Ache 1978; Ache, Gleeson & Thompson 1986), Alternatively, it may be the concentration of betaine that is responsible for its attractant qualities, although this could not be quantified due to the unknown concentration of betaine in Finnstim

A new feeding attractant based upon a yeast extract was tested in this study Information

on the type of yeast that was extracted to produce this attractant was not available However,

it apparently acted as a potent feeding attractant for P monodon, being preferred (P < 0-05)

over all diets except the one containing taurine

Feed intake and assimilation

Data on feed intake and assimilation can serve as a guide to whether the attractants were also acting as feeding stimulants The diet containing taurine promoted a high and consistent feed intake This diet was also shown to have excellent attractant qualities from the behavioural study, Taurine thus seems to be effective as both a feeding attractant and as a mild feeding

stimulant for P monodon, as was found for P japonicus (Deshimaru & Yone 1978) and M.

rosenbergii (Smith et al 1987),

The best overall feed intake and assimilation results, although not significantly different (P < 0-05), were obtained for the diet containing AMP, which was shown to be a poor attractant in the behavioural studies AMP may thus have limited stimulant qualities, without

being attractive to P monodon In contrast, AMP has been shown to be a chemoattractant for the caridean shrimp, Palaemonetes pugio (Carr & Thompson 1983), and M rosenbergii

(Harpaz, Kahan & Galun 1987; De Proenca 1990), The diet stimulating the next highest increase in food intake was that containing an amino acid mixture A similar result was

achieved by Deshimaru & Yone (1978) with P japonicus fed diets supplemented with the

same amino acid mixture In previous trials, diets containing TMAH were found to result in a

30-38% increase in food consumption when fed to M rosenbergii (Costa-Pierce & Laws

1985), This was suggested to be due to their strong faecal odour and the coprophageous feeding habits bf caridean prawns In this trial, however, TMAH was not found to increase feed ingestion or growth rate significantly, although it did act as a mild attractant for P

monodon.

Feeding trial

Differences in the growth rate of shrimp, although insignificant, followed similar trends as the attractiveness of the diets during the behavioural trials, Taurine and the amino acid mixture promoted the best growth rates of the attractants tested Good growth was also achieved on diets containing the yeast extract, TMAH, Finnstim and AMP, all greater than the control, while the betaine/glycine mixture promoted the poorest growth

These results indicate that in addition to the attractant qualities of taurine, the yeast extract and the amino acid mixture, these substances may have stimulant qualities, resulting not in increased ingestion or assimilation of food, but in increased growth Similar results

have been achieved by Pascual (1980) and Murai et al (1981) with P, monodon These

authors found that both shrimp and mussel extracts containing high quantities of both taurine

and a mixture of amino acids were very attractive to P monodon and increased growth and

survival rates due largely to enhanced diet acceptability

The addition of protein and/or essential amino acids to a diet deficient in these nutrients might have a role in metabolic processes as well as acting as incitants (Heinen 1980) The ability of these diets to enhance growth rate may thus have been due to enhanced nutritional quality, particularly because the test diets used purified ingredients which may lack essential nutrients and attractive qualities However, taurine is not regarded as an essential amino acid (Coloso & Cruz 1980) and other attractants containing amino acids, such as Finnstim and the betaine/glycine mixture, were not able to promote such a response Additionally, the amino

acid mixture attractant, based on the short-necked clam (a known favourite food for P.

japonicus, Deshimaru & Yone 1978), contained a mixture of essential and non-essential

amino acids but was not able to promote a better growth or feeding response than taurine alone

Analysis of the FCR of shrimp fed the test diets revealed similar results to those with the growth rate This measurement thus seemed to be a more sensitive indicator of shrimp feeding efficiency than assimilation efficiency, and again confirmed results from the behavioural trial section of this investigation Only the diet containing taurine promoted better survival than the control diet, although there were no significant differences between any of the diets

A comparison of data from the behavioural and growth trials revealed similarities in the ranking of the attractiveness of the test diets This suggests that behavioural studies may be able to predict subsequent ongrowing performance Extending the duration of the feeding trial stage of this study may have resulted in a closer correlation between the two stages of this investigation and deserves further attention

Acknowledgments

The authors wish to thank the following staff from the Institute of Aquaculture, University of Stirling: Dr S Smith, Mr D Milroy and Mr A Porter for their technical support, Dr B J MacAndrew for assistance with the statistical analysis and Dr J H Brown, Dr C J Fox and

G S Haylor for their critical reviews of the manuscript

We would like to thank the Ministry of Education and Culture, Indonesia and The British Council for supporting R Hartati during this work

These trials were part of work funded by the Overseas Development Administration (ODA), under Research Project Number R4443 NRG 522/832/8A

References

Ache B.W., Gleeson R.A & Thompson H.D (1986) Mechanism of interaction between odorants at olfactory

receptor cells Chemical Senses 11, 575.

Carr W.E.S (1978) Chemoreception in the shrimp, Palaemonetes pugio: The role of amino acids and betaine in elicitation of a feeding response by extracts Comparative Biochemistry and Physiology 61 A, 127-131.

Carr W.E.S & Thompson H.W (1983) Adenosine 5'-monophosphate, an internal regulatory agent, is a potent

chemoattractant for a marine shrimp Journal of Comparative Physiology 153, 47-53.

Carr W.E.S & Derby C D (1986) Chemically stimulated feeding behaviour in marine animals Journal of Chemical

Ecology 12,989-1001.