ORIGINAL PAPER

S.-R. Birkely Æ O. Grahl-Nielsen Æ B. Gulliksen

Temporal variations and anatomical distributions of fatty acidsin the bivalve Mya truncata, L. 1758, from Isfjorden, Spitsbergen

Received: 2 May 2002 /Accepted: 30 September 2002 / Published online: 20 November 2002� Springer-Verlag 2002

Abstract The annual variation and anatomical distri-bution of fatty acids in the burrowing bivalve, Myatruncata, were studied at five different times betweenMay 1997 and July 1998, in Isfjorden, Spitsbergen.Twenty-eight fatty acids were determined by directmethanolysis of samples from various tissues, gaschromatography and statistical, principal componentanalysis of the chromatographic results. Gill, foot andmantle tissue, as well as gonads and stomach contents,were subjected to analysis. The branched fatty acid,4,8,12-trimethyltridecanoic acid occurred in about 3%relative abundance in the stomach contents in all ani-mals, originating from ingested phytoplankton. Thisacid was either not assimilated but excreted directly or, ifassimilated, it was completely catabolised. The fatty-acid composition of the stomach contents varied sig-nificantly with time of sampling, indicating variation inthe phytoplankton community in the bottom water. Thegill, foot, mantle and gonad tissue had mutually differentfatty-acid composition, all significantly different fromthe composition of the stomach contents. This showsthat the tissues have typical fatty-acid composition, andthat assimilated fatty acids are subject to modificationand/or selection before being incorporated into tissuelipids. This was further substantiated in that the ratherlarge temporal changes in fatty-acid composition of thestomach contents were not mirrored in the tissues.However, a minor effect of the changes in the diet wasdetected in the gill tissue. Male and female gonads hadsignificantly different fatty-acid compositions, with C20and C22 acids more abundant in the male gonads andC16 and C18 acids more abundant in the female gonads.

Fatty-acid analysis may be used in identification of thesex of the bivalves.

Introduction

The use of lipid analysis has increased the understandingof both trophic relationships between species and foodpathways in Arctic marine communities (Falk-Petersenet al. 1990). An important link here is the bivalve mol-luscs, particularly the filter-feeding species, with theirability to accumulate and incorporate lipids from pre-vious links (Voogt 1983). In the Arctic food chain, manybivalve species serve as food both for seals (Hjelset et al.1999) and walruses (Fisher and Stewart 1997), with theclams Mya truncata and Hiatella arctica as importantspecies.

M. truncata, L. 1758, occurs in mixed muddy andsandy substrates (burrowed down approximately8–30 cm.), from the lower shore to about 70 m depth. Itis a habitant off all British coasts, circumboreally, in thenortheast Atlantic (including the North Sea and thecoast of Norway), extending as far south as Biscay(Hayward and Ryland 1990; Brattegard and Holthe1997), and is a common species in the waters aroundSvalbard (Gulliksen et al. 1999).

Lipids play an important role in the metabolicpathways in bivalves, in the formation of gametes(Beninger 1984) and oocytes (Gabbott 1983; Beningerand Lucas 1984). Further lipids serve as energy storageunder critical nutritional conditions (Walne 1970; Ben-inger and Lucas 1984). Zandee et al. (1980) suggestedthat lipids might be saved for gametogenesis from au-tumn until spring. A study of two populations of biv-alves (Tapes decussatus and T. philippinarum) indicatesthat reserves of glycogen may be converted to lipids forgamete utilisation prior to spawning (Beninger and Lu-cas 1984). The adductor muscle and the mantle aresuggested organs for storing complex carbohydrates(mostly glycogen), while lipids are accumulated in the

Polar Biol (2003) 26: 83–92DOI 10.1007/s00300-002-0450-3

S.-R. Birkely (&) Æ B. GulliksenNorwegian College of Fishery Science,University of Tromsø, 9037 Tromsø, NorwayE-mail: stenr@nfh.uit.no

O. Grahl-NielsenDepartment of Chemistry,University of Bergen, 5007 Bergen, Norway

digestive gland (Napolitano and Ackman 1992). An-other energy reserve, at least in adult bivalves, is protein,which can be used during gametogenesis (Beninger andLucas 1984). A limited number of reports present thespatial distribution of total lipids in bivalves (Swift et al.1980; Klingensmith and Stillway 1982; Kluytmans et al.1985; Jarzebski et al. 1986). Studies of lipids and totalfatty acids in the myid M. truncata have not been pub-lished.

Some fatty acids are known to be characteristic formarine organisms, especially the (n-3) polyunsaturatedfatty acids. In bivalves, fatty acids 20:5n3 and 22:6n3 arepredominant (Voogt 1983) and have an essential dietaryrole, together with other n-3 PUFAs (Whyte 1988).Ackman (1980) found that filter-feeding invertebrateshave phospholipids rich in n-3 PUFA.

Interpretation of data of fatty-acid composition frombiological samples poses a problem due to the largenumber of fatty acids usually analysed. By looking atone fatty acid at a time (monovariate), there is a ten-dency to focus on the most abundant ones. Multivariatetreatment techniques, such as principal componentanalysis (PCA), are powerful tools to retrieve the totalinformation from a data set, even with small variationsand large numbers of fatty acids.

The purpose of the present investigation was toconduct a chemometric study of the tissue fatty-acidcomposition in Arctic M. truncata. We wanted to studythe variation in the composition with season, amongtissues and between sexes. By determination of the fatty-acid composition of the stomach contents, the dietaryinfluence on the fatty-acid composition of the tissuesshould be elucidated. By comparison of the fatty-acidcomposition of the gonads with the composition in thevarious tissues and in the diet, we wanted furthermore tostudy the mobilisation of lipids to the gonads.

Materials and methods

Between May 1997 and July 1998, blunt gapers (Myidae; M.truncata) were collected (five collection periods, Tables 1, 2) byhand from one locality near Longyearbyen, Svalbard (78�15¢N15�20¢E). The sampling was done by SCUBA diving at depthsbetween 15 and 30 m. Mussels were wrapped in aluminium foil andfrozen at –84�C prior to dissection.

In the laboratory, the gill, foot and mantle tissue from eachindividual were dissected out. Contents from the stomach andgonad were also sampled. The sex was determined by microscopy.The tissues were carefully rinsed in distilled water, and dried onfilter paper to remove excess water. Sub-samples weighing between80 and 150 mg were transferred to 15-ml thick-walled glass tubes;0.5–1.0 ml of anhydrous methanol (2 M HCL) was added and thetubes sealed with teflon-lined screw caps, after exchanging the at-mosphere in the bottles with pure nitrogen. Methanolysis andsubsequent gas chromatography of the methyl esters of 28 fattyacids were carried out as described elsewhere (Joensen and Grahl-Nielsen 2000). To monitor the performance of the gas chromato-graph, a standard mixture of 20 fatty-acid methyl esters in knownconcentrations (GLC-68A from NuChekPrep, Elysian, Wyo.) waschromatographed for every tenth sample. The chromatograms wererecorded by the labdata system VG-Multichrome. The peaks wereidentified by comparison with retention times in the standard

mixture and by mass spectrometry. The nonmethylene-interrupteddienes (NMID) with 20 and 22 carbons, were identified by con-version of the methyl esters to picolinylesters (Harvey 1982), fol-lowed by mass spectromety. Using peak areas, empirical responsefactors relative to the fatty acid 18:0 were computed for the other19 methyl esters in the standard mixture. The response factors forthe methyl esters not present in the standard mixture were esti-mated according to their structure. The areas of the selected peakswere corrected by the response factors and converted to relativeamounts, expressing each fatty acid as percentage of the total.

The samples were then subjected to principal componentanalysis (Wold et al. 1987), using the software package Sirius(Kvalheim and Karstang 1987). The data were first normalised byexpressing the area of each methyl ester as percentage of the sum ofthe areas for all methyl esters in the sample. The normalised datawere then logarithmically transformed to avoid domination of themethyl esters present in highest amounts. In the principal compo-nent analysis, the samples were placed in a 28-dimensional space,i.e. 1 coordinate for each fatty acid. New coordinates, principalcomponents (PCs), were generated through the centre of gravity ofall samples, in the multidimensional space, in the direction of thelargest and second-largest variation of the samples. In this manner,the dimensionality was reduced from 28 to 2. The relation amongthe samples was then displayed by projecting them on the plane,spanned by the two principal components, i.e. PC1-2-plot.

The original variables, i.e. the fatty acids, are also displayed inthe PC1 versus PC2 coordinate system, resulting in so-called bi-plots. The correlation among samples and fatty acids is therebyshown. Fatty acids with high, positive or negative, values along aPC in a biplot have high importance for that PC and, consequentlyfor the position of the samples in the plot. A sample lying close to afatty acid in a biplot contains relatively higher amounts of thatfatty acid than a sample lying far from that fatty acid. A closepositioning of two fatty acids in a biplot means that they arepositively correlated, while two acids on opposite sides of the originwill be negatively correlated. The more perpendicular the directionsfrom the origin towards two fatty acids are, the less they are cor-related.

Results

The relative amounts of 28 fatty acids have been deter-mined in gill, foot and mantle tissue, in the gonads, aswell as in the stomach contents in the bivalve M. trun-cata, collected on five different occasions (Tables 1, 2). Inaddition to the large amount of numbers, the large in-dividual differences among mussels sampled at the sametime make a direct evaluation of the data in the tablefutile. There is, however, one clearcut systematic differ-ence apparent in the table: the presence of the branched16:0 acid (16:0b) in the stomach contents, while it is ab-sent from all tissues. The methyl ester of this acid elutedfrom the column just after the methyl ester of 14:0, andwas identified by mass spectrometry to 4,8,12-trimeth-yltridecanoic acid (4,8,12-TMTD) (Ackman 1989).

To interpret the results in the table, it is necessary touse multivariate statistics, which extract the compositeinformation embedded in all the variables, i.e. fatty ac-ids. A principal component analysis of the total datamatrix in Tables 1 and 2 shows that the main feature inthe data is the distinction of the samples of stomachcontents from the tissue samples (Fig. 1a). The cause forthis distinction is the presence of 16:0b in the stomachcontents while it is absent in the tissues. This is mani-fested in the biplot in Fig. 1a, where 16:0b is located

84

Table

1Fatty-acidcomposition,expressed

asmeanrelativeamountsin

percentoftotalfattyacid±

SD

a,aswellassum

saturatedfattyacids,sum

monounsaturatedfattyacids,sum

n3

polyunsaturatedfattyacidsandsum

n6polyunsaturatedfattyacids,

invarioustissues,stomach

contents

andgonadsofMyatruncata

collectedin

Isfjorden,Spitsbergen

onfive

occasions(m

male,ffemale,ndnotdetected)

FattyacidsGill

Foot

Mantle

May97

Sept97

April98

June98

July

98

May97

Sept97

April98

June98

July

98

May97

Sept97

April98

June98

July

98

n=

4n=

10

n=

9n=

6n=

10

n=

11

n=

10

n=

10

n=

6n=

10

n=

4n=

10

n=

10

n=

6n=

10

14:0

1.1±

0.1

1.2±

0.2

0.8±

0.6

1.5±

0.1

1.4±

0.3

2.0±

0.4

1.7±

0.2

1.6±

0.2

1.5±

0.7

1.6±

0.2

1.1±

0.8

1.4±

0.1

1.3±

0.5

1.4±

0.6

0.9±

0.8

15:0

0.9±

0.2

1.1±

0.2

1.2±

0.2

0.8±

0.1

0.8±

0.1

1.0±

0.1

0.9±

0.1

0.9±

0.1

0.9±

0.0

0.8±

0.1

0.7±

0.1

0.7±

0.1

0.8±

0.1

0.8±

0.0

0.8±

0.1

16:0

22.2±

0.8

24.1±

1.9

23.2±

0.9

21.9±

0.9

21.6±

0.8

19.4±

2.2

20.0±

0.7

19.4±

0.9

18.9±

1.0

19.0±

0.4

23.3±

1.1

22.6±

1.1

22.6±

1.2

21.4±

0.6

21.5±

0.9

16:0b

nd

nd

nd

nd

nd

nd

nd

nd

nd

nd

nd

nd

nd

nd

nd

17:0

1.0±

0.1

1.2±

0.1

0.2±

0.0

0.2±

0.0

0.2±

0.0

0.9±

0.1

0.9±

0.1

0.9±

0.2

0.7±

0.2

0.5±

0.1

1.0±

0.1

0.8±

0.1

0.3±

0.1

0.2±

0.1

0.2±

0.1

17:0i

3.3±

0.5

3.9±

0.8

4.2±

0.4

4.3±

1.1

3.8±

0.5

2.9±

0.6

2.9±

0.3

3.7±

0.3

3.1±

0.4

2.9±

0.2

3.2±

0.2

2.8±

0.3

3.3±

0.3

3.2±

0.6

3.0±

0.3

18:0

3.3±

0.5

4.2±

0.7

3.7±

0.4

3.5±

0.3

3.7±

0.3

4.2±

0.5

4.5±

0.7

4.7±

0.4

3.9±

0.4

4.0±

0.4

5.3±

2.2

4.7±

0.4

5.0±

0.6

4.3±

0.4

4.7±

0.4

16:1n7

2.3±

0.2

2.4±

0.3

2.5±

0.4

4.1±

0.5

3.9±

0.5

3.3±

0.5

3.0±

0.5

2.8±

0.2

3.5±

0.4

3.4±

0.3

2.8±

0.2

3.0±

0.2

2.7±

0.3

3.5±

0.2

3.6±

0.4

18:1n7

1.7±

0.2

1.7±

0.3

1.6±

0.2

2.0±

0.2

2.2±

0.3

1.3±

0.2

1.4±

0.3

1.2±

0.1

1.3±

0.2

1.4±

0.2

2.0±

0.5

1.8±

0.2

1.6±

0.2

1.7±

0.3

1.8±

0.2

18:1n9

2.8±

0.3

3.0±

1.1

4.0±

0.5

3.4±

0.4

3.6±

0.3

3.2±

0.3

3.3±

0.9

4.0±

0.3

4.0±

0.4

3.9±

0.4

4.5±

1.5

3.6±

0.3

5.1±

0.3

4.7±

0.4

4.5±

0.3

18:1n11

0.5±

0.1

0.7±

0.2

0.7±

0.2

0.5±

0.1

0.5±

0.1

0.4±

0.1

0.4±

0.1

0.4±

0.1

0.4±

0.1

0.3±

0.1

0.3±

0.0

0.4±

0.1

0.4±

0.1

0.5±

0.2

0.4±

0.1

20:1n7

1.9±

0.1

2.0±

0.3

2.3±

0.2

1.8±

0.2

1.7±

0.2

2.5±

0.3

2.4±

0.3

2.9±

0.2

2.5±

0.2

2.4±

0.2

1.8±

0.1

2.3±

0.3

2.5±

0.3

2.3±

0.2

2.2±

0.2

20:1n9

3.5±

0.1

3.4±

0.4

3.7±

0.4

3.1±

0.1

3.2±

0.3

4.1±

0.3

4.1±

0.4

4.6±

0.4

4.4±

0.4

4.3±

0.3

4.0±

0.5

4.4±

0.4

4.7±

0.6

4.3±

0.3

4.2±

0.4

20:1n11

3.8±

0.7

3.8±

0.4

3.4±

0.4

2.4±

0.4

2.9±

0.3

2.2±

0.3

2.9±

0.5

2.0±

0.2

2.3±

0.4

2.7±

0.3

2.3±

0.2

2.3±

0.4

2.0±

0.4

1.8±

0.3

2.1±

0.2

22:1n7

0.8±

0.2

1.2±

0.4

1.0±

0.3

1.5±

0.6

1.0±

0.3

0.4±

0.1

0.4±

0.1

0.4±

0.3

0.4±

0.1

0.4±

0.1

0.5±

0.1

0.6±

0.2

0.5±

0.2

0.8±

0.4

0.6±

0.2

18:4n3

1.6±

0.2

1.1±

0.2

1.4±

0.3

1.2±

0.2

1.5±

0.2

1.3±

0.3

0.9±

0.2

1.2±

0.4

1.2±

0.1

1.3±

0.2

1.5±

0.2

1.3±

0.1

1.5±

0.2

1.3±

0.1

1.3±

0.1

20:4n3

0.4±

0.1

0.4±

0.2

0.2±

0.1

0.3±

0.0

0.3±

0.1

0.3±

0.1

0.4±

0.1

0.2±

0.0

0.2±

0.0

0.3±

0.0

0.5±

0.2

0.4±

0.1

0.2±

0.0

0.2±

0.1

0.2±

0.0

20:5n3

11.3±

1.0

9.0±

1.1

8.6±

1.2

13.0±

0.3

12.8±

0.8

11.5±

1.3

10.6±

0.9

10.0±

0.7

11.6±

0.8

11.9±

0.5

9.9±

1.1

9.5±

0.8

8.2±

0.8

10.7±

0.6

10.8±

0.5

21:5n3

2.7±

0.5

2.3±

0.3

2.3±

0.4

2.1±

0.2

2.1±

0.3

2.0±

0.2

2.0±

0.2

1.8±

0.3

2.1±

0.3

2.1±

0.2

2.2±

0.2

2.1±

0.2

2.5±

1.7

2.1±

0.3

2.0±

0.2

22:4n3

0.6±

0.2

0.6±

0.1

0.6±

0.2

0.5±

0.1

0.5±

0.1

0.6±

0.2

0.6±

0.2

0.6±

0.1

0.6±

0.1

0.5±

0.1

0.5±

0.1

0.6±

0.1

0.6±

0.1

0.6±

0.1

0.6±

0.1

22:5n3

1.6±

0.1

2.1±

2.0

1.6±

0.1

1.4±

0.2

1.4±

0.1

1.7±

0.1

2.8±

2.9

1.9±

0.2

1.8±

0.2

1.7±

0.1

1.7±

0.3

1.9±

0.1

1.8±

0.1

1.7±

0.2

1.7±

0.2

22:6n3

20.0±

0.5

16.6±

2.2

17.6±

2.0

17.0±

1.3

17.7±

0.9

22.3±

2.5

21.2±

2.0

21.3±

1.3

20.5±

1.4

20.6±

1.2

19.4±

2.5

20.0±

1.4

18.4±

1.4

18.1±

2.0

18.8±

1.0

18:2n6

0.7±

0.7

0.1±

0.0

1.9±

0.4

1.7±

0.2

1.8±

0.2

0.6±

0.6

0.1±

0.0

1.7±

0.3

1.9±

0.2

1.8±

0.3

0.4±

0.6

0.1±

0.0

2.3±

0.2

2.2±

0.2

2.0±

0.3

20:2n6

1.0±

0.1

1.6±

1.3

1.0±

0.1

0.9±

0.1

0.9±

0.0

0.9±

0.1

1.1±

0.1

0.9±

0.1

0.8±

0.0

0.8±

0.1

1.1±

0.1

1.3±

0.2

1.0±

0.1

0.9±

0.1

0.9±

0.1

20:4n6

3.2±

0.4

3.0±

0.4

3.0±

0.3

2.8±

0.4

2.6±

0.4

3.7±

0.4

3.3±

0.6

3.3±

0.3

3.3±

0.3

3.2±

0.4

2.5±

0.3

2.8±

0.3

2.7±

0.3

2.9±

0.4

2.9±

0.2

22:4n6

1.0±

0.2

1.2±

0.4

1.1±

0.2

1.1±

0.1

1.0±

0.1

1.2±

0.3

1.5±

0.5

1.2±

0.4

1.5±

0.2

1.4±

0.2

1.0±

0.2

1.2±

0.1

1.1±

0.1

1.2±

0.1

1.2±

0.1

20:2

D5,11b

6.3±

0.3

7.3±

0.9

7.4±

0.6

6.2±

0.5

6.2±

0.7

5.6±

0.4

5.8±

0.8

5.7±

0.4

5.4±

0.4

5.5±

0.3

5.5±

0.7

6.5±

0.6

6.1±

0.4

6.3±

0.4

6.5±

0.5

22:2

D7,13b

0.6±

0.1

0.6±

0.1

0.7±

0.2

0.6±

0.1

0.6±

0.1

0.8±

0.2

1.0±

0.2

0.8±

0.2

1.1±

0.1

1.2±

0.3

0.9±

0.1

0.8±

0.2

0.9±

0.2

0.8±

0.1

0.8±

0.1

SSFA

32

36

33

32

32

30

31

31

29

29

35

33

33

31

31

SMUFA

17

18

19

19

19

17

18

18

19

19

18

18

20

20

19

Sn3PUFA

38

32

32

36

36

40

39

37

38

38

36

36

33

35

35

Sn6PUFA

66

76

66

67

87

55

77

7

aThepercentages

andSD

ofthefattyacidsare

allgiven

withonedecim

al,even

ifinsignificantfiguresin

somecasesare

included

inthenumbers

bThenon-m

ethylene-interruptedfattyacid20:2

D5,11ispositivelyidentified,while22:2

D7,13istentativelyidentified

85

Table

2Fatty-acidcomposition,expressed

asmeanrelativeamountsin

percentoftotalfattyacid±

SD

a,aswellassum

saturatedfattyacids,sum

monounsaturatedfattyacids,sum

n3

polyunsaturatedfattyacidsandsum

n6polyunsaturatedfattyacids,

invarioustissues,stomach

contents

andgonadsofMyatruncata

collectedin

Isfjorden,Spitsbergen

onfive

occasions(m

male,ffemale,ndnotdetected)

FattyacidsStomach

Gonad

May97

Sept97

April98

June98

July

98

May97

Sept97

April98

June98

July

98

n=

7n=

10

n=

10

n=

6n=

10

n=

2n=

3n=

5n=

5n=

6n=

2n=

3n=

3n=

7n=

3m

fm

fm

fm

fm

f

14:0

4.6±

1.0

1.7±

0.5

0.8±

0.9

2.8±

0.2

2.6±

0.2

1.3±

0.1

2.5±

0.5

1.3±

0.3

3.2±

0.4

0.8±

0.6

2.9±

0.2

0.9±

0.7

2.9±

0.4

0.9±

0.6

3.0±

0.1

15:0

0.4±

0.1

0.6±

0.2

1.2±

0.3

0.4±

0.1

0.3±

0.0

0.4±

0.1

0.3±

0.0

0.4±

0.0

0.3±

0.0

0.6±

0.1

0.5±

0.0

0.5±

0.1

0.4±

0.1

0.5±

0.0

0.4±

0.1

16:0

20.3±

3.7

18.6±

2.4

23.4±

2.7

15.0±

0.9

15.3±

0.6

21.3±

0.2

19.7±

1.5

21.9±

0.9

20.3±

0.9

22.4±

1.8

20.6±

0.5

22.1±

2.4

20.0±

0.7

21.5±

0.9

19.8±

1.2

16:0b

3.0±

1.4

2.6±

0.8

2.3±

0.7

3.4±

0.6

2.9±

0.6

nd

nd

nd

nd

nd

nd

nd

nd

nd

nd

17:0

0.6±

0.2

0.8±

0.2

0.1±

0.1

0.8±

0.2

0.7±

0.1

0.8±

0.1

0.4±

0.1

0.8±

0.1

0.4±

0.0

0.1±

0.0

0.1±

0.0

0.1±

0.1

0.2±

0.0

0.1±

0.0

0.2±

0.0

17:0i

1.1±

0.2

1.5±

0.3

2.7±

0.4

1.1±

0.3

1.0±

0.2

1.6±

0.1

0.7±

0.2

2.0±

0.6

0.9±

0.2

2.3±

0.9

1.7±

0.5

2.5±

1.5

1.0±

0.2

1.8±

0.3

0.9±

0.1

18:0

3.8±

0.8

4.0±

1.0

5.1±

0.4

2.9±

0.5

2.9±

0.4

4.4±

0.5

2.3±

0.9

4.5±

0.7

2.2±

0.4

4.8±

0.2

2.2±

0.3

4.1±

0.4

2.0±

0.3

4.4±

0.4

1.8±

0.1

16:1n7

8.8±

0.9

6.3±

2.5

3.8±

0.9

18.1±

3.2

17.4±

2.6

4.3±

0.9

9.5±

1.3

3.9±

0.8

8.8±

0.9

3.7±

0.7

7.5±

2.4

4.6±

0.4

9.6±

1.2

4.4±

0.5

10.1±

0.6

18:1n7

4.6±

0.6

3.8±

1.1

2.8±

0.6

5.1±

0.9

5.4±

0.7

2.8±

0.7

6.0±

0.7

2.9±

0.6

4.7±

0.7

2.9±

0.7

3.4±

0.3

2.7±

0.7

4.5±

0.7

2.7±

0.2

4.7±

0.5

18:1n9

2.3±

0.3

2.7±

0.8

3.9±

0.6

1.7±

0.3

1.8±

0.3

1.7±

0.0

6.0±

1.5

1.8±

0.2

5.4±

0.9

2.7±

1.0

4.6±

0.3

2.2±

0.6

5.3±

1.1

2.3±

0.2

5.4±

1.0

18:1n11

0.6±

0.1

1.0±

0.2

1.0±

0.2

0.3±

0.1

0.4±

0.1

0.3±

0.1

1.3±

0.1

0.4±

0.1

1.2±

0.2

0.3±

0.1

1.1±

0.2

0.3±

0.1

1.0±

0.1

0.3±

0.0

1.0±

0.3

20:1n7

2.1±

0.3

2.0±

0.3

2.7±

0.3

1.8±

0.3

1.8±

0.2

2.9±

0.0

1.7±

0.5

2.5±

0.2

1.5±

0.2

2.7±

0.3

1.7±

0.0

2.4±

0.8

1.7±

0.2

2.4±

0.5

1.4±

0.1

20:1n9

2.3±

0.2

2.7±

0.4

4.4±

0.8

1.4±

0.4

1.4±

0.3

3.3±

0.6

2.0±

0.5

3.4±

0.3

2.1±

0.2

3.5±

0.4

2.1±

0.4

3.6±

0.5

2.0±

0.2

3.3±

0.3

1.8±

0.2

20:1n11

2.3±

0.6

2.7±

0.7

3.8±

0.8

0.9±

0.3

1.2±

0.4

1.4±

0.2

2.0±

0.2

1.6±

0.4

1.9±

0.3

1.5±

0.4

2.4±

0.2

1.1±

0.2

1.9±

0.2

1.3±

0.4

1.7±

0.1

22:1n7

0.4±

0.1

0.6±

0.3

0.9±

0.3

0.5±

0.3

0.3±

0.1

0.3±

0.0

0.2±

0.1

0.4±

0.2

0.2±

0.0

0.4±

0.4

0.3±

0.1

0.6±

0.8

0.4±

0.2

0.3±

0.1

0.2±

0.1

18:4n3

8.0±

1.5

2.5±

1.1

1.3±

0.2

3.3±

0.7

3.1±

0.4

1.5±

0.2

2.5±

0.9

0.7±

0.4

2.8±

0.2

0.9±

0.3

2.3±

0.5

0.9±

0.1

2.3±

0.0

1.0±

0.1

2.1±

0.1

20:4n3

1.9±

0.3

1.2±

0.4

0.6±

0.2

0.9±

0.1

0.8±

0.1

0.9±

0.1

1.0±

0.1

0.8±

0.2

1.1±

0.1

0.7±

0.3

0.7±

0.0

0.6±

0.2

0.8±

0.1

0.6±

0.1

0.8±

0.0

20:5n3

12.7±

3.7

15.9±

1.9

10.0±

2.9

23.9±

2.1

24.0±

1.1

21.5±

1.9

22.2±

1.0

19.9±

2.0

20.6±

1.7

18.4±

3.8

21.4±

0.4

20.9±

3.1

22.6±

1.5

22.7±

1.5

23.3±

2.1

21:5n3

1.8±

0.3

1.8±

0.3

1.9±

0.2

1.3±

0.2

1.2±

0.1

2.0±

0.1

1.5±

0.1

1.6±

0.1

1.5±

0.2

1.5±

0.1

1.4±

0.0

1.8±

0.1

1.6±

0.3

1.6±

0.2

1.6±

0.1

22:4n3

0.2±

0.1

0.6±

0.3

0.5±

0.2

0.2±

0.0

0.2±

0.0

0.4±

0.0

0.2±

0.0

0.4±

0.1

0.2±

0.0

0.4±

0.1

0.3±

0.1

0.3±

0.1

0.2±

0.1

0.4±

0.1

0.2±

0.0

22:5n3

0.7±

0.2

1.0±

0.3

1.0±

0.2

0.6±

0.1

0.7±

0.1

1.1±

0.3

0.7±

0.3

1.0±

0.2

0.6±

0.1

1.0±

0.1

0.8±

0.2

0.9±

0.0

0.7±

0.2

1.1±

0.1

0.6±

0.0

22:6n3

10.8±

2.4

16.5±

4.3

13.2±

3.8

9.6±

1.6

10.3±

1.5

19.5±

1.2

11.7±

1.6

21.1±

1.6

13.9±

1.1

20.8±

2.1

14.9±

1.1

20.2±

1.3

12.5±

1.5

20.1±

1.6

13.3±

0.3

18:2n6

1.1±

0.1

0.5±

0.2

1.8±

0.4

0.6±

0.2

0.7±

0.2

0.8±

0.1

0.8±

0.0

0.2±

0.1

0.7±

0.1

1.2±

0.6

0.9±

0.2

1.0±

0.4

0.7±

0.1

1.1±

0.1

0.7±

0.0

20:2n6

2.1±

0.2

2.2±

0.3

1.4±

0.3

0.8±

0.1

0.9±

0.1

1.6±

0.4

1.6±

0.1

1.8±

0.3

1.8±

0.1

1.6±

0.4

1.2±

0.2

1.2±

0.4

1.2±

0.2

1.2±

0.2

1.3±

0.2

20:4n6

1.0±

0.2

1.7±

0.6

2.0±

0.3

0.8±

0.3

0.7±

0.2

1.5±

0.3

0.8±

0.2

1.6±

0.2

0.9±

0.1

1.6±

0.2

1.2±

0.1

1.6±

0.4

1.1±

0.5

1.6±

0.1

0.9±

0.1

22:4n6

0.3±

0.1

0.5±

0.3

0.7±

0.1

0.3±

0.1

0.2±

0.1

0.4±

0.1

0.2±

0.0

0.5±

0.1

0.3±

0.1

0.4±

0.1

0.3±

0.0

0.5±

0.2

0.3±

0.1

0.5±

0.2

0.3±

0.0

20:2

D5,11b

2.1±

0.2

3.5±

0.8

5.8±

1.2

1.4±

0.7

1.5±

0.5

2.0±

0.3

2.1±

0.3

2.3±

0.8

2.5±

0.3

2.3±

1.5

3.4±

0.6

2.1±

1.2

2.9±

0.7

1.8±

0.8

2.4±

0.4

22:2

D7,13b

0.3±

0.1

0.4±

0.1

0.7±

0.3

0.2±

0.1

0.2±

0.1

0.3±

0.1

0.2±

0.1

0.2±

0.1

0.1±

0.0

0.3±

0.1

0.2±

0.0

0.2±

0.1

0.2±

0.1

0.2±

0.1

0.2±

0.0

SSFA

34

30

36

26

26

30

26

31

27

31

28

30

27

226

SMUFA

23

22

23

30

30

17

29

17

26

18

23

17

26

17

26

Sn3PUFA

36

40

29

40

40

47

40

46

41

44

42

46

41

47

42

Sn6PUFA

45

63

34

34

35

44

34

3

aThepercentages

andSD

ofthefattyacidsare

allgiven

withonedecim

al,even

ifinsignificantfiguresin

somecasesare

included

inthenumbers

bThenon-m

ethylene-interruptedfattyacid20:2

D5,11ispositivelyidentified,while22:2

D7,13istentativelyidentified

86

close to the samples of the stomach contents and farfrom the other samples, while all the other fatty acids areclustered in the vicinity of the tissue samples.

The large difference in 16:0b between the stomachsamples and the other samples subdues the effect theother fatty acids might have on the difference amongthe samples, as they are clustered tightly together in thebiplot in Fig. 1a. Therefore, a recomputation of the datawithout 16:0b was carried out, resulting in quite a dif-ferent biplot (Fig. 1b). The samples from the femalegonads appear together with the stomach samples in thisplot, while the samples from the male gonads show upcloser to the tissue samples. The difference between themale and female gonads is even more evident when aPCA is carried out on the gonad samples alone (Fig. 2).The fatty acids 16:1n7, 18:1n9, 18:4n3, 18:1n7, 18:1n11,14:0 and 18:4n3 are more abundant in the female than inthe male gonads, while 18:0, 17:0i, 22:6n3, 20:1n9,20:1n7 and 22:4n6 are most abundant in the male go-nads (Fig. 2, Tables 1, 2).

Differences in the fatty-acid composition of thestomach contents from the five collection periods wereevident. These differences were studied more closely bycomputing the stomach samples alone (Fig. 3). Aclearcut distinction in four groups is evident, but thebivalves sampled in June and July had non-distinguish-able stomach contents. The fatty acids 18:4n3, 20:4n3and 20:2n6 are located close to the samples from May1997 in the biplot in Fig. 3. Likewise, the fatty acids15:0, 17:0i, 22:1n7 and 22:4n6 are close to the samplesfrom April 1998, and the acids 16:1n7 and 20:5n3 areclose to the samples from June and July 1998. Thesefatty acids occur in higher relative amounts in thesamples from the respective collection periods (Tables 1,2), and are the ones most responsible for the distinctionamong the groups of samples.

The large differences in the stomach contents amongthe various collection times are not reflected in the fatty-acid composition of the tissues, as displayed in Fig. 1.However, a possible effect of the changes in diet cannot

Fig. 1a, b Biplot of samples oftissue (gill, foot and mantle),stomach and gonads, togetherwith fatty acids in a coordinatesystem made by the first andsecond principal component.The plot in a is based on all 28fatty acids, while the branchedfatty acid 16:0b (4,8,12-TMTD)has been omitted from thecomputations in b. The indi-vidual samples are not shown inthe plots, but all samples fromthe respective tissues are locatedinside the encircled areas. Thesamples from the male andfemale gonads, respectively, arelocated inside the areas encir-cled by dashed lines. The per-centage of the total variationamong the samples that occuralong PC1 and PC2, respec-tively, is given

87

be ruled out on the basis of these plots, since all sampleshere mutually influence their positions. Possible inter-actions of stomach contents on the tissue compositionwere therefore investigated by projecting the tissuesamples onto the PC-plot of the stomach samples.The shaded area in Fig. 3 represents the samplesfrom the gill, foot and mantle tissues. They all grouptogether with the stomach samples from April 1998.There are no temporal differences among the foot andmantle samples. A small temporal variation, although

insignificant, is, however, present among the gill sam-ples; i.e. the samples from June/July 1998 are displacedsomewhat in the direction of the stomach samples fromthe same period, and the samples from May 1997 aredisplaced somewhat in the direction of the correspond-ing stomach samples (Fig. 3).

Even if there is no apparent difference between thesamples of the gill, foot and mantle tissues when they areplotted together with the samples of the stomach con-tents (Fig. 3), there is a systematic difference amongthem on a smaller scale when they are subjected toseparate PCA (Fig. 4). The gill tissue, on one side, isdistinct from the foot and mantle tissue on the other.This is mainly caused by the relatively higher levels of22:1n7 in the gill tissue. The monoenes 18:1n11 and20:1n11 are also systematically more abundant in the gilltissue, while 22:2n9 has higher abundance in the footand mantle tissues. Overall, the patterns of the fattyacids from the five collection periods are similar (Fig. 4).On an even smaller scale, there is also a distinctionbetween the foot and mantle samples.

Fig. 2 Biplot of the gonadsamples and fatty acids. Theindividual samples, not shown,are located within the encircledareas

Fig. 3 Biplot of samples of stomach contents from the differentsampling periods and fatty acids. The individual samples, notshown, are located within the encircled areas. The shaded areashows where the gill, foot and mantle samples from all samplingperiods are located when they are plotted into the coordinatesystem of the first and second principal component, generated fromthe PCA of the samples of stomach contents. Gill samples from thefour sampling periods are encircled separately; May 1997 sampleswithin the encircled area marked 1, September 1997 samples withinthe encircled areamarked 2, April 1998 samples within the encircledarea marked 3, and June and July 1998 samples within the encircledarea marked 4

88

Discussion

The single most prominent feature in the data is thepresence of the branched fatty acid, 4,8,12-trimethyltri-decanoic acid, in the stomach contents, in significantamounts between 2.2 and 3.4% (Tables 1, 2). It was notdetected in the tissues. This branched fatty acid has beenfound in many molluscs (Joseph 1982). For example,Ackman et al. (1971) detected the 4,8,12-TMTD in highamounts in the mud snail, Littorina littorea. It was

postulated that the mud snail efficiently degrades phytol,pristanic and phytanic acid, arising from metabolism ofdietary chlorophyll, to 4,8,12-TMTD. This fatty acidwas detected in the foot and gonad tissues of the gas-tropod Patella peroni, and in the foot, gills and vicera ofthe chiton Pionerplax costata (Johns et al. 1980). Ittherefore appears that M. truncata, belonging to anotherclass (Bivalvia) than Patella peroni (Gastropoda) andPionerplax costata (Polyplacaphora), has a more selec-tive metabolism: Even if dietary chlorophyll is degradedto 4,8,12-TMTD in the stomach, it is not incorporated

Fig. 4 Biplots of samples of gill, foot and mantle tissues and offatty acids from bivalves sampled in a May 1997, b September1997, c April 1998, d June 1998, e July 1998. The individualsamples, not shown, are located within the encircled areas

89

into the tissues. This is in accordance with a findingthat oysters, Ostrea edulis, do not accumulate in thetissues certain algal fatty acids, 16:3n3 in particular,from the diet, indicating a species-oriented metabolism(Watanabe and Ackman 1974).

Other fatty acids with uncommon structures detectedin the tissues are the 20 and 22 carbon NMIDs. The firstwas positively identified as 20:2D5,11 by mass spect-rometry of its picolinyl ester. The second was present inamounts too small for certain identification, but is ten-tatively identified as 20:2D7,13, arising from chainelongation of the former. These NMIDs were detectedearlier in the mud snail and the moon snail (Lunatiatriseriata) (Ackman and Hooper 1973) and in the oyster,Crassostrea vireginica (Paradis and Ackman 1975). Laterthey were found in many molluscs, echinoderms andcrustaceans (Joseph 1982). They may be formed by D5desaturation of 20:1n9 to give 20:2D5,11 and chainelongation of this to 22:2D7,13 (Ackman and Hooper1973). These fatty acids may be more useful in eluci-dating food-web relationships than the more commonlyobserved fatty acids. Therefore, it would have been in-teresting to see if they are present in the tissues of wal-ruses that prey on M. truncata.

The energy transfer as lipids and the role of carbo-hydrates/glycogen and proteins during gametogenesisare presented in a number of papers (Swift et al. 1980;Zandee et al. 1980; Gabbott 1983; Beninger 1984;Beninger and Lucas 1984; Kluytmans et al. 1985;Jarzebski et al. 1986; Napolitano and Ackman 1992,1993). Developing gametes is an energy-demandingprocess. It requires mobilisation of nutrients eitherdirectly from the stomach and ingested food or from thestorage and subsequent use of reserves from the sur-rounding somatic tissue. Vassallo (1973) documentedtransport of lipid material from the digestive gland tothe gonads in females of Chlamys hericia. Zandee et al.(1980) studied Mytilus edulis and found that, during thewinter, protein content exceeds that of glycogen, and itwas suggested that lipids are saved for gametogenesisfrom autumn until spring, whereas protein and glycogenare presumably used for both energy production andgametogenesis. Kluytmans et al. (1985) indicated thatthe fluctuations in the contents of lipids are rather smallcompared to the changes in protein and glycogen.Nevertheless, lipids have important functions both asstructural constituents in membranes and in metabolism.Glycogen reserves (and possible proteins) are convertedto lipids prior to and during gametogenesis, causing thesmall lipid-content fluctuation (Beninger and Lucas1984; Kluytmans et al. 1985).

Our investigation reveals a significant, season-inde-pendent difference in fatty-acid composition betweenfemale and male gonads (Tables 1, 2, Fig. 2). Sexualdifferences have also been observed in the gonads of thegastropod Subninella undulata (Johns et al. 1980) and inthe gonads of Mytilus californianus (Rodegker andNevenzel 1964). However, the differences occurred indifferent fatty acids in the two species. The sexual

difference in Mya truncata is mainly due to higher rel-ative amounts of 14:0 and the monoenic acids with 16and 18 carbons in the female gonads, and higher relativeamounts of polyunsaturated fatty acids in the male go-nads. These differences show some similarity with thedifference in S. undulata. However, the gonadal differ-ences appear to be species dependent.

As indicated in Fig. 1b, the fatty-acid composition ofthe female gonads reveals a somewhat closer resem-blance to the composition of the stomach contents thanthe composition of the male gonads do. However, thecomposition of the male gonads is more like the com-position in the tissues. This suggests that the lipids in thefemale gonads are mobilised more directly from thestomach contents than the lipids in the male gonads.Thus, the female gonad and egg production are moredependent on reliable food resources than male gonadsare. The benefit of this feature lies in the possibility ofhaving large amounts of lipids from blooming phyto-plankton available for developing female gonads. Thenthe fatty-acid composition in the female gonads may bemore influenced by the composition in the diet than thecomposition in the male gonads.

In Placopecten magellanicus, the amount of triacyl-glycerols in the lipids in the female gonads was ap-proximately 30% versus approximately 2% in the lipidsof the male gonads (Napolitano and Ackman 1992).However, the level of phospholipids was very high, ap-proximately 85% in the male gonads. Triacylglycerolsare mainly used for energy storage and may be moredirectly mobilised from the diet, and are less unsaturatedthan the structurally more determined phospholipids.Our findings for Mya truncata are in accordance withthis.

Tissue fatty acids in benthic animals are dominatedby the polyunsaturated fatty acids 20:5n3 and 22:6n3(Ben-Mlih et al. 1992), in addition to the saturated fattyacid 16:0 (Graeve et al. 1997). This is also the case formolluscan bivalves (Voogt 1983). These acids are pre-dominant in phospholipids, an important membraneconstituent (Graeve et al. 1997). Similarly, earlierstudies have shown that the phospholipids of filter-feeding benthic invertebrates are enriched in n3 PUFA(Ackman 1980). These fatty acids originate fromorganic material from the photic zone, i.e. from phy-toplanktonic primary production (Ben-Mlih et al. 1992).This illustrates an aspect of an ecosystem in a fjord:abundance of n3 PUFA in phytoplankton is ‘‘trans-ferred’’ to herbivorous animals, including benthic her-bivores (Bell and Sargent 1985). Many marine animalshave very limited abilities to chain-elongate and desat-urate C18 n3 PUFA to long-chained C20 and C22PUFA (Dunstan et al. 1996). This means that theseanimals must obtain 20:5n3 and 22:6n3 from the diet,i.e. they are ‘‘essential fatty acids’’.

Mya truncata is a filter feeder (and a suspensionfeeder), and the stomach content reflects what has beenprecipitated in the water column, i.e. phytoplankton butalso zooplankton. The two fatty acids, 20:5n3 and

90

22:6n3, are associated with different classes of phyto-plankton: the former, plus 16:1n7, is typical for the di-atoms, and the latter, plus 18:4n3, is typical for thedinoflagellates (Graeve et al. 1994, 1997; Falk-Petersenet al. 1998; Scott et al. 1999). In June and July 1998,there were higher levels of 20:5n3 and 16:1n7 in the diet,similar to ice algae/diatoms. At that time the samplingsite had recently been covered with ice. The elevatedamount of 18:4n3 in the May 1997 collection indicatesthe uptake of the more pelagic flagellate phytoplankton.This corresponds to the ice-free conditions that had beenat the sampling site at that time. This indicates the fol-lowing change in availability of dietary fatty acids overthe year: (1) from a spring state, i.e. newly ice-free water,with diatomic 20:5n3 and 16:1n7; (2) via summerand autumn with flagellatic fatty acids 18:4n3 and20:4n3; (3) to a winter state with ice cover and low lightconditions, containing a predominance of 15:0, 17:0i,22:1n7, 22:4n6 and 22:2D7,13. These indications areconfirmed by several studies on the fatty-acid composi-tion in diatoms, dinoflagellates and bacteria (Ackmanet al. 1968; Chuecas and Riley 1969; Claustre et al. 1989;Napolitano et al. 1990).

The foot, mantle and gill tissues ofMya truncata havemutually different fatty-acid compositions (Fig. 4). Butthey are temporally stable compared to the rather largefluctuations in the composition of the stomach contents(Fig. 3). This indicates a controlled selection and/ortransformation of fatty acids from the time of digestion,until they are incorporated into the different tissues. Thisdoes not correspond with the suggestion by Graeve et al.(1994) that Arctic herbivorous copepods respond tochanges in fatty-acid composition in the diet with asimilar change in their own fat.

The diet-independent composition of the tissue fattyacids in Mya truncata is in contrast to the disclosures byGraeve et al. (1997) that the fatty acids of the availablefood sources dominated the fatty-acid composition ofbenthic species from the shelves off Greenland, Spits-bergen and the western Barents Sea. Graeve et al. (1997)found that the influence of diet masked potential taxon-specific patterns. It thus appears that molluscs havemore species-dependent composition of tissue fatty acidsthan anthozoans, polychaetes, pantopods, crustaceansand echinoderms, studied by Graeve et al. (1997).

Beninger and Stephan (1985) also found that changesin total lipid fatty-acid composition of marine inverte-brates, earlier expected to be a response to temperaturevariation, were caused by dietary shifts. They did not,however, study any molluscs. Saether et al. (1986)studied krill, and found that storage lipids (triacylgly-cerols) are far more receptive to dietary shifts than tis-sues dominated by structural lipids.

However, despite the large differences between dietand the tissues that we detected, a certain influence ofthe diet upon the composition of tissues cannot becompletely ruled out. The ratios between some fattyacids may be maintained, even if the absolute amount ischanged. Thus, it appears that the large variance in

fatty-acid composition in the stomach samples results ina small shift in the fatty-acid composition of the gilltissue (Fig. 3).

In conclusion, sexual differences in fatty-acid com-position of the gonads are evident, as well as the sourceof energy for gonadal products. The male gonads have afatty-acid composition that resembles the compositionof the tissues, while female gonads have a fatty-acidcomposition that is more similar to the composition ofthe diet. The fatty-acid composition in the stomachcontents can be utilised for investigating the content ofthe bottom water, since indicator fatty acids can betraced to lower trophic levels. The branched fatty acid4,8,12-TMTD in the stomach contents is indicative of aphytoplanktonic diet. This acid is not incorporated inthe tissues. Dietary shifts were evident through the sea-sons. These shifts are not ‘‘reflected’’ in the foot ormantle tissues, and only to a minor degree in the gilltissue. This shows that a controlled selection and/ortransformation of fatty acids from time of ingestionuntil incorporation into the tissues is taking place. Thisstudy demonstrates the advantage of multivariate sta-tistical treatment of large data matrices. However, tofully understand the metabolism and fluctuation of en-ergy as lipids in Mya truncata, a closer study of themajor lipid classes in various tissues, gonads and stom-ach content is needed.

Acknowledgements The following three institutions were involvedin accomplishment of this research: The University Courses onSvalbard, The Norwegian College of Fishery Science (University ofTromsø) and Department of Chemistry (University of Bergen).Thanks are also due to Jussie Evertsen and Ole Jørgen Lønne fordiving assistance, and Terje Lygre for technical assistance. Thisresearch received financial support from TotalFinaElf ExplorationNorge AS.

References

Ackman RG (1980) Fish lipids. In: Connell J (ed) Advances in fishscience and technology. Fishing News Books, Farnham, pp 86–103

Ackman RG (1989) Lipids. In: Ackman RG (ed) Marine biogeniclipids, fats, and oils. CRC Press, Boca Raton, Fla, pp 123–125

Ackman RG, Hooper SN (1973) Non-methylene-interrupted fattyacids in lipids of shallow-water marine invertebrates: a com-parison of two molluscs (Littorina littorea and Lunatia triseri-ata) with the sand shrimp (Crangon septemspinosus). CompBiochem Physiol 46B:153–165

Ackman RG, Tocher CS, McLachlan J (1968) Marine phyto-plankter fatty acids. J Fish Res Board Can 25:1603–1620

Ackman RG, Hooper SN, Ke PJ (1971) The distribution of satu-rated and isoprenoid fatty acids in the lipids of three species ofmolluscs, Littorina littorea, Crassostrea virginica and Venusmercenaria. Comp Biochem Physiol 39B:579–587

Bell M, Sargent J (1985) Fatty acid analysis of phosphoglyceridesfrom tissues of the clam Chlamys islandica (Muller) and thestarfish Ctenodiscus crispatus (Retzius) from Balsfjorden,Northern Norway. J Exp Mar Biol Ecol 87:31–40

Beninger P (1984) Seasonal variations of the major lipid classes inrelation to the reproductive activity of two species of clamraised in a common habitat: Tapes decussatus L. (Jeffreys, 1863)and T. philippinarum (Adam & Reeve, 1850). J Exp Mar BiolEcol 79:79–90

91

Beninger P, Lucas A (1984) Seasonal variations in condition, re-productive activity and gross biochemical composition of twospecies of adult clam reared in a common habitat: Tapesdecussatus (Jeffreys) and Tapes philippinarum (Adam & Reeve).J Exp Mar Biol Ecol 79:19–37

Beninger P, Stephan G (1985) Seasonal variation in the fatty acidsof the triacylglycerols and phospholipids of two populations ofadult clam (Tapes decussatus L. and T. philippinarum) reared ina common habitat. Comp Biochem Physiol 81B:591–601

Ben-Mlih F, Marty J-C, Fiala-Medioni A (1992) Fatty acid com-position in deep hydrothermal vent symbiotic bivalves. J LipidRes 33:1797–1806

Brattegard T, Holthe T (1997) Distribution of marine, benthicmacro-organisms in Norway: a tabulated catalogue. Director-ate for Nature Management, Trondheim

Chuecas L, Riley J (1969) Component fatty acids of the total lipidsof some marine phytoplankton. J Mar Biol Assoc UK 49:97–116

Claustre H, Marty J, Cassiani L, Dagaut J (1989) Fatty acid dy-namics in phytoplankton and microzooplankton communitiesduring a spring bloom in coastal Ligurian Sea: ecological im-plications. Mar Microb Food Webs 3:51–66

Dunstan GA, Baillie HJ, Barrett SM, Volkman JK (1996) Effect ofdiet on the lipid composition of wild and cultured abalone.Aquaculture 140:115–127

Falk-Petersen S, Hopkins C, Sargent J (1990) Trophic relationshipsin the pelagic, arctic food web. In: Barnes M, Gibson R (eds)Trophic relationships in the marine environment. AberdeenUniversity Press, Aberdeen, pp 315–333

Falk-Petersen S, Sargent JR, Henderson J, Hegseth EN, Hop H,Okolodkov YB (1998) Lipids and fatty acids in ice algae andphytoplankton from the Marginal Ice Zone in the Barents Sea.Polar Biol 20:41–47

Fisher KI, Stewart REA (1997) Summer foods of Atlantic walrus,Odobenus rosmarus, in northern Foxe basin, northwest territo-ries. Can J Zool 75:1166–1175

Gabbott PA (1983) Developmental and seasonal metabolic activi-ties in marine molluscs. In: Hochachka P (ed) The Mollusca,vol 2. Environmental biochemistry and physiology. AcademicPress, New York, pp 165–217

Graeve M, Kattner G, Hagen W (1994) Diet-induced changes inthe fatty-acid composition of arctic herbivorous copepods –experimental evidence of trophic markers. J Exp Mar Biol Ecol182:97–110

Graeve M, Kattner G, Piepenburg D (1997) Lipids in Arctic ben-thos: does the fatty acid and alcohol composition reflect feedingand trophic interactions? Polar Biol 18:53–61

Gulliksen B, Palerud R, Brattegard T, Sneli J-A (1999) Distribu-tion of marine benthic macro-organisms at Svalbard (includingBear Island) and Jan Mayen: figures and tabulated cataloguesof species-distributions, literature and synonyms. Directoratefor Nature Management, Trondheim

Harvey DJ(1982) Picolinyl esters as derivatives for the structuraldetermination of long chain branched and unsaturated fattyacids. Biomed Mass Spectrom 9:33–38

Hayward PJ, Ryland JS (1990) Bivalvia and Cephalopoda. In:Hayward PJ, Ryland JS (eds) The marine fauna of the BritishIsles and North-West Europe. Clarendon Press, Oxford,pp 731–793

Hjelset AM, Andersen M, Gjertz I, Lydersen C, Gulliksen B (1999)Feeding habits of bearded seals (Erignathus barbatus) from theSvalbard area, Norway. Polar Biol 21:186–193

Jarzebski A, Wenne R, Habermehl G (1986) Anatomical distribu-tion of lipids and sterols in Macoma baltica L. Comp BiochemPhysiol 85B:135–137

Joensen H, Grahl-Nielsen O (2000) Discrimination of Sebastesviviparus, Sebastes marinus and Sebastes mentella fromFaroe Islands by chemometry of the fatty acid profile in heartand gill tissues and in the skull oil. Comp Biochem Physiol126:69–79

Johns RB, Nickols PD, Perry GJ (1980) Fatty acid components ofnine species of molluscs of the littoral zone from Australianwaters. Comp Biochem Physiol 65B:207–214

Joseph JD (1982) Lipid composition of marine and estuarineinvertebrates. Part II: Mollusca. Prog Lipid Res 21:109–153

Klingensmith JS, Stillway LW (1982) Lipids composition ofselected tissues of the hardshell clam, Mercenaria mercenaria.Comp Biochem Physiol 71B:111–112

Kluytmans JH, Boot JB, Oudejans RCHM, Zandee DI (1985)Fatty acid synthesis in relation to gametogenesis in the musselMytilus edulis L. Comp Biochem Physiol 81B:959–963

Kvalheim OM, Karstang TV (1987) A general-purpose programfor multivariate data-analysis. Chemometr Intell Lab 2:235–237

Napolitano GE, Ackman RG (1992) Anatomical distributions andtemporal variations of lipid classes in sea scallops Placopectenmagellanicus (Gmelin) from Georges Bank (Nova Scotia).Comp Biochem Physiol 103:645–650

Napolitano GE, Ackman RG (1993) Fatty acid dynamics in seascallops Placopecten magellanicus (Gmelin, 1791) from GeorgesBank, Nova Scotia. J Shellfish Res 12:267–277

Napolitano GE, Ackman RG, Ratnayake WMN (1990) Fatty acidcomposition in three cultured algal species (Isochrysis albana,Chaetoceros gracilis and Chaetoceros calcitrans) used as foodfor bivalve larvae. J World Aquacult Soc 21:122–130

Paradis M, Ackman RG (1975) Occurrence and chemical structureof nonmethylene-interrupted dienoic fatty acids in Americanoyster Crassostrea virginica. Lipids 10:12–16

Rodegker W, Nevenzel JC (1964) The fatty acid composition ofthree marine invertebrates. Comp Biochem Physiol 11:53–60

Saether O, Ellingsen TE, Mohr V (1986) Lipids of North Atlantickrill. J Lipid Res 27:274–285

Scott CL, FalkPetersen S, Sargent JR, Hop H, Lonne OJ, Polter-mann M (1999) Lipids and trophic interactions of ice fauna andpelagic zooplankton in the marginal ice zone of the Barents Sea.Polar Biol 21:65–70

Swift ML, White D, Ghassemieh MG (1980) Distribution of neu-tral lipids in the tissues of the oyster Crassostrea virginica.Lipids 15:129–132

Vassallo MT (1973) Lipid storage and transfer in the scallop Chla-mys hericia (Gould). Comp Biochem Physiol 44A:1169–1175

Voogt PA (1983) Lipids: their distribution and metabolism. In:Hochachka P (ed) The Mollusca, vol 1. Metabolic biochemistryand molecular biomechanic. Academic Press, New York,pp 329–370

Walne PR (1970) The seasonal variation of meat and glycogencontent of seven populations of oysters, Ostrea edulis L, and areview of the literature. Fish Invest Lond 26:1–35

Watanabe T, Ackman RG (1974) Lipids and fatty acids of theAmerican(Crassostrea virginica) and European flat (Ostrea ed-ulis) oysters from a common habitat, and after one feeding withDicrateria inornata or Isochrysis galbana. J Fish Res Board Can31:403–409

Whyte JNC (1988) Fatty-acid profiles from direct methanolysis oflipids in tissue of cultured species. Aquaculture 75:193–203

Wold S, Esbensen K, Geladi P (1987) Principal component anal-ysis. Chemometr Intell Lab 2:37–52

Zandee DI, Kluytmans JH, Zurburg W, Pieters H (1980) Seasonalvariation in biochemical composition of Mytilus edulis withreference to energy metabolism and gametogenesis. Neth J SeaRes 14:1–29

92