Cell Tissue Res (1982) 222:563-577 Cell and Tissue

Research

(1:;) Springer-Verlag 1982

Muscle fibre differentiation and vascularisationin the juvenile European eel (Anguilla anguilla L.)

s. Egginton and I.A. J ohnstbnDepartment of Physiology, University of St. Andrews, St. Andrews. Fife. Scotland, Great Brital

Summary. The differentiation of the lateral musculature in the elver stage of the

European eel (Anguilla anguilla, L.) has been studied using histochemical

staining techniques.. Extracellular lipid deposits constitute 12% of the bodycross-sectional area.

Two fibre types may be di~tinguished on the basis of myofibrillar A TPase

activity. Fibres with an alkaline-labile (pH 10.2) ATPase occur as a two-fibre

layer around the trunk circumference, with invaginations along the horizontal

septum and fin insertions. These pale-yellow fibres correspond to the slow

("red") fibres of other fish and comprise around 7% of the body cross-section

(mean size, 167 ~m2). Slow fibres show a moderate staining for PAS (glycogen),but a relatively weak reaction for Sudan black (lipid) and the aerobic enzyme

markers succinic dehydrogenase (SDH) and cytochrome oxidase (COX).The bulk of the trunk muscle is composed of fast fibres (68 %; 328 ~m2).

These are characterised by an alkaline (pH 10.2)-stable A TPase activity, and

their innervation. Each fibre is innervated by a single "en-plaque" type endplateat one myoseptal end. Fast fibres adjacent to the slow fibre layer (2-4 fibres

deep) show a moderate staining for PAS and a weak reaction for SDH and COX.

Deeper regions show a wide range of fibre size, with about 5 % being> 1,400 J.UD2. Fibres > 200 J.UD2 show no significant staining for glycogen, lipid,

SDH, or COX. Small fast fibres < 120 ~m2 (up to 5% of the white muscle mass)

showa strong staining reaction for PAS and a slight reaction for SDH and COX

activities.

Parameters of vascularisation were calculated from low-magnificationelectron micrographs ( x 1,900). The number of capillaries/fibre and % fibre

perimeter in contact with capillaries were, respectively, 0.98, 6.33% (slow

fibres); 0.33, 1.96% (superficial fast fibres); and 0.12,0.71 % (deep fast fibres).These values are low in comparison with other fish species. It is suggested thatthe low aerobic capacity of elver slow muscle reflects a relatively restricted

aerobic scope for activity associated with the anguilliform mode of locomotion.

Key words: Elver, eel -Skeletal muscle -Histochemistry -Capillarisation

Send offprint requests to: Dr. Ian A. Johnston, Department ofPhysiology , University ofsi. Andrews, St.Andrews. Fife KY 16 9TS. Great Brit,.;n

()1()'-7f\nx/R2/0222/0563/$03.00

~(i4 s. E~ginton and I.A. Johnston

The European eel (Anguilla anguilla, L.) has an unusually long post-larval(leptocephalus) stage, migrating from the Sargasso Sea to the West Europeancontinental shelf over a period of 11/2-21/2 years (Schmidt 1922). Followingmetamorphosis, the juveniles (glass eels) migrate to freshwater. This activeupstream migration is accompanied by a progressive increase in pigmentation(Stubberg 1913), growth (Deelder 1970), and various biochemical changesassociated with adaptation to freshwater life (Tesch 1977). The post-metamorphictransformation from the partially pigmented (elver) stage to the young (yellow) eelcontinues for up to one year after arrival at the estauaries (Strubberg 1913; Boetius1976; Tesch 1977).

There have been few studies on the histology and biochemistry of muscle fromadult and sub-adult stages of the European eels (Willemse and de Ruiter 1979;Bostrom and Johansson 1972) and American eels (Hulbert and Moon 1978). Grosseffects of growth on fibre number and size have been described for both elver andyoung adult stages of Anguilla anguilla (Willemse 1976). The present study extendsthese observations to a detailed description of the elver muscle.

Most previous reports of fish muscle histochemistry and fine structure haveinvolved random, or semi-random, sampling of the different trunk muscle regions.The small size of the elver ( < 6 mm2 cross section) allows a more detailedexamination of regional differences in fine structure, fibre types and vascularisationwithin the myotome. Such a detailed description is essential to an understandingof the developmental processes (fibre proliferation and differentiation) underlyingmetamorphosis and post-embryonic muscle growth in this species. To the authors'knowledge this study also contains the first quantitative description of capillari-~:ltinn in :I teleost with fnca.\lv innervat~n f:l~t fihres.

Materials and methnds

Fish

Elvers were trapped on their upstream migration on the lower reaches of the River Sevem, A von, U .K.,during February and March 1979; transported on ice and maintained in recirculated, filtered tapwater at20 :!: 1°C for six months prior to sampling. Freshwater life was therefore 10-11 months and age around21/2-3 years. Qualitative changes were investigated in samples throughout the migratory period, and inlaboratory-maintained animals up to a period of 12months. Weight and length were, respectively,0.14:!: 0.10 g; 73.0:!: 3.5 mm (mean:!: S.D., N = 16). Lyophilised tubifex was fed to satiation 2-3 times

per week and feeding was stopped two days before sampling. Photoperiod was maintained at"nnrnximatelv 12L:12T)

Preparation of material

Fish were pithed and the sample region located (Fig. 1 a); this includes the point ofmaximum flexure anda relatively constant ratio of red:white muscle. The right side was carefully excised before severing thebackbone and, in routine examination, dorsal and ventral fin masses were removed to assist sectioning.Tissue blocks ( < Smm3) were mounted on chilled cryostat chucks, with a filter paper interface, in aninert embedding medium (OCT Compound, Lamb, London) and immediately immersed in semi-solidiso-pentane (2-methylbutane) cooled in liquid nitrogen. Blocks were allowed to equilibrate to the cuttingtemperature of -22 to -18°C, and serial sections cut at 8-10~. Replicate s,ets of sections wereincubated, supported on dry l!Jass coverslips.

565Light microscopy of elver muscle

All incubations were performed on unfIXed sections at room teperature (20° C); fixation or lowertemperatures having been shown to offer no improvement .in section quality. Sections were mounted

(without dehydration) in glycerol jelly on glass slides.

Myofibrillar ATPase

Sections were stained by a modification of the method of Guth and Samaha (1970). Preincubation

(7min) in 18mM CaCI2, 100mM 2-amino-2-methyl-1-propanol (221 buffer, Sigma, Poole), pH 10.2,was used to differentiate different fibre types. Controls were performed in which A TP was omitted fromthe incubation medium or had sodium azide (55 mM) added to inhibit mitochondrial A TPase activity.

Acid preincubation (pH 4-5) failed to give a staining reversal.

Glycogen ,

Sections were incubated for 30-90min in 1% periodic acid and stained with Schiff reagent (pASreaction). Control sections were treated with 1% amylase in loomM phosphate buffer, pH 6.3, for

60min prior to incubation, or had the periodic acid incubation omitted.

Succinic dehydrogenase (SDH)

Incubation was carried out using 80 mM sodium succinate, 50 mM potassium phosphate buffer, pH 7.4,and 1 mgml-l nitroblue teti:azolium (NBT) as the electron aceptor. Long incubation times, up to threehours, were required to give a significant staining reaction. In controls, succinate was replaced by

m"lnn"t.e in the incubation medium.

Cytochrome oxidase (COX)

Sections were incubated in freshly prepared incubation medium (10ml) containing: cytochrome-c(10mg), catalase (30~g), sucrose (100mg), and 3'3'-diaminobenzidine (15mg) in O.15M phosphatebuffer, pH 7.4, for 15-30min. Control incubations lacked cytochrome-c.

LipidStaining was carried out using Sudan Black B, saturated solution in propylene glycol, for 30-60 min.

Control sections were rinsed in acetone prior to staining.

A ce tylcholinesterase

Strips of muscle were pinned i)n cork strips and fixed using 10% neutral formalin in 0.1 N sodiumacetate-acetic acid buffer, pH 5.2, for 4h at 4° C. Incubation (12h, 20° C) was in a freshly preparedsolution containing: 2.5% CuSO4.5H2O (0.2m1); 3.7% glycine (0.2ml); 0.1 N buffer (8.8ml); and thesupernatant from 15 mg acetylcholine iodide, 0.3 ml copper sulphate and 0.7 ml water. Muscle was

cleared and mounted in j!;jycerol.

Electron micrographs

Tissue was prepared as for the ultrastructural study (see Egginton and Johnston 1981

Semithin sections

Araldite-embedded tissue was sectioned at 0.5-1.0 11In, flattened on wann slides, dried overnight andstained with 1 % toluidine blue in 1 % sodium tetraborate at 60-70° Cor p-phenylenediamine (PPDA) in1 : 1 isopropanol: methanol at room teperature. After washing, sections were dried and mounted in DPX.

()ro(>c:;lI\-~Q \Anatomical parameters

Fibre areas were detennined from photomicrographs (X50) ofPAS-stained cyostat sections projected( x 17.5) onto paper and quantified by digital planimetry (Summagraphics digitiser interfaced with anOlivetti P6060 minicomputer). With set magnifications, all muscle fibres around the lateral line triangle

566 s. Egginton and I.A. Johnston

were sampled within a standard area (about 1 mm2); this differs from the usual count of a standardnumber of fibres, and is thought to give a more representative sample accounting for regional variation.The percentage of extracellular lipid, red and white muscle was determined in a similar manner fromprojections of whole-body cross sections.

Capillarisation

Parameters of capillarisation were quantified using tracings oflow magnification electron micrographs( x 3,300- x 4,800 final magnification), the regions of sampling being chosen after a preliminary

investigation using semithin sections. Many indices used elsewhere in the literature were found to beinappropriate to muscle with a low, and inhomogeneous, capillary supply. The parameters used here arederived by simple manipulation of the presented data (Table 2).

A

O.5CM

B

Fig.1A. Outline of an elver used in the present study; samples were taken at a position between thetwo arrows. B Diagrammatic cross section of an elver trunk at the point of sampling, showing theanatomical details used in mapping each section. Muscle was taken from the epaxial myotome adjacentto the lateral line triangle, just posterior to the cloaca. The numbers along the transect refer to thenumber of fibres deep from the skin, and are used in the delineation of the muscle I;egions under study.R red muscle; S superficial white muscle; M mid white muscle; D deep white muscle; NC nerve chord;LTlateralline triangle: LN lateral line nerve: HS horizontal sentum. L linid. M." mvo.ent" .."K .1cin

Light microscopy of elver muscle 567

Results

Macroscopic appearance

The ~stinct difference in colour usually found between fast and slow fibres inteleosts is absent. Slow fibres in the juvenile eel have a pale yellow appearance; thishas led some workers to sample elver muscle as a whole, having failed to recognisethe existence of different muscle types (Bostrom and J ohansson 1972). Slowfibres occur all around the trunk circumference as a two-fibre deep layer,invaginating along the horizontal septum and fin insertions. The myosepta areclearly defined (Figs. 16,20), as is the lateral line nerve located ca. 1/3 along thehorizontal septum from the skin (Fig. 15). The distinct separation of the "red" andwhite muscle masses is evidenced by a small space appearing between the two layers,especially in Araldite sections (Figs. 17, 18). Such a connective tissue fascia is moredistinct in the adult (Willemse and de Ruiter 1979). Both muscle types failed to stainto any extent with PPDA, a result that is at variance with the adult (Willemse andde Ruiter 1979) and other fish species studied in this laboratory.

Subcutaneous lipid deposits are widespread, but often accumulate at themyoseptal insertions with the skin (Fig. 18). The largest extracellular deposits arefound adjacent to the vertebral column, chiefly along the horns of the haemal arch,at the head of the lateral line triangle, and around the apex of vertebrae. However,this pattern is not uniform along the body. For example, the apical deposits areelongate andheart-shaped, with the broadest portion located in the inter-vertebralregions. Using these anatomical features as reference points, specific regions of thetransect can be described and individual fibres located on different sections(Fig. 1 B).

A qulaitatively similar picture is seen in elvers from the earliest availablespecimens, and those maintained in the laboratory for up to 12 months. Toward theend of this period, however, there appears to be a proliferation of fibres within the"red" muscle, giving a three-fibre deep layer.

Table 1. Summary of the histochemical staining profIle of elver trunk muscle

Enzyme ormetabolite

Myotomal fibre type and position

RF0-2

SWl2-4

SW22-4

MW6-8

DW10-12

SDW6-12

MyofibrillarA TPase

SDH

COX

PAS

Lipid

0 ++++++++++++++++++++ +++++

oooo

oooo

+++ ++++++ +++++++ +++++ +

++++

++++n

O background stain only, + light stain, + + +'+ + heaviest stain, The red fibre category (RF) includesall parts of the slow fibre system. SW 1 refers to the superficial white muscle adjacent to the lateral linetriangle, and SW2 to the superficial white fibres from peripheral niyotomes. MW and DW refer to themid- and deep-white fibres, respectively. SDW indicates the small fibres of the fast fibre system, found inthe MW and DW regions, The numbers at the head of each column refer to the number of fibres deep

from the skin (see Fig. 18)

56R s. Egginton and IA. Johnston

Figs. 2-7. Histochemical staining, frozen sections

Fig.2. Myofibrillar A TPase. Region around the lateral line triangle (L1). Homogeneous staining can beseen in both the slow (pale) and fast (dark) fibres. Skin and melanophores are visible on the left, withmyosepta and the lateral line nerve (LN) toward the Centre. 10 ~m section

Fig.3. Myofibrillar A TPase.. Detail showing the distinct segregation of the slow and fast fibres..Adjacent, epaxial, myotome to that shown in Fig.. 2.. 8IUU section

Fig.4. SDH.. Area around L T , showing the moderate staining for slow (dark) fibres and the slightreaction found in the superficial white fibres (SW1, see Table 1).. 10 IUU section

Fig.5. SDH. Detail of adjacent myotome showing the intermediate reactivity of the superficial whitefibres (SW2, see Tablet) relative to the slow (dark) and fast (pale) fibres. 8~ section

Fig.6. COX. Area around L T showing of the SW 1 fibres. 10 I!m section.

Fig. 7. COX. Peripheral myotome showing the more intense staining reaction of slow than fast fibres.R Il.m .ect.inn

569Light microscopy of elver muscle

Figs.8-11. Histochemical staining, frozen sections

Fig. 8. PAS. Area around L T showing the variability of glycogen deposits within the fast fibres, relativeto the slow (dark) fibres. Skin, myosepta and lateral-Iine nerve are also stained. Note the depth of the

SW1 zone. 10111n section

Fig.9. PAS. Adjacent myotome showing the progressive reduction of stain from the SW 1 to SW2

regions. 8 ~m section

Pig. 10. PAS. Detail of SDW showing the more intense staining than the adjacent DW fibres. 8 ~m

!;ection

Fig. 11. Sudan black. Detail of peripheral myotome showing the relative difference in lipid deposits

between the slow (dark) and fast (pale) fibres. 8 ~ section

Fig. 12. SDH slow fibres at the level of the transect (see Fig. 1 b) showing even staining along the length of

the fibres. L.S., 10 I1In section

Fig. 13. Teased fibre bundle stained for acetylcholinesterase activity, showing an "en-plaque" endplate

from the MW region

570 s. Egginton and I.A. Johnston

FREO.

0 600 1200 1800 2400 3000

AREA(~M2)

Fig. 14A-E. Frequency vs fibre size histograms: A Slow muscle (RF, n = 378). B Superficial whitemuscle (SW1 + SW2). C Deep white muscle (MW +DW +SDW). D Combined data for all fast fibres(n = 1729). E Expanded view of the small fast fibre data (SDW). Note the contribution of the SDW to

the bimodal distribution of fast fibre size, extending the large range seen within the fast system

Fibre size

Slow fibres show an even spread of size around a modal category of 240-360 !lm2,with few fibres found > 800 !lm2 (Fig.14a). Fast fibres show a bi-modaldistribution ofsizes (Fig. 14d) with those fibres < 120 !lm2 all being of intermediatestaining intensity for the aerobic enzymes (Fig. 14e, Table 1); only three profiles ofthis type appeared outside this size category. The rest of the fast muscle shows aneven spread across a large size range, with about 5% of fibres being > 1,400 !lm2

(Fig.14c).

HistochemistrySlow fibres of the eel have a multiple-axonal, multiple-temiinal pattern ofinnervation (Bone 1978), whereas fast fibres are innervated by a single "en-plaque"

Light microscopy of elver muscle

~.~

-nu1BlL---=~- _C.U&.u1Jm -,,",,_-:cc~,-

Fig. 15. Semithin section of the LT. Note the slow muscle layer (darker fibres) and large lipid depositsassociated with the invagination. Nuclei, capillaries and lateral-Iine nerve are also clearly visible.Toluidine blue stain: 0.5 LlIn section

Fig. 16. Semithin section of the fast muscle (MW + DW) showing the absence of capillaries, the smallfast fibres and the close association of some fibres with the myosepta. Toluidine blue stain; O.5111n!;ection

endplate at one myoseptal end (Fig. 13). The contraction speed qf fish muscle hasbeen shown to be related to biochemical measurement of myofibrillar A TPaseactivity (J ohnston et al. 1972; Flitney and Johnston 1979) and the histochemicalstaining reaction following alkaline (pH 10.4) pre-incubation (Johnston et al. 1974).Slow fibres are alkaline labile and fast fibres are alkaline stable (pH 10.2). Bothtypes of fibre show homogeneous staining for A TPase, irrespective of position orfibre size (Figs.2, 3). No true intermediate, or "pink", fibres could be detected(Johnston et al. 1974).

The slow muscle layer also shows an even distribution of staining for the aerobicenzymes and metabolites studied (Figs. 4-7). In all cases, the intensity of staining isgreater than any of the fast fibres (Table 1), but required considerably longer todevelop a significant reaction product than other species previously investigated.This long staining period is indicative of a lower capacity for aerobic metabolismthan is usually associated with fish slow fibres (Bone 1978; Johnston 1981). Fastmuscle fibres show a staining profile for enzymes and metabolites that is dependenton both position and size (Figs. 2,4,6,8, 10). Large fibres show a homogeneous lowreactivity with all staining procedures tested, whereas the smallest fibres in the deepregions, and the most superficial layer of fast muscle, show both mo'derate SDHand COX activities and a moderately high PAS staining reaction (Figs. 4-10). The

j

"""C_.~,"cc¥c,"_'[1~-Fig. 17. Transect of elver muscle from the skin toward the vertebral column, showing the relationship offibre size within the various regions under study. Stipple RF, hatching SW fibres. Camera lucida tracingof semithin section

Fig. 18. Semithin section showing one half of an elver trunk. The vertebral column, lateral-Iine triangle,lipid deposits and blood vessels are visible. O.5-l1m section .

Fig. 19. Semithin section of slow muscle (R1') and the superficial white muscle (SW2). Note the relativesize of the lipid deposits within the fascia (1'). O.5-11In section

Fig.20. Semithin section of MW muscle illustrating the insertion of fast fibres into the myoseptal sheet(MS) and the associated lipid. Darkly staining regions are nuclei. O.5-11In section

Light microscopy of elver muscle ~71

RED SUP.W DEEPW

100

%

FIBRES

50

Fig.21. Frequency ( %) of fibres vs numberof capillaries per fibre for the three mainfibre cat.e"nrie.

01 23 O 1 23 O 1

NUMBER OF CAPILLARIES / FIBRE

Table 2. Indices of vascularisation in the elver trunk muscle sampled at three positions. Mean :t SoD

Fibre

type

No. offibres

(capil-laries)in the

analysedsample

Mean

fibre

areaa

(I1m2)

Numberof

capil-laries

perfibrea

Mean Vascu-capillary larisedcontact fibre

length peri-per meterfibrea ( %)

(L1In)

Mean

capillaryarea

(J.Un2)

Slow

fibres

(RE)

1~1 190.5:t193.5

(179.3:t 66.9)

0.98:t 0.78

(1.35:t 0.58)

52.6

:t 17 00

(55.2:t1707)

4.8+3.2

6.33 15.6:t3.3

14~~ti OOQ

(59)

Super-ficialwhitefibres

(SW)

117 274.4

:t211.1

(269.5

:t 232.9)

0.33:t 0.63

(1.26:t 0.57)

70.3

::!:93.0

(60.6::!: 26.2)

4.4

:!:3.7

196 17.1::!: 16.5

6.606.2 OOIi

(17)

Deepwhitebfibres

(DW)

1~7 286.7:t 240.7

(349.9:t 296.5)

0.12

:to.33

(1.0)

63..4:t 32.0

(70.1:t 32.5)

3.3+2.0

0.71 10.9:\:2.8

34-3n 7 OOIi

(9)

a Figures in brackets indicate values for the vascularised portion of the muscle only, i.e., fibres in direct

contact with a capillaryb Includes SDW fibres. No significant difference was detected between MW and DW fibres, for any parameter ;

nor could any correlation be found with fibre size

homogeneous reaction along the length of fibres (Fig. 12) is in apparent agreementwith the quantitative results found in mammalian muscle (pette et al. 1980).

Capillarisation

Figure 17 shows a camera lucida tracing of fibres and associated capillaries along atransect from the skin to the vertebral column (see Fig, 1 b), The various indices of

Mean

fibre

circum-

ference"

(J.lm)

Area ofmuscle

suppliedby 1 J.lm2of

capillary(I1m2)

Mean

capillaryarea

supplying

1~2of fibre

(11ffi2)

574 s. Egginton and IA. Johnston

vascularisation show an increase in capillary supply from the fast to slow systems:deep white < superficial white < "red" muscle (Table 2). No difference can bedetected between fibres of different sizes within any group. Deep white fibres have avery anaerobic character with 88% having no capillary contact; however, thesuperficial white muscle is slightly better supplied with some fibres having contactwith two capillaries (Fig.21). Interestingly, there is a large portion of the slowmuscle without any direct capillary contact (Fig.21), clearly showing that elverslow fibres are poorly vascularised in comparison with other fish (see Johnston1981a).

Discussion ,

In order to locate more precisely the various metabolic sub-populations, and as anaid to parallel ultrastucture studies (Egginton and Johnston 1982), the charac-teristic morphology of each cross-section was noted (Fig. 1 b ). Two distinct classesof fibres can be identified in the elver trunk by staining for myofibrillar A TPaseactivity, both having a homogeneous reaction throughout their size range andposition within the myotome. On the same basis, two fibre types have beenidentified in the brook trout (Johnston and Moon 1980a), three in carp (Johnstonand Maitland 1980), and five in the dogfish (Bone and Chubb 1978).

Elver fast muscle can be further sub-divided into three metabolic sub-groups onthe ba~is ofstaining for PAS, lipid, COX, and SDH activities (Table 1). In the caseof carp (Johnston et al. 1977) and flathead (Mosse and Hudson 1977), the smallPAS-positive fast fibres may be further differentiated on the basis of their stabilityto alkaline (pH 10.4) pre-incubation prior to staining for myofibrillar ATPaseactivity (patterson et al. 1975). No such "pink", or true intermediate, fibres aredetectable in either the elver or in adult eels of either Atlantic species (Willemse andde Ruiter 1979; Hulbert and Moon 1978).

Fast fibres are focally innervated by a single "en-plaque" typeendplate (Fig. 13);this type of fast fibre innervation appears to be typical of primitive teleosts (Bone1978). The nature of innervation offish fast fibres appears to be correlated with thedivision of labour between fast and slow muscles during swimming. In focallyinnervated fish, elasmobranchs and some teleosts such as the herring, it has beenshown that the fast glycolytic fibres are reserved exclusively for burst activity (Boneet al. 1978). The low aerobic capacity and capillary supply to deep fast fibres in theeel is compatible with a similar pattern of fibre recruitment. However, Grillner andKashin ( 1976) have reported electrical activity in the white muscle of adult eels, evenat low swimming speeds. This apparently resembles the pattern of fibre recruit-ment found in higher teleosts, with multiply innervated fast fibres, where it is oftenpossible to record electromyograms over a wide range of swimming speeds(Hudson 1973; Johnston and Moon 1980b). The superficial white muscle in theelver shows a more aerobic character than the deep white fibres, and may representa functional sub-type of the fast fibre system. There is some evidence, in the coalfish,that the superficial fast fibres can be recruited at lower swimming speeds thandeeper fast fibres (Johnston and Moon 1980b); and some differences in theinnervation of these two regions has been shown in the cod (Altringham andJohnston 1981). Electromyographical studies in the carp have shdwn the order ofrecruitment of fibre types with increasing swimming speeds to be slow > fastaerobic (intermediate) > fast glycolytic (Johnston et al. 1977). Further e.m.g.

J

Light microscopy of elver muscle 575

studies would be required to discover whether the pattern of fibre recruitment ineels differ from that reported for dogfish (Bone 1966) and herring (Bone et al. 1978),as the possibility remains that electrical activity recorded in eels at low swimmingspeeds originates from these superficial aerobic fibres. The histochemical profile ofslow muscle fibres indicates a more aerobic metabolism than that of the adjacentfast fibres; however, the long incubation time required suggests a low, absolute,aerobic potential.

Comparative data for the vascularisation of fish muscle is limited to a fewteleosts and elasmobranchs (Mosse 1979; Flood 1979) and for a chondostean,Acipenser ste/latus (Kryvi et aI1980). In the elver, the proportion of fibres in directcontact with the capillaries within the red, superficial white, and deep white regionsare, respectively, 72.5 %, 26.5 %, and 12,4% (Table 2). The average number ofcapillaries per fibre for the three categories are 0.98,0.33,0.12 (RF, SW, DW). Thewhite muscle appears to have only slightly fewer capillaries/fibre than that of otherspecies: Acipenser, 0.2 (Kryvi et al. 1980); Australian salmon, 0.14-0.27 (Mosse1979); Myxine and the flathead having similar values, around 0.6-0.7 (Flood 1979;Mosse 1979). However, the slow fibres, with 0.98 capillaries/fibre, has significantlyless than that found in the above species: 2.3, 1.9-4.2, 2.76, and 5.3-6.6,respectively. The range of values found in fish is further illustrated by the pelagicanchovy, which has 12.9 capillaries per slow fibre, resulting in over 50% of the fibresurface being vascularised (J ohnston 1982 b ). Only 70% of elver slow fibres are indirect contact with capillaries. It would appear, therefore, that elver slow musclehas a lower aerobic capacity than many other fish (Bone 1978; Johnston 1981).

The superficial white fibres have a capillary density between that ofred and deepwhite fibres, and is in agreement with the situation reported for the intermediatefibre layer in Acipenser and Myxine. This correlates well with the relative aerobiccapacity, as shown by histochemistry.

The downstream, sexual, migration of the silver eel is associated with increasesin activity of muscle aerobic enzymes, intracellular lipid content, and red musclemass (Bostrom and Johansson 1972; Lewander et al. 1974). This pattern ofincreased aerobic metabolism would appear to be a common feature of migration(Fontaine 1975). In the yellow eel, the low respiration rate (Berg and Steen 1965)and moderate gill surface areas (Byczkowska-Smyck 1958) are reflected in its slowswimming and bottom-dwelling habits (Deelder 1970). The present study, however,indicates that a somewhat different situation occurs in the elver. Migration involvesprolonged active swimming, and even under laboratory conditions elvers arealways active, especially if flowing water is present. Although the slow musclesystem is moderately well developed, the histochemical and capillary data indicatesa suprisingly low aerobic capacity. The propagated locomotory wave inanguilliform locomotion is characteristically of high amplitude and low velocity(Gray 1933). It would appear that this form of swimming is energetically mostefficient at low speeds (Webb 1978), and the range of efficient sustained swimmingspeeds will therefore be restricted, relative to carangiform and sub-carangiformlocomotion. This would suggest a limited and rather inflexible aerobic scope foractivity, which is consistent with the observed low aerobic potential of the slowmuscle.

Acknnwl"d""m"n/" The authnrs are f!rateful to the Natural Environment Research Council for suDDort

576 So Egginton and IA. Johnston

References

Altringham ID, Johnston lA (1981) Quantitative histochemical studies of the peripheral innervation ofcod (Gadus morhua) fast myotomal muscle fibres. I Comp Physiol143: 123-127

Berg G, Steen IB (1965) Physiological mechanisms for aerial respiration in the eel. Comp Biochem

PhysioI15:469-484Bone Q (1966) On the function of the two types of myotomal muscle fibre in elasmobranch fish. I Mar

BioI Ass UK 46:321-349Bone Q (1978) Locomotor muscle. In: Hoar WS, Randall Dl (eds) Fish physiology 7: Academic Press,

New York, pp 361-424Bone Q, Chubb AD (1978) The histochemical demonstration ofmyofibrillar ATPase in elasmobranch

muscle. Histochem I 10:489-494Bone Q, Kiceniuk I, Jones DR (1978) On the role of the different fibre types in fish myotomes at

intermediate swimming speeds. Fish Bull 76:(2):691-699Boetius I (1976) Elvers, Anguilla angui/la and Angui/la rostrata from two Danish localities. Size, body

weight, developmental stage and number of vertebrae related to time of ascent. Medd fra Dan Fiskog Havunders NS Vol 7:199-220

Bostrom SL, Iohansson RG (1972) Enzyme activity patterns in white and red muscle of the eel (Angui/laangui/la) at different developmental stages. Comp Biochem PhysioI42B:533-542

Byczkowsk-Smyk W (1958) The respiratory surface of the gills in teleosts. Acta BioI Cracoviensia SrZool1 :83-97

Deelder CL (1970) Synopsis of biological data of the eel. FAG Fish Synop No 80. 82 pp, RomeEgginton S, Johnston lA (1982) A morphometric analysis of regional differences in myotomal muscle

ultrastructure in the juvenile eel, Angui/la angui/la. L. Cell Tissue Res 222:579-596Flitney FW, Johnston lA (1979) Mechanical properties of isolated fish red and white muscle fibres. I

PhysioI295:49-50PFlood PR (1979) The vascular supply of three fibre types in the parietal trunk muscle of the Atlantic

hagfish (Myxine glutinosa, L.). Microvascular Res 17:55-70Fontaine M (1975) Physiological mechanisms in the migration of marine and amphihaline fish. Adv Mar

BioI 13:241-355Gray I (1933) Studies in animal locomotion. I Exp BioI 10:88-104Grillner S, Kashin S (1976) On the generation and performance of swimming in fish. Adv Behav BioI18:181-201 .

Guth L, Samaha FI (1969) Qualitative differences between actomyosin ATPase of slow and fastmammalian muscle. Exp Neurol 25:138-152

Hudson RCL (1973) On the function of the white muscle in teleosts at intermediate swimming speeds. IExp BioI 58:509-522

Hulbert WC, Moon TW (1978) A histochemical, light and electron microscopic examination of eel,Angui/la rostrata, red or white muscle. I Fish BioI 13 (5):527-533

Johnston lA (1981) Structure and function of fish muscles. In: Day MH (ed) Vertebrate locomotion.Symp Zool Soc Lond 48:71-113

Johnston lA (1982) Quantitative analyses of ultrastructure and vascularisation of the slow musclefibres of the European Anchovy (Engraulis encrasicolus). Tiss Cell (in press)

Johnston lA, Maitland B (1980) Temperature acclimation in Crucian carp: a morphometric analysis ofmuscle fibre ultrastructure. I Fish BioI 17:113-125

Johnston lA, Moon TW (1980a) Exercise training in skeletal muscle of brook trout (Salvelinusfontinalis). I Exp BioI 87: 177-194

Johnston lA, Moon TW (1980 b )Endurance exercise training in the fast and slow muscles of a teleost fish(Pollachius virens). I Comp Physiol135B: 147-156

Johnston lA, Frearson N, Goldspink G (1972) Myofibrillar ATPase activities of red and white musclesof marine fish. Experentia 28:713-714

Johnston lA, Patterson S, Ward P, Goldspink G ( 1974) The histochemical demonstration of myofibrillaradenosine triphosphatase activity in fish muscle. CanI Zool52 (7):871-877

Johnston lA, Davison W, Goldspink G (1977) Energy metabolism of carp swimming muscles. I Comp

PhysioI114:203-216Kryvi H, Flood P, Guljaeu D (1980) The ultrastructure and vascular supply to the different fibre types in

the axial muscle of the sturgeon Acipenser stellatus. Cell Tissue Res 212: 117-126

~77Light microscopy of elver muscle

Lewander K, Dave G, Iohansson M-L, Larsson A, Lindman V (1974) Metabolic and haematologicalstudies on the yellow and silver phases of the European eel. Anguilla anguilla, L. Comp Biochem

PhysioI47B:571-581Mosse PRL (1979) Capillary distribution and metabolic histochemistry of the lateral propulsive

musculature of pelagic teleost fish. Cell Tissue Res 203:141-160Mosse PRL, Hudson RCL (1977) The functional roles of different muscle fibre types identified in the

myotomes of marine teleosts: a behavioural, anatomical and histochemical study. I Fish BioI

11:417-430Patterson S, Johnston lA, Goldspink G (1975) Histochemical study of the lateral muscles of five teleost

species. I Fish BioI 7: 159-166Pette D, Wimmer M, Nemeth P (1980) Do enzyme activities vary along muscle fibres? Histochemistry

67:225-231Schmidt I (1922) The breeding pla~ of the eel. Phil Trans R Soc Lond (BioI) 211: 1-79-208Strubberg A (1913) The metamorphosis of elvers as influenced by outward conditions. Some

experiments. Meddr Kornrn for Havunders Ser Fiskeri 4 (3):1-11Tesch F-W (1977) The eel. Chapman and Hall, LondonWebb PW (1978) Hydrodynamics: nonscombroid fish. In: Hoar WS, Randall DI (eds) Fish physiology

7: Academic Press, New York, pp 189-237Willemse II (1976) Characteristics ofmyotomal muscle fibres and their possible relation to growth rate

in eels Anguilla anguilla. Aquaculture 8:251-258Willemse II, De Ruiter A (1979) A quantitative identification of four types of fibres in the lateral

musculature of immature European eel, Anguilla anguilla, L. Aquaculture 17:105-111

Accepted September 3, 1981