189-202.pdf (525.6Kb)

Flodevigee rapportse-c*, 1 , 1904. ISSN 0233-2594 The Propagation of Cod G n d u n rno.?huo L.

TRYPSIK AKD TRYPSIYOGER AS INDICES OF GROWTH JND SURVIVAL POTENTIAL OF COD ( G a d u s mozhrtu L.) LARVAE

X. Hjelmeland T

,

use^,

1 Instrtute of Fisheries, Unrversrry of Tromsiz N-9000 TROMSD Norway

2 Institute of Marrne Research, 4ustevoll Marlne Aquaculture Station, N-5392 STOREBm Uorway

ABSTRACT

Fi~eZmeland, R., Fiilse, I., Jmrgeqsen, T., hlolvik, G. an3 Raa, J., 1984. Trypsrn and trypsrroge~ as rndrces of growth and survival potentral of cod [ G a d i i ~ m o ' i h u o L.) iarvae. In:

E. Dahl, D.S. Danielssen, E. Moksness a ~ d P. Solemdal (Editor%), The Propagatron of Cod Gadud rncahua L.

Fl~devrgen rapportser., I , 1984: 189-201.

Uslng a radrormmu~oassay 1c could be demonscrated thac the content of trypsrn and crypsinogep rn cod larvae increased markedly durlng the frrst 4 days after hatching. In the same perlod the trypsin acrrvlry, measured by the conventronal qethod, was conscanc, but rncreased conslderably on day 5. At that 21me the larx~ae were reaay for the frrsc feed rntaue.

After day 5 che total quancitq of trypsrn and rrypslnogen aropped agarn, independent of wnecher che larvae received any food or ?ot.

Larvae offered food which sdpported growth, started to nroduce trypsin and trypsrnogen, whereas, rn starvlng larvae and those ofrered rncomplete alets, che level of crypsrn and trypslnogen remained Jcry low. We have concluded chat tne radrormmunoassay is a convenient and sensrtlve method for the quantltatlve determination of trypsrn and trypsinogen In very small samples, and chat the level of crypsln ard trypsinoyen may be o s e d as an Index of tne feedrng starus and growth potential of tne larvae.

INTRODUCTION

Cod larvae are only 1 - 5 mm long when the y o l ~ rs exhaasted and they become dependent upon exogenous nutrients (Thellacker and Dorsey, 1980). Even though larvae actrbely Ingest dead partrcles ar thls stage, artrfrcral food has proved to be Inadequate to sustarn growth throughour the remalnrng larval stages and beyond metamorphosrs (Howell, 1979; Ruse, 1980;

T h o m p s o ~ and Rlley, 1981).

Many theorles have been proposed to explarn the lnabrllty of most marlne larvae to utrllze dead food (Therlacker, 1980).

One suggestlon 1s that the drgestlve system of the larva 1s unable to transform the food par-clcles lnto molecules, which the larvae can use for growth (Dabrowskr, 1979; Hogendoorn, 1980; Huse et al., i982; Szlamrnska, 1980).

The flrst step rn the drgestron process In che larva 1s the enzymatlc exrracellular degradatron of rhe food partrcles rn the alrrrentary rract. Most ilsh larvae have n o morphologrcal (F:g. 1: or nrsrologrcai stomach (Tanaka, 1969, 1971). The entlre extracellular drgestlon rs therefore llkely to occar rn tne rntestine by pancreatic enzymes. The ~ e x t srep rn the transformation of tne ingested food 1s absorptron of the degradatron products through the rntesrrnal wall. It has been demonstrated rnat lntestrnal cells of fisn larvae are able to absorb protern macromolecules ~y means of prnocytosrs (Iwal and Taraka, 1968; Srroband ana Krooq, 1981). Thrs has led to the suggestron that plnocyrosrs and rnrracellular digestion In the larva compensate for a posslale rncomplece ex~racellclar hydro- lysls, ~n rhe ahserce of a functional stomach (Iwal, 1969:

hoarllac-Depeyre, 1976; Stroband an2 Van der Veen, 1981 j.

Tne pancreas produces and srores dlgesrr-~e enzymes In the form of rnactrve proforms, whrcn are then actrvrated when secreted llto tne cwt. Srnce rrypsln 1s che only pancreatic protease whrch cap actlvate rts own proform as well as the proforms of other proteases secrcted froir the Fancreas (Corrrrg,

19801, this enzyme has a ~ e y positron ~n conrrollrng tne

actrvrty of the pancreatrc 2roteases. For tnrs reason analysrs ol crjpsln con5ent ln the lar,ae was c n o ~ g h t to be a sultaale

191

tool for stcdying t h e digestive facctioa and growtk o f larvae.

- .

i . 1 Histornicrogra~i? of a cod lerva. The picture sk~ows a longitudinal section of a larva.

Trypsin-like activity is usually assayed witn a model protein as the S-~bstrnte at pi! 7-3, or with a s T ~ r . t k e c i c sub- strati. !Rec;c, 19741- i'owe~ier, a crui1,e enzyn:e-extract Eron the digestive tract, or of the whole organisic, will consain other proteases tEsa? pancreatic trypsiri which are able TO react with both the model protein a!nd tile syrthetic sizbstrate a r 7-9 (Barret, 1979). The trypsin level may therefore be over- estinated. But two factors might caase ar 7~clerestirnarior of trypsin assayed enzynatically; first, =he ~ r e s e ? c e of zrypsin inhibitors i n blood acd t i s s ~ e fiaids, and s e c u n d ? ~ , because

t r y p 5 i . r has co ertzyxe activity ;v."en existing: r n irs oroform,

v , -

LLjp~j.nocjen. These problems are circurnve?ted by thc use of an

immunological assay based on antibody recognition of both active and inhibited enzyme, as well as the inactive proform

[Hjelmeland, et al., In prep.).

This paper presents the results of using this method for the analysis of trypsin content in cod larvae given living food organisms and dead artificial feed. The aim was to examine whether the amount of trypsin may be a useful index of growth and survival potential of the larvae, and whether rhey respond to different diets by producing different quantities of trypsin.

Vl.TE3IALS AND METHODS

Srarr-feeding of cod larvae

Fertiilzed eggs were collected from a pen with spawnlng cod and r~cubated and hatched rr polychylene cylinders as described by Huse, et al. 119821. Frve day old larvae were transferred to four 200 1 coqical experlmenra: tanks Mrtn ap lnltial larvae denslty of 20 per L. The ex~erlmental ranKs v?ere s ~ ~ o p l ~ e d with flltered and UV-treated sea water pumped from a depth of 55 m. The temperatare varied between 5 0 C and 8OC durlnq the experimental period. Feedrng was srarted immedrately after rhe first sampling 00 day 5. Larvae i n group I were offered a standard dlet based on bens egg, proreose c e ~ t o n e , cod liver oll andfrsk- prctelii ahtollsate ~rlt? visarnins aqd m ~ n c r a l s adned

(Huse, et al., 1982), grodp I1 a mixture of celt-~ated rozrfers and wrid lrvlng plaqkton, arid group III a cod roe dler (Molvlc et al., 1984). These larlae were fed rhree c r ~ s daliv.

Larvae rn group IV recelved ro food.

Sampllng and preparatron of sample

Samples were taken from two different hatching groups. One group covers the period from day 0 to day 5 after hatching, while the second gro.Jp represenrs larvae f r c ~ 5 days on after hatching; the Latter group was dsed in rho feeding trial.

The random samples from tkc? experimental tanks were ":set3 as

:allows. Ten larvae From each sample here used for the estrrna- tion of trypsrn oy radrormmanoassay and by enzyme a c t i ~ l t y assay. The adherrng sea water was gently removed from che larvae whlch were then frozen at -30°c In one small plastlc tube. Later, a phosphate buffer sallne wlth 0.2 % bovrne albumeu (PBS%) was added to the tube to glve an average volume of 30 v1 per larva. Thereafter, rhe sample was homogenized usrng a Branson Sonlfler B-12X. The remarnlng larvae ln the orlgrnal sample (30-40 larvae) were preserved In 4 % korwalrn and used for determlnrng dry welght.

Radioimmunoassay of trypsin in cod larvae

Detalls of the estlpatlon of trypsln content In cod by radrolmrnunoassay wlll be descrrbed elsewhere (Blelmeland, In prep.). Brlefly, 50 ul of the sample (dlluted 10 to 10.000- fold in PBSAI was mlxed wrth 20 p1 (0.5 ng) of 1251-labelled purrfred cod trypsrn (Hjelmeland and Raa, 1982) and 40 p1 raohrt antlbody agalnst cod crypsrn (diluced 20.000-fold ~n 2 % normal rabblt seruw/PBSA). The precrprtatron anrlbody used was sheep antl-rabbit Ig drlated 5-fold In PBSA. All assays were done In duplicate.

Enzymatic activity assay of trypsin in cod larvae

Alrqbots of 10 h1 of che samples were mlxed wrth 25 p i of 0.2 M Trrs-baffer (OH 7 . 8 ) and 100 ~l 1 m74 s o l u t ~ o n of the chromoge~lc substrake Carbobenzoxy-Val-Gly-Arg-p-nlrranlld- acetate (Boehringer Manrnelm G m o H , d. Germany) rn wells of a mlcrotirer plate. Afcer ~ncubarron at 2 3 O ~ for l h the llght absorption at 405 nm was recorded, sing a Tltertek Multrscan photometer (Flow).

RESULTS

The average amount of trypsrn and rrs proform trypsi~ogen, meas~red oy radlolrnm~~noassay, in cod larvae increased markedly

during the first 4 days after hetching (Fig. 2). Before the first feeding betweendays 5 and 6 the csntent decreased. The larvae had a slight trypsin-like activity on the day of hatch- ing, and there was no significant change in this activity during the following 4 days. On day 5, however, a sharp increase was found.

Fig. 2. Trypsin-like enzymatic activity ( 0 )

and trypsin/trypsinogen content (G) in cod larvae.

Absclssa expresses days after hatching.

0 2 4 6

DAYS

Frg. 3 shows the results of trypsln/trypsrnogen analyses by radlolmmunoassay of larvae In feedrng experrments. Also

lncluded are values for cod larvae from a pond where the larvae were feedlng on natural plankton (Kvenseth and Blestad, 1 9 8 4 ) .

In all groups of larvae a drastlc decrease In trypsrn/trypsln- ogen conteqt occurred durlng the flrst 3-4 days of feedlng.

The value of trypsrn,/trypsrnogen on day 5 represents the aver- age of as many as 60 larvae, whlle the other values are each based on samples contalnrng 1 0 larvae. Larvae rn the group

grven a drer of lrvrng organlsrns [ C a l a n ~ s and Roratorlai had an almost constant level of trypsrn/trypsrnogen from day 9 untrl day 18, when zc started to rlse very sreeply.

.--.

= = COD ROE DIET

.----A CALANUS +ROTATORIA

0 2 f6 10 14 18 22 26 30 FIRST FEEDING

DAYS

Frg. 3. Trypsln/trypsreogen content, estrmated by rad~ormmuno- assay, rn cod larvae from a feedrng experiment. Also rnciuded rn the flgure are values for larvae from a pond where tney were feedrng. on natural p'Lanic~cri. The frrst sample (day 53 was taken 3 h before the larvae were offered any dret.

Abscrssa expresses aavs after Patchrna.

DAYS

Fig. 4. Trypsin-like activity, estimated by an enzymatic assay, in cod larvae from a feeding experiment. Absclssa expresses days after hatching.

l ^.--.

= STANDARD DIET

140 • COD ROE DIET

A -

. cALANus+RoTAToRIA

i

5 10 15 20

DAYS

Flg. 5. Dry werght of cod larvae from a feedlng experiment.

Each value represents the average of 30-40 larvae, welghed lndlvldually. Fbsclssd expresses days after hatching.

The trypsin/trypsinogen content of the Larvae which were given the artificial diets dropped to a stable low level on day

12. From then on and until day 26 the values were close to the lower detection limit of the method (0.3 ng).

At the end of the feeding trial 8 larvae remained in the group fed live food and 40-50 larvae in both groups fed artifi- cial feed. The larvae in the starvation group were ail dead on

day 1 7 after hatching.

The trypsin-like activity in cod larvae was almost the same in all feeding groups during the period from day 5 until day 18 (Fig. 4). On day 18 there was a very steep rise in the trypsin- like activity in larvae feeding on live food. A slight increase in activity was also observed for the two other feeding groups.

It is worth mentioning that the group given live food had a lower trypsin-like activity at the end of the experiment (day 29) than at day 20.

Fig. 5 shows the average dry weight of the larvae sampled from the four experimental tanks at Austevoll. Each value represents the average of 30-40 Larvae, weighed individually.

Until day 14 no significant difference in weight between larvae given live and dead food was registered, but after day 14 there was a significant increase in weight of those given live food.

The larvae which were offered dead food did not gain any weight.

DISCUSSION

Due to its high sensitivity, the radioimmunoassay can be used to determine trypsin and trypsinogen in very small samples, for example a few cod larvae. Since the method is based on the very specific interaction between antibody and antigen (trypsin/

trypsinogen), the enzyme content can be quantitatively deter- mined even in crude extracts. This is unlike direct enzyme activity measurements, which will detect neither the proform of the enzyme nor any trypsin bound to tissue and serum inhibitors.

Such inhibitors may be mixed with the gut trypsin during prepar- ation of the sample (Hjelmeland and Raa, 1980; Hjelmeland, In press).

Using the radiairnrnunoassay it could be shown that the larvae rapidly mobilize trypsin/trypsinogen during the first 4 days after hatching. Since direct enzyme assays showed an almost constant low activity in the same period, it is likely that the trypsin recorded by the immunological assay was the inactive proform, trypsinogen. This is physiologically sound, since the larvae in this period are not feeding but conditioning them- selves for the first feed intake a few days later. It might be of course that the constant low level of active trypsin during the first 4 days is due to inhibition by tissue inhibitors which become mixed with pancreatic trypsin during preparation of the sample. This, however, seems an unlikely supposition since a marked increase in trypsin activity was recorded on day 5, just at the time when the larvae had reached a stage of development when they were ready for their first feed intake (Ellertsen, et al., 1980). It is premature to speculate on the physiological significance of the increase in trypsin-like activity on day 5, and the simultaneous decrease of trypsin/

trypsinogen as measured by the radioimmunoassay, before more experimental studies with larvae at this stage of development have been carried out. At any rate, on day 5 the larvae had no access to feed and trypsin activity can therefore not have been induced as a result of active feeding. Thus, the larvae appar- ently start to produce trypsin activity at a certain stage in development, somewhere between day 4 and 5, whether or not there is food present in the gut.

It is a significant phenomenon, demonstrated with all larval groups, that the total quantity of trypsinjtrypsinogen,

measured by radioimmunoassay, decreases sharply after day 5, independent of whether the larvae are offered feed or not (Fig.

3 ) . Accordingly, we fee: confident that the high trypsinj

trypsinogen content in the larvae on day 5 reflects an imprinted physiological event in their development, and that this occurs independently of their later destiny. A drop in average trypsin/trypsinogen level in a sample of larvae would also have been recorded i f a high proportion of the larvae after days 4-5 were non-functional, or "losers". But accepting this as the main reason for the drop in trypsin/trypsinogen

after day 5 (Fig. 3 1 , would be to imply that the majority of the larvae were actually physiologically dead. This was definitely not the case with the larvae from the pond, which were also hatched in the laboratory, where a survival beyond metamorphoses of 70 % was obtained (Kvenseth and Biestad, 1984). Another interpretation of the decline in trypsln/

trypsinogen after day 5 could be that the larvae actually were beginning to starve at that time, and that pancreatic proteins like trypsin therefore were degraded to mobilize energy

(O'Connell, 1976). However, since the successful pond experi- ment (Kvenseth and Diestad, 1984) was also done with larvae hatched artificially and transferred to the pond on day 5, any starvation before day 5 cannot seriously have affected the survival potential of the larvae. Nevertheless, our experi- ments should be supported with measurements of trypsini trypsinogen in larvae hatched under natural conditions, where soluble nutrients and small particles may be taken up

passively by the larvae before they start active feeding. In addition, experiments where feed is offered before day 5 should be carried out.

There was a fairly good correlation between dry weight of the larvae and the analytical figures on trypsin activity and trypsin/trypsinogen content. When the larvae did not commence growth they did not produce trypsin; when they grew they did.

With the radioimmunoassay it was possible to record a significant increase in trypsinitrypsinogen in growing larvae when the weight gain was significant. Trypsin/trypsinogen is accordingly a useful and significant index of the nutritional status as well as the growth capability of young larvae. It also seems to be a useful tool in screening potential starter feeds for larvae, since good feeds should be able to induce an increase in trypsin/trypsinogen. The only artificial feed which so far has supported growth of the cod larvae is one based on cod roe, and the larvae responded to this feed by producing more trypsin/trypsinogen (Molvik, et al., 1984).

Since cod larvae have no stomach, it is debatable whether the larvae are able to digest feed protein completely to amino acids in the gut. Trypsin and chymotrypsin alone are,

accordrng to therr sonstrate specrficity (Walsh and h ~ l c o x ) , 1970), deflnrtely not able to degrade protelns to free arnrno aclds. This poses the questron of whether tne larvae take up peptldes and even complete particles and perform rntracellular drgesrron (Srroband and Kroon, 1981). Our data do nor relate dlrectly to thrs questron, but rndrrectly by showlng that the trypsln coqtent present ln one larva of 40 pg 1s sufflclent to degrade 25 pg protern an hour. Thls flgure derlves from the fact that 1 ng trypsln from flsh 1s able to degrade about 8 pg muscle proteln at 25OC and pH 7.2 (Grldberg, 1982), or about 0.8 vg at 6 - 8 O ~ . Such a hlgh trypsrn actlvrty rn the gut suggests a major role for extracellular dlgestron In rendering the feed available for growth of the larvae, but that complete hydrolysis to arnlno aclds may lnvolve ~nrracellular dlgestron.

ACKNOWLDGEMENTS

Thrs work 1s supporced by t5e Norweqla~ Research Cou?cll of Flsnerles (NFFR) and by a grant under the joint research and development prolect between the Federal Republlc of Germany

(BMFT) and Norway INTNF), ref. BMFT-09.08.1982, 427-7291-BCT 362/0.

REFERENCES

Barret, A.J., 7979. Introduction to the hrstory and classlflca- tron on tlssue protelnases. In: A.J. Barret (Edrtor), Protelnases rn inammallan cells and tlssues. North-Holland Publlshrng Co., Amsterdam, pp. 1-55.

Corrlng, T., 1980. The adaptation of digestive enzymes to the dlet: Irs physlologrcal srgnlfrcance. Reprod. Nutr. Develop., 20: 1277-1235.

Dabrowskr, R.K., 1979. The role of proteolytlc enzyres rn frsh dlgestron. In: European ffarrculture socrety, No. 4.

Cultlbaclon of frsh fry and rts llve food, pp. 107-126.

Ellertsen, B., Moksness, E., Solemdal, P., S t r ~ m m e , T., Tilsetn, S., WestgArd, T. and @restad, V., 1980. Some blologlcal aspects of cod larvae ( G a d u s moahua L.). Flsk Dir. Sxr. Ser. HavUnders., 17: 29-47.

Grldberg, A., 1982. Aotolysrs of flsh tlssue - General aspects.

Tnesrs, iirilbftrsity of Tromsa, I 1 2 pp.

A]elmeland, K. and Qaa, J., 1 9 8 0 . Flsh tlssue degradation ay trypsln cype enzymes. Iq: J.J. Coaneil (Edrtori, Advances In flsb sclence and rechnology. Frshlng News BooRs Ltd., Farnham, Surrey, England, pp. A56-459.

Hjelrneland, K. and Raa, J., 1982. Characterlszics of two crypsln cype lsozymes lsolated from the arctlc flsh capelrn

( M a l l o t u n v ~ l l o n u n l . Comp. Blochem. Physlol., Vol. 71B, 4:

557-562.

Hjelmeland, K., 1983. Proteinase lnnlbitors In the muscle and serum of cod tGadun m o t h u a i . Isolarlon and characterlzatlon.

Comp. Blochem. Physlol.

Hogendoorn, H., 1980. Controlled propagatlon of the Afrrcan ca-cfrsh, C l a n i a 5 l a z e 4 0 (C. & V . ) 111: Feedlng and growth of fry. Aquaculture, 21: 233-241.

Howell, B.R., 1979. Rearing larval cod ( G a d u d monhua L.) on cultured foods. Coun. Meet. Int. Coan. Explor. Sea, 1979 (F:17) 1-4, 2 Flgs (Mlmeo.).

Huse, I., 1980. An attempt to start feed cod larvae wlth artlflclal dlets. Coun. Meet. Int. Coun. Explor. Sea, 1982

(F:14), 1-3 (Mimeo.).

Huse, I., Opstad, I., Klungsmyr, L. and Walther, B., 1982.

Endocrine and nutrltlonal factors affecting the flrst food uptake by cod larvae. Coun. Meet. Int. Coun. Explor. Sea, 1982 (F:17), 1-7, 5 Frgs (Mlmeo.).

Iwai, T., 1969. Fine structure of gut eplthellal cells of larval and juvenlle carp durlng absorptio~ of fat and protern. Arch. hlstol. lap. 30: 183-199.

Iwai, T. and Tanaka, H., 1968. The comparatrve study of the drgestlve tract of teleost larvae - 111. Eplthelral cells In the posterior gut of halfbeak larvae. Bull. Jap. Soc. sclent.

Fish., 34: 44-48.

Kvenseth, P.G. and Olestad, V., 1984. Large scale rearlng of cod fry In an enclosed pond on the natural food production.

In: E. Dahl, D.S. Danlelssen, E. Moksness and P. Solemdal (Editors), The propagatlon of Cod Gadun mofihua L. Flmdevlgen rapportser., 1, 1984: 645-655.

Molvlk, G., Hjelmeland, K., Rlnga, E. and Raa, J., 1984.

Properties of a new artlflclal dlet for flsh larvae, lncludlng cod ( G a d u n moahua L.). In: E. Dahl, D.S.

Danlelssen, E. Moksness and P. Solemdal (Editors), The Propagation of Cod Gadun monhua L. Fladevlgen rapportser.,

1, 1984: 203-211.

Noalllac-Depeyre, J. and Gas, N., 1976. Electron microscopic

study on gur eplthellum of the tench ( T ~ n c a t 4 n c a L.) wlth respect to ~ t s absorptive functions. Tlssde 6 Cell, 8:

51 1-530.

O'Connell, C.P., 7976. Histological crlterla for dlagnosrng the starving condition In early post yolk sac larvae of the northern anchovy, E n g a a u l ~ n moadax Glrard. J. exp. mar. Biol.

Ecol., 25: 285-312.

Reck, W., 1974. Trypsln. In: U.H. Bergmeyer (Editor), Methods of Enzymatrc Analysis, Vol. 11, Academlc Press, New YorK, pp. 1013-1024.

Stroband, H.W.J. and Kroon, G.A., 1981. The development of the stomach ln U t a a ~ u n t a z e a a and the lntestrnal absorption of protern macromolecuies. Cell 6 Trssue Res., 215: 397-415.

Stroband, H.W.J. ana Van der Veen, P.R., 1 9 8 1 . Locairzatlon of 9roterD aosorptron durrog rransporiz of food In xhe rntestine bf the grasscarp C t e . n u p h a ~ y n y c d o n ~ d e t f a (Val

.

,

Zool., 2 1 8 : 1 4 9 - 1 5 6 .

Szlamrnska, H., 1 9 8 0 . A hlstochemlcal study of dlgestrve enzymes rn prke larvae. Frsh. Mgmt., 1 1 : 1 3 9 - 1 4 0 .

Tanaka, M., 1 9 6 9 . Studles on the structure and functlon of the dlgestrve system rn teleost larvae - 11. Characterrstlcs of the dlgesrlve system rn larvae at the stage of frrst feedlng.

Jnp. J. Ichthyol., 1 6 : 4 1 - 4 9 .

Tanaka, M., 1 9 7 1 . Studles on the structure and functlon of the dlgesrrve system rn teleost larvae - 111. Development of the dlgestrve system durrng postlarval stage. Jpn. J. Ichthyol.,

1 8 : 1 6 4 - 1 7 4 .

Thellacker, G.H., 1 9 8 0 . Rearrng container srze affects morph- ology and nutrltlonal condltron of larval jack mackerel T z a c h u t u n n y m m e t z ~ c u n . ~ l s h . ~ u l l . , U.S., 7 8 : 7 8 9 - 7 9 1 . Thellacker, G. and Dorsey, K., 1 9 8 0 . Larval flsh dlverslty, a

summary of laboratory and fleld research. In: G. Sharp (Editor), Workshop on the effects of env~ronmental varlatlon on the s u r v ~ v a l of the larval pelaglc flshes. UNESCO,

Interaovernmental Oceanoara~hlc < & Commlsslon, Workshop Report

- -

2 8 , pp. 1 0 5 - 1 4 2 .

Thompson, B.M. and Riley, J.D., 1 9 8 1 . Egg and larval develop- ment studies in the north sea cod ( G a d u n mozhua L.). Rapp.

P.-v. Reun. Cons. int. Explor. Mer, 1 7 8 : 5 5 3 - 5 5 9 .

Walsh, K.A. and Wllcox, P.E., i 9 7 0 . Serrne proteases. In: E.G.

Perlmann and L. Lorand (Edl~ors), Methods rn Enzymology, Vol. XIX, Academlc Press, New York, pp. 3 1 - 4 1 .