-- ~--­ ~ I i ,rl . J' This paper riot to be eited ..\ithout referenee to the authors International Council for the Exploitation of the Sea .. Shellfish cOmrIUttee C.l\L 19951K:40 Ref: F The Azorean squid L'!lig() jorbesi (Cephalopoda: Loliginidae) m captivity: reproductive behaviour Filipe M. Porteiro. Joao M. Gon~ves. Frooerieo Carcligos & Helen R. MartinS Departarnento de Oceanognma e PescaS da Universidade dos A~ores. PT 9900 Hortl Abstract Reproductive beha\iour of L. förbesi in captivity iS described. Male·riiate agonistic internctions increaS'ed when a female is introduced between triaIes. Mate pair formation is quickly established with acorisequent hier.irchy of dominance bet\veen males. During such beha\iour the male coristrained the female againstthe tank wall. isolating her from theother maIe present Exchange of the partnerS in~olved in courtship was fe,,, times observed. A chromatic pattern associated with courtship was diSi>layed continuously by the dominant male. Thfee females laid eggs on the tank PVC walls. One of them SUI'\ived 12 days after spanning. Tbe reproductive orgails of post·spa,..ning feInales bad mature. m:uuring :md immature oOc)1e. Spa,..ning b7haviour :ind :i briefdescnption of the si:iaWn is presented. An update ofthe components ofbeba..iour is provided. Introduction Squid ai-e schooIing soeial invene1>cites ..\ith high developed system' of comniimicacion :uid camouflage. They ure able to exhibit many ritualised beha\icumI patterns tbat can be described as a coinbinatiori of chromatic. postumI and mcvement componerits (Packard & Sanders, 1971; Packard and Hcehberg, 1977; Haßlon. 1982). Chromatic componentS, rriade up by seleetive nervous e~eiciticn of units cf chromatcphores and iridophores (P~ekard. 1982) are the eore of the visual commumcation amcng cephalopods. Body patterns are associated with specific behaviour as feedirig, predator avoidarice. courtship. ete. and they can alter during the omogeny of one spccies (e.g.• Hanlcn & Woltering, 1989). The accurate descripticn of the eomponents of pattemmg is required for several biologieal approaehes as quantitative ethOIcgy. b~haviotimI taxcnomy and phylogem:tic analysis (see Hanlcn. 1988 a and Hanlon et. al. 1994). However, the description of tbe behaviour is somewhat complicate due .to the multiplicity arid ambivalence of the sigrials combined with the dependence cf their expression on extemaJ. environment (Moynihan & Rodaniche. 1982). Obviously, observations in the field ure the ideal when this is possible. but when squid gi-ounds ure unkno\\n or not accessible. studies in capti\;ty are an alt~n1ative. Also. the proximity to the squid and the confuted space can turn some behavioural patterns conpiscous. . Oetopods and sepiids were the cephalopods preferred for beha\ioural sriidies, both in situ and in labor.ltory. Nevertheless, the behaviours of some myopsid were extensively investigated (in field e.g.• :\Ioymhan & Rodaniche. 1982~ Sauer & Srßale. 1993: Hanlon er al. 1994 and iIi capcivity e.g.. Hanlon 1982, Anlold, 1990). Lvligo forhesi is the oilly loliginid species '. lhing arourid the Azores islands. It inhabits inshore water from the surfuce. ai night. -- - -- ---- ., ·" dO\\n to 400 meters during da)time. A daylight squid jigging fishery was kno\\n sinee the last eentury (Drouet, 1858). Mainly maturing and mature squid are caught (Martins, 1982). However, the spa\\ning grounds of this population and spa\\ning migrations bave never been deteeted. Naturallaid eggs have been found only once (porteiro & Martins, 1982) in the Azares. The Azorean L. forbesi proved to be effieient for culture e:<:perirnents, at least for maintenance and reanng (sensus Boletzky & Hanlon, 1983). Squid embryos (laid by captive females) were transported to Marine Biomedical Institute, Texas arid two successful reanng. trials were carried" out (Forsythe et al., 1989; Hanlon et al., 1989). During maintenance preliminary studies of adult squid behaviour" were condueted (Porteiro et al., 1990). There, 17 chromatic, 6 poStural and j movement components were described as weil as feeding arid agonistic behaViour. Ho\vever, reproductive behaviour was not extensively emphasised. Presently, we bave a closed sea-water system designed specifically for squid culture (see Gon~ves et al., 1995) which allow to maintairi simultrineously' a signifieant number oflarge squid. In this study we present additional qualitative information of the behaviour of captive squid mainly related with courtship and spa\\nmg. recorded daily and several ..vater parameters controlled regularly (see Gon~ves et al. 1995, for a complete description ofthe system). The number of squid inside eaeh tank was variable (1 to 7) and different combinations of sexes were mamtained (e.g., 1 female and two males, 2 feIriaIe and 2 lna1es, 1 female"and 4 males, etc.). Individual squid were· distinguished thrOught natural skin rilarks cr artificial tags Observations v."ere performed during day time, from the top of the tanks and laSted from 10 minutes to ca. 2 haurs. While observing, the black plastic cover was partially elevated and the exterior light held at minimum to avoid major disturbance. The water influx was, also, suspended. Squid were fed once or twice a day. \\-:lS Trachurus picturatus, Pagellus bogaraveo and Scoinber japonicus were the main food items. Feeding took place in the marning and atnoon. The hierarchical oi-ganisation of bemiviouf as described by PackaCd & Sanders (1971), Packard & Hochberg (1977), and Hanlon (1982) arid the behavioUraI components terminology previously established for" loliginids (Hanlon, 1982, 1988; Porteiro et aL, 1990;SaueretaL~ 1992, 1993;Hanlon al., 1994) were adoptoo. er Results Ari Material aod :\'Iethods Several squid were rriainciined during t\Vo sets of experiments (16 Oetober to 27 December 1994 and 2 February to 18 April). The squid size ranged was between 275 to 639 DML. Squid were eaught by jigging, tr.l.nsported ashore ::md kept in a elose sea water system. Basically this arrangement consist af three circular t:lnks (3.6 m 0 arid 0.90 high: ca 80001 of natural sea \\-ater) painted with verrical dark stripes on the \\-alls and ca. ..lern of gravel an bartom Light was kept at dim during all the experiments. temperature mm m average of 7.6 squid were maintained simultaneously in captiviiy and the avernge survival waS 17.7 days (up to 46 days) in the first experiment. During the second trial the average survival \\-:lS longer (22.8 days) niainly due to the preferential maintenance of mature, maturing squid and sexual segregation in same tanks. Behaviour .Wale-male ü1teractions ~[ale-male agamsttc be.'taviour was observed many times. In the tanks a hierarchy af domirumce was eSt:lblished. ~crmaIly, the weal<er male die with major damaged areas after few d::i.ys in captivity 2 1 and successively all squid but the dominant male die. The most obvious damaged areas were usually the same: the latero anterior part of the mantle, funnel and head. Injuries on ehe fins and on ehe posterior tip of mantle were caused by abrasion and collision on the tank walls, respeetively. Lesions on the anterior dorsal tip of the mantle were commonly observed but eould not be related with any type of eontaec. Agonistie interactions oceured between male of markedly different sizes are maintained together and between males of the same size or of different size when competing for reproductive dominanee,. Fig. 1 - Dorsal view of counship body pattern of adult male squid Loligo forbesi in capnvlty. Composed by the chromatie components, White arm tip. Accentuated tesris and .rlccentuaced digestive gland. All the movement eomponents preVlous deseribed for the speeies (Parallel posirioning, Fin hearing, Forward rosh, Chase and Flee) were observed. Some variation on Forward rosh was reeorded: the aggressor eharges ehe subordinate male by jetting against him the tentacles resulting often in eontaer. TItis component was preeeded by Chase (by the dominant male) and culminate with Flee by th~ subordinate male). All dark is the common body pattern in th se intemctions (Dark flashing was ::Uso observed several times). Perpendic:liar positiomng, a new movement componenr.. was recorded dunng low stress situations. TItis is characterised by the presence of several males elose together moving on different vertic::u levels mainuining between them perpendicular positions. The chronie chromatlc pattern was predominantly observed during such behaviour (Clear. Dorsal scripe). lvfaie- emale courtshiv behavlOur (mare paIr/ormanon Mature male-femaie inter:lctiv~ benaviour expressed by captlve squid was characterised by Pair formation. When a female was in a tank with males, one of the latter would shortly afterwards change his behaviour. The male exhibits his dominance isolating the female from the other male squid. He constrained her against the tank wall limiting her 'swimming space. A short distance was maintained continuously between the two and prolonged fin to fin contaet occurred. A circular motion around the tank wall was prevalent. in opposition to the normal forward backward radial trajectory. It was always the male who ~nforced ehe pair jormation behaviour. However, the preservation of the eouple depends of the male aptitude to ;lvoid the evasive action of the female (unhealthy males are unable to maincain a pair with a hea1th female). Sometimes male sexual dominance exchange was observed (i. e.. , the dominant male is replaced by another). Also, the dominant male may change his partner preference (observed when a new female was introduced in the tank). Most of ehe pairs were suble for se'/eral days. Jockeying J.lld Parrymg (sensus Hanlon et ai. 1994) were observed several tunes and often culminated in violent male-male 3 agonistic behaviour (see above). Stress also inereased in the tank also when a male interfered with the trajeetory of the pair. Subordinate squid tend to avoid such ................................ .. Figure 2.Top plan of the tank and detail of the area selected by squid to lay eggs (see text). interactions whieh is diffieult given the eonfined space ofthe tanks. The courrship behaviour is characterised by a specific chromatic pattern. That pattern is more obvious in the male. At least, three dorsal chromatic components are ~xpressed to fonn the Courtship body pattern (Fig. 1): Accentuated testis, Accentuated digestive gland and White arm tips. Strong green iridescence on the dorsal mantle areas was exhibited. During all watehing periods the dominant male frequently showed this pattern (it ean pt:rsist during a long time). However. it was more conspieuous when another male was in the proximity of the pair. The female display was of a less obvious chromatic pattern. but also some distinet components were consistently present (e.g.. Don'al stripe. Ivfantle spots, Dark fin fine). Whlte arm fips (as in males). Accentuated ovary, Accennlated nidamental oalands and Asvmmetric ocular • spot (on the male side) were also observed duling more intense interactions. As....m metrv of the ehromatic eomponems w~ 5ho~ (stronger on the adjacent squid side} by the MO parmers. Spawning behaviour Three spawns were laid by captive females during the period of observation (Tablel.). All squid initiated spawning during aftemoon and continued through the night. Two of the females (Al, B2) that started egg-Iaying, finished one day 1ater and then died. One female (B 1) lived for 12 days after spawning. During that period she ate 4 small meals (average daily feeding rate of 2.1 % against 6.5% before spawning). The female Al bad paired with a dominant male (two male squid were in the tank) and B 1 forrned a stable pair with a sole male in the tank. The squid 82 were all the time alone. Sinee light seemed [0 influence behaviour of the squid, observations of the spawning process were gready limited by the dark condition in the tank. However, some postural and movement components were observed. The female approached the egg mass (previously laid) with eonie arms and when in contact with them she spiayed the arm tips and attached a new str.lI1d (Esg holding). This activity took around _0 seconds or less. During this operation the squid unduiated the fins vigorously in a forward movement. The body was kejJt parallel to the substrarum. 4 .. { am Frequently,the female sW fast to theegg strands handled them and jetted ,water violently toward the spa\\n in back",..ard mo\'ement. This beha\iour Y,as observed manv tirries. Male also made such be~viour, but less frequently. ThemaIe was alwavs swimming elose to the fernale, showing the courtship body pattern ~ith high iridescence (Mate guarding) . At least the first strands laid by the squid were fixed elose to the surfaee on the tank wallover a PVC liner, (fig.2). The squid B2 iaid eggs on the same place where the prrnous strandS were laid by BI. Water flow WaS at a maxm'l1.uu at spa\\n location (oxrsen level \Vas, elose to saturation unifonnly in the tmks). Later some capsures were disperse and others were in a burich, throughout the gcivel bottom. Few of thern were cemented on the basalt gravel. It was not cIear to wh::ii extent these strings v,'ere displaced from the PVC wall by the jenmg and manipulation behaviour of the female. Spawn The spa\\ns were eomposed by well-sh:iped egg strands (normal strands) and by a few with redueed dimensions and with fewer eggs or none (balloon Strands). Fourtythree common strands were laid by both squid Al arid BI. The squid B2 lay significandy less (15) of whieh 50% of them were empty. The remairiderS were shorter and bad fe\v eggs thanthose from the other. squid. The norIDal sti':inds. bad bet\~,een 25 and 90 eggs and the percentage of fertilisation was 64%. balloon strands bad a lower percent:age of fertilisation. Reeentlv fixed strings, (during the fust 2-3 houes), • were more siretchilble, thin and long, but afterwards they decrease in size, and the gelatinous ninics became more finii The. Table 1. Refereneeof L. jorhesi female squid timt spa,\n in captivity. DML, i - I?0rsai man~e i~gth when captured; BW i-Total body weight when captured; OVW f - Ov:uy welght after spawnmg; ODPW r- Proximal oviduet weighl after spawning. .Ref. Date of capture Days captive before Days fram DMLi sp'a\Vnmg , (nun) to dead B\Vi OVW (g) f f (g) (g) ODPW spawnimz. AI BI B2 05.11.94 16 09.02.95 03.03.95 8 7 Reproducti....e organs /emales 0/ 1 12 1 31 330 993.0 279 835.0 607.9 33.7 16.8 55.4 56.9 0,8 6.0 post-spawning Post-spa"ning ferilales still had ~ture, maturing and immature eggs m the proxinial oviduct and ovary (see Tabl~ 1)~ All had rnatcd but fe,v sperm were in the bursa colmlatr:x. The ovary of the fe=nale BZ lud a high quantity of ocC:.1e c!C:,~ to rnaturation but less few eggs in the oviducJ.1 comple::.. The ferI1aIe BI bad i few hundredS mature eggs in the oviduct and ooc:-te in v::l.nous devdcpm.:nul .rt::l.ges in the ovar)'. Discussion Agonistic behaviour be:'veen males ~~ ~e main reason for mortallty both by colliSlon and abrasion on the tank walls or by direet charge. This behaviour \vas.obs,erved ,in several species kept in captiVlty but bodily contact has oDly been rarely observed (see Arnold. 1990; Hanlon et al. 1983). Prcbablv. the coniined space in the tanks ccmbin~d \\;th the big size ?f the squid :l2!!1"::l.v::l.te .menistie interactiens ccmpared t;-animals in the wild. In the wild the 5 , , subordinate squid can run away before being touched during an attack. These interactions are more evident when different sizes male squid are mairitained together or when males compete for females (see Hanlon, 1990). No cannibalism was recorded probably due to small differences bet\veen the size of captive squid. Pair formation 'was observed for the firsi time for L. forbesi. HO\vever, it has been described for several other squid held in captivity (e.g.• L opaleseens by Hurley, 1977; L. pealei by Amold, 1990; L (D.) plei by Hanlon, 1981) as weIl as in the natural spa\\ning grounds (e.g.• Sepioteuthis sepioidea by Moynihan & RodaIiiche, 1982; L. v. reynaudii by Sauer & Smale, 1993). During our observations, captive pairs seem 10 be quite stable and male dominant, except when a weak ll1a1e is not abte to contain the evaSive behaviour of the female. Tbe latter behaviour can support the assumption by Hanlon (1994) that female choice and evaluation of the partner is an iritPOrtaIlt reproductive str.:i.tegy bet\veen squid. However, llla1e dominance has been described by several authors based on the displacenierit of ari intruder male by the paired male and the maintenance of the originill pair. Pair stability on field is not kno\\n but it probably split during riight (see Sauer & Smale. 1993). The long duration of the pairs obsenoed in this stUdy can be an arrifice due to thecaptive condition. Although, we did not observe ccpulation, we know that sperm is stored only in the bursa copulatnx (Porteiro & Martins. 1994) which indicate that just head-to-head mating occurs. Very little sperm was found in the bursa copularnx of post-spa\Vnmg females. This suggests that the amount of sperm in this organ is not enough for fertilisation if a prolonged spa\\ning period occur or if one intensive sp:mning is l1lade. We :lgree with HanIcn (1994) that squid reprcductive tactics would be better understood if anaIysed in a sperin competition contest. The existence of small mature r"sneaker'') males \"ithin this spec:es (Martins. 1982: Boyle & Pierce. as 1994) weIl as Jockey;ng and Parrying movernent componenis cari indicate an alternative reproduCtive str3.tegy, aS observed for L v. reyruludii (Hanlon et al. 1994). The meclUiriiSiris of sperm competition are not lciIö,,,n for the majority of cephalopodS. Arnold (1990) suggest thai L. pealei pair formation and consequently reproductive doininaßce by the Iriale. is established by a visUa1 Stimulus (i.e.•. the presenee of egg masses m the aquarium). Hurley (1977) found some evidence that egg deposition is visually Stirriulated by the preSence of preVious spa\Vns but he could not ioelate the male reproduCtive dominance 10 presence of eggs. In our case pair formation occurred without any visua! externaI stizllulus, it eStablished soon as a feIriale was mtrOduced among ritales. Male chromatie eomponents related 10 reprOductive dOßUnance behaviour have not been deseribed previously for this species. White arm tips cOuld be associated 10 the importance .of the arms in rrlating and egg mass h3ndling. The Accentuated testis is a component very common arnong 10liginidS and interpreted as a sign of mature coridition. Tbe component Accentuated digestive gland is described hen: for the tirst time alld it could be a sign related to theeondition of the squid. since this organ is used as storage resen'e. Similaicourtship body pattern is not found iri the literatUre. Is accepted that acute body patterns (expressed during few seeondS or minutes) are more useful for behavioural ta.'conomy and phylogenetie approaches tfcin chrome patterns (whieh can be displa):ed during lang periods). However. courtship body pattern (as, a chrome pattern) is stereotyped, species-specific and orie whieh proVide more information about the species chafaeteristies (Hanlon, 1988). It would have ethological interest to compare the behaviour between Azorean and Europe::m populations of L. for~esi smce they are genetically separated (Brierley et al. 1993; Pierce et aI.• 1994) and body patterns c::m be used probably as a eharaeter which shows variations at the population or sub- as was as 6 • . ( • species level (Hanlon, 1988; Hanlon et al. 1994). One example of different reproduetive behaviour benveen these nvo populations is show in the nurnber of eggs within each str:md. Lurn-Kong (1993), reported a mean vaIue of 109 eggs per string against less than 90 found by us in several spa\'fns. Also, the percenmge of mated immature females in Azores seems to be higher than L. forbesi from elsewhere (Porteiro & Martins, 1994). In order to provide basic information for further ethologic::u companson it is presented an update of body components recorded for the Azorean population of L. forbesi (Table 3). Table 3. Update of the behavioural components recorded for Lo/igo forbesi (based on Hanlon, 1988; Poneiro et al., 1990; this scudy; and unpublish observations). Chromatic components: Dark components: • 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 'J,,: All dark Later::l1 mantle streaks Ring Fin and mantle spots Infraocular spot Dark fust or third arms Darkarms Mid-ventr.ll ridge Later::l1 mantle stripe Dark fin line Mantle margin stripe Dorsal stripe Shaded eye Postur::l1 components: 1. 2. ... .J. 4. 5. 6. 7. 3. Two raised arms DO\mward curling Extended tentac!es Splayed arms Ellipsoidal eyes Bottom sitting Egg holding Perpendicular positioning Fresh laid egg strings were sirnilar to those described for L. pealei by Roper (1965) (leng :md soft). The duration ef softness ef the string tunics can presumably be rehlted \"ith the fertilisation process. Tnis me::..:..:mism remains co be studied. Variation in the way that squid make their sp:l\\ning bed is knC\\n (see Vechicne. 1983: Sauer er al. 1992). From our Light components: 1. 2. 3. 4. 5. 6. 7. 8. 9. Clear Dorsal mantle eollar iridophores Dorsal iridophores splotches Gold irideseent sheen Aceentuated testis Accentuated nidamental gland Accentuated ovary Accentuated digestive gIand White tip ofarms Movement eomponents 1. 2. ... ~. 4. 5. 6. 7. 3. 9. Par::l1lel positioning Fin beating Forward rush Chase Flee Inking Jetting Jocking and Parrymg Mate guarding observations it is difficult to say how this species makes their spa\'.ns in nature. The loeation where the squid fi.'ced their eggs in the tanks is probably related to the water current or to the existenee of a PVC fold (see Fig.l). The later supposition. if correct. can suggest that cre"lices or smaIl caves are used fcr spa\ming as discussed before by Porteiro & ~[artins (1992). L. 7 vulgaris in Madeira spa\..n in such bottom 1-- structures (p. Winz. pers. comm.) Death of captive squid after spa\..ning haS been recorded by several sttidies and accepted as a natural behaviour. The presence of mature, maturing and immature eggs in the reproduetive system of post-spa\'.ning females and the survival of one female for several days (12) after egg- laying could support the assumption of a prolonged intermittent spa\..ning before exfuluStion. Several questions arise when this problem is examined: v,'hy does the squid die befoIe all the eggs are laid? is this ecologically justified? have the "üd femaIe the same behaviour or is it influenced. by captive condition? does she need high food input after spa\..ning one batch of eggs? This should be investigated carefully to 6stablish the reproductive behaviour ofthis loliginid. a AcknowledgementS This work could not be done ",ithout the assistance of Paulo Martins. Thanks to Ricardo S. Santos, to commentS and revision of the manuscript. . This ?''ork was funded by the European Commission within the frame of EC research programme in the Agriculture and Fisheries seetor(AIR. Contract No. AIR l-CT92-573). We also thank to the Region::U. Secretariat of Agriculture and Fisheries of the Azores Government who gave us permission to use their installations at Porto Pim beach (Rorta) and supported its running costs. References Arnold. I. M 1990. Squid mating beruivior. Pp 65-75, in: Squid as Experimental Animals, D.L. Gilbert, W. I. Adelman Ir &. I. M AInold (Eds). New York. Plenum Press, 516pp Brierley, A. S. ,. 1. P. Thorpe. M. R.. Clarke &. H. R. Martins, 1993. A preliminary biochemic:ll genetic investigation of population stnleture of Loligo förbesi Steenstrup. 1856 frOm British Isles 'and the Azores. Pp, 61-69, in: Recent Advances in Cephalopod Fisheries Biology, T. Okutani. R.K. O'Dor & T. Kubodera.. Eds. Tokai UDiversity Press. Tokyo, Iapan. 752 pp. Boletzky, S. v. &. R. T. Hanlon., 1983. A review ofthe'laboratory maiIitenance, rearing, and culture of cephalopod molluses. •\femoirs ofthe .Vational.\luseum ofVictoria. +t:147-187. Drouet H., 1858. Mollusques manns des lIes Acores. ....Iemoires de la Societe, Academique de I:-lube. 22: 53 p. . Forsythe, 1. W. &. R. T. Hanlon. 1989. Gro'\th of the Eastem Atlantic squid. Loligo forbesi Steenstrup ~follusca: CephaJopoda).Aquac'.llture and Fisheries .\fanagement, 20: 1-14. Gom;aJves. 1. M.• F. M. Porteiro. F. Cardigos & H. R.Martins, 1995. Adult squid Loligo forhesi (Cephalopoda: Loliginidae) in c:lptivity: maintenance arid transport International Council for the Exploitation ofthe Sea. C.M. 19951F:14 Ref. K Hanlon. R. T., 1981. Courting and rnating beha\ior in the squid Loligo (Doryteuthis) plei (abstract). International Society oflnvercebrate Reproduction.l p. Hanlon. R. T., 1982. The functional orgWzation of chrolnatophores and iridiscent cells in the body patterning of Loligo plei (Cephalopoda: Myopsida). Malacologia, 23(1): 89-119. Hanion. R. T.. 1988. Beha,ioraJ and body p:iuernig charncters useful in ta.xonomy and field indentific:luon of cephalopods. .\lalacologia, 29( 1): 247-264. HanJon. R. T., 1990. Maintenance. reanng, and cUIture ofteuthoid and sepioid squids. pp. 35-62. in: Squid as Experimental Animals, D.L. Gilbert, W. J. AdeIxii.an Ir &. I. ~l Arnold (Eds ). New York. Plenum Press. 516 pp. HanJon. R. T., ·1994. Mating systems of cephalopods: contrasting strategies and the possible iruluerices of. sperm campetition behu\ior. Programme and AbstraCts of the Napoli CIAC Symposium "The Behaviuur and .Vatural History of Cephalvpocis" (Vico Equense. 5-11 June. 1~'94): 14. 3 • • • r • Hanlon, R. T. ,. R. F. Hixon &. W. H. HulIet. 1983. SU1'\ival. grO\\th and beha\ior of the Ioliginid squids Lo/igo p/ei. Lo/igo pea/ei and Lo//iguncu/a brevis (Mollusca: Cephalopoda) in closed water systems. Bi%gica/ Bulletin. 165(3): 637-685. Haßlon. R. T.•• W. T. Yang. P. E. TUrk, P. G. Lee &. R. F. Hixon. 1989. Laboratory and estimated life span of the Eastem Atlantie squid, Lo/igo jorbesi Steenstrup. 1856 (Mollusca: , C:phalopoda) A.quacu/ture and Fisheries .\lanagement, 20: 15-34. Hanfon, R. T.•. M J. SffiaIe &. W. H. H. Sauer. 1994. An ethogram ofbody patterning behavior in the squid Lo/igo vu/garis reynaudii on spa\'i.ning grounds in South Africa. Bi%gica/ Bulletin, 187: sea culture 363-372. ,, , , " , Hudey, A. C., 1977. Mating behavior of the squid Lo/igo opa/escens. Marine Behaviour and Physi%gy, 4: 195-203. , , LUni-Kong. A.,1993. Oogenesis. fecundity arid pattern of spaWning in Lo/igo jorbesi (Cephalopoda: Loliginidae). Ala/ac%gica/ Review, 26: 81-88. Martins. H. R.. 1982. Biologica1 studies of the e.xploited stock of Lo/igo forbesi (MoIltisca: Cephalopoda) in the Azores. Jouma/ oj}.,larüie Bi%gica/ Association ojthe United Kigdom, 62: 799-808. Moynihan, M &. A. Rodaniehe. 1982. Tbe beha\ior and natural bistor)' of the Caribbean reef squid Sepioteu/his sepioidea. Am'ances in Ethology, Supplements to JoiimaI of Comp:irative Eth%gy, • 25: 113 p. , OTIor:'R. K., J. A. Hoar. D. M Webber. F. G. <:arey. s. Tanalca. H. R. Martlns & F. M. Porteiro• 1994. Squid (Lo/igo !orbesi) performanCe and metabolie rates in nature. Alarine and Freshwater Behaviour and Physi%gy. 25: 163-177. Packard, A. &. G. Sanders. 1971. Body patterns of OctopUs vu/garis and rnaturation ofthe resPonse to disiurbänce. Animal Behaviour, 19: 780-790. Packard. A.1982. Morphogenesis of chfClinatophore patterns in cephalopods: are morphologica1 , physiologica1 "unitS" the same? .\fa/ac%gia. 23(1): 193-201., ' Packard, A. &. F. G. Hoeheberg, 1977. Skin patterning in OetopuS arid other Genera. PP. 191-231. in: The Bi%gy oj Cepha/opods, M.N. Nixon & J.B. Messenger (Eds.). London.: Academie Press, , Zo%gical Society ojLondon.: 1977:.615 p. , , " ,' Porteiro. F. ~l &. H. R. MartinS. 1992. First finding of natUral Iaid eggs frOm Lo/igo forbesi Steenstup. 1856 (Mollusca: Cephalopoda) in the Azores. Arquipe/ago. Life and Earth Sciences. and 10: 119-120. • ",.,. poneiro",F. ,M & H. R. MartinS. 1994: 'Biology of Lo/igo jorbesi SteeDstrUp. 1856 (MoIiusca: C.:phalopOda) in 'tbe Azores: sampie composition and rnaturation of squid caught by jigging. Fisheries Research. 21: 103-114. Porteiro. F. M.•. H. R. ~1artins &. R. T. Hanfon. 1990. Some obsemitionS on the beha\iour of adult squids. Lo/igo jorbesl in eapti\;ty. Joumal oj.\larme Bi%gical Association oj rJ.K.• iO: ~594i2. Porte:ro. F. M.• J. ~l Gon<;:l1ves. F. Cardigos. P. ~lartins & H. R. Martins. 1995. Tbe squid Loligo jorbesi (C~phalopoda: Loliginidae) in captivity: Feeding Gromh. Internationa/ CcJunci/ jor Exploitation ofthe Sea.C.M. 19951K: ~1 Roper. C.• 1965. A note on .:gg deposition by DOf)teuthis plei (BlairiVille. 1823) and its comparision with other nonh americ:m loliginid squid. Bui/elin oj.\laril1e Science. 15 (3): 589-598. Sauer. W. H. H. &. Mo 1. SffiaIe. 1993. Spa\\ning beha\ior of Lo/igo vu/garis reynaudii in shallow coästal waters of the south~tem Cape. South Africa. Pp ~89498. in Recent Advances in C.!pha/opod Fisneries Bi%gy. T. Okutani. R.K. O'Dor & T. Kubodera. Eds. Tokai University . ' , Press. Tok)·o. 752 pp. Sauer. W. H. H. .. ~1. 1. Smale &. M R. Lipinski. 1992. Tbe Ioeation of spa\\ning grounds. spaWnirig arid schooling beha\iour of of the squid Lo/igo vu/garis reynaudii (Cephalopoda: Myopsidae) off the Eastem Cape Coast. South Africa. .\fanne Bi%g)!. 11~: 97-107. Vecchione. M.. 1985. In-situ observations on a large squid-sp:l\\ning bed in the Eastem Gulf of Mexico. •\la/ac%gui. 29( 1): 135-1~ 1. .md 9