Thesis (MSc)--Stellenbosch University, 2001. / ENGLISH ABSTRACT: Smoothhound sharks Mustelus mustelus (Chondrichthyes: Triakidae) were conditioned to associate a compound auditory and electrical conditioned stimulus (CS) using the presentation of food as the unconditioned stimulus (US). This was done to investigate the general conditioning process of sharks, specifically the use of conditioned stimuli, the time frame required for conditioning, the retention of conditioned responses and the individual learning capabilities of sharks in classical conditioning experiments. Conditioning was done in two CS experiment replicates, in which the CS and the US were paired to elicit a conditioned response (CR: entering of a feeding area). Shark behaviour in these replicates was compared with those of sharks in a CS control. In the CS control, the CS and US were not paired; consequently no CS-US association was formed. In contrast, half of the sharks in the 1st and 2nd CS replicates formed the conditioned CS-US association after five and six days of conditioning training, respectively. This conditioned association was also retained over a two-month rest period. However the rate of conditioning of sharks varied, which was thought to be due to differences in motivational drives, particularly hunger. These experiments demonstrated that classical conditioning could provide a mechanism whereby smooth hound sharks can adapt to utilise new food sources. However these experiments could not demonstrate whether auditory or electrical cues are more biological relevant signallers of the presence of food for smooth hound sharks. According to the "belonging ness" phenomenon, for conditioning to be successful, the CS should be biologically relevant to the US. Therefore by comparisons of the conditioning success with different conditioned stimuli, the biological relevance of each CS can be determined. To determine the biological relevance of each component of the compound CS, an auditory or electrical CS was paired with the presentation of food (US), in separate experiments. The conditioned CS-US association formed in the majority of the sharks after 60 and 57 conditioning trials in the auditory CS and electrical CS experiment, respectively. This suggests that both auditory and electrical cues are biologically relevant signallers of food, facilitating the oppurtunistic feeding behaviour of smoothhound sharks. However, after five days of conditioning, conditioning with the electrical CS had proceeded further and the electrical CS-US association was stronger than the auditory CS-US association. This indicates that electrical cues are more biologically relevant due to past feeding experiences. Thus, classical conditioning during past feeding experiences can influence the feeding behaviour of smooth hound sharks. / AFRIKAANSE OPSOMMING: Gladde-hondhaaie Muste/us muste/us (Chondrichthyes: Triakidae) is gekonditioneer om "n gekombineerde klank en elektriese konditionerings stimulus (CS) met voedsel, die ongekonditioneerde stimulus (US) te assosieer. Dit is gedoen om die konditionerings proses, spesifiek die gebruik van konditionerings stimuli, tyd benodig vir konditionering, behoud van die gekonditioneerde gedrag en individuele leer-vermoëens van haaie in klassieke konditionerings eksperimente te ondersoek. Konditionering is gedoen in twee CS eksperiment replikate, waarin die CS en die US saam aangebied is, om "n konditionerings reaksie (CR: om in 'n voedings area in te beweeg) teweeg te bring. Die gedrag van haaie in die CS replikate is vergelyk met die van haaie in "n CS kontrole. In die CS kontrole is die CS en die US nie saam aangebied nie en gevolglik is geen gekonditioneerde CS-US assosiasie gevorm nie. Daarenteen is die CS-US assosiasie by die helfte van die haaie in die 1st en 2de CS replikate gevorm, na vyf en ses dae, onderskeidelik. Hierdie gekonditioneerde CS-US assosiasie het selfs behoue gebly na 'n twee maande rus periode. Die tempo van konditionering het egter verskil, moontlik a. g. v. motiverings verskille, veral hongerte. Hierdie eksperimente het gedemonstreer dat klassieke konditionering kan dien as 'n meganisme waarmee gladde-hondhaaie kan aanpas om nuwe voedsel bronne te benut. Hierdie eksperimente kon egter nie demonstreer óf klank stimuli óf elektriese stimuli meer biologiese relevante seine van die teenwoordighied van voesel is vir gladdehondhaaie nie. Volgens die "belonging ness" verskynsel, moet die CS biologies relevant to die US wees, vir konditionering om suksesvol te wees. Dus kan die biologiese relevansie van verskillende konditionerings stimuli bepaal word deur die konditionerings sukses van elke CS te vegelyk. Om die meer biologiese relevante komponent van die gekombineerde CS te bepaal, is 'n klank of elektriese CS saam met voedsel aangebied in aparte eksperimente. Die gekonditioneerse CS-US assosiasie is gevorm by die meeste van die haaie na 60 en 57 konditionerings proeflopies in die klank en elektriese CS eksperimente, onderskeidelik. Hierdie resultate dui daarop dat beide klank en elektriese stimuli biologies relevante siene van voedsel is, wat die oppurtunistiese voedings wyse van gladde-hondhaaie fasiliteer. Konditionering met die elektriese CS het egter verder gevorder en die elektriese CS-US assosiasie was sterker as die klank CSUS assosiasie na vyf dae van konditionering. Dus, mag elektriese stimuli meer biologies relevant wees weens vorige voedings ervaringe. Dit illustreer dat klasseke konditionering tydens vorige voedings ervaringe die voedings gedrag van gladde-hondhaaie kan beinvloed.