Thesis (MScAgric)--University of Stellenbosch, 2004. / ENGLISH ABSTRACT: Cape hake (Merluccius capensis and M. paradoxus) is commercially the most important trawl-caught fish off the South African, coastline and due to current intensive fish processing procedures Cape hake contributes the most to the total fishwaste production. Besides its commercial importance fish is also regarded as one of the single most important consumable natural resources, either in the raw or frozen form. Most of South Africa's commercially trawled demersal fish has already been partially cleaned (i.e. headed and gutted) before landing with non-marketable bycatch and hake-waste normally disposed of as discards, resulting in a waste of a potential protein source. This study was thus aimed at fulfilling several objectives namely: observing the current large-scale commercial Cape hake harvesting procedure; constructing prediction models for several morphological parameters (whole hake mass, headed & gutted hake mass, hake head mass, hake head length, hake head breadth and hake head height) of Cape hake (Merluccius ssp.), using whole hake length as the independent variable; and determining the chemical composition (moisture, protein, fat, ash, macro and trace elements) of several hake head sections (clean head, neck flesh, tongue, tongue cartilage, jaw, gills, heart, intestines, gut, kidney, kidney & kidney bone and gut & gall); determining the effect that storage has on the fatty acid profile of both the clean head and neck flesh sections. The results obtained would supply necessary data required for techno-economic investigations in the use of hake heads. For each of the six prediction models constructed, there was an increase in the variance of the data points of categories 3 (64-80 cm) and 4 (>80 cm) as opposed to categories 1 (30-46 cm) and 2 (47-63 cm). This could be attributed to a smaller sample set for both categories 3 and 4 or due to an expected increase in the variance when investigating larger biological samples. There was also a clustering of data in the three areas for each prediction model namely, within category 1 and across categories 2 and 3 and 3 and 4. This emphasised the latitudinal stratification of the Cape hake population by age, hence their stratification by size. The prediction models constructed for both boat trips 2 and 3 differed significantly (p<0.01) from that of boat trip 1, with the exception of the hake head length (cm) prediction model. The constructed prediction models, for each of the three respective boat trips, showed good predictive abilities as was indicated by the low Mean Square Error (MSE) values for the test sets, and high Pearson's correlation coefficient (r) values. These prediction models can be used in the fishing industry with confidence for Cape hake within the time frame each respective boat trip was carried out. The neck flesh could be regarded as the most important concerning chemical composition whereas the jaw could be seen as the most important when one considers mineral content. This therefore means that the jaw section, once appropriately processed is a potential Ca, Na and Fe source for supplementing diets of people suffering from a Ca, Na or Fe deficient diet. With regard to chemical status the neck flesh section is seen as a good potential source of both protein and fat, which could be attributed to the fact that hake muscle constitutes a major portion of this section. This section could thus be used to supplement the protein and fat of an existing food product, which is protein and fat deficient for people suffering from a protein and fat deficient diet. Similarly, a market could be created for the production of an economical food product with the neck flesh section being the main ingredient. Once this have been accomplished, fishing vessels may be persuaded to retain their Cape hake fish-waste for further processing due to the value of the prepared food products and thereby maintain profitability while abiding to governmental law. In conclusion non-government scientists should have more input in the decision-making process concerning matters affecting South Africa's marine biodiversity in order for future key policy and legislation drafts to be effective. Improvement of current fish preservation techniques and the known chemical composition of currently discarded material will result in informed decisions of future matters concerning its disposal. / AFRIKAANSE OPSOMMING: Kaapse stokvis (Merluccius capensis en M. paradoxus) is kommersieel Suid-Afrika se belangrikste vis spesie. Aangesien die Suid-Afrikaanse visprosesseringsbedryf baie intensief is, dra die Kaapse stokvis verwerkingsindustrie die grootste gedeelte by tot die totale visafval produksie. Die meeste van Suid-Afrika se visvangste word gedeeltelik skoongemaak voor landing terwyl nie-kommersiële byvangste en visafval gewoonlik oorboord gegooi word tydens die vangproses. Dit lei tot die vermorsing van 'n potensïele proteïen bron. Hierdie studie was dus gemik om: die huidige grootskaalse kommersiële Kaapse stokvis visvangsproses waar te neem; voorspellingsmodelle vir verskeie morfologiese parameters (heel vis massa, vis massa sonder kop en binnedele, stokvis kop massa, stokvis kop lengte, stokvis kop breedte en stokvis kop hoogte) vir Kaapse stokvis (Merluccius ssp.) te ontwikkel deur die hele lengte van die vis te gebruik as die onafhanklike veranderlike; die chemiese samestelling (vog, proteïen, vet, as, makro en spoor elemente) van verskillende dele van die viskop (skoonkop, nekweefsel, tong, tong kraakbeen, kaak, kiewe, hart, ingewand, derm, nier, nier & nierbeen en derm & gal); sowel as die effek van opberging op die vetsuurprofiel van beide die skoonkop en nekweefsel dele van die Kaapse stokvis kop. Hierdie resultate sal dan gebruik word vir die tegnies-ekonomies ondersoek in die gebruik van Kaapse stokvis koppe. Vir elk van die ses voorspellingsmodelle ontwikkel, was daar 'n vermeerdering in die variansie van die datapunte vir kategorieë 3 (64-80 cm) en 4 (>80 cm) teenoor kategorieë 1 (30-46 cm) en 2 (47-63 cm). Dit kan moontlik wees as gevolg van die kleiner monster trekking vir beide kategorieë 3 en 4 of as gevolg van verwagte toename in variansie wanneer groter biologiese monsters ondersoek word. Daar was ook 'n groepering van data in drie plekke vir elke voorspellingsmodel naamlik; binne in kategorieë 1 en oor kategorieë 2 en 3 en 3 en 4. Dit beklemtoon die geografiese breedte van die Kaapse stokvis populasie op grond van ouderdom, en dus die geografiese breedte op grond van grootte. Die voorspellingsmodelle ontwikkel vir beide die tweede en derde bootvangs het betekenisvol verskil (p<0.01) van die eerste bootvangs, behalwe die vir die stokvis kop lengte (cm) voorspellingsmodel. Die voorspellingsmodelle vir elk van die bootvangste het goeie voorspellingsvermoë getoon wat bewys is deur die lae Gemiddelde Kwadraat Fout waardes vir toetsgroepe en hoë Pearson's korrelasie koeffisiënt (r) waardes. Hierdie voorspellingsmodelle wat ontwikkel is, kan dus met vertroue in die Kaapse stokvis visvangsbedryf gebruik word mits dit ooreenstem met die periode waarin elke bootvangs uitgevoer was. Die nekweefsel gedeelte is die mees belangrikste met betrekking tot chemiese samestelling en die kaak die belangrikste in terme van minerale samestelling van die verskeie viskop dele. Die kaak is dus, as dit voldoende geprosesseer word, 'n goeie potensïele bron van Ca, Na en Fe en kan dus gebruik word om die dieet van mense wat 'n gebrek het aan hierdie minerale aan te vul. Met betrekking tot die chemiese samestelling van die nekweefsel gedeelte kan dit beskou word as 'n goeie potensiële bron van beide proteïen en vet, wat toegeskryf kan word aan die feit dat spierweefsel 'n groot deel uitmaak van hierdie viskop gedeelte. Hierdie viskop gedeelte sal dus uitstekend wees om die proteïen- en vetinhoud van 'n voedselproduk wat van nature 'n lae proteïen- en vetinhoud het te verhoog en hierdie produk sou dan geteiken word op daardie gedeelte van die gemeenskap wat 'n proteïen en vet tekort in hul dieet het. As dit eers alles in plek is, dan sal die visvangs bedryf hul Kaapse stokvis afval behou vir verdere prosessering deurdat dit gebruik word om die voedingsinhoud van bestaande voedsel soorte sal verbeter en terselfdertyd sal hulle aan wetgewing voldoen. Gevolglik sal nie-regerings navorsers meer betrokke moet wees by die besluitnemingsproses met betrekking tot sake wat die Suid-Afrikaanse mariene lewe affekteer en wat toekomstige wetgewing meer effektief sal maak. Die verbetering van huidige vis preserveringstegnieke gepaardgaande met die kennis van die chemiese samestelling van die Kaapse stokvis koppe sal lei na beter toekomstige besluite oor die afset daarvan.