Thesis (MScAgric)--University of Stellenbosch, 1994 / 127 Leaves printed single pages, preliminary pages i-viii and numberd pages 1-118.Includes bibliography,tables and figures. / Date on t.p.: Dec. 1994. / ENGLISH ABSTRACT: Effects of pH,sucrose, boric acid and temperature on in vitro germination of pollen of Protea repens (L.) L. cv. 'Embers' were investigated in hanging-drop culture to establish optimum conditions for germination. Optimum values were found within ranges pH: 5 - 8, sucroseconcentration:0.4 - 0.7 M, boric acid concentration:50 - 500 mg.e-1 , and incubation temperature: 5 - 30°C. Storage temperature and humidity on pollen viability was studied in four Pro tea clones. Pollen was stored at a range of temperatures and relative humidities for up to one year and tested for ability to germinate in vitro. Pollen of P. repens cv. 'Sneyd', P. eximia cv. 'Fiery Duchess' andP. magnifica clone 'T 84 07 OS', stored in liquid nitrogen (-196°C) and in a freezer (-14° to -18°C), retained a germination percentage as high as that of fresh pollen regardless of humidity. The study showed that long-term storage of protea pollen is not feasible at temperatures above O°C. The correlations between the fluorochromatic reaction (FCR) and germinability were found to be low and nonsignificant. Fifteen month old cryopreserved 'Sneyd' pollen was shown to retain its ability to fertilize and set seed equal to that of fresh pollen. 'Sneyd', 'Fiery Duchess' and 'T 84 07 OS' pollen could be repeatedly thawed and frozen in liquid nitrogen before its germinability in vitro decreased. The morphology and size of Protea pollen was studied, using light and scanning electron microscopy. Polymorphic grains were observed in two interspecific hybrids. Very small differences in pollen grain size were recorded between clones/species. The male fertility of 25 interspecific Pro tea hybrids, based on in vitro pollen germinability, was investigated. The majority of hybrids were found to be sufficiently fertile to be used in a breeding programme. Pistil structure and pollen tube pathways were investigated in 'Sneyd' using light and scanning electron microscopy. The pistil had four distinct regions, consisting of the stigma, the vertebra-shaped upper style, the heart-shaped lower style, and the ovary. The pistil had a stylar canal along its entire length and this canal was also the route by which pollen tubes grew to the ovary. Very low numbers of pollen tubes reached the ovary. The breeding system of 'Sneyd' and 'Fiery Duchess' were determined from pollen tube and seed set data, after controlled hand-pollinations. Both clones were found to be fully selfcompatible. Very low percentages autogamous seed set were recorded. Interspecific crosses had a low success rate. An incompatibility reaction probably occurred on the stigma and/or in the upper style regions.The attainment of maximum stigma receptivity of two Protea cultivars was investigated by means of seed set experiments, pollen tube growth observations and measurement of the degree of opening and closing of the stigmatic groove. Both cultivars were found to be protandrous. The maximum stigmatic groove width of both cultivars never exceeded the pollen grain diameter. It was concluded that Protea spp. must be hand-pollinated two to six days after anthesis in order to obtain maximum seed set; while for the observations of pollen tubes in the ovary, inflorescences must not be harvested before seven days after pollination. / AFRIKAANSE OPSOMMING: Ten einde 'n optimale medium vir die in vitro-ontkieming van Protea-stuifmeel te ontwikkel, is die effek van pH, sukrose, boorsuur, en temperatuur op die in vitro-ontkieming van Protea repens (L.) L. cv. 'Embers'-stuifmeel deur middel van die hangdruppel-metode ondersoek. Die volgende reekse van veranderlikes wat getoets is, is as optimaal gevind; pH: 5 - 8, sukrosekonsentrasie: 0.4 - 0.7 M, boorsuurkonsentrasie: 50 - 500 mg.e-1 en inkubasietemperatuur: 5 - 30°C. Die invloed van bergingstemperatuur en humiditeit op stuifmeel-Iewenskragtigheid is in vier Protea-klone ondersoek. Stuifmeel is gestoor by 'n reeks temperature en relatiewe humiditeite vir tot eenjaar, en vir in vitro-ontkiemingsvermoe getoets.' Stuifmeel van P. repens <?v. 'Sneyd', P. eximia cv. 'Fiery Duchess', en P. magnifica kloon'T 84 07 OS', in vloeibare stikstof (-196°C) en in 'n vrieskas (-14° tot - 18°C) geberg, het 'n ontkiemingspersentasie gelykstaande aan die van vars stuifmeel gehandhaaf, ongeag van die humiditeit. Hierdie studie het verder aangetoon dat langtermynberging van Protea-stuifmeel bokant O°C me die moeite werd is me. Die korrelasie tussen die fluorochromatiese reaksie (FCR) en ontkieming was laag en me betekemsvol me. 'Sneyd' -stuifmeel wat vir 15 maande in vloeibare stikstof gestoor is, het die bevrugtings- en saadsetvermoe gelykstaande aan vars stuifmeel behou. 'Sneyd', 'Fiery Duchess' en 'T 84 07 OS'-stuifmeel kon herhaaldelik in vloeibare stikstof gevries en ontdooi word voordat hul ontkiemingsvermoe afgeneem het. Die morfologie en grootte van Proteastuifmeel is deur middel van lig- en skandeerelektronmikroskopie bestudeer. Polimorfiese stuifmeelkorrels is in twee interspesie-hibriede waargeneem. Baie klein verskille in stuifmeelkorrelgroottes het tussen klone/spesies voorgekom. Die manlike vrugbaarheid van 25 Protea-interspesiehibriede, gebaseer op die in vitro-ontkiembaarheid, is ondersoek. Dit is gevind dat die meerderheid hibriede 'n voldoende graad van vrugbaarheid het om in 'n teelprogram te gebruik. Die stamperstruktuur en stuifmeelbuiswee in P. repens is deur middel van lig- en skandeer-elektronmikroskopie ondersoek. Die stamper bestaan uit vier kenmerkende gebiede, naamlik die stempel, die werwelvormige bo-styl, die hartvormige onderstyl, en die vrugbeginsel. Die stamper het 'n stylkanaal regdeur die totale lengte van die stamper, en hierdie kanaal is ook die weg waarvolgens stuifmeelbuise na die vrugbeginsel gegroei het. Min stuifmeelbuise het die vrugbeginsel bereik. Die teelsisteem van 'Sneyd' en 'Fiery Duchess' is deur middel van stuifmeelbuis- en saadsetdata na gekontroleerde handbestuiwings ondersoek. Beide kIone was ten volle selfverenigbaar. Die persentasie outogame saadset was baie laag. Interspesiekruisings het 'n baie lae sukses gehad. Dit is voorgestel dat die onverenigbaarheidsreaksie in die stempel en/of in die bopunt van die styl plaasvind. Die bereiking van maksimum stempelontvanklikheid van twee Protea-cultivars is deur middel van saadseteksperimente, stuifmeelbuisdata en waarnemings van die oop- en toemaak van die stempelgroef ondersoek. Beide cultivars was protandries. Die maksimum stempelgroefwydte het nooit die stuifmeelkorreldeursnee oorskry nie. Dit is afgelei dat Protea-spesies twee tot ses dae na antese handbestuif moet word vir optimale saadset. Vir die waarneming van stuifmeelbuise in die vrugbeginsel, moet bloeiwyses nie voor sewe dae na bestuiwing geoes word nie.