Thesis (PhD(Agric))--University of Stellenbosch, 2000. / ENGLISH ABSTRACT: Little is understood regarding flowering in the genus Protea. The information available on inflorescence initiation and development in the family Proteaceae was reviewed and discussed. A number of experiments were conducted to investigate inflorescence initiation and development, and their manipulation for commercial production, in selected Protea cultivars, in the Western Cape, South Africa (33°S, Protea species can be allocated into groups according to similar times of flower initiation and of harvest. The stages occurring during flower initiation, and their synchrony relative to shoot growth were investigated for three cultivars, viz. Protea cv. Carnival (P. compacta x P. neriifolia), Protea cv. Lady Di (P. compacta x P. magnifica) and Protea cv. Sylvia (P. eximia x P. susannae), when flower initiation occurred on the spring growth flush. For all three cultivars the spring flush was preformed and enclosed in the apical bud before spring budbreak. During elongation of the spring flush the apical meristem produced floral primordia which differentiated into involucral bracts. After completion of the spring flush meristematic activity continued, to produce floral bracts with florets in their axils. The three cultivars showed differences and similarities in the time of budbreak, and the rates of shoot growth, appendage formation and flower development. The presence of mature leaves on an over-wintering shoot is essential for inflorescence initiation on the spring growth flush of 'Carnival'. Inflorescence initiation in 'Carnival' started at spring budbreak, and production of involucral bracts occurred concurrently with spring flush elongation. Shoots were defoliated at different degrees of severity at intervals from pre- to post- spring budbreak. Total defoliation applied earlier than 6-7 weeks before spring budbreak prevented flowering. Defoliation closer to spring budbreak affected characteristics of the spring flush and the inflorescence subtended by the spring flush. Effects were most marked following total defoliation and diminished with less severe treatments imposed by partial defoliation. Total defoliation applied before spring budbreak resulted in slower inflorescence development and lead to later anthesis. Defoliation treatments applied after completion of spring flush elongation had no effect on either vegetative or reproductive spring growth. The requirement for mature overwintering leaves to effect inflorescence initiation in 'Carnival' suggests that environmental factors, such as low temperature and daylength may play an inductive role. Shoots were in the induced state and committed to flowering 6-7 weeks before spring budbreak. A change in source size and position subsequent to different severalties of defoliation in 'Carnival' lead to reduced dry mass accumulation and altered partitioning. Mature leaves on the overwintering shoot supported growth of the spring flush and the early stages of inflorescence development. When these leaves were removed by total defoliation dry mass accumulation in the spring flush was reduced. A hierarchy of priorities between competing sinks was revealed by defoliation during growth of the spring flush and concomitant inflorescence development: formation of involucral bracts> leaf growth> stem elongation. Dry mass accumulation of the inflorescence subtended by the spring flush was supported by the spring flush leaves and was only indirectly affected by defoliation. Treatments which resulted in the production of a weaker spring flush lead to a reduction in dry mass accumulation of the inflorescence. Different severalties of partial defoliation, whereby either upper or lower leaves were removed from a shoot, indicated that the position of leaves relative to the active sink is more important, with respect to source availability, than the number of leaves on the shoot. Mature overwintering leaves are essential in 'Lady Di' for shoots to achieve the induced state for flowering; and are also crucial to the early stages of inflorescence initiation. Defoliation applied before formation of involucral bracts was complete prevented flowering. Defoliated shoots either remained vegetative or produced inflorescences which aborted. Reserve carbohydrates in the stem and leaves of overwintering shoots were low, and early growth and development of both the spring flush and inflorescence were, therefore, supported by current photosynthates from the overwintering leaves. Likewise, reserve carbohydrates available in the flowering shoot were insufficient to account for the dry mass increase during the major portion of growth of the spring flush and inflorescence. This rapid increase in dry mass occurred after elongation of the spring flush was complete and was supported by current photosynthates from the leaves of the spring flush. Defoliation treatments that did not prevent inflorescence initiation, had no effect on inflorescence development, and flowering time of 'Lady Di' was not delayed by defoliation. 'Sylvia' has an open window for inflorescence initiation and can initiate flowers throughout the year. Despite the 'open window' inflorescences are initiated more readily on the spring flush, when it is subtended by one or more overwintering shoots. This may be the expression of a facultative response to inductive conditions for which 'Carnival' and 'Lady Di' have an obligate requirement. The date of pruning affected flowering time of 'Sylvia' by influencing on which flush inflorescence initiation occurred, and the harvest could be manipulated to fall within the optimum marketing period for export to Europe. Flowers initiated on the spring flush reach anthesis in January and February; on the first summer flush predominantly in April and May; on the second summer flush in July and August; and on the autumn flush in November and December. Thus, shoots harvested within the optimum marketing period (September to February) initiated inflorescences on the autumn and spring flushes. Due to the readiness of shoots to initiate inflorescences on the spring flush many shoots harvested in January and February (following initiation in the previous spring) were short and were rendered unmarketable. For commercial production pruning in July is recommended. Long flowering stems will be harvested in October to November of the following year. Since the vegetative and reproductive cycles necessary to produce inflorescences on long stems span more than a year, a biennial cropping system is recommended. / AFRIKAANSE OPSOMMING: Bloeiwyse-inisiasie en -ontwikkeling, en die manipulasie daarvan, van geselekteerde cultivars van die genus Protea. Min word verstaan van blomvorming in die genus Protea. Die beskikbare inligting oor die bloeiwyse-inisiasie en -ontwikkeling in die familie Proteaceae is nagegaan en bespreek. 'n Aantal eksperimente is uitgevoer waarin geselekteerde Protea cultivars van die Wes-Kaap, Suid-Afrika (33°S, 19°0) se bloeiwyse-inisiasie en -ontwikkeling, asook die manipulasie daarvan vir kommersiële produksie ondersoek is. Protea spesies kan in groepe ingedeel word op grond van blominisiasietye en oestye wat ooreenstem. Die verskillende stadiums van blominisiasie en hulle sinchronisering relatieftot stingelgroei is ondersoek vir drie kultivars, naamlik Protea cv. Carnival (P. compacta x P. neriifolia), Protea cv. Lady Di (P. compacta x P. magnifica) en Protea cv. Sylvia (P. eximia x P. susannae) tydens blominisiasie op die lentegroeistuwing. By al drie die kultivars was die lentegroeistuwing reeds gevorm en omsluit in die apikale knop voor die lente-knopbreking. Gedurende die verlenging van die lentegroeistuwing het die apikale meristeem blomprimordia, wat in bloeiwyseomwindselskutblare gedifferensieer het, geproduseer. Na voltooiing van die lentegroeistuwing, het meristematiese aktiwiteit voortgeduur en blomskutblare met blommetjies in hulle oksels is gevorm. Die drie kultivars het verskille en ooreenkomste vertoon tydens die periode van knopbreking, asook in die tempo van stingelgroei, aanhangselformasie en blomontwikkeling. Die teenwoordigheid van volwasse blare op 'n oorwinteringstingel is noodsaaklik vit bloeiwyse-inisiasie op die lentegroeistuwing van 'Carnival'. Bloeiwyse-inisiasie in 'Carnival' het met lente-knopbreking begin en die produksie van bloeiwyseomwindselblare het gelyktydig met lentegroeistuwing verlenging plaasgevind. Stingels is met tussenposes, van voor tot na die lente-knopbreking, en met verskillende grade van felheid, ontblaar. Algehele ontblaring vroeër as 6-7 weke voor die lente-knopbreking het blomvorming verhoed. Ontblaring nader aan die lenteknopbreking het 'n invloed gehad op die eienskappe van die lentegroeistuwing asook die bloeiwyse gedra deur die lentegroeistuwing. Die effek was die duidelikste sigbaar by algehele ontblaring en het verminder namate die behandeling minder fel geword het by gedeeltelike ontblaring. Algehele ontblaring wat voor die lente-knopbreking gedoen is, het gelei tot stadiger bloeiwyse-ontwikkeling en later antese. Ontblaringsbehandelings wat na die voltooiing van die lentegroeistuwing verlenging toegepas is, het geen effek op die vegetatiewe of die reproduktiewe lentegroei gehad me. Die nodigheid van volwasse oorwinteringsblare vir bloeiwyse-inisiasie in 'Carnival' dui daarop dat omgewingsfaktore soos lae temperature en daglengte 'n induktiewe rol kan speel. Stingels was in die geïnduseerde toestand en verbind tot blomvorming 6-7 weke voor die lente-knopbreking. 'n Verandering in oorspronggrootte en -posisie as gevolg van verskille in die felheid van ontblaring by 'Carnival', het gelei tot verminderde droë-massa-akkumulasie en veranderde verdeling. Volwasse blare op die oorwinteringstingel het die groei van die lentegroeistuwing en die vroeë stadiums van bloeiwyse-ontwikkeling ondersteun. Toe hierdie blare verwyder is in 'n algehele ontblaring, het die droë-massa-akkumulasie in die lentegroeistuwing verminder. 'n Hiërargie van prioriteite tussen kompeterende sinke is blootgelê tydens ontblaring gedurende die lentegroeistuwing en saamlopende bloeiwyse-ontwikkeling: vorming van bloeiwyse-omwindselblare > blaargroei > stamverlenging. Droë-massa-akkumulasie van die bloeiwyse onderspan deur die lentegroeistuwing is ondersteun deur die blare van die lentegroeistuwing en is slegs op 'n indirekte wyse deur ontblaring geaffekteer. Behandelings wat tot die produksie van 'n swakker lentegroeistuwing gelei het, het tot 'n vermindering in die droë-massaakkumulasie van die bloeiwyse gelei. Verskille in die felheid van gedeeltelike ontblaring, waartydens óf die boonste óf die onderste blare van 'n stingel verwyder is, het aangetoon dat die posisie van die blare relatief tot die aktiewe sink belangriker is, met betrekking tot die beskikbaarheid van die oorsprong, as die aantal blare op die stingel. By 'Lady Di' is volwasse oorwinteringsblare noodsaaklik VIr stingels om die geïnduseerde stadium van blomvorming te bereik en hulle is ook van die uiterste belang in die vroeë stadiums van bloeiwyse-inisiasie. Waar ontblaring gedoen is voordat die vorming van bloeiwyse-omwindsel voltooi was, het blomvorming nie plaasgevind nie. Ontblaarde stingels het ófvegetatief gebly ófbloeiwyses geproduseer wat geaborteer het. Reserwe-koolhidrate in die stam en blare van die oorwinteringstingels was laag en die vroeë groei en ontwikkeling van beide die lentegroeistuwing en die bloeiwyse is dus deur die bestaande fotosintate van die oorwinteringsblare onderhou. Net so was die reserwe-koolhidrate beskikbaar in die blomdraende stingels nie voldoende om die toename in droë massa gedurende die grootste deel van die groei van die lentegroeistuwing en die bloeiwyse te verklaar nie. Hierdie vinnige toename in droë massa het plaasgevind nadat die verlenging van die lentegroeistuwing voltooi was en is deur die bestaande fotosintate van die blare van die lentegroeistuwing onderhou. Ontblaringsbehandelings wat nie bloeiwyse-inisiasie verhoed het nie, het geen effek op bloeiwyse-ontwikkeling gehad nie en die blomtyd van 'Lady Di' is nie deur ontblaring vertraag nie. 'Sylvia' beskik oor 'n oop venster vir bloeiwyse-inisiasie en kan regdeur die jaar blomme inisieer. Ten spyte van die 'oop venster', word bloeiwyses tog meer geredelik in die lentegroeistuwing geïnisieer, wanneer dit deur een of meer van die oorwinteringstingels gedra word. Dit mag die uitdrukking wees van 'n fakultatiewe respons op induktiewe toestande wat vir 'Carnival' en 'Lady Di' 'n verpligte vereiste is. 'Sylvia' se blomtyd is deur die snoeidatum geaffekteer omdat die snoeidatum 'n invloed gehad het op die keuse van by watter groeistuwing bloeiwyse-inisiasie plaasgevind het. Die oestyd kon gemanipuleer word om binne die optimum bemarkingstydperk vir uitvoer na Europa te val. Blomme wat op die lentegroeistuwing geïnisieer is, bereik antese in Januarie en Februarie; dié wat op die eerste somergroeistuwing geïnisieer is, bereik antese hoofsaaklik in April en Mei; dié wat op die tweede somergroeistuwing geïnisieer is, bereik antese in Julie en Augustus en dié wat op die herfsgroeistuwing geïnisieer is, bereik antese in November en Desember. Stingels wat in die optimum bemarkingsperiode (September tot Februarie) geoes is, het dus bloeiwyses op die herfs- en lente-groeistuwings geïnisieer. As gevolg van die gereedheid van stingels om bloeiwyses op die lentegroeistuwings te inisieer, was baie van die stingels wat in Januarie en Februarie geoes is, kort en kon nie bemark word nie. Vir kommersiële doeleindes word snoei in Julie aanbeveel. Lang blomdraende stingels sal in Oktober en November van die volgende jaar geoes word. Aangesien die vegetatiewe en reproduktiewe siklusse wat nodig is om bloeiwyses met lang stingels te produseer oor meer as fn jaar strek, word fn tweejaarlikse oesinsamelingstelsel aanbeveel.

Dissertation (PhD(Agric))--University of Stellenbosch, 2006. / ENGLISH ABSTRACT: Advancement of the flowering time of Protea cv. Carnival by approximately three months, without compromising the product quality, was achieved by the application of 6- benzyladenine-containing plant growth regulators to three-flush shoots in autumn. This earlier flowering time coincides favourably with the prime European marketing period (November-January). The percentage three-flush shoots initiating an inflorescence following the brush application of the 6-benzyladenine (BA)-containing regulators, ABG- 3062 (active ingredient: BA 2% w/w) and Accel® (active ingredients: BA 1.8% w/w; gibberellins A4A7 0.18% w/w) on dormant terminal buds, increased with later application dates and flowering percentages as high as 90% was achieved. No inflorescences were initiated on flushes induced by Promalin® (active ingredients: BA 1.8% w/w; gibberellins A4A7 1.8% w/w). Phenological phase progression of green point, flush expansion and inflorescence development of 'Carnival' shoots as induced by BA was calculated to have base temperatures of 8°C, 6°C and 1°C respectively. The days required from application of the BA-containing growth regulator until green point stage increased progressively over the six consecutive treatment dates in autumn (14 March - 22 May 2003). In contrast, the days required to complete inflorescence development decreased with each successive treatment date. The days required between the respective stages were mostly negatively correlated with temperature, except for the phase 'green point to flush expansion', where the relationship was unclear. For three-flush shoots of eight-year old plants, between 13-57, 39-65 and 121-177 days were required to reach green point, to achieve full flush expansion following green point and to complete inflorescence after flush expansion respectively. BA application enhanced budbreak in most dormant shoots, irrespective of plant age, BA concentration, decreasing temperature over time or shoot characteristics. However, twoflush shoots treated in late May had low budbreak and hence low flowering percentages. Shoots varied considerably in their responsiveness to BA treatments. BA application (500mg·L-1) as MaxCelTM (active ingredients: BA 1.9% w/w) to terminal buds alone of mature three-flush shoots from less vigorous growing plants resulted in the highest flowering percentages. Applications were most effective when applied to the terminal bud in the dormant state or up to the ‘green point’ stage. Shoot characteristics such as flush length, leaf area, shoot dry mass, number and proximity of the leaves to the terminal bud were all positively correlated with the propensity of shoots to initiate inflorescence under BA induction. Terminal flush intercalation shoot diameter (>7mm) was identified as the most important variable influencing the likeliness of flowering and can effectively serve as a nondestructive estimation of a shoot's propensity to flower. The presence of developing inflorescences or possible floral inhibiting factors derived from the previous flowering season is suggested to be inhibitory to inflorescence initiation following BA application. Synchronisation of shoot growth by pruning plants in late winter appears to be an essential step to ensure high percentages inflorescence initiation with BA treatment the following autumn. The use of BA as a management tool to control flowering times in Protea for better market opportunities is shown to hold considerable commercial potential. / AFRIKAANSE OPSOMMING: Protea cv. Carnival se blomtyd is met ongeveer drie maande vervroeg sonder om produkkwaliteit prys te gee. Hierdie vervroegde blomtyd wat gunstig saam val met die optimale Europese bemarkingstyd van November-Januarie is bewerkstelling deur die herfstoediening van 6-bensieladenien-bevattende plantgroei-reguleerders op lote bestaande uit drie groeistuwings. Die persentasie lote met drie groeistuwings wat 'n bloeiwyse geïniseer het na 'n kwas-aanwending met die 6-bensieladenien (BA)-bevattende groeireguleerders, ABG-3062 (aktiewe bestandeel: BA 2% w/w) en Accel® (aktiewe bestandele: BA 1.8% w/w; gibberellins A4A7 0.18% w/w), het toegeneem met latere behandelingsdatums en blompersentasies so hoog as 90% is behaal. Geen bloeiwyses is geïnisieer op groeistuwings wat deur Promalin® (aktiewe bestandeel: BA 1.8% w/w; gibberellins A4A7 1.8% w/w) teweeggebring is nie. Basis temperature van 8°C, 6°C en 1°C respektiewelik is bereken vir fenologiese fasevordering vanaf groeireguleerder toediening tot by groenpunt, groeistuwing-voltooing en bloeiwyse-ontwikkeling van 'Carnival' lote soos geïnduseer deur BA. Die dae wat benodig was vanaf toediening van die BA-toediening totdat groenpunt stadium bereik is, het progressief toegeneem oor die ses opeenvolgende herfsbehandelingsdatums (14 Maart-22 Mei 2003). In teenstelling met bostaande, het die vereiste aantal dae om bloeiwyseontwikkeling te voltooi afgeneem met elke opeenvolgende behandelingsdatum. Die aantal dae wat benodig was vir die onderskeie fases was meestal negatief gekorreleer met temperatuur, behalwe vir die fase 'groenpunt tot groeistuwing-voltooing', waar die verhouding onduidelik was. Vir lote van agt-jaar-oue plante met drie groeistuwings was tussen 13-57, 39-65 en 121-177 dae respektiewelik benodig om groenpunt te bereik, volledige groeistuwingverlenging te bewerkstellig en om bloeiwyse-ontwikkeling wat volg na groeistuwing verlenging, te voltooi. BA-toediening het knoprusbreking bevorder in die meeste dormante lote, ongeag plant ouderdom, BA konsentrasie, afname in temperatuur met tyd of loot eienskappe. Lote met twee groeistuwings wat laat in die herfs behandel is, het egter lae rusbreking en dus gevolglik ook lae blompersentasies getoon. Lote varieer aansienlik in hul reaksie op BA behandeling. BA toediening (500mg·L-1) as MaxCelTM (active ingredients: BA 1.9% w/w) op die terminale knop van afgeharde lote met drie groeistuwings en afkomstig van minder groeikragtige plante het tot die hoogste blompersentasies gelei. Die effektiwiteit van die behandeling was die hoogste met toedienings aan dormante terminale knoppe tot en met groenpuntstadium. Loot eienskappe soos groeistuwinglengte, blaaroppervlakte, loot droë massa, asook die aantal en nabyheid van die blare relatief tot die terminal knop was almal positief gekorreleerd met die vermoë van die loot om 'n blom te inisisieer in reaksie op BA induksie. Terminale groeiverstuwing interkalasie-lootdikte (>7mm) is geïdentifiseer as die belangrikste veranderlike wat die vermoë om te kan blom kan beïnvloed en kan gebruik word as 'n nie-destruktiewe voorspeller vir blom-inisiasie. Die teenwoordigheid van ontwikkelende bloeiwyses of potensiële blom-inhiberende faktore aanwesig in die loot na die vorige blomperiode, word moontlik beskou om inhiberend te wees vir BA-geïnduseerde blom-inisiasie. Sinchronisering van lootgroei deur die snoei van plante in laat-winter blyk krities te wees om 'n hoë blompersentasie met BA behandeling te verseker in die daaropvolgende herfs. Die aanwending van BA as 'n bestuurstegniek om die blomtyd van Protea te posisioneer vir beter bemarkingsgeleenthede toon aansienlike kommersiële potensiaal.