Dormancy and germination of the Manketti nut, Ricinodendron rautanenii.
The distribution of Ricinodendron rautanenii trees is confined to a fairly distinct band across southern Africa. This stretches from South West Africa in the west, through Botswana and Zimbabwe, to Mozambique in the east. These plants are a potential source of both timber and food. In this respect, the fruits and seeds of this species are highly nutritious and the latter, by virtue of their high lipid content, represent an excellent source of oils for both domestic and industrial uses. For these reasons consideration is being given to cultivating this species on a commercial scale. One problem, however, is that the seeds of these plants are dormant and in the first part of this study an attempt was made to establish the cause of this dormancy and how it could be overcome. A thorough investigation of all the possible causes of dormancy revealed that ethylene was the only naturally occurring stimulus that could relieve this condition. Exogenously applied gibberellin (GA3 ) and ethrel were found to be equally as effective as ethylene, but such treatments cannot be regarded as natural. It was also found that all the dormancy breaking treatments were only effective once the endocarp had been removed. This indicated that R. rautanenii seeds had a combined coat imposed-physiological dormancy where ethylene was only able to stimulate germination once the endocarp had been removed. Scarification treatments showed that the endocarp most probably had its effect by restricting embryo enlargement since this structure did not inhibit water uptake or gaseous exchange. Once the ger- mination requirements of these seeds had been established a more detailed investigation was carried out to determine their general sensitivity to ethylene, as well as the actual role this gas played in breaking dormancy. Ricinodendron rautanenii seeds were found to exhibit a high degree of sensitivity to ethylene. The threshold concentration at which a response was obtained was approximately 10¯³ microlitres per litre and this is the lowest yet recorded for any species prior to any additional seed treatments. At concentrations above this, the response was saturated indicating that the seeds are well adapted to the ethylene concentrations most likely to occur in the field. In addition to this, manketti seeds also responded to ethylene after only very brief exposures to the gas and optimum germination was recorded after 30 minutes incubation in an ethylene saturated atmosphere. The temperature range over which an optimum response to this phytohormone was obtained was found to be between 25 and 35°C. One of the most striking features regarding the sensitivity of these seeds was the apparent ability of dry and partially imbibed seeds to perceive the dormancy breaking stimulus. Furthermore, once treated, the seeds retained the dormancy breaking effect of ethylene even when subjected to almost complete re-dehydration. In this instance, 50 per cent germination was recorded for ethylene treated seeds which had lost approximately 97 per cent of their moisture content between dormancy breaking and re-incubation. It was thus concluded that, not only could Ricinodendron rautanenii seeds respond to very low ethylene concentrations but could probably also retain the effects of this gas during adverse environmental conditions. The effects of imbibition and dormancy breaking were followed separately at the ultrastructural and biochemical level. The ultrastructure of dry embryonic axes of these seeds was characterized by massive stores of food reserves in the form of lipid and protein. Upon imbibition the number and size of spherosomes decreased and protein and globoid hydrolysis was clearly evident. Polysomes and microbodies (including mitochondria) were also visible prior to dormancy breaking but there was no evidence of any endoplasmic reticulum or dictyosomes. Imbibition also resulted in the expans ion of the nuclei and there were indications of an increase in the granular content of the associated nucleoli. The number of nucleolar vacuoles, however, remained unchanged. These features indicated that nuclear activity had commenced albeit limited. The ultrastructure ·of untreated seeds which were maintained in the imbibed state for an extended period of time (six days) was also examined. Cells of the embryonic axes of these seeds showed no further changes with regard to their nuclei and protein hydrolysis appeared to have ceased. At this time spherosomes resembled those in freshly imbibed tissue in terms of their size and numbers, suggesting that the lipid reserves had been resynthesized. No immediate ultrastructural changes were observed after ethylene treatments. However, 24 and 48 hours after dormancy breaking further expansion of the nuclei was noted. At the same time the nucleolar vacuoles disappeared and the granular content of this region increased markedly. This suggested that an increase in the synthesis of various RNA fractions was taking place. Vigorous protein hydrolysis was also observed after the ethylene treatment whereas spherosome numbers increased. Three days after the dormancy breaking treatment, the first signs of germination were visible. Externally this was characterized by a splitting of the testa in the region of the radicle. At this time, endoplasmic reticulum and dictyosomes were still not visible but from this point onwards the ultrastructural changes observed were typical of those recorded during the germination of other species. Thus, no single ultrastructural feature could be associated with the breaking of dormancy and the most notable changes which occurred during this period took place in the nucleus. Biochemical changes occurring during imbibition resulted in an overall decrease in the levels of extractable food reserves present in the embryonic axes. During this period, lipid levels were found to decrease by 44 per cent, protein levels by 12 per cent, sucrose levels by 68 per cent and glucose, fructose and starch levels by 100 per cent. These levels were found to return to their original values when seeds were incubated under moist conditions in the absence of ethylene for extended periods of time. Ethylene treatments, on the other hand caused a further, marked decrease in sucrose levels, whereas protein and lipid levels increased. Hydrolysis of the endosperm reserves commenced three days after the application of ethylene and this was characterized by a decrease in lipid levels and an overall increase in soluble carbohydrates. The timing of this event suggested that the endosperm was not involved in the actual process of dormancy breaking. The importance of protein synthesis in dormancy breaking was also investigated. It was found that seeds incubated with a protein synthesis inhibitor, cycloheximide, failed to germinate, confirming the view that protein synthesis is an essential pre-requisite for germination. Inhibition of RNA synthesis with actinomycin D, on the other hand, did not prevent germination. This suggested that the materials necessary for early protein synthesis were already present in the dry seeds. Actual measurements of protein synthesis showed that this process took place in the embryonic axis, cotyledons and endosperm of seeds imbibed for as little as two hours. Protein synthetic abilities increased considerably in most instances after 48 hours imbibition but then decreased upon application of ethylene. At the same time, however, a marked increase in the uptake of 14C-Ieucine was noted in ethylene treated axes. This may indirectly reflect an effect of ethylene on membrane permeability. Protein synthesis in cycloheximide and actinomycin D treated embryonic axes was also measured. No consistent trends were evident but it was found that after ethylene treatments, protein synthesis was generally lowest in those seeds which were destined to germinate. In addition, these seeds also exhibited the greatest uptake of C-Ieucine. Ricinodendron rautanenii seeds incubated with compounds known to stimulate the pentose phosphate pathway failed to germinate. This indicated that dormancy in this species was probably not the result of a block in alternate respiration. The possible involvement of endogenous phytohormones in the overall process of dormancy breaking was also in- vestigated. In this regard, the role of gibberellic acid appeared to be enigmatic. This is based on the observation that applied gibberellins could stimulate germination whereas inhibitors of endogenous gibberellin synthesis applied to ethylene treated seeds had no effect. It was concluded from this that the effects of ethylene are not mediated via an enhancement of endogenous gibberellin synthesis. A preliminary investigation carried out on the endogenous cytokinins showed that this hormone was absent from dry and imbibed seeds. A transient increase in zeatin levels was observed 24 hours after the ethylene treatment. A similar transient increase was noted in non-induced seeds maintained under moist conditions for six days. In this latter instance, however, the peak co-chromatographed with the biologically less active cytokinin, zeatin glucoside. A basal level of endogenous ethylene production was recorded in all imbibed Ricinodendron rautanenii seeds. Ethrel, ethylene and gibberellin treatments caused an initial, transient increase in this ethylene production after which no further significant changes were recorded. It is suggested that dormancy breaking in this species is not related to enhanced endogenous ethylene synthesis. The results of the biochemical and ultrastructural studies are discussed in relation to what is known regarding features associated with dormancy and its removal and on the known effects of ethylene on seed tissues.