Dormancy characteristics and optimum conditions for germination of Cleome gynandra seeds have not been explained. Seed storage proteins were extracted, analysed with SDS-PAGE and sequenced. Seed proteins of Cleome were characterised by comparison with those of wild mustard (Brassica kaber). Wild mustard showed seed proteins composed of two α-chains of molecular weight (24-32 kDa) and another two β-chains of 18-22 kDa. The seed proteins of Cleome comprised two α-chain polypeptides of molecular weight (25-30 kDa), two β-chain polypeptides of molecular weight (18-20 kDa) and a smaller β-chain of 13-15 kDa. The storage proteins occurred in the seeds as dimeric complexes of molecular weight 40-65 kDa, which were broken into polypeptide chains of approximately 20 and 30 kDa by the reducing . action of DTT. Comparison with proteins in the proteome library and similarity index further confirmed that the seed proteins of Cleome had similarities with those of wild mustard. Two dimensional SDS-PAGE showed that the two species have nine similar polypeptides and four different ones. Events associated with dormancy release during seed germination still require explanation. Seeds of Cleome are characterised by low germination and there has been no explanation for this. Changes in protein expression during germination of Cleome in the presence or absence of light and at constant or alternating temperatures were examined. The germination of Cleome seeds at 20 degrees C was inhibited by light, but it was improved at 20 degrees C in darkness. There was no photoinhibition when seeds were germinated at constant 30 degrees C or alternating 20/30 degrees C (16 h night and 8 h day) for 10 days. Four proteins were observed to decrease in expression as germination progressed, but remained unchanged during photoinhibition. Photoinhibition was expressed more in seeds that were harvested late, after the pods had turned brown. These seeds showed a fifth, low molecular weight protein (13 kDa) that was absent from the immature seeds and embryos. Photinhibition is a pseudo-dormancy condition during which seed storage proteins are not utilised and the seed coat could partially play a role in it. The temperatures for the germination of Cleome in darkness have been determined. However, prior to this study the effects of temperature, light and pre-germination treatments (chilling, scarification, hydration and germination in the presence of KN0(3) or GA(3) on the germination of the seeds of this species have not been investigated. Seeds were germinated for 10 days and the final count of germination was used to determine seed performance. The highest germination percentage (60% and 80%, for a 2-year old and a l-year old seed lot, respectively) of untreated seeds was achieved when alternating temperatures of 20/30 degrees C (16 h/S h) in the dark or constant 30 degrees C in the dark were used. Among the pre-germination treatments, only scarification (puncturing of seeds at the radicle end) improved germination. Seeds were found to be negatively photoblastic, and the phenomenon was more pronounced when they were germinated at 20 degrees C and 12 h photoperiod or longer. Germination of photoinhibited seeds was, however, improved by treatment with GA(3) It is recommended that the germination of Cleome be undertaken under conditions of darkness and at either alternating 20/30 degrees C or continuous 30 degrees C. Seed lot vigour and seedling vigour are two important seed quality aspects that are used in defining the seed germination process. Seed germination is appropriately characterised by radicle protrusion and the attainment of normal seedling structures. However, the international rules for testing seeds combine radicle protrusion and normal seedling attainment in separating seed germination into the first and final counts. The challenge to a seed analyst testing the germination of a species whose first and final counts are unknown is that there is no statistical guideline to determine these important stages of seed germination. Cauliflower and broccoli, for which the first and final counts are published in the international rules for testing seeds and Cleome, for which there is no data on the first and final counts, were examined to determine the statistical significances of the first and final counts. Analysis of variance, logistic regression, 'broken-stick' regression models and survival analysis procedures were used. Analysis of variance showed that there were no differences between the germination percentages on the fourth, fifth and seventh days of germination. Low and stable standard deviations were recorded when evaluating germination after the fourth day. The germination curves of broccoli and cauliflower did not fit the Gompertz curve but fitted the exponential curve. The broken-stick model 'broke' the cumulative germination curve for the Cleome seed lots into two linear curves that were significantly different, but failed to break those for broccoli and cauliflower. However, this study confirmed the first and final counts for broccoli and cauliflower as determined by the international rules for testing seeds. Broken-stick modelling and life table analyses confirmed the fourth day as being appropriate to determine the first count for Cleome germination. There was no evidence of further seed germination after the seventh day as shown by probability density and hazard rate. It is suggested that for Cleome, the 'first count' and 'final count' be performed on the fourth and seventh day of the germination, respectively.

Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2005.