Avocado sunblotch disease is a graft-transmissible disorder known for over 60 years and has now been recorded in at least eight countries around the world. Affected trees develop yellow, depressed streaks on young stems and fruit, marked rectangular cracking of the mature bark and a decumbent style of growth. Often a tree with symptoms produces completely symptomless shoots, termed 'recovery' growth, which are latently infected. There is a reported 95 to 100% transmission of sunblotch through the seed of such branches, and "the resultant seedlings are themselves symptomless. Indexing for sunblotch to ensure that scion and, in view of seed transmission, especially rootstock material is free of the disease is very important . The standard method used for many years has been to graft tissue onto healthy indicator seedlings and observe for symptom development for 18 months to two years. One aim of the study presented in this thesis was to develop more rapid methods for detecting the sunblotch agent. By conducting the standard indexing method in a glasshouse at controlled high temperatures of 30/28º C (day/ night) and by cutting back the indicator plants every three months, the time was reduced from two years to eight months. While this represents a considerable saving in time, the ideal must be to develop a laboratory diagnostic test that requires no more than a few days, at most, to complete. A comparative study was therefore initiated on the phenol metabolism of healthy and sunblotch-infected avocados to determine whether infection causes any major change that may reliably serve as a marker for diagnostic purposes. Significantly increased peroxidase (PO) and phenylalanine ammonia-lyase (PAL) activities, decreased indoleacetic acid (IAA) oxidase activity and higher sinapic acid levels were detected in bark tissue showing sunblotch symptoms, but not in symptomless 'recovery' growth. In contrast, increased polyphenoloxidase (PPO) activity and isoenzymes, total soluble protein levels, water soluble phenols and reduced ferulic acid levels were found in the bark of all infected trees tested, both with symptoms and symptomless. However, these latter changes have been associated with other plant-virus systems and are therefore not necessarily specific for sunblotch. Neither is any sufficiently large to be definitive as a diagnostic test. Two unidentified phenols were detected in infected, mature bark, but not in infected young bark and leaves. introduced the possibility of rapid disease detection by polyacrylamide gel electrophoresis (PAGE) of extracted RNA's as used for known viroids. In this study the presence of previously reported small molecular weight sunblotchassociated RNA's was confirmed using PAGE methods requiring two to four days to complete. This thesis presents as a further development a more rapid method of PAGE detection of RNA's enabling indexing for sunblotch to be completed in under six hours. Whilst the biochemical studies did not reveal diagnostically meaningful differences between healthy and infected avocados, there were tendencies towards differences between healthy and symptomless carrier tissues, further investigation of which may lead to a future understanding of symptom development and the symptomless condition. These include apparent higher PO and lower PAL activities in symptomless carrier tissue, as well as higher PO isoenzyme a[1] and lower IAA oxidase isoenzyme a[1] activities. General studies on sunblotch-infected avocados showed that fruit from symptomless 'recovery' growth branches are significantly larger and have a higher oil content than those from healthy or diseased branches, the latter finding possibly indicating a more advanced state of maturity of 'recovery' growth fruit due to earlier flowering. The avocado sunblotch agent was shown to have an in vivo thermal inactivation point of 55º C, a temperature higher than the avocado tissue can withstand thereby eliminating the possibility of thermotherapy of infected twigs. In a host range study four lauraceous plant species, Persea Schiedeana, Cinnamomum zeylanicum, C. camphora and Ocotea bullata, were successfully infected with sunblotch by grafting from infected avocado. This is the first demonstration of any host other than avocado. A phanerogametic member of the same family, Cassytha filiformis, was shown to be able to transmit the disease from avocado to avocado. No hosts from other families were found. During an electron microscope study of sunblotch-infected avocado leaf tissue, gross alterations of the chloroplasts in the yellow areas were observed. These changes included organelle swelling, loss of grana and stroma lamellae, rearrangement of remaining membranes and presence of vesicles. Also in the yellow areas paramural bodies were encountered in higher numbers and displaying altered structure than in healthy and symptomless infected leaf tissue. This study on avocado sunblotch disease was successful in both of its aims. Firstly with regard to quicker indexing techniques, the standard method using indicator plants was shortened from two years to eight months, while a rapid, six-hour test based on PAGE analysis, was developed. Secondly, more light has been shed on the biochemical and ultrastructural effects of sunblotch on its host, the avocado, as well as providing information regarding the thermal sensitivity and the host range of the agent. / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 1980.

The ecology of mangrove forests on Potential Acid Sulphate soils at Bintuni Bay (132° 55' - 134° 02' E, and 2° 02' - 2° 97' S), Irian Jaya, Indonesia was studied. The annual rainfall is 3000mm and there is mixed semi-diurnal tide of 1-5.6m amplitude. The water has a varying salinity of 0-27%. Forest structure was studied in plots of 10m x 10m along three transects across Sikoroti Island and in three 50m x 50m plots in mixed Rhizophora - Bruguiera forest. All trees > 10cm trunk diameter were enumerated, measured and identified. Of the nine tree species, Rhizophora apiculata was the most dominant, followed by Bruguiera gymnorrhiza and Ceriops decandra. There was good regeneration seven years after clear-felling in plots on Amutu Besar and Amutu Kecil Islands. Small litterfall production, measured using ten 1m x 1m litter traps in each of the three 50m x 50m plots, was 11.09 t ha-1 y-1, and greatest during the wet season (December, 1.29 t ha-1 ) and least during the dry season (July, 0.61 t ha-1). The annual litterfall mineral element accession was (kg ha-1 y-1): N 240.4, P 6.1, K 43.2, Na 136.2, Ca 204.4 and Mg 48.7. Leaf decomposition of five species was studied in litterbags on the forest floor under trees of the same species. Sonneratia alba decomposed quickly with a half-life of 24 days, and Bruguiera parviflora was the slowest with a halflife of 124 days. Decomposition rates of all species followed a single exponential decay model. Leaf herbivory of young stands of Rhizophora apiculata and Bruguiera gymnorrhiza was significantly different among sites, species, plant height and leaf-age. Seed predation on six species was studied in three different sites. Twenty propagules of each species were secured by string in each of six 10m x 10m sub-plots within each of five 10m x 60m plots. On average 62.1% of the seeds were dead after 36 d. Predation was higher in the lower intertidal zone, and Avicennia alba and Bruguiera parviflora were the most predated. The results suggested that the Bintuni Bay mangroves can be sustainably managed but the best silvicultural system needs to be determined.

Septoria blight of celery (Apium graveolens) caused by Septoria apiicola is a common disease. This work dealt with the development of a simple forecast model to time fungicide application to manage the blight. Two studies were conducted to examine the effect of relative humidity (65, 88, 92, 96%, and wet) temperature (15, 20, 25, and 28 C) and exposure period (6, 12, 24, 48, 72, and 96 hr) on spore production. Spores were produced at all levels of relative humidity and temperature tested. Overall spore production increased with increasing relative humidity. The greatest and the least were produced after a 48 hr wet period at 20 and 28 C, respectively. Two regression models were developed to predict spore production as functions of relative humidity and temperature. In order to validate the infection function model a field experiment was conducted in 1989 to quantify latent period. A mean latent period beginning of 12 days, a mean latent period 50% of 14 days, and a mean latent period ending of 19 days were established. Two regression models were developed to predict latent period beginning and latent period ending as functions of mean maximum and mean minimum temperatures. In order to develop a simple forecast model field experiments were conducted in 1989 and 1990. From these studies two simple forecast models are proposed to initiate fungicide applications: (1) Initial disease occurrence. (2) A cumulative disease severity value (CDSV) of 23. Using the forecasts proposed it is possible to reduce five to seven applications of fungicides during the celery growing season in Quebec.

Late blight, caused by Septoria apiicola Speg., is the most important disease affecting celery in Quebec. Biological control was investigated as an alternative to conventional chemical control of late blight. Two hundred and four bacterial isolates were collected from celery leaves, and muck and mineral soils of celery fields in south-western Quebec. Two experiments were conducted to screen the bacteria for antagonism toward Septoria apiicola: one on agar to test for inhibition of pycnidial formation, and the other on leaf disks to test for inhibition of germination of conidia. From these two experiments, 18 isolates were selected that prevented pycnidial formation in an inhibition zone $ ge$1.0 cm wide and reduced germination to below 30% of the control. These isolates were re-evaluated for inhibition of germination on leaf disks. A bacterial suspension (10$ sp7$ cells/ml) was incubated on leaf disks for 24 hours before a suspension of S. apiicola conidia (150,000 spores/ml) was applied, and the disks were incubated for 25 hours. Four isolates reduced germination to $ le$19% of the control. These isolates were tested on plants in a greenhouse. None of the isolates was able to reduce the number of late blight lesions compared to the control.

No description available.

Forecasts for the first fungicide application to manage cercospora blight of carrot (Daucus carota L. subsp. sativus (Hoffim) Arcang) caused by Cercospora carotae (Passerini) Solheim were evaluated during the summers of 1991 and 1992. In the forecast based on infection, the blight severity values (BSV) were calculated for each day using duration of leaf wetness, relative humidity and temperature. The first fungicide was applied when the cumulative blight severity value reached 14 (CBSV 14) and 18 (CBSV 18). The forecasts were compared with the 50% disease incidence threshold method where the first fungicide was applied when the middle leaf of 50% of plants was diseased, and the conventional method where the first fungicide application was made when the plant height reached 15 cm. For the threshold of CBSV 14 and CBSV 18, no loss was observed in yield and fewer fungicides were applied as compared to conventional and 50% disease incidence threshold methods. Three years of weather data from experimental plots and 155 commercial field data were used to study the behaviour of CBSV at 50% disease incidence threshold. Only in 3% of the fields, the disease incidence of 50% was observed before the CBSV reached 11. The increase from CBSV 11 to 20 was rapid. Depending on environmental conditions the first fungicide at a CBSV range of 11 to 20 is proposed rather than a critical threshold since no loss in yield was observed at 18 in the experimental plots. No fungicide is applied until the CBSV reaches 11 and the application must not be delayed after CBSV reaches 20.

Graduation date: 1944

Typescript. / Thesis (Ph. D.)--University of Hawaii at Manoa, 1977. / Bibliography: leaves 150-154. / Microfiche. / xvi, 154 leaves ill

Typescript. / Thesis (Ph. D.)--University of Hawaii at Manoa, 1973. / Bibliography: leaves [104]-111. / x, 111 l graphs, tables

Mode of access: World Wide Web. / Thesis (Ph. D.)--University of Hawaii at Manoa, 2005. / Includes bibliographical references. / Electronic reproduction. / Also available by subscription via World Wide Web / xv, 222 leaves, bound col. ill 29 cm