Cladistic analyses of plastid DNA sequences rbcL and trnL-F are presented separately and combined for 48 genera of Amaryllidaceae and 29 genera of related asparagalean families. The combined analysis is the most highly resolved of the three and provides good support for the monophyly of Amaryllidaceae and indicates Agapanthaceae as its sister family. Alliaceae are in turn sister to the Amaryllidaceae/Agapanthaceae clade. The origins of the family appear to be western Gondwanaland (Africa), and infrafamilial relationships are resolved along biogeographic lines. Tribe Amaryllideae, primarily South African, is sister to the rest of Amaryllidaceae; this tribe is supported by numerous morphological synapomorphies as well. The remaining two African tribes of the family, Haemantheae and Cyrtantheae, are well supported, but their position relative to the Australasian Calostemmateae and a large clade comprising the Eurasian and American genera, is not yet clear. The Eurasian and American elements of the family are each monophyletic sister clades. Internal resolution of the Eurasian clade only partially supports currently accepted tribal concepts, and few conclusions can be drawn on the relationships of the genera based on these data. A monophyletic Lycorideae (Central and East Asian) is weakly supported. Galanthus and Leucojum (Galantheae pro parte) are supported as sister genera by the bootstrap. The American clade shows a higher degree of internal resolution. Hippeastreae (minus Griffinia and Worsleya) are well supported, and Zephyranthinae are resolved as a distinct subtribe. An Andean clade marked by a chromosome number of 2n = 46 (and derivatives thereof) is resolved with weak support. The plastid DNA phylogenies are discussed in the context of biogeography and character evolution in the family.

This study suggests that the essential oils from Cinnamomum pubescens Kochummen and Etlingera elatior could be potentially used as a new source of natural antioxidant and antibacterial in the food and pharmaceutical industries.

No abstract

Enroute to mapping QTLs for yield components in oil palm, we constructed the linkage map of a FELDA high yielding oil palm (Elaeis guineensis), hybrid cross. The parents of the mapping population are a Deli dura and a pisifera of Yangambi origin. The cross out-yielded the average by 8–21% in four trials all of which yielded comparably to the best current commercial planting materials. The higher yield derived from a higher fruit oil content. SSR markers in the public domain - from CIRAD and MPOB, as well as some developed in FELDA - were used for the mapping, augmented by locally-designed AFLP markers. The female parent linkage map comprised 317 marker loci and the male parent map 331 loci, both in 16 linkage groups each. The number of markers per group ranged from 8–47 in the former and 12–40 in the latter. The integrated map was 2,247.5 cM long and included 479 markers and 168 anchor points. The number of markers per linkage group was 15–57, the average being 29, and the average map density 4.7 cM. The linkage groups ranged in length from 77.5 cM to 223.7 cM, with an average of 137 cM. The map is currently being validated against a closely related population and also being expanded to include yield related QTLs.

Increased modern farming of superior types of the oil palm, Elaeis guineensis Jacq., which has naturally efficient oil biosynthesis, has made it the world's foremost edible oil crop. Breeding improvement is, however, circumscribed by time and costs associated with the tree's long reproductive cycle, large size and 10-15 years of field testing. Markerassisted breeding has considerable potential for improving this crop. Towards this, quantitative trait loci (QTL) linked to oil yield component traits were mapped in a high-yield population. In total, 164 QTLs associated with 21 oil yield component traits were discovered, with cumulative QTL effects increasing in tandem with the number of QTL markers and matching the QT? alleles for each trait. The QTLs confirmed all traits to be polygenic, with many genes of individual small effects on independent loci, but epistatic interactions are not ruled out. Furthermore, several QTLs maybe pleiotropic as suggested by QTL clustering of inter-related traits on almost all linkage groups. Certain regions of the chromosomes seem richer in the genes affecting a particular yield component trait and likely encompass pleiotropic, epistatic and heterotic effects. A large proportion of the identified additive effects from QTLs may actually arise from genic interactions between loci. Comparisons with previous mapping studies show that most of the QTLs were for similar traits and shared similar marker intervals on the same linkage groups. Practical applications for such QTLs in marker-assisted breeding will require seeking them out in different genetic backgrounds and environments.Electronic supplementary material The online version of this article