The plant density and spatial arrangement of kenaf is an important aspect in kenaf fiber production. A field plant density experiment was conducted at Ledesma (Jujuy, Argentina) in 2001 on a sandy loam soil. The treatments used were different combinations of 35 and 70 cm row spacings with lineal sowing densities of 25 and 40 plants m-1, and were applied to the Cuba 108, Endora and Tainung 1 cultivars. Two indices (bark content and bark index) related to fiber yield and useful for the individual selection of plants were measured. The combination of 35 cm row spacing and 25 plants m-1 lineal sowing density, representing an initial density of 714,286 plants ha-1, resulted in the best dry bark yield for the three cultivars. As a result of strong intraspecific competition, height and diameter of the plants decreased while plant density increased. The initial lineal density of 40 plants m-1 was not advantageous in comparison to 25 plants m-1, because plant survival rates were reduced at 40 plants m-1 and yield did not increase linearly.

AFLP analysis was carried out to assess genetic variability and determine the population structure of the sugarcane rust Puccinia melanocephala in northwest Argentina. Molecular data were also used to clarify whether genetic variation was correlated with host variation and ⁄ or the geographic distribution of the disease. Bulk rust uredospores were collected in the field, and both the geographical area and the infected host sugarcane cultivar were recorded. A total of 538 AFLP markers generated with 20 primer combinations were used to perform the genetic analysis. The percentage of polymorphic loci was quite high (85.7%), considering that P. melanocephala only reproduces asexually. Cluster analysis (UPGMA) and principal co-ordinate analysis (PCoA) grouped populations from distinct geographic and host origins, suggesting that neither geographical region nor sugarcane variety constrains the relationships among the populations. This finding was corroborated by a lack of significant correlation between genetic distance and geographic distance (r = 0.057; P = 0.285). The non-significant differences found between rust populations collected from distinct sugarcane varieties (PhiT = 0.026; P = 0.06) also support these results. Analysis of Molecular Variance Approach (AMOVA) analysis attributed most of the variation (95%) to differences within populations. No genetic structure was detected, and the populations behaved as a large undifferentiated population with high level of genetic variability.

No abstract

No abstract