ABSTRACT. The genetic divergence of 38 melon accessions from traditional agriculture of the Brazilian Northeast and three commercial hybrids were evaluated using fruit descriptors and microsatellite markers. The melon germplasm belongs to the botanic varieties cantalupensis (19), momordica (7), conomon (4), and inodorus (3), and to eight genotypes that were identified only at the species level. The fruit descriptors evaluated were: number of fruits per plant (NPF), fruit mass (FM; kg), fruit longitudinal diameter (LD; cm), fruit transversal diameter (TD; cm), shape index based on the LD/TD ratio, flesh pulp thickness, cavity thickness (CT; cm), firmness fruit pulp (N), and soluble solids (SS; °Brix). The results showed high variability for all Genetic diversity of Cucumis melo from Brazilian Northeast descriptors, especially for NPF, LD, and FM. The grouping analysis based on fruit descriptors produced eight groups without taxonomic criteria. The LD (22.52%), NPF (19.70%), CT (16.13%), and SS (9.57%) characteristics were the descriptors that contributed the most to genotype dissimilarity. The 17 simple sequence repeat polymorphic markers amplified 41 alleles with an average of 2.41 alleles and three genotypes per locus. Some markers presented a high frequency for the main allele. The genetic diversity ranged from 0.07 to 0.60, the observed heterozygosity had very low values, and the mean polymorphism information content was 0.32. Molecular genetic similarity analyses clustered the accessions in 13 groups, also not following taxonomic ranks. There was no association between morphoagronomic and molecular groupings. In conclusion, there was great variability among the accessions and among and within botanic groups.

Beetroot (Beta vulgaris L.) is a tuberous vegetable that is under experimentation in the vegetable production system in Rio Grande do Norte state (Brazil). It is quite demanding, in nutritional terms. One of the alternatives to meet this need is green manuring using spontaneous species of the Caatinga biome. The aim of the present work was to optimize the agro-biological performance and economic profitability of the beetroot 'Early Wonder', fertilized with different doses of hairy woodrose (Merremia aegyptia (L.) Urb.) in this semi-arid environment. Hairy woodrose is a spontaneous species of the Caatinga biome that is widely distributed, being found in forests, fences, forest clearings and fields, growing in soils of different textures, producing biomass with a high content of macronutrients. This material was collected from spontaneous vegetation, native to the region, and used as green manure.The experimental design was a randomized complete block, with four treatments and five replicates. The treatments consisted of four amounts of hairy woodrose biomass incorporated into the soil (10, 25, 40 and 55 t ha-1 of dry matter). The following agronomic characteristics were evaluated in each beetroot treatment: plant height, fresh mass of shoots, number of leaves per plant, dry mass of shoots and dry mass of roots, total and commercial root productivity, and root productivity classified into scrap and extra roots, extra A roots, extra AA and great roots. Apart from these characteristics, the following economic indicators were also determined for each treatment: gross income, net income, rate of return and profit margin. The maximum agro-biological efficiency of beet root productivity (22.68 t ha-1) was obtained from 44.92 t ha-1 of hairy woodrose biomass added to the soil. The maximum economic efficiency of the yield of beet roots was achieved as a net income of AU$2,491.00 (R$6,406.36) ha-1, provided by the production of 22.04 t ha-1 of commercial roots, using 37.03 t ha-1 of hairy woodrose biomass incorporated into the soil. The use of hairy woodrose as green manure presents agro-economic feasibility in beet cultivation in this semi-arid environment.

Intercropping of beet with cowpea-vegetable is increasing in the State of Rio Grande do Norte, Brazil. However, its management requires scientific information and technologies pertaining to space arrangement and adequate fertilization with roostertree. The present study evaluated the agronomic/biological performance of the beet and cowpea-vegetable intercropping system with different amounts of roostertree and spatial arrangements of the component cultures. We established a 4 × 3 factorial completely randomized design with four blocks. The first factor was the amount of roostertree biomass incorporated into the soil (20, 35, 50, and 65 t ha-1 in dry matter), while the second was the spatial arrangement of beet rows alternated with cowpea rows (2:2, 3:3, and 4:4). We evaluated the total, commercial, and classified productivities of beet roots, as well as the number of green pods per m2, green and dry pod productivities, the number of green grains per pod, the weight of 100 green grains, and green and dry grain productive of cowpea. The agronomic efficiency of the intercropping system was evaluated by the land equivalent ratio, productive efficiency index, and score of the canonical variable z. The results showed no significant interactions among the amounts of roostertree biomass and spatial arrangements for any evaluated trait or agronomic index. The highest agronomic efficiency of the intercropping system was obtained with 65 t ha-1 of roostertree biomass and the 2:2 spatial arrangements.