Among the forage species cultivated in South America, the genus Urochloa is the most used, and the cultivar Marandu of U. brizantha is the most widely planted in Brazil. The objective of this study was to evaluate forage performance in association with Azospirillum brasilense, combined with nitrogen (N) fertilisation. The study was conducted under field conditions in Araguaína, Tocantins, in the central region of Brazil, between March 2016 and March 2017. Four N fertiliser rates (0, 12.5, 25 and 50kg/ha of N per cutting cycle) were combined with two inoculation treatments (inoculated and non-inoculated), with evaluations carried out in three periods of the year (transition, dry and wet seasons). Marandu grass plants inoculated with A. brasilense had greater plant height, number of tillers and forage production than non-inoculated plants, regardless of the N rate. Inoculation with A. brasilense allowed a 20% reduction in N fertilisation. Our results indicate that inoculation with A. brasilense in Marandu grass, as well as increasing forage production, can help to mitigate the stresses caused by the dry season.

In the SE periphery of Brazilian Amazonia, low-input agriculture systems on sandy loam soils have very low nutrient use efficiency. In a low-input alley cropping system, we measured residue decomposition dynamics and the yield and nutrient uptake of a maize crop associated with the following treatments: Clitoria + Pigeon pea; Acacia + Pigeon pea; Leucaena + Clitoria; Leucaena + Acacia, Leucaena + Pigeon pea and no residue input (control). The acacia treatments provided better soil coverage throughout the whole corn cycle. Potassium was released faster than nitrogen from the residues; N concentration in corn leaves in the residue treatments were below critical levels. The Leucaena + Acacia treatment was the most effective in increasing posttasseling N and K assimilation and K use efficiency.This resulted in corn productivity 3.5 times greater (7.3 Mg ha −1 ) than the control without residue application. In the Amazonian sandy loam soils, which are susceptible to hardsetting and nutrient leaching, efficient N and K use should be priorities for soil management. Although no-till alley cropping of leguminous trees constitutes an important option for lowinput farming, its efficiency depends on using a mixture of residues that keeps soil covered and have high rates of both N and K release during the entire crop cycle.

RESUMOA revegetação com espécies arbóreas é uma estratégia adequada, quando se pretende a reabilitação de terras degradadas. Entretanto, o estabelecimento e o crescimento destas em solos pobres em nutrientes dependem de diversos fatores como a capacidade de formar e beneficiar-se das micorrizas. No presente estudo, realizado em casa de vegetação na Universidade Federal de Lavras, Minas Gerais -Brasil, avaliaram-se aspectos da relação fungos micorrízicos arbusculares e espécies arbóreas do sudeste brasileiro. Os fungos estudados foram: Scutellospora pellucida, Acaulospora scrobiculata, Entrophospora colombiana, Gigaspora gigantea, Gigaspora margarita, Glomus etunicatum, Scutellospora gregaria, Glomus clarum e isolados oriundos de agrossistemas e de mata, os quais foram inoculados em dezesseis espécies vegetais: Luehea grandiflora, Cecropia pachystachya, Schinus terebinthifolius, Machaerium nyctitans, Senna macranthera, Senna spectabilis, Solanum granuloso-leprosum, Caesalpinea ferrea, Tabebuia serratifolia, Maclura tinctoria, Guazuma ulmifolia, Acacia polyphylla, Mimosa caesalpiniaefolia, Enterolobium contortisiliquum, Trema micrantha e Cedrela fissilis. Estas espécies apresentaram comportamento diferenciado em relação à susceptibilidade aos fungos e às respostas em crescimento decorrentes da inoculação. A amplitude de eficiência simbiótica dos fungos variou muito, sendo o Gl. clarum, E. colombiana, S. pellucida e Gl. etunicatum os fungos de maior amplitude, beneficiando mais de 80 % das espécies vegetais estudadas, e a A. scrobiculata a espécie de comportamento mais restrito em relação aos hospedeiros. A análise da compatibilidade fungo-hospedeiro e o comportamento geral das

Using biological inputs to improve the efficiency of nitrogen fertilizers represents an alternative for the cultivation of grasses in tropical regions. Azospirillum brasilense is a species of plant growth promoting bacteria widely studied and used in inoculants. Thus, this study aimed to evaluate the performance of Mombasa grass (Megathyrsus maximus) in association with A. brasilense and nitrogen (N) fertilization.The study was conducted under field conditions in Araguaína-Tocantins State, between December 2017 and May 2018. The treatments were arranged in randomized blocks, in a 5 × 2 factorial arrangement, with five doses of N fertilization (0, 25, 50, 75, and 100 kg ha -1 ) combined with two inoculation treatments (inoculated and non inoculated), in four replicates. For the number of tillers and root production, the inoculation efficiency varied as a function of the supplied N doses. However, the percentage of leaf N was higher for inoculated plants regardless of the application of nitrogen. In the absence of nitrogen fertilization, it was possible to increase forage production by up to 36 % with inoculation.

O método aqui proposto, denomidado de Área Foliar – Universidade Federal do Tocantins (AFUFT) baseia-se em três medidas da folha (comprimento total, largura na base do limbo e largura no meio do limbo), decompondo-a em um triângulo e um trapézio para o calculo da AF. O AFUFT (método II) foi comparado com três métodos: método I – indireto, com uso de aparelho de interceptação luminosa (SunScan®); método III – segmenta as folhas em partes de 10 cm, seguido de medida da largura destes segmentos; método IV – utiliza-se scanners e programa computacional (aqui considerado como método padrão). As quatro metodologias foram testadas em blocos casualisados, sendo os blocos as 24 parcelas (4,0 m x 1,0 m) de Urochloa brizantha cv. ‘Marandu’. O método AFUFT apresentou os maiores coeficientes de correlação (r= 0,99; p