Pratylenchus brachyurus is widespread in Brazil and there are reports of its occurrence in many crops (^{MACHADO et al., 2006}; ^{ALVES et al., 2011}; ^{INOMOTO, 2011}). In soybean, its importance has increased in recent years, especially in the Midwest region, in areas of maize second crop after soybean in no tillage production systems under irrigation and in sandy soils (^{GOULART, 2008}). Corn is highly susceptible to nematodes, especially P. brachyurus (^{LORDELLO et al., 1985}; ^{INOMOTO, 2011}).

Among the main measures to control nematodes, the use of genetic resistance stands out. Although some preliminary research results regarding the reaction of soybean and corn genotypes to this nematode exist, data are not conclusive (^{ALVES et al., 2011}; ^{INOMOTO, 2011}). Studies have been conducted under greenhouse and in field conditions in different locations in Brazil; however, few advances have been achieved, and, in some cases, with contradictory results. This can be attributed to different methods of evaluation and to the variability of the nematode present in different regions. For ^{MACHADO et al. (2006}) variability occurs in nematode populations leading to differences in the aggressiveness of P. brachyurus. The fact that the interaction of P. brachyurus with plants does not induce specialized feeding cell formation makes it difficult to find resistance and to elucidate the mechanisms involved (^{GOULART, 2008}).

In order to evaluate commercial soybean and corn genotypes looking for resistance to P. brachyurus two trials were carried out: one with soybeans in Vicentinópolis/GO and the other with corn in Edéia/GO. Both experiments were installed in commercial crops under conventional sowing systems naturally infested by the nematode, with a history of high nematode populations and the occurrence of stunted spots in the previous harvest. The experimental design was a randomized complete block with split plot over time, with 50 treatments (genotypes) and eight replications for soybean and 38 treatments (genotypes) and ten replications for corn. The evaluations were performed at 30 and 60 days after emergence (DAE). Each plot consisted of a 50cm row, allowing a ten seed condensed sowing. Each row contained all genotypes ordered at random, composing a block.

For each evaluation, samplings were drawn collecting three plants from each plot. The shoots were discarded and the root systems taken to the laboratory for nematode extraction according to methodology described by ^{COOLEN & D'HERD (1972}). Nematode density data were subjected to tests for normality and homogeneity of variance. Once the statistic assumptions were complied, the analysis of variance was performed with the Scott-Knott test at the 5% level of significance. Statistical tests were performed with data transformed into y'= y^0.20 for soybean and y' = y^0.057 for corn, according to ^{BOX-COX (1964}) and subjected to the analysis of variance, using ^{SAS statistical software (1999)}.

The mean test (P≤0.05) separated the soybean genotypes into two groups in both evaluations (Table 1). The analysis showed interaction between genotypes and evaluation periods (P≤0.05) for 12 soybean genotypes; ten with population reduction (BRSGO Chapadões, BRSGO Paraiso, M-Soy 7211 RR, M-Soy 8008 RR, Emgopa 313 RR, M-Soy 8411, BRSGO Juliana RR, Emgopa 316, BRSGO Luziania RR and TMG 103 RR), ranging from 46% to 70%. A possible explanation for the decline observed in the nematode population in the second evaluation is that the tested soybean cultivars have determined growth habit that paralyzes their growth after flowering, before entering the reproductive stage (^{NOGUEIRA et al., 2009}). As roots stop growing, there is greater competition for food among nematodes, forcing their output from the plant or reducing their multiplication.

Table 1 Population density of P. brachyurus (n. of individuals 10g^-1 of roots) at 30 and 60 days after emergence (DAE) in soybean genotypes. Vicentinópolis, GO. UFG, 2009. 

Means followed by the same letter in the column do not differ by the Scott-Knott test at the 5% level of significance. After the analysis of variance data were transformed into y' = y ^0.20.

^*Significant interaction between evaluation periods.

Other authors also tested some soybean genotypes tested in this study and, in some cases, results were confirmed. ^{ALVES et al. (2011}) also observed that cultivars 'Emgopa 313 RR' and 'M-Soy 8411', considered potentially resistant in this study, had low reproduction factor (RF) under controlled conditions. ^{RIBEIRO et al. (2007}) and ^{MACHADO (2009}) also found similar results for cultivar BRSGO Chapadões under greenhouse conditions. Conversely, 'TMG 103 RR' and 'BRSGO Paraiso' cultivars, which in the present study were considered resistant, presented high RF in a study by ^{RIBEIRO et al. (2007}).

The corm genotypes tested were divided by the mean test (P≤0.05) in three groups, in both evaluations (30 DAE and 60 DAE) (Table 2). From the corn genotypes tested, 16 significantly increased the nematode population density at 60 DAE, most with an increase of over 100%. Similar results were found by ^{LORDELLO et al. (1985}), who also observed a gradual increase of nematodes P. brachyurus and P. zeae during the testing period in a trial in field with evaluations at 39, 59 and 90 DAS. ^{INOMOTO (2011}) studied the reaction of corn hybrids to P. brachyurus under controlled conditions and observed RF ranging from 4.0 to 15.4.

Table 2 Population density of P. brachyurus (n. of individuals 10g^-1 of roots) at 30 and 60 days after emergence (DAE) in corn genotypes. Edéia, GO. UFG, 2009. 

Means followed by the same letter do not differ by the Scott-Knott test at the 5% level of significance. For the analysis of variance data were transformed into y' = y 0.057.

^*Significant interaction between evaluation periods at 5%.

In this study, only four corn genotypes showed low population density in both evaluations, which can indicate genetic resistance (Table 2). They were P 30F80, GNZ 2500, DKB 350 and NK Impacto. Hybrid Agromem 30A06, despite being among those that showed high population densities in the first evaluation, was the one that had the density reduced by 58% at 60 DAE.

Diverse results reported in different studies for soybean and corn genotypes may be due to differences in the methods used for conducting and evaluating experiments. Extraction and counting of nematodes from roots may still be the most appropriate method to check nematode population density as well as to evaluate crop yield related to the existence of tolerance to P. brachyurus under naturally infested field conditions. It is possible that the genotypes that remained with low nematode population density along the evaluation period may present mechanisms to make difficult the nematode penetration. Conversely, genotypes that presented significant reduction on the nematode population at the second evaluation may have some mechanism that works after the nematode has entered the roots. This suggests that new studies should look for mechanical or biochemical modifications in the plants due to the nematode parasitism.

Another hypothesis to explain the diversity of results when compared to other studies, is that there may be differences of aggressiveness among nematode populations of P. brachyurus from different regions of Brazil (^{FALLAS et al., 1996}; ^{MACHADO et al., 2006}), suggesting that studies to characterize the behavior of soybean and corn genotypes to this nematode should be sustained.

The genotypes of soybean and corn that remained with the lowest population densities in the two evaluation periods or significantly reduced the population in the second evaluation can be considered moderately resistant and be good choices for sowing in areas infested by the nematode. They can also be targeted for further investigations identifying genetic resistance.