Control of Jointvetch ( Aeschynomene spp . ) , Establishment and Productivity of Rice as a Function of [ Imazapic + Imazapyr ] Doses

We aimed with this work to evaluate the efficiency of the herbicide [imazapic + imazapyr] in the weed control, especially of the jointvetch, and its reflex in rice plant establishment and grain yield. A field study was conducted, where treatments were: control without herbicide application; Kifix 140g ha (24.5 ga.i. ha imazapic + 73.5 ga.i. ha imazapyr) and Kifix 280 g ha (49 ga.i. ha imazapic + 147 ga.i. ha imazapyr). We evaluated crop emergence up to 25 days after sowing, crop plant height 12 days after emergence (DAE), jointvetch plant density 31 and 62 DAE, jointvetch dry mass 62 DAE and general weed infestation 62 DAE. At the end of the crop cycle we evaluated grains per panicle, panicle and 1000 grain weight, panicle density and crop grain yield. There is a risk of 15% reduction in the establishment of Clearfield rice plants when the maximum dose of the herbicide [imazapic + imazapyr] is applied pre-emergence, but under appropriate conditions, the crop development tends to compensate for this reduction in establishment. The herbicide [imazapic + imazapyr] is effective on jointvetch, but under high infestation conditions, as in the present study, a complementary herbicide application is required to achieve adequate control of jointvetch and avoid rice grain yield reduction due to competition.


Introduction
Rice is one of the most important cereals produced in the world.It is cultivated mainly in paddy fields using a continuous flooding.In Brazil, Rio Grande do Sul (RS) and Santa Catarina (SC) States are the main producers, being responsible for approximately 80% of the national production, where the average productivity is 6,837 and 7,139 kg ha -1 , respectively (CONAB, 2017).
Paddy rice cropping fields in southern Brazil are generally highly productive, although yield averages are still lower than those achieved in areas that adopt recommended cropping practices such as experimental fields.One of the main reasons for yield reduction in this crop is related to unsatisfactory weed control (Galon et al., 2007;Agostinetto et al., 2008).The importance of weed infestation on rice grain yield has been increasing due to the adoption of minimum till and the low efficacy of non-selective herbicides against them.
Weedy rice and Echinochloa sp. are the most important weeds in paddy rice in RS and limit crop productivity due to their competitive ability and wide occurrence throughout the state (Agostinetto et al., 2001).Another weed commonly found in rice fields is the jointvetch (Aeschynomene rudis and A. denticulata), which impacts rice cultivation by reducing its grain yield, causing trouble to the harvesting operation, and affect seed/grain quality.In addition, jointvetch produces large amounts of seeds, which contributes to the increase of soil weed seed bank, interfering in subsequent crop cycles as well as in crops planted in succession to rice.It is a dicotyledonous that often requires herbicide applications specially directed to its control (Ferreira, 2007).
Currently, with the introduction of Clearfield ® technology, herbicides belonging to the chemical group of imidazolinones are primarily used for weed control.These herbicides inhibit the Acetolactate Synthase enzyme (ALS) and are highly efficient.However, the repeated application of these herbicides at doses above the recommended, causes prolonged residual effect.In addition, the absence of mechanisms of action rotation over time in the same area increases the selection pressure (Vargas, 2017) and favors the appearance of resistant biotypes to imidazolinones.Thus, several biotypes of weedy rice have already been found to be resistant to ALS-inhibiting herbicides in Brazil (Heap, 2017).
In rice fields planted with Clearfield ® rice varieties, there are frequent complaints that the herbicide [imazapic + imazapyr], associated to the technology, does not provide sufficient control of jointvetch.This situation is aggravated when the herbicide is applied post-emergence, when most jointvetch plants are taller, and some growers choose to use the maximum herbicide registration dose of [imazapic + imazapyr] with no clear information about the impact of this decision on Clearfield ® rice establishment, or on the emergence of non-Clearfield ® cultivars planted in succession.Some studies report the potential damage of Clearfield-related herbicides, as imazethapyr, on non-Clearfield ® rice varieties (Kraemer et al., 2009;Dornelles et al., 2010;Bundt et al., 2015;Ceolin et al., 2017), and other crops planted in succession to Clearfield rice (Alister & Kogan, 2005;Pinto et al., 2009).In addition, some studies reported possible damages to Clearfield rice after application of Kifix (Sousa, 2010;Galon et al., 2012) but with no conclusive data.Thus, we aimed to evaluate the efficiency of the herbicide [imazapic + imazapyr] in the weed control, especially of jointvetch, and its effects on plant establishment and rice grain yield.
The vegetation burndown prior to planting was done with 1440 g a.e.ha -1 of glyphosate, seven days before sowing the cultivar Guri INTA CL, on 11/09/2016, with row spacing of 0.17 m.The base fertilization consisted of 300 kg ha -1 of the formula N-P-K 5-25-25 applied to the planting row.
The emerged rice seedlings were counted every other day to obtain the emergence curve of the crop up to 25 DAP, in two samples per plot, each sample consisting of two planting rows with 60 cm.Rice plant height was assessed 12 days after emergence, by measuring the distance from the soil surface to the tip of the longest leaf, with a ruler.Ten plants were measured per plot.
The density of jointvetch plants was assessed on 12/19/2016, to quantify the effect of the treatments with Kifix, and again 35 days after quinclorac application, on 01/19/2017.Jointvetch plants present in 4 m 2 samples per plot, were counted and cut to soil level, being dried into oven with forced air circulation 65±5 °C, until constant weight.Also during this period, the general infestation of the area by weeds (composed mostly by jointvetch, with a few individuals of barnyardgrass and other aquatics) was assessed.
Rice grain yield was assessed at the end of the cycle, when two samples of 4 m 2 per plot were harvested by hand and threshed, subjected to oven drying with forced air circulation at 45±5 °C, after which the grains were weighed and their mass corrected to 13% humidity.The weight of one thousand grains and the weight of grains per panicle were evaluated by collecting five panicles per plot.We also assessed the panicle density and the number of grains per panicle, in two subsamples per plot, each measuring 0.25 m 2 (0.5 × 0.5 m).At harvest, the weed infestation was assessed again.
We processed the experimental data through descriptive statistics by presenting treatment means with their respective 95% confidence intervals, according to Cumming et al. (2004).By this method, the comparison between treatments is done based on an expected response interval for similar cropping situations, instead of only on the response of the treatments in the experiment.For rice emergence curve, we fitted a 2 nd degree polynomial surface as a function of both days after planting, and dose of Kifix.All analyzes were performed into the statistical environment "R".

Results
The Figure 5.