Ranking techniques was investigated the influence of independent factors (the content of the clay fraction and humus, pH, CEC, etc.) On the dependent variable (the change of the rate of decomposition of the herbicide - k).
The biological activity of the arable layer of farmland depends on the physico-chemical properties of plant protection products used in agriculture, and also determines the behavior of xenobiotics in soil. We have investigated the degradation of herbicides preparations depending on soil properties. Ranking techniques was investigated the influence of independent factors (the content of the clay fraction and humus, pH, CEC, etc.) On the dependent variable (the change of the rate of decomposition of the herbicide - k). The result set different contribution of soil conditions on the degradation of the active substances in the soil herbicide formulations (Table. 1). For imidazolinone herbicides observed binding activity of their protonated molecules soil absorbing complex, regardless of the content of humus in acidic soil. Indicator for sorption imazamox ranged Kd = 1,5-2,4, and in neutral or alkaline soils not exceed Kd = 0,5-0,8. For imazethapyr regardless of the acidity of the soil and humus content index Kd ranged 0,08-0,76 and determines the differences in the mineralogical composition of the soil. Found high rates of sorption for atrazine, imazethapyr and other herbicides in the case of worsening of the mechanical composition of the soil (clay mineral content greater than 42%). In general, increasing the rate of degradation studied herbicides (benzoic acid derivatives tiadiazinony, sulfonylureas, imidazolinones, chloroacetamides) associated with the value EKO, humus soil and silt. So, especially the influence of different factors on the degradation of herbicides in agro-ecosystems is important to know. With the right choice of adaptive strategies of agriculture can reduce environmental risks in the system of soil-herbicide.

An experiment was conducted on transplanted Aman (monsoon) rice at the Sher-e-Bangla Agricultural University farm Dhaka (90033´ E longitude and 23077´ N latitude), Bangladesh during June-November, 2005. There were seven different weed control treatments viz. T1= Amchlor® 5G @ 15 kg ha-1 + IR5878® 50 WP @ 120 g ha-1, T2= Ronstar® 25EC @ 1.25 L ha-1 + IR5878® 50 WP @ 120 g ha-1, T3= IR5878® 50 WP @ 150 g ha-1, T4= IR5878® 50 WP @ 120 g ha-1, T5= Set-off® 20WG @ 50 g ha-1 + IR5878® 50 WP @ 120 g ha-1, T6= Two hand weeding and T7= Weedy check. There were 14 different weed species infested the field among which Panicum repens was the most important. Among the herbicides T2 was the most efficient with the lowest weed population and weed dry weight. The yield and the yield contributing characters (plant height, number of effective tillers per hill, panicle length and no. of filled grains) were influenced according to the effectiveness of the treatments, with T2 being the highest yielding herbicide treatment, reaching the yield level of the hand weeding treatments (T6). Maximum benefit-cost ratio with T2 suggested that this herbicidal treatment can be used as an alternative when labour is a limiting factor in producing transplant Aman rice.