**3. Materials and methods**

The experiment was established in April 2009 at the State University of Ponta Grossa Farm (Fazenda Escola Capão da Onça) in the city of Ponta Grossa, Paraná, in southern Brazil (25° 05′ S and 50° 05′ W). The climate is classified as Cfb according to the Köppen system [19], with cold and humid winters and occasional frosts between May and July. The annual mean precipitation during the experimental period of 44 years is 1545 mm, with higher precipitation levels in the summer and no dry period defined. The mean maximum temperature is 24°C, and the minimum is 13.3°C. The soil is classified as Haplic Cambisol with medium texture, and it represents 27% of the region [35]. The results of the soil fertility analysis performed before the experiment were pH (CaCl2, 1M), cation exchange capacity = 11.2 cmolc dm−3, soil density = 1.35 Mg m−3, total organic C = 11.9 g kg−1, total organic nitrogen = 16.12 g kg−1, available P = 38.1 mg kg−1 and available K = 0.24 cmolc dm−3.

The experimental design was of completely randomized blocks with six treatments and three replications. The following treatments were performed: T1 = control, with no slaughterhouse organic residue (SOR) or synthetic mineral fertilizer (SMF) applications; T2 = 100% SMF, with all plant nutrient supply applied via synthetic mineral fertilizer; T3 = 100% SOR, with all plant nutrient supply applied via slaughterhouse organic residue; T4 = 75% SOR + 25% SMF; T5 = 50% SOR + 50% SMF; and T6 = 25% SOR + 75% SMF. For T4, T5 and T6, the rates of SORs applied were equivalent to 75%, 50% and 25% of the residues used in T2, respectively. The rates of SMFs applied were based on the soil fertility analysis and the recommendation for the crops used in the region.

In T3 (100% SOR), 94, 21 and 19 kg ha−1 of N, P and K, respectively, were applied. These were equivalent to 2 Mg ha−1 bio‐fertilizer. The crop sequence from 2009 to 2012 was the alternation of crops used in the summer and winter seasons in the region.

Crop yield was determined by harvesting 5 m of the three central rows in the summer and winter season crops. The grains were submitted to a cleaning process to remove impurities and then dried for humidity correction. Grain weight was corrected to 14% humidity for beans and 13% for the other crops. The unit was converted to kg ha−1 and then to Mg ha−1. For black oat, we determined the dry mass production collecting two points with 0.17 m2 at each plot. The soil total organic carbon (TOC) content was determined through the dry combustion method using an elementary C/N analyzer (TruSpec CN LECO® 2006, St. Joseph, EUA).

The results for content and the TOC stock in whole samples, the particle‐size fractions and the SOM labile compartments were subjected to analysis of variance (ANOVA). The means that were significantly different from the *F* test were compared using the LSD test at 5% probability (*I* = 0.05) using SISVAR 5.1 [36].
