Abstract: Overuse of chemical fertilizers is one of the important factors that cause the agricultural nonpoint source pollution to aggravate in precise agriculture. The middle and lower reaches of the Yangtze River are the main producing areas of oilseed rape in China. In sowing, the mixed and extensive application of chemical fertilizers with low utilization has become a severe restriction on the sustainable production of rape in green agriculture. Layered and quantitative fertilization is an alternative way to reduce the excessive amount, while increasing the efficiency of chemical fertilizers. In layered fertilizing, the proportion of fertilizers in the upper and lower layer can be accurately controlled with a high utilization rate for the later advantageous growth of crops in different areas. However, the traditional extensive mode of operation has posed a great challenge on the direct seeding of rapeseed in the middle and lower reaches of the Yangtze River. In this study, a layered and proportional fertilizing device was proposed for the uniform distribution of fertilizer flow in a rape seeder, according to the planting and growth pattern of the root system. A proportional control system was applied to adjust the amount of fertilizer in the upper and lower layers, while separately fertilizing at the left and right of the upper layer. A quadratic regression orthogonal test was conducted to take the diameter of blocking rod, the number of the blocking rod groups, and the distance between blocking rods as the experimental factors. The optimal structural parameters of the device were determined: the rod diameter was 3 mm, the number of rod groups was 5, and the distance between the rods was 8.9 mm. A performance test of fertilizer discharge was carried out under the optimal combination of parameters. The evaluation indexes were set as the error between the target fertilizing ratio and the actual fertilizing ratio, the variation coefficient of stable discharge volume for the upper and lower pipes, and the variation coefficient of stable discharge volume for the left and right sides of the upper pipe. The test results showed that the maximum error between the actual and target fertilizing ratio was 4.1 percentage points for the upper and lower layers, indicating very little influence on the later growth of rape. The variation coefficient of stable fertilizer discharge in the upper and lower layers was lower than 3.9%, indicating a stable distribution of fertilizer ratio for the standard requirements in the layered fertilization of rape. The error between the actual and target fertilizing ratio was less than 4.1 percentage points in the upper pipe, meaning an excellent fertilization effect for the standard requirements in the upper layer. The variation coefficient of stable discharge volume was lower than 4.8% in the upper pipe. There was no obvious relationship with the rotational speed of the discharge device, indicating excellent stability of discharge volume for the left and right sides. The field test showed that the fertilizing device was stable for the high requirements in the layered fertilization of rape. This finding can provide potential technical support for the proportional layering of fertilizer during large-scale production of oilseed rape.