The aims of this study were to explore the evolution of pH in topsoils under long-term different fertilization regimes, and further to understand the mechanisms and methods against soil acidification in croplands. We collected historical data on six 18–30 years long-term fertilization experiments in typical Chinese croplands of red soil in Qiyang and Jinxian, purple soil in Chongqing, black soil in Gongzhuling and Harbin with conventional and high fertilization rates, and then analyzed the differences in soil pH and acidification rate among experimental stages (six years each) and treatments, i.e., no-fertilizer control (CK), sole chemical nitrogen fertilizer (N), chemical nitrogen plus phosphorous and potassium fertilizers (NPK) and NPK fertilizers amended with manure (NPKM). Results show that under control, topsoil pHs declined somewhat and acidification rates averaged 0.013 pH a ^-1. Under the N treatment, topsoil pHs of the experiments in Qiyang, Gongzhuling and Harbin with conventional fertilization rate were lowered by 0.32–0.55 units at the initial 6 years and all the experiments by 0.64–1.46 units at the recent 6 years as compared with those under control. Topsoil pHs under the NPK treatment were obviously lower than those under the N treatment over the 1–12 years experimental period. However, topsoil pHs under the NPKM treatment were 0.30–0.53 units higher than those under the NPK treatment, and the two had a similar pH evolution curve. Paired T test between the treatments shows that in terms of soil acidification rate, the treatments followed the order of N > NPK > NPKM ≈ CK. The mean soil acidification rate under the N and NPK treatments was 4.6 and 3.2 times, respectively as high as that under control. Taking well-drained upland or upland-paddy rotation croplands in the humid or semi-humid zone as examples, the processes of soil acidification in the croplands were analyzed. The evolution of pH in topsoils of croplands was characterized as a common response to fertilization, regardless of certain regional differences.