【Objective】Being the kernel of the traditional agriculture of China, and one of the major practices in developing ecological agriculture, intercropping still occupies an important position in modern agriculture in China as well as other parts of the world. One of the advantages of intercropping is overyielding relative to monocultures, which consequently removes more soil nutrients from the soils. However, in respect of high nutrient removal, few researches have been done on intercropping as on monocultures on a long-term scale. 【Method】To find a solution to this issue, a long-term field experiment was therefore established in 2009 at Baiyun, Wuwei City, Gansu Province, Northwest China, and carried out to investigate changes in soil chemical properties in response to P application and cropping system. The experiment was laid out in a two factorial design with three replicates, where the main factor was P application rate, designed to have three rates (0, 40 and 80 kg hm^-2 of triple superphosphate) and the second factor was cropping system, including maize/faba bean, maize/soybean, maize/chickpea, maize/rapeseed and their corresponding monocultures, maize, faba bean, soybean, chickpea and rapeseed. In 2013 (the 5th year) and 2014 (the 6th year) yields were measured and some soil chemical properties (soil organic matter, soil total N, Olsen P, available K, and pH were examined after the crops were harvested to explore changes in soil fertility as affected by cropping system. To estimate yields, two rows of maize and three rows, each of the other 4 crops, regardless of cropping pattern, were selected at random and harvested at their respective harvest seasons. To collect soil samples, 0 ~ 20 cm in length, for analysis, an auger was used to minimize disturbance of the soil structure after harvest. Eight cores and four cores were collected from intercropping plot (4 for each crop species) and monocultures, separately. And then the soil samples from the same plot under the same crop were blended separately together as one for analysis. 【Result】Results show that by averaging the yields of the plots the same in cropping system regardless of P application rates, 1) The grain yield of treatment, chickpea/maize, faba bean/maize, soybean/maize, and rapeseed/maize was 38.2%, 32.6%, 34.0% and 38.4% higher than their respective monocultures treatment; 2) Intercropping significantly increased the content of soil organic matter; 3) Neither P application nor cropping pattern had any effect on soil total nitrogen concentration; 4) In 2013, intercropping decreased soil Olsen P by 5.2% in treatment chickpea/maize, by 6.9% in treatment faba bean/maize, by 15.9% in treatment soybean/maize and by 11.3% in treatment rapeseed/maize, while in 2014, it did not affect much; 5) Intercropping significantly depleted available K, by 10.3% in treatment chickpea/maize, by 14.1% in treatment faba bean/maize, by 8.5% in treatment soybean/maize and by 13.2% in treatment rapeseed/maize over the two years; 6) There was no significant differences in soil pH between P applications and cropping systems in 2013 and 2014. To sum up, intercropping enhances productivity for at least five to six years. No matter P application or not, it could increase grain yield, and the effect intensified with P application rate rising till 80 kg hm^-2. Moreover, intercropping tended to increase soil organic matter and reduce soil Olsen P and available K. But neither P application nor intercropping had much effect on soil total nitrogen or pH in 2013 and 2014. 【Conclusion】In conclusion, intercropping enhances not only crop productivity over time, but also improves stability and sustainablility of soil fertility if coupled with proper applications of organic manure, P and K fertilizers.