The paper presents the results of a 3-year field experiment designed to evaluate the content of organic carbon in brown soil (Haplic Cambisol Dystric) developed from a light loamy sand under legumes cultivation. Experimental factors were: species of legume crop (colorful-blooming pea (Pisum sativum), chickling vetch (Lathyrus sativus), narrow-leafed lupin (Lupinus angustifolius), methods of legumes tillage (legumes in pure culture and in mixture with naked oats) and mineral N fertilization (0, 30, 60, 90 kg N·ha−1). Cultivation of legumes on sandy soil did not result in an increase of organic carbon content in the soil after harvest as compared to the initial situation, i.e. 7.39 vs. 7.76 g·kg−1 dry matter (DM), on average, respectively. However, there was the beneficial effect of this group of plants on soil abundance in organic matter, the manifestation of which was higher content of organic carbon in soils after legume harvest as compared to soils with oats grown (7.21 g·kg−1 DM, on average). Among experimental crops, cultivation of pea exerted the most positive action to organic carbon content (7.58 g·kg−1, after harvest, on average), whereas narrow-leaved lupin had the least effect on organic carbon content (7.23 g·kg−1, on average). Pure culture and greater intensity of legume cultivation associated with the use of higher doses of mineral nitrogen caused less reduction in organic carbon content in soils after harvest.