The decline of arable species characteristic of winter cereal fields has often been attributed to different factors related to agricultural intensification but most importantly to herbicide use. Herbicide phytotoxicity is most frequently assessed using short-term endpoints, primarily aboveground biomass. However, short-term sensitivity is usually not sufficient to detect actual effects because plants may or may not recover over time following sublethal herbicide exposures. Therefore, it is important to assess the long-term effects of herbicide applications. Annual species rely on renewable seed production to ensure their persistence; hence, assessment of herbicide sensitivity is more accurately estimated through effects on reproduction. Here we aim to assess the phytotoxicity of two commonly used herbicides: tribenuron and 2,4-D on eight plant species belonging to four families, each with one rare and one more common species. Specifically we examined the pattern of sensitivity using short-term and long-term endpoints (total aboveground biomass, total seed biomass and number of seeds) of these species; we determined the levels of and time to recovery in terms of stem length and fruit number, and assessed whether their rarity relates to their sensitivity to herbicide application. Our results suggest that although differences in herbicide sensitivity are not a direct cause of rarity for all species, it may be an important driver of declining arable plants.