Purpose The main goal of this study was to determine if ancient wheat varieties could store more carbon than modern ones in the presence or absence of inputs, due to a likely bigger and deeper root system and a slower mineralization rate. Methods We conducted a field experiment with four modern and four ancient varieties (released before 1960 and often grown without inputs), with and without chemical inputs (nitrogen, herbicide and fungicide taken as a single factor). Root morphology was assessed by image analysis, potential catabolic activities of fructose, alanine, citric acid by MicroResp™ and overall CO2 emissions by incubating soil and roots from each modality for 60 days. Results The breeding type did not affect root traits, substrates respiration nor CO2 emissions in our environmental conditions. The application of inputs did not affect root traits but influenced the respiration of specific substrates and CO2 emissions. The most noticeable response was due to the “breeding type x inputs” interaction: inputs increased CO2 emissions from soil and root tissues of ancient varieties by 19%, whereas no effect was observed for modern varieties. Conclusion Taken together, our results did not support the hypothesis that ancient varieties could be more performant than modern ones in storing carbon in our experimental conditions. Increased CO2 emissions by ancient varieties in the presence of inputs showed that ancient and modern varieties differed in their phenotypic plasticity.