The objective was to study the interaction effects of rumen nitrogen balance (RNB) and dietary protein source on feed intake, apparent total-tract digestibility (ATTD), eating and ruminating activity, milk yield (MY), and milk composition in lactating dairy cows. Twenty-four lactating Holstein cows were divided in 4 groups, which were randomly assigned to the dietary treatments included in a replicated 4 × 4 Latin square experimental design that consisted of four 20-d periods, each with 12 d of adaptation to the experimental diets and 8 d of sampling. The dietary treatments followed a 2 × 2 factorial arrangement with 2 main protein sources, faba bean grain (FB) and SoyPass (SP; Beweka Kraftfutterwerk GmbH), offered at 2 dietary RNB levels: RNB0 (RNB of 0 g/kg of dry matter) and RNB- (RNB of -3.2 g/kg of dry matter; i.e., 4 treatments). The composition of concentrate mixtures was adjusted to create diets with the desired RNB levels. Each of the protein sources supplied ≥35% of the total dietary crude protein (CP). Both diets within a protein source had similar forage sources and forage to concentrate ratios and were iso-energetic, but differed in CP concentrations. The main effects of RNB, protein source, and their interactions were tested by PROC MIXED in SAS 9.4 (SAS Institute Inc.). Interaction effects were observed for daily dry matter intake and energy-corrected MY, which were lower for RNB- than RNB0 in diets containing FB (23.5 vs. 24.4 kg dry matter/d; 28.6 vs. 30.6 kg milk/d), but similar in diets containing SP (24.2 vs. 24.3 kg dry matter/d; 31.3 vs. 31.7 kg milk/d). The ATTD of NDF was lower for RNB- compared with RNB0 in the FB (44.9 vs. 49.1 g/100 g) and SP (48.5 vs. 51.9 g/100 g) diets, and greater for the SP than for FB diets. There were interaction effects for ATTD of CP and concentrations of milk urea nitrogen, which were lower for RNB- compared with RNB0 in both, FB (55 vs. 63.1 g/100 g of CP; 5.65 vs. 11.3 mg/dL milk) and SP diets (60 vs. 64.4 g/100 g of CP; 8.74 vs. 13.4 mg/dL milk). However, differences between RNB levels were greater for FB than for SP diets. Furthermore, proportions of milk fatty acids followed the same pattern as that of dietary fatty acids, but biohydrogenation appeared to be greater for RNB- than RNB0 for both protein sources and in FB than in SP diets for both RNB levels. There was an interaction effect on total number of chews per unit of NDF intake, which was greater for RNB- compared with RNB0 for both protein sources. However, the differences between RNB levels were greater in FB than in SP diets. Overall, differences in the animal responses to negative RNB between FB and SP diets suggest a need to better understand the effect of negative RNB levels with different dietary ingredients at similar utilizable CP supply.