Sixteen lactating Holstein cows were randomly assigned to treatment sequences in a 4 × 4 changeover experiment. Experimental periods were 28 days in length separated by 7 days for changeover. Cows were fed corn silage free choice and gradually increasing amounts of one of four dairy concentrates containing 0, 11, 22, or 34% rapeseed meal (RSM) of a Canadian, experimental, low glucosinolate–low erucic acid variety. Concentrate intake (kg dry matter/day), milk yield (kg/day) and body weight change (kg/28 days) were 8.45a, 8.71a, 8.37a and 7.17b; 24.40a, 24.03ab, 22.92bc and 21.96c and 7.6b, 22.4b, 13.0ab, and 3.0b for 0, 11, 22 and 34% RSM mixtures, respectively. There was no influence of treatment on milk composition. Ration dry matter (DM) and nitrogen digestibilities declined (P < 0.05) as the RSM in the concentrate mixture increased. The amount of nitrogen excreted in the urine was greatest for cows fed 0% RSM and least for cows fed 34% RSM. Cows fed concentrate mixtures containing 34% RSM had a lower blood thyroxine level than others (P < 0.05), but RSM-containing concentrates had no apparent effect on the conversion of thyroxine to triiodothyronine. Results of the trial indicated that this experimental variety of RSM had no detrimental effect on concentrate intake, DM digestibility, milk composition or thyroid hormone metabolism when included at levels up to 22% in the concentrate mixture. However, milk yield and apparent digestibility of ration nitrogen were decreased (P < 0.05) by the 22% RSM mixture. It was not certain from the current findings whether the depression in digestibility was due largely to the high oil content of the meal or due to the higher fiber content of the RSM-supplemented rations.