Non-perforating abomasal lesions are common in Danish Holstein cows, but the impact on production has not been studied. Our objective was to compare milk yield, carcass weight, days to first service and initiation of breeding among cows with different non- perforating abomasal lesion types and locations to cows with no such lesions. The occurrence, type and localisation of non- perforating abomasal lesions were characterised at slaughter in Danish Holstein cows, and first insemination dates (n = 592), milk yield (n = 999) and carcass information (n = 1,317) was also available. Production and reproduction data were retrieved from all other cows from the herds of origin. This information was used to evaluate the expected production performance in the herds. Of the 1,317 cows investigated at slaughter, 572, 428, 677, and 91 cows had at least one lesion Subtype 1a (mucosal erosion), 1b (acute deeper lesion), 1c (chronic deeper lesion) or 1d (scarring following a lesion), respectively. Cows with lesion Subtype 1c and lesions in the pyloric part of the abomasum were generally found to be culled later in lactation than their herd mates. Milk yield and days to first service were not associated with any lesion types. The occurrence of lesion Subtype 1c and lesions in the pylorus were associated with a higher weight at slaughter compared to cows without lesions of this subtype or in this location. Most Subtype 1c lesions was found in the pylorus. Initiation of breeding before 150 days in milk was positively associated with the number of lesions in the corpus of the abomasum. We discovered two sources of selection bias. First, cows with lesions had a higher carcass weight but also remained in the herds for longer after calving than cows without lesions. This would give them more time to overcome the period of negative energy balance in early lactation. Second, although the causes of culling were not known, cows with no lesions generally had a relatively low milk yield and could have been culled for this reason.”. While non- perforating abomasal ulcers may have an effect on the production and reproduction data, the observational data presented show no evidence of such a relationship. It may be necessary to employ other study designs to identify any effects that might exist, but if the relationship cannot be demonstrated in a dataset of this magnitude, it is not deemed to be of practical relevance.