Repeated interactions within individual human and animal dyads can lead to the establishment of human?animal relationships (HARs), which may vary in quality from good to bad, defined in terms of the positivity (e.g., friendly contact, play) or negativity (e.g., aggression) of the interactions on which they are based. Particularly good HARs can be regarded as Human? Animal Bonds (HABs) if they are reciprocal and promote wellbeing in both parties. Although there is extensive evidence of the effects of HARs in agricultural animals and HABs in companion animals, there has been less investigation of these relationships in zoos, even though the development of HARs/HABs between zoo animals and their keepers could have important consequences for the welfare of both. Here we apply a modified version of the Lexington Attachment to Pets Scale (LAPS) in a zoo setting to quantify the strength of attachment of a sample of 22 keepers to the animals in their care at the zoo (ZA), in comparison with their attachment to their companion animals at home (PA). Results showed that mean PA scores (47.54 ± 3.6) were significantly higher than mean ZA scores (32.89 ± 2.6; t = ?5.16, df = 13, p < 0.001), indicating stronger attachment to the companion animals. PA scores were lower in keepers who thought it inappropriate to have a bond with a zoo animal, compared with those who deemed it appropriate. Thus, HABs do appear to occur in the zoo context, though they are weaker than those developed in the home. This work also shows that a modified LAPS questionnaire is a suitable instrument for further investigation of HABs in zoos.